An open channel model predictive control method considering fixed boundary flow dynamics

Abstract

The application discloses a kind of open channel model predictive control methods considering fixed boundary flow dynamics, belong to canal water conveyance engineering automation control technical field.This method includes: identifying fixed opening gate, and establishing the prediction model of gate outflow dynamic change under disturbance;For the canal section comprising at least one controllable gate and fixed opening gate, a local one-dimensional hydrodynamic model is constructed to predict the gate flow in the future prediction time domain;The predicted flow is introduced into the state space equation of the model predictive controller as a known disturbance;In each control cycle, the optimal control sequence is iteratively solved according to the flow prediction result;The first control quantity of the optimal control sequence is converted into gate opening instruction and executed.By converting the fixed boundary flow from unknown disturbance to predictable disturbance, the application can improve the accuracy and robustness of open channel water level control under abnormal conditions, reduce the steady-state error and integral absolute error.

Description

Technical Field

[0001] This invention relates to the field of automated control technology for open channel water conveyance projects, and specifically to a model predictive control method for open channels that considers the dynamics of fixed boundary flow. Background Technology

[0002] Long-distance water transfer projects are crucial engineering measures for addressing the uneven spatial and temporal distribution of water resources and ensuring regional water supply security. Open channel water conveyance systems, as an important form of water transfer in long-distance projects, directly impact the safety, stability, and economy of the project's operation due to the accuracy of their water level and flow control. Model predictive control, capable of handling multivariate coupling, explicitly considering constraints, and predicting future system behavior, has been widely researched and applied in the field of open channel control in recent years.

[0003] When an open channel water conveyance system has uncontrollable gates, existing model predictive control methods have the following shortcomings: First, fixed-opening gates alter the control boundary conditions of the original water conveyance system, degenerating it from a series controllable system to a partially uncontrollable system, with the downstream channel water supply entirely dependent on the outflow from the upstream fixed-opening gate. Second, state-space models based on conventional controllable operating conditions struggle to accurately characterize the input-output relationship under fixed-opening gate conditions. Third, existing methods typically treat the flow rate through the fixed-opening gate as an unknown disturbance or simply assume it to be constant, failing to reveal and utilize its unique nonlinear dynamic characteristics. Through extensive hydrodynamic simulations, the inventors have discovered that when a step change in flow occurs upstream of the fixed-opening gate, the flow rate through the gate is not monotonically changing, but exhibits a strong nonlinear dynamic characteristic: an initial oscillating surge followed by a stable, slow increase. Existing methods ignore this dynamic characteristic, relying solely on feedback correction, leading to control lag and steady-state deviation, making it difficult to meet water level control requirements under abnormal operating conditions.

[0004] Therefore, for open channel water conveyance systems where some gates are at a fixed opening, it is still necessary to propose an open channel model predictive control method that can predict the dynamic changes of fixed boundary flow and incorporate them into the model predictive control process. Summary of the Invention

[0005] To address the aforementioned shortcomings in the prior art, this invention provides a predictive control method for open channel models that considers the dynamics of fixed boundary flow.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: A predictive control method for open channel models considering the dynamics of fixed boundary flow includes the following steps: S1. Identify fixed-opening gates in open channel systems; S2. Construct a local one-dimensional hydrodynamic model for a channel section containing controllable gates and fixed-opening gates, and predict the change process of the flow through the fixed-opening gates in the future prediction time domain. S3. The overpass flow rate predicted in step S2 is introduced as a predictable disturbance into the state-space equation of the model predictive controller. S4. Perform hydrodynamic simulation and optimization solution iteratively in each control cycle to obtain the optimal control sequence; S5. Extract the first control quantity of the optimal control sequence, convert it into a gate opening command, and then execute it.

[0007] Furthermore, S1 specifically includes: defining a fixed-opening gate as a gate whose opening remains constant due to equipment failure, power outage, maintenance, or manual intervention.

[0008] Furthermore, step S2 includes the following steps: S21. A local one-dimensional hydrodynamic model is established based on the one-dimensional Saint-Venant equations. The model is discretized using the Presman implicit difference scheme with a spatial discretization interval of 50~200m. The time step is smaller than the control step of the model predictive controller, which is 60~1800s. The model covers a local channel section of 1~3km upstream and downstream of a fixed-opening gate. The time taken for a single prediction calculation is less than 0.5s. S22. Set the upper boundary of the model to the future multi-step flow sequence given by the model predictive controller, and the lower boundary to the target water level boundary or free outflow boundary of the downstream control section. The initial conditions are the measured water level and flow data at the time of the current call. When there is water diversion in a local channel section, the water diversion boundary is given according to the water diversion plan in the prediction time domain.

[0009] Furthermore, step S3 specifically includes the following steps: S31. Construct the discrete state-space equations based on the integral time delay model and set the objective function of the model predictive controller; S32. For a series multi-channel pool system, the state-space equations of each channel pool are combined to construct a global prediction model, and the outflow of the upstream channel pool is used as the inflow of the downstream channel pool. S33. Establish an online correction mechanism for the parameters of the integral time delay model, and dynamically update the area of ​​the backwater zone and the time delay based on real-time changes in upstream flow. S34. Establish an adaptive tuning mechanism for weight coefficients, dynamically adjust the corresponding water level deviation weight coefficients and control action weight coefficients according to the sensitivity indicators of each canal and pool, and set the weights to decay in the prediction time domain.

[0010] Furthermore, the discrete state-space equation based on the integral time-delay model in S31 is expressed as follows:

[0011] In the formula, for The state vector at any given time; for The control input increment vector at each moment; for The perturbation input vector at time step; The state matrix; For control matrix; The disturbance matrix; the disturbance input vector It includes at least the change in the diversion flow rate and the change in the flow rate through a fixed-opening gate, which is predicted by a local one-dimensional hydrodynamic model.

[0012] Furthermore, the objective function of the model prediction controller in S31 is expressed as:

[0013] In the formula, The objective function is... For prediction in the time domain; To control the time domain; Number of controlled canals and pools; The number of controllable gates; In order to be in The time of prediction based on the current state A canal pool in The water level at that moment; For the first A canal pool in The target water level value at that moment; In order to be in The first time optimization obtained A controllable gate in Incremental flow adjustment at any given moment; For the first Weighting coefficients for tracking errors in individual canals and ponds; For the first Weighting coefficients for each control input increment; For discrete time intervals.

[0014] Furthermore, step S4 includes the following steps: S41. Initialize the control sequence. For the first control cycle, take the steady-state flow rate value as the initial control sequence. For subsequent control cycles, take the optimal control sequence of the previous control cycle as the initial control sequence. S42. Using the current control sequence as the upper boundary inflow sequence of the local one-dimensional hydrodynamic model, and combining the downstream water level boundary and the water distribution plan, solve the Saint-Venant equations to obtain the sequence of predicted flow rates of the gate with a fixed opening in the future prediction time domain. S43. Combine the change in flow rate through the gate relative to the steady-state value with the measured or planned change in the diversion flow rate to form a disturbance input vector. ; S44, Based on the extended perturbation term The state-space equations and objective function are used to solve the quadratic programming problem, resulting in the optimized control sequence for this iteration. S45. Calculate the Euclidean distance between the current iterative control sequence and the previous iterative control sequence. If the distance is less than a preset threshold or the number of iterations reaches 5, terminate the iteration and output the optimal control sequence. Otherwise, use the current iterative control sequence as the new upper boundary condition and return to step S42 to continue the iteration. The present invention has the following beneficial effects: (1) It reveals the nonlinear overcurrent law of a fixed opening gate under disturbance (initial surge - oscillation - uniform slow increase), and based on this, the fixed boundary flow is transformed from an unknown disturbance into a predictable disturbance, overcoming the defect of the traditional MPC that treats it as a constant value or a black box disturbance, and significantly improving the pertinence of feedforward compensation.

[0015] (2) A fixed boundary flow prediction method based on a local one-dimensional hydrodynamic model is proposed. Compared with the construction of a global hydrodynamic model, the number of calculation nodes is reduced by more than 80%, and the time for a single prediction is less than 0.5s. While ensuring the prediction accuracy, it meets the real-time control requirements of engineering.

[0016] (3) A tightly coupled iterative strategy for hydrodynamic simulation and MPC rolling optimization was designed, which enables the flow prediction and controller optimization to converge iteratively and achieve active compensation for flow disturbances at fixed boundaries. The results of the embodiments show that, compared with the existing methods, the steady-state error (StE) of the method of the present invention is reduced by about 51.93%, the integral absolute error (IAE) is reduced by about 79.59%, and the settling time is shortened by more than 50%.

[0017] (4) It has good engineering adaptability: only a hydrodynamic model needs to be established for the local channel section where the fixed opening gate is located, without changing the overall framework of the original MPC controller, and it is easy to deploy and upgrade in the existing open channel automation system. Attached Figure Description

[0018] Figure 1 This is an overall flowchart of the open channel model predictive control method considering the dynamic characteristics of fixed boundary flow according to the present invention. Figure 2 A schematic diagram of the extended state-space model structure considering fixed boundary flows; Figure 3 This is a comparison chart of the water level control effects of the method of this invention and the traditional model predictive control method. Detailed Implementation

[0019] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0020] To achieve the above objectives, the present invention adopts the following technical solution: a predictive control method for open channel models considering the dynamic characteristics of fixed boundary flow, such as... Figure 1 As shown, it includes the following steps: Step S1: Identify fixed-opening gates in the open channel system; the fixed-opening gates are those that cannot participate in real-time adjustment due to equipment failure, power outage, maintenance, or manual intervention and whose opening remains unchanged.

[0021] Step S2 involves constructing a local one-dimensional hydrodynamic model for a channel segment containing at least one controllable gate and a fixed-opening gate. This model is used to predict the flow rate change process of the fixed-opening gate in the future prediction time domain. The local one-dimensional hydrodynamic model is based on the one-dimensional Saint-Venant equations and discretized using the Pressmann implicit difference scheme. The spatial discretization interval is 50-200m, and the time step is smaller than the control step of the model prediction controller (typically 60-1800s). Compared with constructing a global hydrodynamic model, the local modeling strategy of this invention only covers the local channel segment where the fixed-opening gate is located (typically 1-3km upstream and downstream of the gate), reducing the number of computation nodes by more than 80%, and controlling the calculation time of a single prediction to within 0.5s, thus meeting the requirements of real-time control. The upper boundary of the model is the future multi-entry flow sequence given by the model predictive controller, and the lower boundary is the target water level boundary (or free outflow boundary) of the downstream control section. The initial conditions are the measured water level and flow data at each time the predictive control algorithm is called. When there is water diversion in a local canal section, the water diversion boundary is given according to the water diversion plan in the future prediction time domain.

[0022] Step S3: The flow rate through the fixed-opening gate predicted in step S2 is introduced as a predictable disturbance variable into the state-space equation of the model predictive controller, so as to transform the unknown disturbance caused by the fixed boundary conditions into predictable feedforward disturbance information. Figure 2 A schematic diagram of the extended state-space model considering fixed boundary flow is given.

[0023] Step S4 involves performing a tightly coupled iteration of hydrodynamic simulation and MPC rolling optimization within each control cycle. The specific sub-steps are as follows: Step S4-1, Initialization: Take the optimal control sequence of the previous control cycle as the initial control sequence of the current cycle. If it is the first control cycle, the initial control sequence is the steady-state flow rate value.

[0024] Step S4-2, Flow Prediction: Using the current control sequence as the upper boundary inflow sequence of the local one-dimensional hydrodynamic model, and combining the downstream water level boundary and water allocation plan, solve the Saint-Venant equations to obtain the predicted flow rate sequence of the gate with fixed opening in the future prediction time domain (k+1 to k+p).

[0025] Step S4-3, Combined disturbance construction: Combine the change in flow rate through the gate (relative to the steady state value) predicted in step S4-2 with the measured or planned change in flow rate to form the disturbance input vector d(k) of the model predictive controller.

[0026] Step S4-4, MPC optimization solution: Based on the extended state-space equation (including the disturbance term d(k)) and the objective function, solve the quadratic programming problem to obtain the optimized control sequence for this iteration.

[0027] Step S4-5, Convergence Judgment: Calculate the Euclidean distance between the current iteration control sequence and the previous iteration control sequence. If it is less than the preset threshold (e.g., 1e-3) or reaches the maximum number of iterations (e.g., 5), terminate the iteration and output the optimal control sequence; otherwise, use the control sequence obtained in the current iteration as the new upper boundary condition and return to step S4-2 to continue the iteration.

[0028] Through the bidirectional coupling iteration between the above hydrodynamic model and the MPC controller, this invention enables the controller to sense the dynamic change trend of the fixed boundary flow in advance and to perform proactive feedforward adjustment, overcoming the lag of traditional methods that rely solely on feedback correction.

[0029] Step S5: Extract the first control quantity of the optimal control sequence and convert it into the corresponding gate opening command for execution.

[0030] In some implementations, the model predictive controller employs a discrete state-space model based on an integral time-delay model, the expression of which is:

[0031] In the formula, x(k) is the state vector at time k; Δu(k) is the control input increment vector at time k; d(k) is the disturbance input vector at time k; A is the state matrix; B is the control matrix; D is the disturbance matrix. The disturbance input vector d(k) includes at least the change in water flow and the change in flow through a fixed-opening gate predicted by a local one-dimensional hydrodynamic model.

[0032] In some implementations, the state vector x(k) includes the downstream water level deviation, the change in water level deviation, and the change in flow rate within the time lag interval; the control input increment vector Δu(k) is the flow rate adjustment increment of each controllable gate.

[0033] In some implementations, the objective function of the model prediction controller is:

[0034] In the formula, The objective function is... For prediction in the time domain; To control the time domain; Number of controlled canals and pools; The number of controllable gates; In order to be in The time of prediction based on the current state A canal pool in The water level at that moment; For the first A canal pool in The target water level value at that moment; In order to be in The first time optimization obtained A controllable gate in Incremental flow adjustment at any given moment; For the first Weighting coefficients for tracking errors in individual canals and ponds; For the first Weighting coefficients for each control input increment; For discrete time intervals.

[0035] In some implementations, for a series multi-channel pool system, the state-space equations of each channel pool can be combined to construct a global prediction model, where the outflow of the upstream channel pool is used as the inflow of the downstream channel pool.

[0036] In some implementations, the method further includes an online correction mechanism for the parameters of the integral time delay model, which updates the area of ​​the backwater zone and the time delay based on real-time changes in upstream flow.

[0037] In some implementations, the method further includes a weight coefficient adaptive tuning mechanism, which dynamically adjusts the corresponding weights according to the sensitivity indicators of each channel pool, and sets the weights to decay in the prediction time domain.

[0038] This embodiment uses the first two series-connected canals and pools between the Gangtou Tunnel and the Beijuma River Control Gate in the Beijing-Shijiazhuang section of the South-to-North Water Diversion Project as application objects to illustrate the open channel model predictive control method considering the dynamic characteristics of fixed boundary flow described in this invention. The main physical parameters of the canals are shown in Table 1, and the integral time delay model parameters of each canal and pool under the initial steady-state condition obtained by hydrodynamic simulation are shown in Table 2.

[0039] Table 1 Main Physical Parameters of the Channel

[0040] Table 2 Parameters of the Integral Delay Model

[0041] To verify the effectiveness of the method of the present invention, the following comparison scheme was set up: MPC1: Traditional model predictive control method, which only takes the water diversion disturbance as a known input, without considering the dynamic prediction of the flow rate through a gate with a fixed opening; MPC2: Based on the method of this invention, MPC1 adds local one-dimensional hydrodynamic model flow prediction for gates with fixed opening, and introduces the prediction results as disturbance terms into the model prediction controller.

[0042] Set the following simulation conditions: Condition A: The canal pool 1 increases by 5m in 12 hours. 3 / s water separation is used to verify the control performance under water separation disturbance; Condition B: Gates 2 and 3 are in a fixed opening state, and only gate 1 is adjusted to verify the control performance under fixed boundary conditions.

[0043] The simulation parameters were set as follows: the calculation step size of the local one-dimensional hydrodynamic model was 60 s, the control step size of the model predictive controller was 3600 s, the total simulation time was 72 h, and the weighting coefficient ratio was set to 100. The local one-dimensional hydrodynamic model adopted the one-dimensional Saint-Venant equations, with a spatial discretization interval of 100 m and an observation section spacing of 1 km.

[0044] The average absolute error (MAE), steady-state error (StE), integral absolute error (IAE), and integral of absolute change in control quantity (IAQ) were used as evaluation indicators. The simulation results are shown in Table 3.

[0045] Table 3. Evaluation Results of Control Effectiveness

[0046] Depend on Figure 3As shown in Table 3, under fixed boundary conditions, compared with the traditional model predictive control method MPC1, the steady-state error StE of the method MPC2 of this invention is reduced by approximately 51.93%, and the integral absolute error IAE is reduced by approximately 79.59%, indicating that the method of this invention has significant advantages in terms of steady-state control accuracy and cumulative error suppression. At the same time, the integral of the absolute change in control quantity IAQ of MPC2 increases, indicating that after introducing fixed boundary flow prediction, the controller takes more proactive adjustment actions to compensate for disturbances in advance.

[0047] Figure 3 Figures (a), (b), (c), and (d) show a comparison of the water level control effects of the method of this invention (MPC2) and the traditional model predictive control method (MPC1) under operating condition B (gates 2 and 3 are fixed at their openings, and only gate 1 is adjusted). As can be seen from these figures: After the water diversion rate of 5 m³ / s was increased in channel pool 1 at time 12h, MPC1 experienced a maximum overshoot of about 0.03m in its water level response because it did not dynamically predict the flow rate of gates 2 and 3 with fixed opening. The adjustment time was as long as 18h. In contrast, MPC2 predicted the "initial surge-oscillation decay" law of the outflow of gates 2 and 3 in advance through a local one-dimensional hydrodynamic model and actively compensated for it in the MPC optimization. The maximum water level deviation did not exceed 0.01m and the adjustment time was shortened to less than 8h.

[0048] also, Figure 3 Subgraph (b) shows a comparison between the actual and predicted flow rates of the gate 2 with the fixed opening after the water diversion disturbance. The results show that the local one-dimensional hydrodynamic model proposed in this invention can accurately capture the nonlinear dynamics of the flow rate (the initial surge peak error is less than 5%, and the oscillation phase error is less than 1h), while the integral time delay model used in traditional MPC cannot characterize this dynamic process and can only output a constant flow rate estimate.

[0049] Furthermore, although the mean absolute error (MAE) of MPC2 increased slightly, its steady-state error and integral absolute error decreased significantly, indicating that the method of the present invention is more conducive to reducing the long-term cumulative deviation of the system and improving the overall control effect under fixed boundary conditions. This is because traditional control methods do not predict the dynamic changes in the flow rate through a fixed-opening gate, relying mainly on feedback to correct errors, which has a certain lag. In contrast, the method of the present invention, by introducing the prediction results of a local one-dimensional hydrodynamic model, enables the controller to sense the trend of fixed boundary flow rate changes in advance and perform feedforward adjustment, thereby improving the water level control performance under abnormal conditions.

[0050] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0051] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0052] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0053] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

[0054] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A predictive control method for open channel models considering the dynamics of fixed boundary flow, characterized in that, Includes the following steps: S1. Identify fixed-opening gates in open channel systems; S2. Construct a local one-dimensional hydrodynamic model for a channel section containing controllable gates and fixed-opening gates, and predict the change process of the flow through the fixed-opening gates in the future prediction time domain. S3. The overpass flow rate predicted in step S2 is introduced as a predictable disturbance into the state-space equation of the model predictive controller. S4. Perform hydrodynamic simulation and optimization solution iteratively in each control cycle to obtain the optimal control sequence; S5. Extract the first control quantity of the optimal control sequence, convert it into a gate opening command, and then execute it.

2. The open channel model predictive control method considering fixed boundary flow dynamics according to claim 1, characterized in that, S1 specifically includes: defining a fixed-opening gate as a gate whose opening remains constant because it cannot be adjusted in real time due to equipment failure, power outage, maintenance, or manual intervention.

3. The open channel model predictive control method considering fixed boundary flow dynamics according to claim 1, characterized in that, S2 includes the following steps: S21. A local one-dimensional hydrodynamic model is established based on the one-dimensional Saint-Venant equations. The model is discretized using the Presman implicit difference scheme with a spatial discretization interval of 50~200m. The time step is smaller than the control step of the model predictive controller, which is 60~1800s. The model covers a local channel section of 1~3km upstream and downstream of a fixed-opening gate. The time taken for a single prediction calculation is less than 0.5s. S22. Set the upper boundary of the model to the future multi-step flow sequence given by the model predictive controller, and the lower boundary to the target water level boundary or free outflow boundary of the downstream control section. The initial conditions are the measured water level and flow data at the time of the current call. When there is water diversion in a local channel section, the water diversion boundary is given according to the water diversion plan in the prediction time domain.

4. The open channel model predictive control method considering fixed boundary flow dynamics according to claim 1, characterized in that, S3 specifically includes the following steps: S31. Construct the discrete state-space equations based on the integral time delay model and set the objective function of the model predictive controller; S32. For a series multi-channel pool system, the state-space equations of each channel pool are combined to construct a global prediction model, and the outflow of the upstream channel pool is used as the inflow of the downstream channel pool. S33. Establish an online correction mechanism for the parameters of the integral time delay model, and dynamically update the area of ​​the backwater zone and the time delay based on real-time changes in upstream flow. S34. Establish an adaptive tuning mechanism for weight coefficients, dynamically adjust the corresponding water level deviation weight coefficients and control action weight coefficients according to the sensitivity indicators of each canal and pool, and set the weights to decay in the prediction time domain.

5. The open channel model predictive control method considering fixed boundary flow dynamics according to claim 4, characterized in that, The discrete state-space equation based on the integral time delay model in S31 is expressed as follows: In the formula, for The state vector at any given time; for The control input increment vector at each moment; for The perturbation input vector at time step; The state matrix; For control matrix; The disturbance matrix; the disturbance input vector It includes at least the change in the diversion flow rate and the change in the flow rate through a fixed-opening gate, which is predicted by a local one-dimensional hydrodynamic model.

6. The open channel model predictive control method considering fixed boundary flow dynamics according to claim 4, characterized in that, The objective function of the model prediction controller in S31 is expressed as follows: In the formula, The objective function is... For prediction in the time domain; To control the time domain; Number of controlled canals and pools; The number of controllable gates; In order to be in The time of prediction based on the current state A canal pool in The water level at that moment; For the first A canal pool in The target water level value at that moment; In order to be in The first time optimization obtained A controllable gate in Incremental flow adjustment at any given moment; For the first Weighting coefficients for tracking errors in individual canals and ponds; For the first Weighting coefficients for each control input increment; For discrete time intervals.

7. The open channel model predictive control method considering fixed boundary flow dynamics according to claim 1, characterized in that, S4 includes the following steps: S41. Initialize the control sequence. For the first control cycle, take the steady-state flow rate value as the initial control sequence. For subsequent control cycles, take the optimal control sequence of the previous control cycle as the initial control sequence. S42. Using the current control sequence as the upper boundary inflow sequence of the local one-dimensional hydrodynamic model, and combining the downstream water level boundary and the water distribution plan, solve the Saint-Venant equations to obtain the sequence of predicted flow rates of the gate with a fixed opening in the future prediction time domain. S43. Combine the change in flow rate through the gate relative to the steady-state value with the measured or planned change in the diversion flow rate to form a disturbance input vector. ; S44, Based on the extended perturbation term The state-space equations and objective function are used to solve the quadratic programming problem, resulting in the optimized control sequence for this iteration. S45. Calculate the Euclidean distance between the current iterative control sequence and the previous iterative control sequence. If the distance is less than the preset threshold or the number of iterations reaches 5, terminate the iteration and output the optimal control sequence. Otherwise, use the current iterative control sequence as the new upper boundary condition and return to step S42 to continue iterating.

Images