CN118131825B

CN118131825B - Water level control method and device for throttle valve

Info

Publication number: CN118131825B
Application number: CN202410546455.7A
Authority: CN
Inventors: 程楠; 陈文龙; 何立新; 雷晓辉; 张峥; 龙岩; 段清; 王二朋; 史博阳; 张宏洋; 刘晓龙; 郭图南
Original assignee: Hebei University of Engineering
Current assignee: Hebei University of Engineering
Filing date: 2024-05-06
Publication date: 2024-07-12
Anticipated expiration: 2044-05-06

Abstract

The invention discloses a method and a device for controlling the water level of a throttle valve, which relate to the technical field of water level control of the throttle valve of hydraulic engineering, and comprise the following steps: the water level of the throttle valve is obtained by a sensor, compared with a preset target water level to obtain a water level error, and the water level is obtained by a feedback control algorithm to obtain a feedback increment; judging the target water discharge amount of the throttle valve according to the water level, updating the predicted water discharge amount to acquire a real-time water discharge correction amount, and obtaining a correction requirement; establishing a water level dynamic model by combining the feedback increment with the correction requirement; predicting the water level change of a future throttle valve by combining a water level dynamic model with a real-time state, and determining a gate opening sequence by setting a water level control problem by using a water level prediction model; and (3) inputting the gate opening sequence as a control instruction into a throttle gate controller to adjust the gate opening so as to realize the water level control of the throttle gate. The invention can realize the water level control of the throttle valve, improve the control efficiency and lighten the manual burden.

Description

Water level control method and device for throttle valve

Technical Field

The invention relates to the technical field of water level control of a hydraulic engineering throttle valve, in particular to a throttle valve water level control method and device.

Background

In recent years, with the development of social economy and the increase of urban population, the contradiction between supply and demand of water resources in China is prominent, and long-distance water regulation and cross-river basin water diversion through channel water delivery are effective ways for solving the water resource problem. In the water diversion engineering, open channel water diversion is a mainstream mode, namely water delivery is carried out through artificial excavation or natural channels, regulating and controlling buildings such as water diversion ports and control gates are arranged along the line, and the hydraulic water quantity is controlled by formulating a reasonable regulating and controlling scheme so as to realize fine regulation and control.

In recent years, along with the development of an automation theory and the practical application of a water diversion project, an automation and intelligent control technology in the water diversion process of a channel is developed to a certain extent, but the hydraulic response in the water diversion process is complex, random fluctuation is easy to generate, so that the instability of the system is caused, and meanwhile, the water diversion project is long-distance water diversion, the propagation speed of water wave motion is slower, and the time lag is larger, so that the automatic control system of the channel gate is established, the hydraulic water quantity management is realized in a refined manner, and the method has important significance for the development of the water diversion project of the channel.

However, in the prior art, as the dynamic process of the channel water delivery system is large in uncertainty, the mechanism process is strong in nonlinearity, and the control target is multidimensional, a complex running state control process can be generated, if the water level state cannot be effectively controlled in the control process, the situation that the channel water delivery system is difficult to control can be caused, and the control effect is reduced.

For the problems in the related art, no effective solution has been proposed at present.

Disclosure of Invention

(One) solving the technical problems

Aiming at the defects of the prior art, the invention provides a water level control method and a water level control device for a throttle valve, which have the advantages of automatically adjusting the opening degree of the throttle valve, realizing the water level control of the throttle valve, improving the control efficiency and reducing the manual burden, and further solve the problem that the water delivery of a channel is difficult to control.

(II) technical scheme

In order to realize the automatic adjustment of the opening of the gate and the water level control of the throttle gate, the invention has the advantages of improving the control efficiency and reducing the labor burden, and adopts the following specific technical scheme:

in a first aspect, the present invention provides a throttle water level control method comprising the steps of:

S1, acquiring the water level of a throttle valve by using a sensor, comparing the water level with a preset target water level to obtain a water level error, and taking the water level error as the input of a feedback control algorithm to obtain a water level quantity to obtain a feedback increment;

S2, judging the target water discharge amount of the throttle valve according to the water level, and updating the predicted water discharge amount based on the target water discharge amount to acquire a real-time water discharge correction amount to acquire a correction requirement;

S3, combining the feedback increment with the correction requirement to obtain the gate flow, calculating the gate opening based on the gate flow, acquiring the mathematical relationship between the water level and the gate opening, and establishing a water level dynamic model;

s4, predicting water level change of a future throttle valve by combining a water level dynamic model with a real-time state, and determining a gate opening sequence by setting a water level control problem by using a water level prediction model based on the water level change;

s5, inputting a gate opening sequence as a control instruction into a throttle gate controller, adjusting the gate opening to control the water level of the throttle gate, and acquiring water level data in a preset time period;

And S6, establishing an evaluation index according to the water level data to estimate the performance index of the water level control of the throttle valve, judging the water level state of the throttle valve based on the performance index, and optimizing and adjusting the control sequence to update the control state.

Preferably, the method for obtaining the gate flow by combining the feedback increment and the correction requirement, calculating the gate opening based on the gate flow, and establishing the water level dynamic model by acquiring the mathematical relationship between the water level and the gate opening comprises the following steps:

S31, comprehensively obtaining flow variation by the feedback flow and the corrected flow, and obtaining the number of sections in the throttle and the contact relation between the bottom end of the throttle and the water surface;

S32, reflecting the overflow capacity of the throttle valve under different conditions based on the combination of the flow variation and the number of sections and the contact relation to obtain the flow rate of the throttle valve, and setting a water flow inlet and a water flow outlet as an outer boundary and taking the water flow in the range of the throttle valve area as an inner boundary;

S33, establishing a control equation by combining the Saint Violet equation with the number of sections, introducing an external boundary and an internal boundary after the establishment is completed to generalize to form an algebraic equation set, and writing the algebraic equation set into a matrix form to establish a water level dynamic equation;

S34, coupling the sluice flow and a water level dynamic equation to achieve the purpose of simulating water condition change in the water delivery process of the inner boundary of the sluice, and setting one hour in the simulation process to acquire water flow and water level data for acquisition frequency;

S35, performing frequency-reducing operation on water flow and water level data to judge each section, water level and flow value, and obtaining a submerged outflow value based on the simulation data to solve a gate flow coefficient;

S36, analyzing the water flow form in the water delivery process according to the gate flow coefficient to obtain the opening form of the gate, and judging the mathematical relationship between the gate flow coefficient and the opening form to establish a water level dynamic model.

Preferably, the method for analyzing the water flow form in the water delivery process according to the gate flow coefficient to obtain the opening form of the gate, and judging the mathematical relationship between the gate flow coefficient and the opening form to establish the water level dynamic model comprises the following steps:

S361, designing a fitting model by adopting a curved surface response method based on the gate flow coefficient to analyze the water flow form in the water delivery process, and judging the opening form of the gate according to the contact relation between the bottom end of the gate and the water surface;

s362, inputting gate flow coefficients with different values into a fitting model, solving a fitting relation between the gate flow coefficients and the opening form, and acquiring a mathematical relation between the gate flow coefficients and the opening form based on a fitting result;

S363, defining water level stability and response time performance indexes, determining constraint conditions of a throttle controller, external disturbance and physical constraints of gate opening, and extracting fitting data in a fitting model after definition is completed;

s364, constructing a water level change coordinate system according to the definition content and the fitting data to obtain the gate opening morphological parameters and the fluctuation, and establishing a water level dynamic model based on digital twin by combining the throttle structure.

Preferably, constructing a water level change coordinate system according to the definition content and the fitting data to obtain the gate opening morphological parameters and the fluctuation, and establishing a water level dynamic model based on digital twin by combining the throttle structure comprises the following steps:

s3641, selecting an average value of gate flow coefficients in fitting data as a reference point of a water level coordinate system, and constructing a water level change coordinate system by taking response time as an abscissa and water level as an ordinate;

s3642, analyzing a water level change rule by using a water level change coordinate system to determine a change trend line, and analyzing the slope and intercept of the water level change based on the change trend line to serve as a gate opening morphological parameter and a variation;

s3643, constructing a twin model capable of reflecting water level dynamics by a throttle structure obtained through a digital measurement technology, and performing separation, extraction, filtering and fitting operations on digital points of gate opening morphological parameters and fluctuation;

S3644, carrying out fitting experiments on the data points by using a least square method and a twin model, adding marks to the data points in the model, solving the least square solution of the whole data points by using a decomposition technology to obtain an end surface model, and combining the end surface model and the twin model to form a water level dynamic model.

Preferably, predicting the water level change of the future throttle valve by combining a water level dynamic model with a real-time state, and determining the gate opening sequence by setting a water level control problem by using a water level prediction model based on the water level change comprises the following steps:

S41, monitoring real-time water level state data by using a sensor in a throttle region, preprocessing the water level data, and inputting the water level data into a water level dynamic model to predict water level change at a future throttle;

S42, setting a water level control target according to a water level change prediction result, and combining water level control with gate opening control to establish a multivariable control problem;

S43, establishing a state space by using a water level prediction model based on the multivariable control problem, calculating a gate opening sequence in the process of realizing the water level control target, and selecting an optimal opening sequence to be applied to water level control.

Preferably, the method for establishing a state space by using a water level prediction model based on a multivariable control problem calculates a gate opening sequence in the process of realizing a water level control target, and selects an optimal opening sequence to be applied to water level control comprises the following steps:

S431, a water level prediction model is used for introducing a mapping relation between flow and water level at a section of a throttle valve and the influence of gate opening increment on the water level and the flow according to a multivariable control problem, a state space model is built, and a quadratic definition cost function is introduced;

s432, establishing a water delivery control optimization model according to the starting speed constraint, the water passing capacity constraint and the water level fluctuation range constraint of the throttle valve in the water level control process and combining the operation characteristics of the throttle valve;

S433, defining input values of water level values at preset time and next time according to the state space model, and setting an expected output value of the state space model to be zero to judge the opening of the gate;

s434, judging that an error value between the preset moment and the gate opening at the next moment is brought into a cost function, and converting the problem into an objective function for solving the minimum value of the cost function by using a quadratic programming solving mode;

S435, carrying out minimum value solving through a secondary objective function solver, obtaining a gate opening sequence according to the solving, wherein a first control action in the opening sequence is transmitted to a throttle controller, and the rest opening sequences are ignored;

And S436, when the water level state variable is updated after the gate acts, entering a model prediction stage of the next stage, carrying out objective function solving, continuously carrying out opening sequence output, and continuously carrying out rolling solving within a certain range to obtain a gate opening sequence to be applied to water level control.

Preferably, the step of establishing an evaluation index according to water level data to estimate the performance index of the water level control of the throttle valve, judging the water level state of the throttle valve based on the performance index and optimizing and adjusting the control sequence to update the control state comprises the following steps:

S61, collecting water level data in a preset time period in the throttle valve, and judging a water level fluctuation range and the time required for reaching a target water level according to the water level data;

S62, defining an evaluation index containing water level stability and control response time as criteria according to the water level fluctuation range and the time required for reaching the target water level;

s63, establishing an evaluation index system based on the evaluation index, performing sorting operation on the evaluation index by adopting a spectral clustering algorithm, and constructing an evaluation model based on the evaluation index system and an index sorting result;

S64, estimating performance indexes of the water level control of the throttle valve by using an evaluation model, dividing the performance indexes into five groups of levels, and making a corresponding adjustment plan according to the level classification;

S65, inputting the water level data in the real-time throttle valve into an evaluation model to obtain an output result, and comparing the output result with the performance index to judge the level of the output result to obtain the water level state adjustment control sequence update control state of the throttle valve.

Preferably, establishing an evaluation index system based on the evaluation index, performing a sorting operation on the evaluation index by adopting a spectral clustering algorithm, and constructing an evaluation model based on the evaluation index system and the index sorting result comprises the following steps:

s631, classifying and layering the evaluation indexes to construct a multi-level evaluation index system, and using a hierarchical analysis method to distribute weights for the indexes to reflect the influence degree of different indexes on an evaluation result;

S632, constructing a similarity matrix according to the evaluation indexes, reflecting the similarity degree among the indexes, and dividing the similarity matrix by using a spectral clustering algorithm to set two groups of index groups;

S633, ordering the evaluation indexes in the index group to reflect the relative importance among the indexes, and designing the structure of an evaluation model according to an evaluation index system and a spectral clustering ordering result to determine the effect and the interrelationship of each index in the model.

Preferably, the expression of the evaluation model is:

；

Wherein L represents the evaluation result; s _Z represents a spectral cluster sequencing result; d _Z denotes a weight vector of the evaluation index; z _ij represents the weight of the i-th evaluation index; NMI represents the amount of mutual information of the evaluation index criteria; w _i denotes the similarity between the i-th index and the remaining indexes.

In a second aspect, the invention also provides a throttle water level control device, which comprises a feedback increment acquisition module, a correction demand acquisition module, a dynamic model establishment module, an opening sequence determination module, a gate water level control module and a performance index evaluation module;

the feedback increment acquisition module is used for comparing the water level of the throttle valve obtained by the sensor with a preset target water level to obtain a water level error, and obtaining the water level quantity by taking the water level error as the input of a feedback control algorithm to obtain a feedback increment;

the correction demand acquisition module is used for judging the target water discharge amount of the throttle valve according to the water level amount, updating the predicted water discharge amount based on the target water discharge amount, and acquiring a real-time water discharge correction amount to obtain the correction demand;

The dynamic model building module is used for combining the feedback increment with the correction requirement to obtain the gate flow, calculating the gate opening based on the gate flow, obtaining the mathematical relationship between the water level and the gate opening, and building a water level dynamic model;

The opening sequence determining module is used for predicting the water level change of the future throttle valve by combining a water level dynamic model with a real-time state and determining the opening sequence of the gate by using a water level prediction model to set a water level control problem based on the water level change;

The gate water level control module is used for inputting a gate opening sequence as a control instruction into the throttle gate controller, adjusting the gate opening to realize water level control of the throttle gate, and acquiring water level data in a preset time period;

the performance index evaluation module is used for establishing an evaluation index according to the water level data to evaluate the performance index of the water level control of the throttle valve, judging the water level state of the throttle valve based on the performance index and optimizing and adjusting the control sequence to update the control state.

(III) beneficial effects

Compared with the prior art, the invention provides a method and a device for controlling the water level of a throttle valve, which have the following beneficial effects:

(1) The invention can accurately adjust the water level by acquiring the error of the water level and the preset target water level in real time and taking the error as the input of a control algorithm, reduces the error, dynamically judges the target water discharge amount through the current water level, updates the water discharge amount based on actual conditions, can more flexibly adapt to environmental changes, optimizes the water discharge efficiency, determines the passing gate flow by combining feedback increment and correction requirement, establishes a mathematical relationship between the water level and the gate opening, predicts the dynamic relationship between the water level and the gate opening, predicts the future water level change of the throttle gate by combining a water level dynamic model with a real-time state, provides scientific basis for water level control, enables a gate opening sequence to be input into a throttle gate controller as a control instruction, automatically adjusts the gate opening, realizes the water level control of the throttle gate, improves the control efficiency and reduces the labor burden.

(2) The invention synthesizes the feedback flow and the correction flow to obtain more accurate flow variation, provides an accurate input value for adjusting the passing gate flow, enhances the basic accuracy of water level control, can more accurately reflect the overflow capacity of the throttle gate based on the combination of the flow variation and the number of sections and the contact relation, provides a more detailed basis for the calculation of the passing gate flow, establishes a control equation by utilizing the san-View process, and introduces an algebraic equation set formed by generalization of an outer boundary and an inner boundary to accurately describe the dynamic process of the water level variation, so that the accuracy and the response speed of the water level control of the throttle gate can be improved, a powerful tool is provided for deeply analyzing and understanding the water flow dynamics, and the efficiency and the safety in aspects of optimizing water resource management, flood control, water supply and the like are facilitated.

(3) According to the invention, the water level dynamic model is utilized to predict the water level change of the future throttle valve in combination with the real-time state, a model prediction control algorithm is utilized to determine the gate opening sequence based on the prediction result, the water level prediction model is utilized to consider the future water level change prediction and the current control target, the optimal gate opening sequence in the process of realizing the water level control target is calculated, the precision and efficiency of water level control are improved, the quality and effect of control measures are ensured, the water level fluctuation and control cost are reduced, and a plurality of control variables and future prediction information are considered based on the water level prediction model, so that the control strategy is more flexible and has strong adaptability, and the high-efficiency and accurate control of the water level of the throttle valve is realized.

(4) The invention can primarily evaluate the stability and response speed of the water level control by judging the water level fluctuation range and the time required for reaching the target water level, define the water level stability and the control response time as evaluation criteria, is beneficial to the evaluation of standardized water level control performance, can evaluate the key performance of the water level control by focusing on the water level stability and the control response time, and ensures the effectiveness and the high efficiency of a control strategy.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method of throttle water level control in accordance with an embodiment of the present invention;

FIG. 2 is a schematic diagram of a throttle water level control device according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a PID control algorithm according to an embodiment of the invention;

Fig. 4 is a schematic diagram of a water level prediction model control algorithm according to an embodiment of the present invention.

In the figure:

1. a feedback increment acquisition module; 2. a correction demand acquisition module; 3. a dynamic model building module; 4. an opening sequence determining module; 5. a gate water level control module; 6. and a performance index evaluation module.

Detailed Description

For the purpose of further illustrating the various embodiments, the present invention provides the accompanying drawings, which are a part of the disclosure of the present invention, and which are mainly used for illustrating the embodiments and for explaining the principles of the operation of the embodiments in conjunction with the description thereof, and with reference to these matters, it will be apparent to those skilled in the art to which the present invention pertains that other possible embodiments and advantages of the present invention may be practiced.

According to the embodiment of the invention, a method and a device for controlling the water level of a throttle valve are provided.

The invention will now be further described with reference to the accompanying drawings and detailed description, as shown in fig. 1, a method for controlling a water level of a throttle valve according to an embodiment of the invention, the method comprising the steps of:

s1, acquiring the water level of the throttle valve by using a sensor, comparing the water level with a preset target water level to obtain a water level error, and taking the water level error as the input of a feedback control algorithm to obtain the water level quantity to obtain a feedback increment.

In this embodiment, the step of obtaining the water level error by comparing the water level of the throttle valve obtained by the sensor with the preset target water level and obtaining the feedback increment by using the water level error as the input of the feedback control algorithm includes the following steps:

And water level sensors are arranged at proper positions on the upstream and downstream of the throttle valve and are used for monitoring the water level condition in real time, continuously collecting real-time water level data of the throttle valve through the sensors, checking whether the collected data are accurate or not, and eliminating possible error readings such as abnormal values or jumps.

And setting a preset target water level according to the running requirement or specific requirement of the water conservancy system, and calculating the difference between the real-time water level and the preset target water level, namely a water level error, according to real-time water level data acquired from the sensor.

The PID control algorithm is selected as a feedback control algorithm, parameters of the control algorithm including a proportional coefficient (P), an integral coefficient (I) and a differential coefficient (D) of the PID control algorithm are set according to actual demands, water level errors are used as input of the control algorithm, and the control algorithm is utilized to calculate water level quantity to be adjusted, namely feedback increment.

As shown in the schematic diagram of the PID control algorithm of fig. 3, it should be explained that the PID control algorithm (pro-port-integrated-DERIVATIVE CONTROL ALGORITHM) is a commonly used feedback control algorithm for adjusting the output of the control system to achieve a desired target, and the PID control algorithm adjusts the output signal of the control system based on the feedback principle according to the difference (i.e., error) between the actual output and the desired output, so as to gradually reduce the error and stabilize the system.

The three main components of the PID control algorithm are:

(1) Ratio (proportial): proportional control acts on the current error, multiplies the error by a proportionality constant, and takes the result as part of the control signal, so that the output varies in proportion to the magnitude of the error, and proportional control can quickly reduce the error, but can cause the system to oscillate or overshoot.

(2) Integral (integrate): the integral control acts on the accumulated value of the error, accumulates the integral of the error, multiplies the integral constant, adds the result to the control signal, and can eliminate steady-state error of the system, i.e. keep the error small for a long time.

(3) Differential (differential): differential control acts on the rate of change of the error, multiplies the rate of change of the error by a differential constant, and adds the result to the control signal. The differential control can predict the future error change trend of the system, thereby inhibiting the system from oscillating and stabilizing the system. However, in the case of more noise, excessive differentiation may introduce more noise.

The specific error e (t) is expressed as the difference between the expected value and the actual output of the system:

；

The output signal calculation formula of the PID control algorithm is as follows:

；

Where e (T) denotes the error, r (T) denotes the expected value of the system, y (T) denotes the actual output value, u (T) denotes the output signal value, K _P is the scaling factor, T _I、T_D is the time constant of integration and differentiation, respectively, the magnitude of which depends on the characteristics and control requirements of the system, K _p e (T) denotes the scaling result, The differential value is represented by a differential value,Representing the integrated value.

In PID control, the proportional control can make the system reach a stable state quickly, the integral control can eliminate the steady state error of the system, the differential control can predict the future error change trend of the system so as to inhibit the system from oscillating and stabilizing the system, but because the differential control has poor effect on the time-lag system, the PI control is generally only adopted for the system with large time-lag characteristics such as water delivery of a throttle.

And S2, judging the target water discharge amount of the throttle valve according to the water level amount, updating the predicted water discharge amount based on the target water discharge amount, and obtaining a real-time water discharge correction amount to obtain the correction requirement.

In this embodiment, the method for determining the target water discharge amount of the throttle according to the water level and updating the predicted water discharge amount based on the target water discharge amount to obtain the real-time water discharge correction amount to obtain the correction requirement includes the following steps:

And continuously monitoring water level conditions by using water level sensors arranged at the upstream and downstream of the throttle valve, recording real-time water level data, analyzing the relationship between the water level and the water discharge amount, determining the water discharge amount corresponding to the water level, and determining the target water discharge amount of the throttle valve according to the current water level and the hydrologic condition.

The method comprises the steps of setting initial predicted water discharge based on the current water level and a set rule, monitoring the actual water discharge of a throttle valve in real time by using equipment such as a flowmeter, recording the real-time water discharge, comparing the real-time water discharge with a target water discharge, analyzing the difference, calculating the difference between the real-time water discharge and the target water discharge, and determining a water discharge correction amount to be adjusted.

And S3, combining the feedback increment with the correction requirement to obtain the sluice flow, calculating the opening of the gate based on the sluice flow, acquiring the mathematical relationship between the water level and the opening of the gate, and establishing a water level dynamic model.

In this embodiment, combining the feedback increment with the correction requirement to obtain the gate flow, calculating the gate opening based on the gate flow, and obtaining the mathematical relationship between the water level and the gate opening to build the water level dynamic model includes the following steps:

S31, comprehensively obtaining flow variation by the feedback flow and the correction flow, and obtaining the number of sections in the throttle and the contact relation between the bottom end of the gate and the water surface.

Specifically, the method for comprehensively obtaining the flow variation by the feedback flow and the correction flow and obtaining the number of sections in the throttle and the contact relation between the bottom end of the gate and the water surface comprises the following steps:

And comprehensively analyzing the corrected flow and the feedback flow, calculating the flow change quantity to reflect the flow adjustment quantity required by the current state to reach the target state, defining the section in the throttle valve, namely the cross section through which the water flows, and determining the number of the sections in the throttle valve based on the design parameters and the actual structure of the throttle valve.

The position of the bottom end of each gate is measured, the relative height of the gate and the water surface is included, the contact relation between the gate and the water surface is analyzed according to the position and the water level of the bottom end of the gate, the gate outflow and the weir flow can be divided according to whether the bottom end of the gate touches the water surface, and the free outflow and the submerged outflow can be divided according to whether the water level downstream of the gate influences the overflow capacity of the gate.

S32, reflecting the overflow capacity of the throttle valve under different conditions based on the combination of the flow variation and the number of sections and the contact relation to obtain the flow rate of the throttle valve, and setting a water flow inlet and a water flow outlet as an outer boundary and taking the water flow in the range of the throttle valve area as an inner boundary.

Specifically, the method for obtaining the flow of the throttle valve based on the flow variation and the overflow capacity of the throttle valve under different conditions by combining the number of sections and the contact relation, and setting the water flow inlet and outlet as an outer boundary, wherein the water flow in the region of the throttle valve is taken as an inner boundary, comprises the following steps:

The required flow change is determined, the flow change which needs to be realized by the throttle valve is determined by combining the feedback flow and the correction flow, the number of sections in the throttle valve, namely the number of channels through which water flows can pass, and the contact relation between the bottom end of the gate in each section and the water surface is analyzed to estimate the overflow capacity of each section.

Based on the number of sections and the contact relation of each section, the total overcurrent capacity of the throttle valve under the current condition is evaluated, the actual throttle valve flow is calculated by combining the flow variation and the total overcurrent capacity, and the upstream water inflow port and the downstream water inflow port of the throttle valve are taken as external boundaries.

S33, establishing a control equation by combining the Saint Violet equation with the number of sections, introducing an external boundary and an internal boundary after the establishment is completed to generalize to form an algebraic equation set, and writing the algebraic equation set into a matrix form to establish a water level dynamic equation.

Specifically, a control equation is established by combining the san-View equation with the number of sections, an algebraic equation set is formed by generalizing an external boundary and an internal boundary after establishment, and the algebraic equation set is written into a matrix form to establish a water level dynamic equation, which comprises the following steps:

According to the number of effective sections in the throttle, the position and the characteristic of each control section are determined, a corresponding water flow control equation is established for each section based on the Save et al equation and the determined control section, the external boundary conditions of the water flow inlet and outlet of the throttle are determined, and the water flow conditions in the throttle are analyzed and defined as the internal boundary, such as flow distribution, flow velocity change and the like of different gate areas.

The control equation and the boundary condition are combined and converted into algebraic equation to reflect the water flow state under the section, the variables and coefficients in the algebraic equation set are extracted and constructed into matrix form, the algebraic equation set is converted into matrix equation, the relation among elements in the matrix equation is analyzed, and the dynamic process of the time and space changes of the variables such as water level, flow rate and the like is understood.

And S34, coupling the sluice flow and a water level dynamic equation, realizing the purpose of simulating the water condition change in the water delivery process of the inner boundary of the sluice, and setting one hour in the simulation process to acquire water flow and water level data for acquisition frequency.

Specifically, the method for coupling the sluice flow and the water level dynamic equation to realize the purpose of simulating the water condition change in the water delivery process of the inner boundary of the sluice, and setting one hour in the simulation process to collect water flow and water level data for collecting frequency comprises the following steps:

and determining a coupling variable between the sluice flow and the water level dynamic equation, connecting the sluice flow equation and the water level dynamic equation through the coupling variable to form a coupling model, and setting water flow conditions and external boundary conditions in the water delivery process of the inner boundary of the sluice.

Determining the time step length in the simulation process to be one hour to control the time precision of the simulation, discretizing the space range of the model, and selecting a numerical solution method of a finite difference method to simulate the water flow process and collect primary water flow and water level data.

S35, performing frequency-reducing operation on the water flow and water level data to judge each section, water level and flow value, and obtaining a submerged flow value based on the simulation data to solve the gate flow coefficient.

Specifically, performing a frequency-reducing operation on water flow and water level data to determine each section, water level and flow value, and obtaining a submerged outflow value based on analog data to solve a gate flow coefficient comprises the following steps:

selecting low-pass filtering to reduce noise brought by data acquisition frequency for a frequency reduction method, extracting main trend, processing acquired water flow and water level data according to the selected frequency reduction method to obtain frequency reduced data, analyzing water flow characteristics of each section, such as flow distribution and water level change, according to the frequency reduced data, determining the relationship between water level and flow of each section, extracting key parameters, such as submerged flow values under different water level conditions, from simulation data, and solving gate flow coefficients by regression analysis or other statistical methods by utilizing the extracted submerged flow values and corresponding water level data.

Specifically, the method for analyzing the water flow form in the water delivery process according to the gate flow coefficient to obtain the opening form of the gate, and judging the mathematical relationship between the gate flow coefficient and the opening form to establish the water level dynamic model comprises the following steps:

The method for constructing a water level change coordinate system according to definition content and fitting data to obtain gate opening morphological parameters and fluctuation, and establishing a water level dynamic model based on digital twin by combining a throttle structure comprises the following steps:

Specifically, assuming that there are L sections in the channel of the throttle, each section has two unknowns of water level and flow, 2L unknowns are total, (L-1) river segments, 2 (L-1) control equations can be established according to the discrete san. View equation set, and the closed algebraic equation set is formed by adding external boundary conditions on the upper and lower sides of the river channel. The outer boundary conditions of the water level dynamic model refer to the upstream boundary (water flow inlet) and the downstream boundary (water outlet) in the water delivery process of the channel.

Boundary conditions in the water level dynamic model are generally divided into three cases: an upstream water level boundary, a downstream water level boundary, an upstream water level boundary, a downstream flow boundary, and an upstream flow boundary, a downstream water level boundary. When the downstream boundary is a regulating building such as a throttle gate, the downstream boundary may be a water level-flow relation boundary, given the flow characteristics of the gate.

The different states of water flow in the water delivery process determine that the opening of the throttle valve has different forms, the throttle valve can be divided into a gate outlet flow and a weir flow according to whether the bottom end of the throttle valve touches the water surface or not, the throttle valve can be divided into a free outlet flow and a submerged outlet flow according to whether the water level at the downstream of the throttle valve influences the overflow capacity of the throttle valve or not, and the throttle valve overflow flow formula is a basic formula for reflecting the overflow capacity of the throttle valve under different conditions. Considering the interaction of the sluice flow and the upstream and downstream water, the sluice flow formula of the sluice is coupled with the discrete of the san Vena process, so that the water condition change simulation of the inner boundary of the sluice is realized. For a general throttle gate, neglecting the flow loss when the gate is crossed, the flow at the section j before the gate is equal to the flow at the section j+1 after the gate, and the continuous equation can be obtained as follows:

；

Wherein Q represents the passing gate flow rate, and Q _j and Q _j+1 represent the flow rates at the gate front section j and the gate rear section j+1, respectively.

For example, a gate sill is processed in the form of a wide top weir, and the flow formula is as follows, taking the outflow of the gate Kong Yanmei as an example:

；

discretization and linearization gate passing flow formula:

；

Wherein e represents the opening of the brake hole; b represents the width of the gate hole; z _j、Z_j+1 represents the water depths of the j and j+1 th sections, namely the section water level before the gate and the section water level after the gate; mu is the integrated flow coefficient, g is the gravitational acceleration, and n and n+1 represent the n and n+1 times respectively.

Establishing a water level dynamic equation:

；

Determining the mathematical relationship between water level and gate opening, and establishing a linear formula with specific valve design and fluid properties: ；

in the method, in the process of the invention, Is the inflow flow; is the outflow flow; h is the water level height, k is a scaling factor, Representing the derivative of the hydrologic high order with time.

And continuously collecting data to perform parameter calibration, determining a gate flow coefficient, defining water level stability and response time performance indexes, setting the maximum allowable value of the water level deviated from the target water level to be 5%, and obtaining the fluctuation frequency through frequency domain analysis. Setting response time performance indexes: the rise time of the system from the initial water level to the target water level was 10 minutes, the transition time of the system from the initial water level to the target water level was 15 minutes, and the time for the system to reach the maximum peak of the water level change was 5 minutes.

And determining system constraint and external disturbance, and setting the maximum opening to be 80% and the minimum opening to be 10% according to the valve structure and design. The maximum allowable flow of the throttle valve is 100 cubic meters per second, external disturbance is disturbance possibly generated to the system such as rainfall, snow melting, upstream water level change, surge and the like, physical constraint of the opening of the throttle valve is determined, and in addition to the valve structure and design, the maximum opening change rate of the valve is set to be 2% per minute in order to avoid impact and vibration.

Therefore, the feedback flow and the correction flow are integrated to obtain more accurate flow variation, an accurate input value is provided for adjusting the passing gate flow, the basic accuracy of water level control is enhanced, the overflow capacity of the throttle gate can be reflected more accurately based on the flow variation combined with the number of sections and the contact relation, and a finer basis is provided for calculating the passing gate flow.

S4, predicting the water level change of the future throttle valve by combining the water level dynamic model with the real-time state, and determining the gate opening sequence by setting the water level control problem by using the water level prediction model based on the water level change.

In this embodiment, predicting a water level change of a future throttle valve by combining a water level dynamic model with a real-time state, and determining a gate opening sequence by setting a water level control problem based on the water level change by using a water level prediction model includes the following steps:

S41, monitoring real-time water level state data by using a sensor in the throttle area, preprocessing the water level data, and inputting the water level data into a water level dynamic model to predict water level change at a future throttle.

Specifically, the method for monitoring real-time water level state data by using the sensor in the throttle region, preprocessing the water level data, and then inputting the water level data into the water level dynamic model to predict the water level change of the future throttle, comprises the following steps:

The method comprises the steps of monitoring water level data of a throttle area in real time by using a sensor, recording the change condition of the water level, checking the collected water level data, removing error data or abnormal values, processing the data by applying proper data smoothing technology (such as moving average, median filtering and the like), reducing noise influence, improving the usability and accuracy of the data, analyzing the trend of the water level data, identifying possible ascending or descending modes, extracting useful characteristics of a prediction model according to a time range required to be predicted, including historical water level change rate, time sequence characteristics and the like, and inputting the latest water level data after pretreatment into the water level dynamic model to execute water level prediction operation.

S42, setting a water level control target according to a water level change prediction result, and combining water level control with gate opening control to establish a multivariable control problem.

Specifically, setting a water level control target according to a water level change prediction result, and establishing a multivariable control problem by combining water level control and gate opening control includes the following steps:

Analyzing the output of the dynamic water level prediction model, ensuring the accuracy and reliability of the prediction result, identifying the trend of the predicted water level, such as ascending, descending or stabilizing, and the like, and setting a proper water level control target according to the water supply requirement.

Analyzing and establishing a relation model between the gate opening and the water level, understanding the influence of different openings on water level control, verifying the accuracy of the relation model between the gate opening and the water level through historical data or small-scale tests, identifying all key control variables such as the gate opening, the water level and the like, designing a multivariable control strategy, and determining how to adjust each control variable to achieve a water level control target.

Specifically, the method for establishing a state space by using a water level prediction model based on the multivariable control problem to calculate the gate opening sequence in the process of realizing the water level control target, and selecting the optimal opening sequence to be applied to the water level control comprises the following steps:

It should be explained that modeling the water level control problem as a water level predictive model control (MPC) optimization problem includes building a state space model:

；

Wherein k is the moment in a discrete state; x (k) is a model state variable; u (k) is model control input; a is a model matrix, and represents the mapping relation between the flow and the water level at the control section; b is a control matrix, and represents the influence of the opening increment of the model control input gate on the water level and the flow; y (k) is the model output; c is the output matrix.

For channel water delivery engineering, in the water delivery and distribution process, the throttle gate has a starting speed constraint and a water passing capacity constraint, the water level has a fluctuation range constraint, and the water delivery control optimization model is established by combining the actual operation characteristics of the throttle valve, and is as follows:

；

Where J represents the optimization result, p represents the predicted time domain step size, m represents the control time domain step size, Q, R is the adjustment matrix, typically the set diagonal matrix, A k +1 time state value representing a prediction of the k time,The k+1 time input value predicted at the k time is represented, k is the time in the discrete state, T represents the transpose of the matrix, and a new matrix is obtained by exchanging the rows and columns of the matrix.

The formula can be developed by:

；

Substituting system state and system input:

；

Wherein X _k represents a system state, Where C represents the output matrix and U _k represents the system input.

The problem is converted into a solution minJ objective function, and the function J is converted into a solution quadratic programming problem to be simplified and arranged to obtain:

；

Where k represents the time of day in the discrete state, The system state variable is defined as x, which is the water level and flow value of each section and building at time n+1, G represents the gate opening, and E represents the error.

It should be explained that:

；

Wherein A represents a system matrix, represents a mapping relation between flow and water level at a control section, B represents a control matrix, B represents an influence of an opening increment of a model control input gate on the water level and the flow, y (k) represents model output, C represents an output matrix, N represents line number, and M represents a matrix.

The control mode of the water level prediction model is the same as PI feedback control, the control target is that the water level before the downstream gate keeps the target water level, the water level before the downstream gate of the ditch is taken as a controlled variable, and the deviation between the measured water level and the target water level value is z (k), namely the controlled variable deviation e (k). The gate opening is used as a control variable, the gate opening value is expressed as G (k), namely the control quantity u (k), the water level prediction model is input as a water level value before the downstream gate of the current time canal, and the water level prediction model is output as a gate opening sequence. In the water level prediction model derivation, the setting error E tends to 0, in the channel water delivery system, the water level deviation E is X (k), at the time of k, the system state X _k can be acquired in real time through a sensor, the water level dynamic model is adopted to replace the known quantity, the water level prediction model is solved and converted into a quadratic programming solution form, and the quadratic objective function solver quadprog (H, f) in MATLAB is used for solving the minimum value of the objective function, wherein the minimum value of the objective function is as follows:

；

u (k) corresponds to x, f corresponds to Expressed as:

；

wherein E represents an error, Representing the system state, f representing the function value.

After the objective function is solved, a control variable U (k) sequence is obtained, a first control action U1 in the control sequence is transmitted to the field executor, other control variables are ignored, after the gate acts, the water level state variable is updated, the model prediction stage of the next stage is entered, the objective function is solved, the control quantity output is continued, and the solution is continuously rolled in a certain range.

It should be explained that, because of the large uncertainty of the dynamic process of the water delivery system of the throttle gate, the nonlinear mechanism process is strong, the control target is multidimensional, and the control process of the running state of the complex system is faced, the single PID control is difficult to ensure that the system performance is kept optimal, and the control effect is limited. Model predictive control (Model Predictive Control, MPC) is an advanced industrial system control algorithm for optimizing performance metrics in dynamic systems. The MPC uses a mathematical model in the control system to simulate the dynamic change process of the system, predicts the future behavior of the system, and makes decisions according to the prediction information to select the best control action at each time step.

The basic principle of operation of an MPC as shown in fig. 4 is that in each time step, based on the current system state and a future predicted time sequence, a series of possible control actions is calculated and one action from among them is selected to be applied in the system that optimizes the predicted performance. After the next time step, the MPC recalculates, feeds back the correction model by using the latest measurement data, and continues to optimize the control action, and the periodic iterative process is rolling optimization, so that the MPC model can adapt to system change and measurement errors, the control robustness is enhanced, and the MPC model is suitable for the control process of a large-time-lag system.

Model predictive control generally consists of three parts: predictive model, rolling optimization, feedback correction, each meaning:

Prediction model: the prediction model is a mathematical model for describing the dynamic behavior of the controlled system, and has the function of predicting the future output of the system according to the state of the controlled object, taking the prediction information as an optimal control input in rolling optimization, and dividing the prediction model into a mechanism model, a black box model, an experimental model and the like according to different construction mechanisms.

Scroll optimization (Rolling Optimization): the rolling optimization is the core of MPC, it refers to optimizing the predictive control in each time step to select the optimal control input sequence, in the limited time, the rolling optimization is online repeated optimization, the optimization problem is to optimize the predictive performance index by adjusting the control input in a future predictive time sequence. Meanwhile, only the first control input in the optimized control input sequence is adopted to be input into an actual system, and recalculated in the next time step, and the optimization is rolled again, and a quadratic objective function is generally adopted as a performance index function.

Feedback correction (Feedback Correction): the MPC optimizes based on the prediction model, but uncertainty errors and noise still exist between an actual system and the model, in order to better adapt to the change of the actual system, the MPC generally comprises a feedback correction mechanism, and in one time step, the MPC corrects the prediction of the prediction model according to the actual measurement quantity, more accurately reflects the current system state, and performs next optimization again after correction, so that closed-loop control is formed, and stability and accuracy of the model are ensured.

Therefore, the water level control method of the throttle valve set by the water level prediction model can control a plurality of variables such as water level, gate opening and the like, effectively process constraint conditions, and more flexibly cope with possible changes, thereby improving control performance, and simultaneously considering the energy consumption of the adjustment of the throttle valve opening, thereby reducing the energy consumption as much as possible and improving the energy efficiency of the system while maintaining the stable water level.

The method comprises the steps of establishing a water level dynamic model, simulating a dynamic change process, predicting the next action of a system, deciding according to prediction information so as to select the optimal control action at each time step, calculating a series of possible control actions based on the current system state and a section of future prediction time sequence in each time step, selecting an action which optimizes the prediction performance from the possible control actions to be applied to the system, recalculating the action after the next time step, feeding back the correction model by using the latest measurement data, continuing to optimize the control action, and performing rolling optimization in a periodical iterative process, so that the MPC model can adapt to system change and measurement errors, the robustness of control is enhanced, the method is more suitable for different operation targets of a water regulation stage, has anti-interference capability when facing uncertainty and disturbance, and is more intelligent and reliable in water level control.

S5, inputting the gate opening sequence as a control instruction into a throttle gate controller, adjusting the gate opening to control the water level of the throttle gate, and acquiring water level data in a preset time period.

In this embodiment, inputting the gate opening sequence as a control instruction into the throttle controller to adjust the gate opening to realize water level control of the throttle, and acquiring water level data in a preset time period includes the following steps:

Based on a water level control target and a multivariable control model, calculating gate opening adjustment instructions, distributing specific execution time points for each opening adjustment instruction, ensuring to meet water level control requirements in a preset time period, ensuring that a throttle controller has the capability of receiving and executing opening sequence instructions, including configuration and detection of software and hardware, carrying out transmission test of control instructions, and ensuring that the control instructions can be transmitted to the controller accurately.

The calculated gate opening sequence is used as a control instruction to be input into a throttle gate controller, after the controller receives the instruction, the gate opening is adjusted according to a preset time point and the sequence, the water level change at the throttle gate is monitored in real time while the gate opening is adjusted, the implementation effect of water level control is ensured to be in line with expectations, real-time water level monitoring data are fed back to a control system, and if a control strategy is required to be adjusted in real time.

In the whole control process, water level data are collected according to a preset collection frequency, and the collected data are recorded and stored so as to facilitate subsequent analysis and evaluation.

In this embodiment, the step of establishing an evaluation index according to water level data to estimate a performance index of the water level control of the throttle valve, determining a water level state of the throttle valve based on the performance index, and optimizing an adjustment control sequence to update the control state includes the steps of:

And S61, collecting water level data in a preset time period in the throttle valve, and judging the water level fluctuation range and the time required for reaching the target water level according to the water level data.

Specifically, collecting water level data in a preset time period in the throttle valve, judging a water level fluctuation range according to the water level data, and the time required for reaching a target water level, wherein the water level data comprises the following steps:

the water level data acquisition frequency is determined according to the starting time and the ending time of a preset time period, so that the continuity and the usability of the data are ensured, the water level monitoring equipment installed in the throttle area is ensured to be in a good working state, and the water level change can be accurately recorded.

According to the plan, water level data are collected in a preset time period, data integrity and accuracy are guaranteed, the collected water level data are checked to find out the highest water level and the lowest water level, and the fluctuation range is calculated.

And analyzing the frequency and the amplitude of the water level change, determining the fluctuation characteristic, determining the specific value of the target water level according to the water conservancy management requirement, and calculating the required time from the time point when the water level starts to change until the target water level is reached.

S62, defining an evaluation index containing water level stability and control response time as criteria according to the water level fluctuation range and the time required for reaching the target water level.

Specifically, defining an evaluation index containing water level stability and control response time as criteria according to the water level fluctuation range and the time required for reaching the target water level comprises the following steps:

Based on the time definition required for reaching the target water level, including the delay time from the control command to the beginning of the change of the water level and the total time from the beginning of the change of the water level to the target water level, data collection and processing are performed, water level data, particularly water level change data after executing the control command, are collected from the water level control system, and the collected data are subjected to cleaning and smoothing processes so as to more accurately calculate the evaluation index.

According to the defined water level stability index, calculating the water level fluctuation amplitude in the control process, simultaneously according to the defined control response time index, calculating the time required from the sending of a control instruction to the water level reaching a target value, comparing the calculated evaluation index with a preset performance standard or past performance data, and evaluating the performance of the current water level control system.

S63, establishing an evaluation index system based on the evaluation index, performing sorting operation on the evaluation index by adopting a spectral clustering algorithm, and constructing an evaluation model based on the evaluation index system and the index sorting result.

Specifically, establishing an evaluation index system based on the evaluation index, performing sorting operation on the evaluation index by adopting a spectral clustering algorithm, and constructing an evaluation model based on the evaluation index system and the index sorting result, wherein the method comprises the following steps:

Wherein, the expression of the evaluation model is:

；

S64, estimating performance indexes of the water level control of the throttle valve by using an evaluation model, dividing the performance indexes into five groups of levels, and making a corresponding adjustment plan according to the level classification.

Specifically, the performance index of the water level control of the throttle valve is estimated by using an evaluation model, the performance index is divided into five groups of levels, and a corresponding adjustment plan is formulated according to the level classification, and the method comprises the following steps:

and (3) estimating key indexes of the water level control performance of the throttle valve by using an estimation model, establishing an estimation standard and a calculation method for each performance index, and collecting data related to performance index calculation, including water level change data, control instruction execution time and the like.

And calculating various performance indexes by using the evaluation model and the preprocessed data, evaluating the performance of the throttle water level control according to the calculation result, dividing the result of the performance indexes into five levels (such as excellent, good, medium, bad and extremely bad), and setting specific standards and thresholds for each level.

According to the performance index result, the water level control performance of the throttle valve is classified into corresponding levels, the reason for the current performance level is analyzed for the performance of the non-excellent level, and specific adjustment measures are formulated for the performance of each level, including improving the control strategy, adjusting the operation mode of the throttle valve, enhancing the monitoring system and the like.

Specifically, inputting water level data in the real-time throttle valve into an evaluation model to obtain an output result, and comparing the output result with performance indexes to judge the level of the output result to obtain the updated control state of the throttle valve water level state adjustment control sequence, wherein the updated control state comprises the following steps:

and (3) monitoring water level data in real time by utilizing a water level sensor in the throttle area, performing preliminary processing on the collected real-time water level data, removing noise and abnormal values, converting or standardizing the data to meet the input requirement of an evaluation model, and inputting the preprocessed real-time water level data into the evaluation model.

The evaluation model calculates the current performance index according to the input data, analyzes the output result of the evaluation model, determines the current performance index, judges the level of the performance of the current water level control according to the preset performance index level standard, and generates a corresponding water level state adjustment control sequence of the throttle valve according to the level of the performance and the preset adjustment strategy.

As shown in fig. 2, according to another embodiment of the present invention, there is further provided a throttle water level control apparatus, which includes a feedback increment acquiring module 1, a correction demand acquiring module 2, a dynamic model establishing module 3, an opening sequence determining module 4, a throttle water level control module 5, and a performance index evaluating module 6;

The feedback increment acquisition module 1 is used for acquiring the water level of the throttle valve by using a sensor, comparing the water level with a preset target water level, obtaining a water level error, and obtaining a water level quantity by taking the water level error as the input of a feedback control algorithm to obtain a feedback increment;

the correction demand acquisition module 2 is used for judging the target water discharge amount of the throttle valve according to the water level amount, updating the predicted water discharge amount based on the target water discharge amount, and acquiring a real-time water discharge correction amount to obtain the correction demand;

The dynamic model building module 3 is used for combining the feedback increment and the correction requirement to obtain the sluice flow, calculating the opening of the gate based on the sluice flow, obtaining the mathematical relationship between the water level and the opening of the gate, and building a water level dynamic model;

The opening sequence determining module 4 is used for predicting the water level change of the future throttle valve by combining a water level dynamic model with a real-time state and determining the opening sequence of the gate by using a water level prediction model to set a water level control problem based on the water level change;

The gate water level control module 5 is used for inputting a gate opening sequence as a control instruction into the throttle gate controller, adjusting the gate opening to realize the water level control of the throttle gate, and acquiring water level data in a preset time period;

and the performance index evaluation module 6 is used for establishing an evaluation index according to the water level data to evaluate the performance index of the water level control of the throttle valve, judging the water level state of the throttle valve based on the performance index and optimizing and adjusting the control sequence to update the control state.

The following examples are provided to further illustrate embodiments of the invention:

example 1

In order to verify the one-dimensional hydrodynamic force numerical simulation model provided in the embodiment, an east Huang Bu throttle is selected for case study, automatic monitoring equipment is adopted for collecting data such as flow and water level, the data such as water level and flow are not obviously changed in adjacent collecting intervals under the influence of the performance of the equipment, and therefore a mode of continuously collecting the data to calculate the average value is adopted for the data, and the data is subjected to frequency reduction for 1 hour and 1 time. The width of the throttle gate bottom is 12.35 meters, the throttle gate is a three-hole gate, the width of a single hole is 3 meters, and the gate overflow coefficient is consistent with the roughness coefficient.

As shown in the simulation results shown in table 1, parameter calibration is adopted, gate flow coefficients are reversely calculated according to actual measurement data and a gate Kong Yanmei outflow formula, and a hydrodynamic model is used for inverting a throttle gate roughness rate, so that whether a water level dynamic simulation model meets daily scheduling requirements and the validity of the calibrated parameters is verified, the simulation duration is 120 hours, the calculation step length is 120s, the output result time step length is 120s, and the boundary condition step length is 7200s.

Table 1: simulation results

Compared with actual measurement data, the average error rate of water level simulation before the gate is 1.34%, the average error rate of water level simulation after the gate is 0.8%, the water level simulation error precision of control sections before and after the gate is controlled within 2%, simulation results of different control sections are as follows, and development trends of simulation values are basically consistent by comparing actual measurement water level and flow data.

Example two

In order to verify a water level control method of a throttle valve provided in the present embodiment, a project including a water level control system of the throttle valve was selected for a case study, the project having a channel length of 545 m, a bottom width of 12.31m, a slope coefficient of 2, a roughness of 0.03, an initial flow rate upstream of the channel of 12.31m ³/s, and a water diversion flow rate increased from 0 to 10m ³/s.

The predicted time domain Np and the control time domain Nc take the same value 25, the model is input into the deviation of the state variable water level and the target water level, the control variable U (k) is output, the control water level is 9.7, the step length is 30min, and the time is 24 hours.

By adopting the control of the water level prediction model, the water diversion change can control the change of the water level in advance, and when the water diversion change occurs, the water level amplitude caused by the water diversion influence can be rapidly controlled, so that the accuracy and stability of channel water delivery are greatly improved.

In summary, by means of the technical scheme, the water level can be accurately adjusted by acquiring the error of the water level and the preset target water level in real time and taking the error as the input of a control algorithm, the error is reduced, meanwhile, the target water discharge amount is dynamically judged by the current water level, the water discharge amount is updated based on actual conditions, the environment change can be flexibly adapted, the water discharge efficiency is optimized, the gate passing flow is determined by combining feedback increment and correction requirements, the mathematical relationship between the water level and the gate opening is established, the dynamic relationship between the water level and the gate opening is predicted, the future throttle water level change is predicted by combining the water level dynamic model and the real-time state, scientific basis is provided for water level control, the gate opening sequence is input into a throttle gate controller as a control instruction, the gate opening is automatically adjusted, the water level control of the throttle gate is realized, the control efficiency is improved, and the manual burden is lightened. The invention synthesizes the feedback flow and the correction flow to obtain more accurate flow variation, provides an accurate input value for adjusting the passing gate flow, enhances the basic accuracy of water level control, can more accurately reflect the overflow capacity of the throttle gate based on the combination of the flow variation and the number of sections and the contact relation, provides a more detailed basis for the calculation of the passing gate flow, establishes a control equation by utilizing the san-View process, and introduces an algebraic equation set formed by generalization of an outer boundary and an inner boundary to accurately describe the dynamic process of the water level variation, so that the accuracy and the response speed of the water level control of the throttle gate can be improved, a powerful tool is provided for deeply analyzing and understanding the water flow dynamics, and the efficiency and the safety in aspects of optimizing water resource management, flood control, water supply and the like are facilitated.

According to the invention, the water level dynamic model is utilized to predict the water level change of the future throttle valve in combination with the real-time state, a model prediction control algorithm is utilized to determine the gate opening sequence based on the prediction result, the water level prediction model is utilized to consider the future water level change prediction and the current control target, the optimal gate opening sequence in the process of realizing the water level control target is calculated, the precision and efficiency of water level control are improved, the quality and effect of control measures are ensured, the water level fluctuation and control cost are reduced, and a plurality of control variables and future prediction information are considered based on the water level prediction model, so that the control strategy is more flexible and has strong adaptability, and the high-efficiency and accurate control of the water level of the throttle valve is realized. The invention can primarily evaluate the stability and response speed of the water level control by judging the water level fluctuation range and the time required for reaching the target water level, define the water level stability and the control response time as evaluation criteria, is beneficial to the evaluation of standardized water level control performance, can evaluate the key performance of the water level control by focusing on the water level stability and the control response time, and ensures the effectiveness and the high efficiency of a control strategy.

The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.

Claims

1. The water level control method of the throttle valve is characterized by comprising the following steps of:

2. The method for controlling a water level of a throttle valve according to claim 1, wherein the step of combining the feedback increment with the correction requirement to obtain the flow rate of the throttle valve, calculating the opening of the throttle valve based on the flow rate of the throttle valve, and obtaining the mathematical relationship between the water level and the opening of the throttle valve to establish the dynamic model of the water level comprises the steps of:

3. The method for controlling a water level of a throttle valve according to claim 2, wherein the step of analyzing the form of the water flow during the water delivery according to the flow coefficient of the throttle valve to obtain the form of the opening of the throttle valve, and determining the mathematical relationship between the flow coefficient of the throttle valve and the form of the opening to establish the dynamic model of the water level comprises the steps of:

4. The method for controlling the water level of the throttle valve according to claim 3, wherein the step of constructing a water level change coordinate system according to definition content and fitting data to obtain the gate opening morphological parameter and fluctuation, and constructing a water level dynamic model based on digital twin by combining the throttle valve structure comprises the following steps:

5. The method of controlling a throttle valve according to claim 4, wherein predicting a future throttle valve water level change by using a water level dynamic model in combination with a real-time state, and determining a gate opening sequence by using a water level prediction model to set a water level control problem based on the water level change, comprises the steps of:

6. The water level control method of claim 5, wherein the calculating the gate opening sequence in the water level control target process by using the water level prediction model based on the multivariate control problem to establish the state space, and selecting the optimal opening sequence to be applied to the water level control comprises the following steps:

7. The method of controlling a water level of a throttle valve according to claim 6, wherein the estimating the performance index of the water level control of the throttle valve based on the water level data establishing evaluation index, judging the water level state of the throttle valve based on the performance index and optimizing the update control state of the adjustment control sequence comprises the steps of:

8. The method according to claim 7, wherein the step of establishing an evaluation index system based on the evaluation index, performing a sorting operation on the evaluation index by using a spectral clustering algorithm, and constructing an evaluation model based on the evaluation index system and the index sorting result comprises the steps of:

9. The method according to claim 8, wherein the expression of the evaluation model is:

；

Wherein L represents the evaluation result;

S _Z represents a spectral cluster sequencing result;

d _Z denotes a weight vector of the evaluation index;

Z _ij represents the weight of the i-th evaluation index;

NMI represents the amount of mutual information of the evaluation index criteria;

W _i denotes the similarity between the i-th index and the remaining indexes.

10. A throttle water level control device for realizing the throttle water level control method according to any one of claims 1-9, wherein the throttle water level control device comprises a feedback increment acquisition module, a correction demand acquisition module, a dynamic model establishment module, an opening sequence determination module, a gate water level control module and a performance index evaluation module;

The feedback increment acquisition module is used for comparing the water level of the throttle valve with a preset target water level by using a sensor to obtain a water level error, and obtaining the water level quantity by taking the water level error as the input of a feedback control algorithm to obtain a feedback increment;

The dynamic model building module is used for combining the feedback increment with the correction requirement to obtain the sluice flow, calculating the gate opening based on the sluice flow, obtaining the mathematical relationship between the water level and the gate opening, and building a water level dynamic model;

The gate water level control module is used for inputting a gate opening sequence as a control instruction into the throttle controller, adjusting the gate opening to realize water level control of the throttle, and acquiring water level data in a preset time period;