Self-adaptive feedback control method for virtual-real synchronization of water conservancy digital twin ends

Abstract

The invention discloses a self-adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin end, and relates to the technical field of water conservancy project operation management. Based on a digital twin architecture, the digital twin architecture comprises a data base plate, a model platform and a knowledge platform; comprising the following steps: collecting state data of a physical entity in real time through a water conservancy sensing network in advance, and carrying out data preprocessing on the state data; the adaptive controller can dynamically optimize the control parameters according to the real-time error and the error change rate, and realizes the real-time adaptive adjustment of the control strategy in cooperation with the rule constraint and reasoning decision of the knowledge platform, thereby reducing the human intervention and improving the decision accuracy and the control timeliness; by means of high-precision synchronization of a physical entity and a virtual model, accurate control over water resource scheduling and water conservancy equipment operation can be achieved, and then comprehensive benefits such as power generation, flood control, irrigation and water supply are optimized.

Description

Technical Field

[0001] This invention relates to the field of water conservancy project operation and management technology, and more specifically, to an adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal. Background Technology

[0002] Digital twin technology, an innovative method for mapping physical entities to virtual models, has been initially applied in water conservancy projects. In existing technologies, digital twin systems typically consist of a data base, a model platform, and a knowledge platform. Through technologies such as water conservancy sensing networks, the Internet of Things (IoT), and big data, they achieve digital mapping of watershed conditions and governance activities. For example, existing systems can simulate and visualize facilities such as reservoirs, sluices, and dikes, and support hydrological simulation and scheduling decisions based on monitoring data. This provides digital and visualized technical support for the operation and management of water conservancy projects, improving the efficiency of water conservancy project monitoring and scheduling to a certain extent.

[0003] However, existing technologies still have many limitations in practical applications, making it difficult to meet the high-precision and high-reliability operation requirements in complex water conservancy environments. On the one hand, the synchronization between physical entities and virtual models relies on fixed algorithms, which are difficult to adapt to dynamic environments such as sudden changes in water flow and equipment aging, resulting in large deviations between the virtual and physical states and insufficient synchronization accuracy. At the same time, when sensors fail or communication is abnormal, the system cannot automatically adjust the data source or model parameters, compromising data authenticity and reliability and exhibiting poor fault tolerance. On the other hand, existing feedback control is mostly based on preset rules and lacks adaptive capabilities, failing to optimize control strategies based on real-time data. This leads to poor performance in complex water conservancy scenarios such as flood control scheduling and water resource allocation. Furthermore, while knowledge platforms can accumulate experience, they are not deeply integrated with real-time feedback control, resulting in lagging decision support. These problems collectively limit the efficient application of digital twin technology in water conservancy projects, necessitating a method that can adaptively adjust the virtual-physical synchronization process to improve the system's robustness and intelligence. Summary of the Invention

[0004] To address the problems in related technologies, this invention proposes an adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal, in order to overcome the problems of low synchronization accuracy, poor fault tolerance, and rigid feedback mechanism in existing technologies.

[0005] The technical solution of this invention is implemented as follows:

[0006] One aspect of the present invention:

[0007] An adaptive feedback control method for virtual-real synchronization in a water conservancy digital twin, based on a digital twin architecture including a data base, a model platform, and a knowledge platform, includes the following steps:

[0008] The status data of physical entities are collected in real time through the water conservancy sensing network in advance, and the status data is preprocessed.

[0009] The physical entity state vector and the virtual model state vector are calibrated, and the instantaneous error vector, cumulative error vector, and synchronization accuracy evaluation index are calculated to evaluate the virtual-real synchronization deviation.

[0010] An adaptive controller based on fuzzy logic and PID control is constructed. The instantaneous error and the rate of change of error are input, the PID parameters are dynamically adjusted through fuzzy logic to generate control signals, and the control strategy is optimized according to the rules and constraints of a preset knowledge platform.

[0011] The system integrates monitoring data, business data, and external data, and uses the extended Kalman filter algorithm to estimate the system state. When the calibration data source is unreliable, the virtual model prediction value is used first.

[0012] Control signals are fed back to the physical entity actuators and the virtual model to achieve two-way synchronization, and decision suggestions and execution are generated based on the reasoning of the knowledge platform.

[0013] Furthermore, the state data includes, but is not limited to, water level h(t), flow rate Q(t), pump speed n(t), sluice gate opening α(t), and dissolved oxygen concentration DO(t).

[0014] Furthermore, the method for evaluating virtual-real synchronization deviation includes the following steps:

[0015] The physical entity state vector is calibrated as follows: The virtual model state vector is represented as: ;

[0016] In the formula, the subscript p represents the physical entity, v represents the virtual model, and t represents the current time.

[0017] Calculate the instantaneous error vector That is, the difference between the physical entity state vector and the virtual model state vector, expressed as:

[0018] ;

[0019] In the formula, This is the instantaneous error in water level. For instantaneous flow error, This refers to the instantaneous error of the sluice gate opening. This refers to the instantaneous error of the water pump speed. This represents the instantaneous error in dissolved oxygen concentration.

[0020] Calculate cumulative error Used to comprehensively assess synchronization deviation, it is expressed as:

[0021] .

[0022] Furthermore, the adaptive controller consists of a fuzzy logic module, a PID control module, and a knowledge fusion module, with the instantaneous error as the input. and error change rate The output is the adjustment amount of the PID control parameters. and control signal u(t).

[0023] Furthermore, the fuzzy logic module includes the following steps:

[0024] Let the error e and the rate of change of error be... The dataset is divided into 7 fuzzy subsets: negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB). The fuzzy membership function adopts the triangular membership function, which is expressed as follows:

[0025] ;

[0026] ;

[0027] ;

[0028] ;

[0029] ;

[0030] ;

[0031] Based on the operational experience and control theory of water conservancy projects, a fuzzy rule base is established.

[0032] The Mamdani inference method is used for fuzzy inference. The membership degree of the fuzzy conclusion is calculated using the maxima-mina combination rule, and then the centroid method is used for defuzzification to obtain the precise value of the PID parameter adjustment, expressed as:

[0033] ;

[0034] In the formula, M represents the number of fuzzy rules triggered. Let k be the membership degree of the k-th rule. This is the output parameter value corresponding to the k-th rule.

[0035] Furthermore, the PID control module includes the following steps:

[0036] Based on the adjustment amount output by the fuzzy logic module, the PID control parameters are updated, as follows:

[0037] ;

[0038] ;

[0039] ;

[0040] In the formula, , and These are the initial values ​​for the PID parameters; , and For adaptive PID parameters that are updated in real time;

[0041] The control signal u(t) is generated using a positional PID control algorithm, expressed as:

[0042] ;

[0043] in, , For water level control signals, For flow control signals, This is the sluice gate opening control signal. This is the water pump speed control signal. This is a dissolved oxygen concentration control signal, used to adjust virtual model parameters or physical entity actuators respectively.

[0044] Furthermore, when the calibration data source is unreliable, the virtual model prediction value is used preferentially, including determining the reliability of the data source, i.e., calibrating the data source reliability index. ;

[0045] in, The variance of the observed data Z(t);

[0046] when When the reliability threshold is reached, the data source is deemed unreliable, and the system uses a virtual model to predict values. This serves as the result of state estimation.

[0047] Furthermore, the implementation of bidirectional synchronization includes the following steps:

[0048] Controlling physical entities involves converting control signals u(t) into control commands for physical actuators, which then drive the physical entity's state adjustment to achieve synchronous updates at the physical end.

[0049] The virtual model is updated based on the control signal u(t) and the state estimation results. Update the parameters and initial conditions of the virtual model to make the virtual model state... It is used to achieve synchronous updates of the virtual terminal.

[0050] Another aspect of the present invention:

[0051] A water conservancy digital twin system is provided for the above-mentioned adaptive feedback control method for virtual-real synchronization of the water conservancy digital twin terminal to realize the virtual-real synchronization of physical water conservancy entities and virtual digital models. The system includes a data base, a model platform, a knowledge platform, and an adaptive controller. The adaptive controller is connected to the data base, the model platform, and the knowledge platform to realize data interaction and collaborative control.

[0052] The beneficial effects of this invention are:

[0053] This invention maintains stable operation and significantly enhances fault tolerance through multiple mechanisms, including automatic data source switching, data interpolation, and virtual model prediction value replacement. The adaptive controller can dynamically optimize control parameters based on real-time errors and error change rates. Combined with the rule constraints and reasoning decisions of the knowledge platform, it achieves real-time adaptive adjustment of control strategies, reduces human intervention, and improves decision accuracy and control timeliness. With the high-precision synchronization of physical entities and virtual models, it can achieve precise control of water resource scheduling and water conservancy equipment operation, thereby optimizing the comprehensive benefits of power generation, flood control, irrigation, and water supply.

[0054] Meanwhile, this method is based on the existing digital twin architecture and can be seamlessly integrated with the data base, model platform and knowledge platform without large-scale modification of the existing system. Moreover, through modular design and parameter configuration, it can be easily extended to various water conservancy scenarios such as reservoir scheduling, flood control and disaster reduction, irrigation management and hydropower station operation, and has good compatibility, scalability and social benefits. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart illustrating an adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to an embodiment of the present invention. Detailed Implementation

[0057] 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. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0058] According to an embodiment of the present invention, an adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal is provided.

[0059] The adaptive feedback control method for virtual-real synchronization of water conservancy digital twin terminals according to embodiments of the present invention is based on a digital twin architecture, which includes a data base, a model platform, and a knowledge platform, and embeds an adaptive feedback control strategy.

[0060] This technical solution employs a distributed database to store basic reservoir data, including reservoir capacity curves, dam parameters, hydrogeological data, real-time monitoring data (water level, flow rate, rainfall), operational data (scheduling plans, water demand), and historical data (hydrometeorological data and scheduling records for the past 10 years). The model platform utilizes a hydrodynamic model as a virtual model, with parameters including roughness coefficient, permeability coefficient, and reservoir discharge curves. Calibration and verification are performed based on historical data to ensure model simulation accuracy. The knowledge platform constructs a reservoir scheduling knowledge graph, containing flood control scheduling rules such as flood control limits, discharge priorities, water resource allocation standards, and equipment operating limits (maximum sluice gate opening, rated pump speed), as well as historical scheduling cases.

[0061] Specifically, such as Figure 1 As shown, the adaptive feedback control strategy includes the following steps:

[0062] Step S1: In advance, the status data of physical entities are collected in real time through the water conservancy sensing network. After anomaly detection, data repair and normalization processing, the data reliability is ensured.

[0063] This technical solution, the water conservancy sensing network, includes water quality sensors, gate pump monitoring equipment, water level gauges, and flow meters;

[0064] Status data, including but not limited to water level h(t), flow rate Q(t), pump speed n(t), sluice gate opening α(t), and dissolved oxygen concentration DO(t); the data acquisition time interval T s It can be configured according to the needs of the scenario.

[0065] The data anomaly detection can employ an anomaly detection algorithm based on the 3σ criterion and a sliding window to identify abnormal data caused by sensor malfunctions, data transmission interference, etc. When an anomaly is detected in a sensor's data, the system automatically switches to a backup data source; if no backup data source is available, linear interpolation is used for data interpolation. Data normalization can be performed using the minimum-maximum normalization method to map the data to the [0,1] interval.

[0066] Step S2: Calibrate the physical entity state vector and the virtual model state vector, calculate the instantaneous error vector, cumulative error vector and synchronization accuracy evaluation index, and evaluate the virtual-real synchronization deviation.

[0067] The physical entity state vector is represented as:

[0068] ;

[0069] The virtual model state vector is represented as:

[0070] ;

[0071] In the formula, the subscript p represents the physical entity, v represents the virtual model, and t represents the current time.

[0072] Among them, the instantaneous error vector is calculated. That is, the difference between the physical entity state vector and the virtual model state vector, expressed as:

[0073] ;

[0074] In the formula, This is the instantaneous error in water level. For instantaneous flow error, This refers to the instantaneous error of the sluice gate opening. This refers to the instantaneous error of the water pump speed. This represents the instantaneous error in dissolved oxygen concentration.

[0075] Calculate cumulative error Used to comprehensively assess synchronization deviation, it is expressed as:

[0076]

[0077] in, This represents the cumulative error in water level.

[0078] This is the cumulative error of the flow rate;

[0079] This refers to the cumulative error of the sluice gate opening.

[0080] This refers to the cumulative error of the water pump speed.

[0081] This represents the cumulative error in dissolved oxygen concentration.

[0082] The synchronization accuracy evaluation metrics include mean absolute error (MAE), root mean square error (RMSE), and maximum error (MaxE), which are expressed as follows:

[0083] ;

[0084] ;

[0085] ;

[0086] In the formula, T is the length of the evaluation time window, used to quantify the effect of virtual-real synchronization.

[0087] Step S3: Construct an adaptive controller based on fuzzy logic and PID control. Input the instantaneous error and the rate of change of error, dynamically adjust the PID parameters through fuzzy logic, generate control signals, and optimize the control strategy according to the rules and constraints of the preset knowledge platform.

[0088] In this technical solution, the adaptive controller consists of a fuzzy logic module, a PID control module, and a knowledge fusion module, with the input being the instantaneous error. and error change rate The output is the adjustment amount of the PID control parameters. and control signal u(t).

[0089] The fuzzy logic module is detailed below:

[0090] Let the error e and the rate of change of error be... The dataset is divided into 7 fuzzy subsets: negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB). The fuzzy membership function adopts the triangular membership function, which is expressed as follows:

[0091] ;

[0092] ;

[0093] ;

[0094] ;

[0095] ;

[0096] ;

[0097] Among them, parameters Calibration is based on the characteristics of the water conservancy scenario and historical data.

[0098] Based on operational experience and control theory of water conservancy projects, a fuzzy rule base is established. Example rules are as follows:

[0099] like and ,but ;

[0100] like and ,but ;

[0101] Specifically, the rule base contains 7×7=49 fuzzy rules, covering different combinations of error states.

[0102] The Mamdani inference method is used for fuzzy inference. The membership degree of the fuzzy conclusion is calculated using the maxima-mina combination rule, and then the centroid method is used for defuzzification to obtain the precise value of the PID parameter adjustment, expressed as:

[0103] ;

[0104] In the formula, M represents the number of fuzzy rules triggered. Let k be the membership degree of the k-th rule. This is the output parameter value corresponding to the k-th rule.

[0105] The PID control module includes the following steps:

[0106] Based on the adjustment amount output by the fuzzy logic module, the PID control parameters are updated, as follows:

[0107] ;

[0108] ;

[0109] ;

[0110] In the formula, , and These are the initial values ​​for the PID parameters, which can be calibrated based on historical data and engineering experience. , and These are adaptive PID parameters that are updated in real time.

[0111] The control signal u(t) is generated using a positional PID control algorithm, expressed as:

[0112] ;

[0113] in, , For water level control signals, For flow control signals, This is the sluice gate opening control signal. This is the water pump speed control signal. This is a dissolved oxygen concentration control signal, used to adjust virtual model parameters or physical entity actuators respectively.

[0114] Among them, the knowledge fusion module, i.e. the controller, optimizes the control signal according to the preset knowledge platform rules;

[0115] Specifically, during implementation, pre-set knowledge platform rules, such as water conservancy scheduling specifications, flood control standards, and equipment operating limit parameters, are used. For example, the sluice gate opening control signal must meet certain requirements. ;in, , This sets the minimum and maximum opening limits for the sluice gate. Additionally, in flood warning scenarios, the flood control priority rule can be invoked to automatically enhance the integral control effect, i.e., increase... To quickly eliminate water level deviations; in water resource optimization allocation scenarios, to strengthen proportional control, i.e., to increase... To improve response speed.

[0116] Step S4: Integrate monitoring data, business data and external data, and use the extended Kalman filter algorithm to estimate the system state. When the calibration data source is unreliable, the virtual model prediction value will be used first.

[0117] This technical solution integrates monitoring data, business data, and external data to construct a fused data vector, represented as follows:

[0118] ;

[0119] In the formula, The monitoring data comes from the water resources sensing network; For business data, such as water conservancy project scheduling plans and water demand data; External data, such as weather forecast data and watershed underlying surface data.

[0120] Among them, the extended Kalman filter (EKF) algorithm is used to estimate the system state, handle the nonlinear characteristics of the hydraulic system, and reduce the impact of noise. This includes state estimation through state equations and observation equations.

[0121] The state equation is expressed as:

[0122]

[0123] In the formula, f(⋅) is the nonlinear state transition function, and w(t) is the process noise, which follows a Gaussian distribution with a mean of 0 and a covariance matrix of Q.

[0124] The observation equation is expressed as:

[0125] ;

[0126] In the formula, h(⋅) is the nonlinear observation function, Z(t) is the observation data, and v(t) is the observation noise, which follows a Gaussian distribution with a mean of 0 and a covariance matrix of R.

[0127] In this technical solution, the prediction step of the extended Kalman filter is expressed as follows:

[0128]

[0129] ;

[0130] In the formula, Let Jacobian be the state transition matrix. This is the state estimation covariance matrix.

[0131] The update step is represented as:

[0132]

[0133] In the formula, To observe the Jacobian matrix, K(t+1) is the Kalman gain.

[0134]

[0135]

[0136] Where I is the identity matrix.

[0137] In addition, determining the reliability of the data source involves calibrating data source reliability metrics. ;

[0138] in, Let Z(t) be the variance of the observed data.

[0139] Specifically, when When the reliability threshold is reached, the data source is deemed unreliable, and the system prioritizes using the virtual model's predicted values. As a result of state estimation, synchronization continuity is ensured.

[0140] Step S5: The control signal is fed back to the physical entity actuator and the virtual model to achieve two-way synchronization, and decision suggestions and execution are generated based on the reasoning of the knowledge platform.

[0141] Among them, controlling the physical entity means converting the control signal u(t) into control instructions for the physical actuator, and driving the physical entity to adjust its state through the physical actuator, thereby achieving synchronous updates at the physical end.

[0142] Specifically, for example, the sluice gate opening adjustment command. Water pump speed adjustment command ,in, , This refers to the actuator gain coefficient.

[0143] Among these, updating the virtual model involves adjusting the control signal u(t) and the state estimation results. Update the parameters and initial conditions of the virtual model to make the virtual model state... This enables synchronous updates on the virtual terminal.

[0144] Furthermore, based on the decision-making suggestions and implementation generated by the knowledge platform, during the implementation process, the knowledge platform constructs a knowledge graph in the water conservancy field based on ontology, including knowledge such as scheduling rules, equipment parameters, and engineering standards. This is combined with error trends. and state estimation results Decision recommendations are generated through rule-based reasoning (RBR) and case-based reasoning (CBR), such as when water level cumulative error occurs. And the rate of change of error The system generates scheduling recommendations to increase the opening of floodgates; based on historical water resource allocation cases and combined with current flow error and water demand data, it optimizes the water resource allocation scheme. Simultaneously, the decision recommendations and control commands are distributed to relevant execution units through the water conservancy project operation management system, and the execution results are fed back to the data base to update the physical entity status data.

[0145] Furthermore, this invention also discloses a water conservancy digital twin system, which uses the above-mentioned adaptive feedback control method for virtual-real synchronization of the water conservancy digital twin terminal to realize the virtual-real synchronization of physical water conservancy entities and virtual digital models. The system includes a data base, a model platform, a knowledge platform, and an adaptive controller. The adaptive controller is connected to the data base, the model platform, and the knowledge platform to realize data interaction and collaborative control.

[0146] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Those skilled in the art, upon considering the disclosure in the specification and embodiments, will readily conceive of other embodiments of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0147] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. An adaptive feedback control method for virtual-real synchronization in a water conservancy digital twin, based on a digital twin architecture, wherein the digital twin architecture includes a data base, a model platform, and a knowledge platform, characterized in that... Includes the following steps: The status data of physical entities are collected in real time through the water conservancy sensing network in advance, and the status data is preprocessed. The physical entity state vector and the virtual model state vector are calibrated, and the instantaneous error vector, cumulative error vector, and synchronization accuracy evaluation index are calculated to evaluate the virtual-real synchronization deviation. An adaptive controller based on fuzzy logic and PID control is constructed. The instantaneous error and the rate of change of error are input, the PID parameters are dynamically adjusted through fuzzy logic to generate control signals, and the control strategy is optimized according to the rules and constraints of a preset knowledge platform. The system integrates monitoring data, business data, and external data, and uses the extended Kalman filter algorithm to estimate the system state. When the calibration data source is unreliable, the virtual model prediction value is used first. Control signals are fed back to the physical entity actuators and the virtual model to achieve two-way synchronization, and decision suggestions and execution are generated based on the reasoning of the knowledge platform.

2. The adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to claim 1, characterized in that, The state data includes, but is not limited to, water level h(t), flow rate Q(t), pump speed n(t), sluice gate opening α(t), and dissolved oxygen concentration DO(t).

3. The adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to claim 2, characterized in that, The method for evaluating virtual-real synchronization deviation includes the following steps: The physical entity state vector is calibrated as follows: The virtual model state vector is represented as: ; In the formula, the subscript p represents the physical entity, v represents the virtual model, and t represents the current time. Calculate the instantaneous error vector That is, the difference between the physical entity state vector and the virtual model state vector, expressed as: ； In the formula, This is the instantaneous error in water level. For instantaneous flow error, This refers to the instantaneous error of the sluice gate opening. This refers to the instantaneous error of the water pump speed. This represents the instantaneous error in dissolved oxygen concentration. Calculate cumulative error Used to comprehensively assess synchronization deviation, it is expressed as: 。 4. The adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to claim 3, characterized in that, The adaptive controller consists of a fuzzy logic module, a PID control module, and a knowledge fusion module, with the instantaneous error as the input. and error change rate The output is the adjustment amount of the PID control parameters. and control signal u(t).

5. The adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to claim 4, characterized in that, The fuzzy logic module includes the following steps: Let the error e and the rate of change of error be... The dataset is divided into 7 fuzzy subsets: negative large (NB), negative medium (NM), negative small (NS), zero (ZO), positive small (PS), positive medium (PM), and positive large (PB). The fuzzy membership function adopts the triangular membership function, which is expressed as follows: ； ； ； ； ； ； Based on the operational experience and control theory of water conservancy projects, a fuzzy rule base is established. The Mamdani inference method is used for fuzzy inference. The membership degree of the fuzzy conclusion is calculated using the maxima-mina combination rule, and then the centroid method is used for defuzzification to obtain the precise value of the PID parameter adjustment, expressed as: ； In the formula, M represents the number of fuzzy rules triggered. Let k be the membership degree of the k-th rule. This is the output parameter value corresponding to the k-th rule.

6. The adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to claim 5, characterized in that, The PID control module includes the following steps: Based on the adjustment amount output by the fuzzy logic module, the PID control parameters are updated, as follows: ； ； ； In the formula, , and These are the initial values ​​for the PID parameters; , and For adaptive PID parameters that are updated in real time; The control signal u(t) is generated using a positional PID control algorithm, expressed as: ； in, , For water level control signals, For flow control signals, This is the sluice gate opening control signal. This is the water pump speed control signal. This is a dissolved oxygen concentration control signal, used to adjust virtual model parameters or physical entity actuators respectively.

7. The adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to claim 6, characterized in that, When the calibration data source is unreliable, the virtual model prediction value is used preferentially, including determining the reliability of the data source, i.e., the calibration data source reliability index. ; in, The variance of the observed data Z(t); when When the reliability threshold is reached, the data source is deemed unreliable, and the system uses a virtual model to predict values. This serves as the result of state estimation.

8. The adaptive feedback control method for virtual-real synchronization of a water conservancy digital twin terminal according to claim 7, characterized in that, The process of achieving bidirectional synchronization includes the following steps: Controlling physical entities involves converting control signals u(t) into control commands for physical actuators, which then drive the physical entity's state adjustment to achieve synchronous updates at the physical end. The virtual model is updated based on the control signal u(t) and the state estimation results. Update the parameters and initial conditions of the virtual model to make the virtual model state... It is used to achieve synchronous updates of the virtual terminal.

9. A water conservancy digital twin system, used in the adaptive feedback control method for virtual-real synchronization of the water conservancy digital twin terminal as described in any one of claims 1-8 to achieve virtual-real synchronization between physical water conservancy entities and virtual digital models, characterized in that, It includes a data base, a model platform, a knowledge platform, and an adaptive controller. The adaptive controller interfaces with the data base, the model platform, and the knowledge platform to achieve data interaction and collaborative control.

Images