Water supply network safety monitoring system based on hydraulic model

Abstract

The invention discloses a water supply network safety monitoring system based on a hydraulic model, and belongs to the technical field of water supply network monitoring. The Caputo fractional order operator introduces historical memory characteristics through an integral term, explicit modeling is carried out on disturbance such as user water consumption fluctuation and sensor noise, and a clear disturbance compensation object can be provided for subsequent multi-source data fusion verification and self-adaptive correction; the pipeline-node coupling relation is embedded into the model through the topological incidence matrix, and structured input can be provided for a subsequent topological consistency verification algorithm; in combination with historical memory characteristics of a Caputo operator, the observer has robust estimation capability on slowly changing leakage disturbance and sudden water impact; the time-varying scaling item decays along with the time index, the gain of the observer is dynamically reduced, and quantization noise and electromagnetic interference of the SCADA sensor can be effectively filtered out in the steady-state stage; and through the positive error convergence index and the negative error convergence index, respectively optimizing the convergence speeds of the positive and negative pressure fluctuations.

Description

Technical Field

[0001] This invention relates to the field of water supply network monitoring technology, specifically to a water supply network safety monitoring system based on a hydraulic model. Background Technology

[0002] Hydraulic model-based water supply network safety monitoring is an intelligent water management method that integrates hydraulic principles, real-time monitoring data, and numerical simulation technology. It is used to assess and ensure the operational safety, reliability, and efficiency of urban water supply systems. Its core is to construct a high-precision hydraulic model of the water supply network and combine it with real-time data such as pressure, flow, and water quality collected by sensors to dynamically simulate the network status, identify anomalies, provide risk warnings, and support emergency decision-making.

[0003] When implementing existing technical solutions, most of them use unstable real-time SCADA data as verification data. If the verification point is located at the end of the pipeline network, it is greatly affected by the user's water consumption, resulting in significant errors in the simulation results. Severe missing or unusable verification data can also lead to model failure. Pipeline topology disconnection, incorrect connection relationships, or incorrect pipe diameter attributes can all cause deviations in simulation results. Misaligned data time scales or inaccurate node elevations can cause longitudinal shifts in verification errors, resulting in poor reliability of model application and verification. Summary of the Invention

[0004] The purpose of this invention is to provide a water supply network safety monitoring system based on a hydraulic model, which solves the technical problems of existing solutions mentioned in the background art.

[0005] The objective of this invention can be achieved through the following technical solutions: A hydraulic model-based safety monitoring system for water supply networks includes: Multidimensional construction module for pipeline network model: Constructs a fractional extended hydraulic model, defines a state vector including nodal pressure, pipeline flow, time-varying friction coefficient and lumped uncertainty disturbance, describes the dynamic characteristics of the pipeline network through Caputo fractional differential operators, and establishes a fractional partial differential equation model; Model configuration optimization module: Design an asymmetric fractional-order finite-time observer. Based on the processed fractional-order dynamic model, reconstruct the observation error through an asymmetric barrier function, inject a time-varying scaling term, and dynamically configure the convergence exponent in the positive and negative error directions to achieve asymmetric compensation for water fluctuations at the end nodes and suppression of high-frequency noise.

[0006] Preferably, based on the physical characteristics of the water supply network, an extended state vector is constructed that includes node states, pipeline attributes, and uncertain disturbances. The corresponding expression is: ; in, Let be the pressure of the i-th node in the pipeline network, where i is the node index, i = 1, 2, 3, ..., n; and n is the total number of nodes. Let be the flow rate of the j-th pipe, where j is the pipe index, j = 1, 2, 3, ..., m; and m is the total number of nodes. Let be the diameter of the j-th pipe; Let be the roughness of the j-th pipe; This refers to lumped uncertainty disturbances, including random fluctuations in user water usage. Sensor measurement noise Time-varying error of pipe friction coefficient .

[0007] Preferably, utilizing the included flow Pressure difference Historical operating data were used to calculate the friction coefficient using the Darcy-Weisbach formula. And calculate and derive the roughness. The expression involved is: ; Where Re is the Reynolds number, , For the density of water, For dynamic viscosity, This refers to the cross-sectional area of ​​the pipe. The pressure loss along the pipeline is g; g is the acceleration due to gravity. This represents the length of the pipe.

[0008] Preferably, an LSTM time-series prediction model is trained based on sample data, and historical water use data is used. Input a trained LSTM time series prediction model and output the flow fluctuation of the terminal nodes. The output will show the flow fluctuation at the terminal node. The conversion calculation yields the random fluctuations in user water usage. The expression involved is: ;in, This is the pressure-to-flow conversion factor.

[0009] Preferably, the time-varying error of the pipe friction coefficient is estimated using a fractional-order expanded state observer. : ; ;in, This is an estimated value for the friction coefficient deviation; This is the initial roughness; It is a Caputo fractional differential operator; Output pressure to the observer; This is the initial hydraulic loss operator.

[0010] Preferably, the Caputo fractional differential operator is used to describe the dynamic characteristics of the pipeline network, overcoming the limitations of traditional integer-order models in modeling transient water usage fluctuations; and, based on the Darcy-Weisbach formula and the pipeline network topology correlation matrix, a hydraulic loss operator is defined. By combining the extended state vector, fractional-order operators, and hydraulic loss operators, a fractional-order dynamic model of the water supply network is constructed.

[0011] Preferably, the traditional symmetric error model is corrected by an asymmetric barrier function to address the asymmetric characteristics of water usage fluctuations. The relevant expression is as follows: ;in, Original observation error; To reconstruct observation errors; is the positive error convergence exponent, with a value range of (0,1); The negative error convergence exponent has a value range of (0,1).

[0012] Preferably, based on the extended state vector and asymmetric error, an r-order fractional-order finite-time observer is designed: ;in, To estimate the derivative of the state vector; To estimate the extended state vector; A is the state matrix; B is the input matrix; u is the control input; K is the time-varying scaling term; K is the observer gain matrix; time-varying scaling term ;in, The attenuation coefficient is... .

[0013] Preferably, when dynamically configuring the convergence exponents for the positive and negative error directions, the positive error direction corresponds to... ≥0, positive error convergence index The value is taken as 0.6, and the error equation is: , The derivative of the reconstruction error, This is the gain coefficient in the positive error direction, determined by the Hurwitz polynomial root collocation, with typical values ​​ranging from 10 to 100. To reconstruct observation errors and quickly track the pressure recovery process; The negative error direction corresponds to <0, negative error convergence index The value is 0.4, and the error equation is: , The negative error direction gain coefficient is determined by the Hurwitz polynomial root configuration, with a typical value of 5 to 50, to avoid pressure oscillations caused by overcompensation. Constrain the perturbation amplitude using the projection operator proj() , This represents the maximum amplitude of the disturbance.

[0014] Preferably, through the Lyapunov function Prove that the observation error converges in finite time. The expression involved in reconstructing the transpose of the error is: ;in, The derivative of the Lyapunov function; It is the smallest eigenvalue of the gain matrix; The minimum convergence index; Ensure that t≤ Internal error convergence is achieved; For convergence time.

[0015] Compared to existing solutions, the beneficial effects achieved by this invention are: In this invention, the Caputo fractional operator introduces historical memory characteristics through the integral term, which can better match the transient fluctuation characteristics of water supply network pressure and flow, effectively reducing dynamic response errors. The extended state vector explicitly models disturbances such as user water usage fluctuations and sensor noise, which can provide clear disturbance compensation objects for subsequent multi-source data fusion verification and adaptive correction, avoiding the compensation lag caused by the traditional model embedding disturbances in parameter errors. By embedding the pipe-node coupling relationship into the model through the topological association matrix, it can provide structured input for subsequent topology consistency verification algorithms, facilitating the automatic identification of pipe topology errors.

[0016] In this invention, by combining the historical memory characteristics of the Caputo operator, the observer has robust estimation capabilities for both slowly changing leakage disturbances and sudden water usage shocks, which can effectively reduce disturbance compensation errors. The time-varying scaling term decays exponentially with time, dynamically reducing the observer gain, and can effectively filter out quantization noise and electromagnetic interference from SCADA sensors in the steady state phase. By optimizing the convergence speed of positive and negative pressure fluctuations through positive and negative error convergence exponents respectively, the performance imbalance problem of traditional symmetric observers during peak and off-peak water usage periods can be solved. Attached Figure Description

[0017] The invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart illustrating the operation of the water supply network safety monitoring system based on a hydraulic model according to the present invention. Detailed Implementation

[0019] 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.

[0020] Example 1, such as Figure 1 As shown, this invention is a water supply network safety monitoring system based on a hydraulic model, including a multi-dimensional network model construction module: constructing a fractional-order extended hydraulic model, defining a state vector containing nodal pressure, pipeline flow, time-varying friction coefficient, and lumped uncertainty disturbance, describing the dynamic characteristics of the network through the Caputo fractional differential operator, and establishing a fractional-order partial differential equation model; the specific steps include: Based on the physical characteristics of the water supply network, an extended state vector is constructed, which includes node states, pipeline attributes, and uncertain disturbances. The corresponding expression is: ; in, Let be the pressure of the i-th node in the pipeline network, including transient pressure changes caused by fluctuations in user water usage, where i is the node index, i=1, 2, 3, ..., n; and n is the total number of nodes. Let be the flow rate of the j-th pipeline, reflecting the pipeline's water transport capacity, where j is the pipeline index, j=1, 2, 3, ..., m; and m is the total number of nodes. Let be the diameter of the j-th pipe; Let be the roughness of the j-th pipe; This refers to lumped uncertainty disturbances, including random fluctuations in user water usage. Sensor measurement noise Time-varying error of pipe friction coefficient , ; Among them, the roughness of the pipe The steps to obtain it include: Based on the pipe material and design parameters, the initial roughness of the pipe is set. Specifically: Metal pipes, such as cast iron pipes, initial roughness The value is 0.2mm; Plastic pipes, such as PVC pipes, initial roughness The value is 0.01 mm; Older pipes, for example, those that have been in service for over 10 years, have a lower initial roughness. The value is 0.5mm, which can be adjusted according to the degree of scaling. Utilizing historical operational data, which includes traffic data... Pressure difference The friction coefficient was calculated using the Darcy-Weisbach formula. And calculate and derive the roughness. The expression involved is: ; Where Re is the Reynolds number, , For the density of water, For dynamic viscosity, This refers to the cross-sectional area of ​​the pipe. The pressure loss along the pipeline is calculated using the pressure difference between upstream and downstream nodes; g is the acceleration due to gravity. This refers to the length of the pipe. Lumped uncertainty disturbance The steps to obtain it include: Based on sample data, train an LSTM time series prediction model and use historical water use data. Input a trained LSTM time series prediction model and output the flow fluctuation of the terminal nodes. The sample data includes several historical water usage data points and corresponding flow fluctuations at the end nodes of the sample. In addition, the construction and training of the LSTM time series prediction model and the data analysis through the LSTM time series prediction model are all existing conventional techniques. The specific implementation steps are not detailed here. Output the flow fluctuation of the terminal node The conversion calculation yields the random fluctuations in user water usage. The expression involved is: ;in, This is the pressure-to-flow conversion factor, with a default value of 0.1. Based on the characteristics of SCADA sensors, the noise is modeled as a zero-mean Gaussian distribution to obtain the sensor measurement noise. : ;in, It follows a Gaussian distribution; For noise variance, , Rated pressure; Time-varying error in estimating pipe friction coefficient using a fractional-order expanded state observer : ; ;in, This is an estimated value for the friction coefficient deviation; This is the initial roughness; It is a Caputo fractional differential operator; Output pressure to the observer; This is the initial hydraulic loss operator; The Caputo fractional differential operator is used to describe the dynamic characteristics of the pipeline network, overcoming the limitations of traditional integer-order models in modeling transient water usage fluctuations, such as sudden water usage shocks. The operator is defined as follows: ;in, The order of the fractional derivative is 0 < α < 1. It is determined offline through historical data. For example, it is set to 0.8 during peak water usage to enhance dynamic response, and 0.3 during off-peak periods to improve steady-state accuracy. For node pressure; This is the derivative of pressure with respect to time. The Gamma function describes the historical memory characteristics of fractional calculus; t is the time variable. Let be the integral variable, representing a historical point in time; Furthermore, based on the Darcy-Weisbach formula and the pipe network topology correlation matrix, the hydraulic loss operator is defined as: ;in, This is the pipeline network topology correlation matrix. This indicates that node i is connected to pipe j. Indicates no connection; By simultaneously applying the extended state vector, fractional-order operators, and hydraulic loss operators, a fractional-order dynamic model of the water supply network is constructed: ; It should be explained that the left side of the equation describes the dynamic evolution characteristics of pressure P(t) through the Caputo operator, while the right side reflects the combined effect of hydraulic loss and uncertain disturbances, thus achieving a unified modeling of the transient and steady-state processes of the pipeline network.

[0021] In this embodiment of the invention, compared to traditional integer-order models that only describe the current state, the Caputo fractional-order operator introduces historical memory characteristics through the integral term, which can better reflect the transient fluctuation characteristics of water supply network pressure and flow, such as the sudden drop and slow recovery process of pressure during peak water use, and can effectively reduce dynamic response errors. The extended state vector explicitly models disturbances such as user water use fluctuations and sensor noise, which can provide clear disturbance compensation objects for subsequent multi-source data fusion verification and adaptive correction, avoiding the compensation lag caused by the traditional model embedding disturbances in parameter errors. By embedding the pipe-node coupling relationship into the model through the topological association matrix, it can provide structured input for subsequent topology consistency verification algorithms, which can facilitate the automatic identification of pipe topology errors, such as reversed connection relationships and abnormal pipe diameter attributes.

[0022] Model configuration optimization module: This module designs an asymmetric fractional-order finite-time observer. Based on the processed fractional-order dynamic model, it reconstructs the observation error through an asymmetric barrier function, injects a time-varying scaling term, and dynamically configures the convergence exponents in the positive and negative error directions. This achieves asymmetric compensation for water usage fluctuations at the terminal nodes and suppression of high-frequency noise. Specific steps include: Based on the fractional-order dynamic model obtained through processing, an extended state vector that integrates state variables and perturbation information is defined: ;in, The instantaneous pressure at the terminal node; This is the fractional differential term of the instantaneous pressure at the terminal node; The total disturbance includes sudden water surges and changes in pipeline friction, etc. By correcting the traditional symmetric error model using an asymmetric barrier function, we can address the asymmetric characteristics of water usage fluctuations, such as a sharp drop in pressure during peak periods and a slow recovery during trough periods. The relevant expression is as follows: ;in, The original observation error corresponds to the deviation between the model prediction and the actual observation, such as pipeline pressure deviation and flow deviation. To reconstruct observation errors; is the positive error convergence exponent, with a value range of (0,1); The negative error convergence exponent has a value range of (0,1). Based on the extended state vector and asymmetric error, design an r-order fractional-order finite-time observer: ;in, To estimate the derivative of the state vector; To estimate the extended state vector; A is the state matrix; B is the input matrix; u is the control input; For time-varying scaling terms; K is the observer gain matrix; Among them, the state matrix A, which incorporates the properties of the Caputo fractional operator, describes the pressure-flow coupling relationship: ;in, All are hydraulic coupling coefficients, and their values ​​are all greater than 0. The disturbance attenuation rate; Time-varying scaling term ;in, The attenuation coefficient is... Initial stage ≈1, retaining fast response, in steady state Approaching 0, suppressing high-frequency noise; The observer gain K is calculated using a Hurwitz polynomial configuration combined with the upper bound L of the Lipschitz constant in the pipe network hydraulic model: ;in, Construct operators for diagonal matrices; These are the pressure gain coefficient, flow gain coefficient, differential term gain coefficient, and lumped disturbance gain coefficient, respectively. , , , , For different calculation factors, , All are determined by Hurwitz polynomial root configuration to ensure that the eigenvalues ​​of the observer error dynamic equation are all located in the left half of the complex plane, with the corresponding real parts being negative, thus guaranteeing system stability; L is the upper bound of the Lipschitz constant. When dynamically configuring the convergence exponents for the positive and negative error directions, the positive error direction corresponds to... ≥0, positive error convergence index The value is taken as 0.6, and the error equation is: , The derivative of the reconstruction error, This is the gain coefficient in the positive error direction, determined by the Hurwitz polynomial root collocation, with typical values ​​ranging from 10 to 100. To reconstruct observation errors and quickly track the pressure recovery process; The negative error direction corresponds to <0, negative error convergence index The value is 0.4, and the error equation is: , The negative error direction gain coefficient is determined by the Hurwitz polynomial root configuration, with a typical value of 5 to 50, to avoid pressure oscillations caused by overcompensation. Constrain the perturbation amplitude using the projection operator proj() , This is the maximum disturbance amplitude, with a default value of 500 to prevent parameter drift. And, through the Lyapunov function Prove that the observation error converges in finite time. The expression involved in reconstructing the transpose of the error is: ;in, The derivative of the Lyapunov function; It is the smallest eigenvalue of the gain matrix; The minimum convergence index; Ensure that t≤ Internal error convergence is achieved; To shorten the convergence time, .

[0023] In this embodiment of the invention, by combining the historical memory characteristics of the Caputo operator, the observer has robust estimation capabilities for both slowly changing leakage disturbances and sudden water usage shocks, which can effectively reduce disturbance compensation errors; the time-varying scaling term decays exponentially with time, dynamically reducing the observer gain, and can effectively filter out quantization noise and electromagnetic interference from SCADA sensors in the steady state phase; by optimizing the convergence speed of positive and negative pressure fluctuations through positive and negative error convergence exponents respectively, the performance imbalance problem of traditional symmetric observers during peak and off-peak water usage periods can be solved.

[0024] Example 2 further includes: a fractional-order adaptive correction processing module: constructing a fractional-order adaptive correction law constrained by the projection operator, dynamically adjusting the parameter correction intensity based on the reconstructed observation error through fractional-order time-varying gain, and using the projection operator to limit the parameter drift range, suppressing model failure caused by topological errors and missing data. Specific steps include: When designing a fractional-order time-varying gain adaptive law, based on the reconstructed observation error, the correction strength of the fractional-order time-varying gain Γ(t) dynamic adjustment parameter is designed, and the formula is: ;in, This is the initial gain matrix; It is a fractional differential operator; This refers to the gain adjustment order; The vector of parameters to be corrected is defined as: Where d is the inner diameter of the pipe, and the initial value is the design value; For pipe roughness; This is the node elevation correction factor, with an initial value of 1; When using a projection operator to limit the parameter drift range, the projection operator is introduced. Limiting parameter drift to prevent parameters from exceeding physical boundaries due to integration accumulation (e.g., pipe diameter cannot be negative) involves the following expression: ;in, The derivative of the parameter vector; The vector of parameters to be corrected; The upper limit vector of physical parameters is preset according to the pipeline design specifications. , These are the maximum values ​​of the pipe inner diameter, the maximum values ​​of the pipe roughness, and the maximum values ​​of the node elevation correction coefficient, respectively. Among them, the operator for dynamically constraining the parameter range outputs 1, 0, or -1 based on the current state of the parameter, implementing three adjustment logics: like and This indicates that the parameters have not exceeded the upper limit and the correction direction is reasonable. Normal updates are allowed. Let i be the physical upper bound vector of parameter i; like and This indicates that the parameter has reached its upper limit and continues to increase. Truncate the update; like , Let i be the physical lower bound vector of parameter i, then Adjust the parameter direction in the opposite direction.

[0025] In this embodiment of the invention, fractional-order time-varying gain is used to achieve real-time optimization of parameters such as pipe diameter and roughness, which can solve the model mismatch problem caused by changes in pipe network topology or missing data. Changes in pipe network topology include valve opening and closing, and missing data includes some sensors going offline. The projection operator strictly limits the parameters within the physical range, avoiding model divergence caused by integral saturation in traditional adaptive algorithms, which can effectively improve the robustness of the system. The fractional-order differential operator integrates historical error information and can adaptively adjust the gain for both peak water usage periods with rapid error changes and trough periods with slow error fluctuations. Compared with integer-order adaptive laws, it can effectively improve the convergence speed.

[0026] In the several embodiments provided by this invention, it should be understood that the disclosed system can be implemented in other ways. For example, the embodiments of the invention described above are merely illustrative; for example, the division of modules is only a logical functional division, and there may be other division methods in actual implementation.

[0027] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0028] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or in the form of hardware plus software functional modules.

[0029] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the essential characteristics of the present invention.

[0030] Finally, 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 water supply network safety monitoring system based on a hydraulic model, characterized in that, include: Multidimensional construction module for pipeline network model: Constructs a fractional extended hydraulic model, defines a state vector including nodal pressure, pipeline flow, time-varying friction coefficient and lumped uncertainty disturbance, describes the dynamic characteristics of the pipeline network through Caputo fractional differential operators, and establishes a fractional partial differential equation model; Model configuration optimization module: Design an asymmetric fractional-order finite-time observer. Based on the processed fractional-order dynamic model, reconstruct the observation error through an asymmetric barrier function, inject a time-varying scaling term, and dynamically configure the convergence exponent in the positive and negative error directions to achieve asymmetric compensation for water fluctuations at the end nodes and suppression of high-frequency noise.

2. The water supply network safety monitoring system based on a hydraulic model according to claim 1, characterized in that, Based on the physical characteristics of the water supply network, an extended state vector is constructed, which includes node states, pipeline attributes, and uncertain disturbances. The corresponding expression is: ； in, Let be the pressure of the i-th node in the pipeline network, where i is the node index, i = 1, 2, 3, ..., n; and n is the total number of nodes. Let be the flow rate of the j-th pipe, where j is the pipe index, j=1,2,3,…,m; and m is the total number of nodes. Let be the diameter of the j-th pipe; Let be the roughness of the j-th pipe; This refers to lumped uncertainty disturbances, including random fluctuations in user water usage. Sensor measurement noise Time-varying error of pipe friction coefficient .

3. The water supply network safety monitoring system based on a hydraulic model according to claim 2, characterized in that, Utilize included traffic Pressure difference Historical operating data were used to calculate the friction coefficient using the Darcy-Weisbach formula. And calculate and derive the roughness. The expression involved is: ; Where Re is the Reynolds number, , For the density of water, For dynamic viscosity, This refers to the cross-sectional area of ​​the pipe. The pressure loss along the pipeline is g; g is the acceleration due to gravity. This represents the length of the pipe.

4. The water supply network safety monitoring system based on a hydraulic model according to claim 3, characterized in that, Based on sample data, train an LSTM time series prediction model and use historical water use data. Input a trained LSTM time series prediction model and output the flow fluctuation of the terminal nodes. The output of the flow fluctuation at the terminal node. The conversion calculation yields the random fluctuations in user water usage. The expression involved is: ;in, This is the pressure-to-flow conversion factor.

5. The water supply network safety monitoring system based on a hydraulic model according to claim 4, characterized in that, Time-varying error in estimating pipe friction coefficient using a fractional-order expanded state observer : ; ;in, This is an estimated value for the friction coefficient deviation; This is the initial roughness; It is a Caputo fractional differential operator; Output pressure to the observer; This is the initial hydraulic loss operator.

6. The water supply network safety monitoring system based on a hydraulic model according to claim 5, characterized in that, The Caputo fractional differential operator is used to describe the dynamic characteristics of the pipeline network, overcoming the limitations of traditional integer-order models in modeling transient water consumption fluctuations; and, based on the Darcy-Weisbach formula and the pipeline network topology correlation matrix, a hydraulic loss operator is defined. By combining the extended state vector, fractional-order operators, and hydraulic loss operators, a fractional-order dynamic model of the water supply network is constructed.

7. The water supply network safety monitoring system based on a hydraulic model according to claim 6, characterized in that, To address the asymmetric characteristics of water usage fluctuations, the traditional symmetric error model is modified using an asymmetric barrier function. The relevant expressions are as follows: ;in, Original observation error; To reconstruct observation errors; is the positive error convergence exponent, with a value range of (0,1); The negative error convergence exponent has a value range of (0,1).

8. The water supply network safety monitoring system based on a hydraulic model according to claim 7, characterized in that, Based on the extended state vector and asymmetric error, design an r-order fractional-order finite-time observer: ;in, To estimate the derivative of the state vector; To estimate the extended state vector; A is the state matrix; B is the input matrix; u is the control input; K is the time-varying scaling term; K is the observer gain matrix; time-varying scaling term ;in, The attenuation coefficient is... .

9. The water supply network safety monitoring system based on a hydraulic model according to claim 8, characterized in that, When dynamically configuring the convergence exponents for the positive and negative error directions, the positive error direction corresponds to... ≥0, positive error convergence index The value is taken as 0.6, and the error equation is: , The derivative of the reconstruction error, This is the gain coefficient in the positive error direction, determined by the Hurwitz polynomial root collocation, with typical values ​​ranging from 10 to 100. To reconstruct observation errors and quickly track the pressure recovery process; The negative error direction corresponds to <0, negative error convergence index The value is taken as 0.4, and the error equation is: , The negative error direction gain coefficient is determined by the Hurwitz polynomial root configuration, with a typical value of 5 to 50, to avoid pressure oscillations caused by overcompensation. Constrain the perturbation amplitude using the projection operator proj() , This represents the maximum amplitude of the disturbance.

10. The water supply network safety monitoring system based on a hydraulic model according to claim 9, characterized in that, via Lyapunov functions Prove that the observation error converges in finite time. The expression involved in reconstructing the transpose of the error is: ;in, The derivative of the Lyapunov function; It is the smallest eigenvalue of the gain matrix; The minimum convergence index; Ensure that t≤ Internal error convergence; For convergence time.

Images