A cooperative adaptive scheduling method for pumping stations based on multi-source data

Abstract

This invention discloses a collaborative adaptive scheduling method for pumping stations and sluice gates based on multi-source data, belonging to the field of smart water conservancy automation control technology. The method includes the following steps: pre-extraction of situational feature parameters, including real-time collection of meteorological and rainfall forecast data, upstream water level, downstream water level, inflow into the reservoir and river, equipment operating status data, and downstream water demand commands; and construction of a comprehensive evaluation function S. This invention significantly improves the predictability and timeliness of flood control scheduling, effectively avoiding flood risks caused by sudden hydrological events. Furthermore, by setting a smoothness penalty term and a stability constraint on the actuator in the objective function, it achieves gradual adjustment of gates and pumping stations, completely solving problems such as water hammer effect and river scouring damage caused by excessively rapid actions in traditional control, ensuring the stability of water flow and the service life of water conservancy facilities.

Description

Technical Field

[0001] This invention relates to the field of intelligent water conservancy automation control technology, and more specifically, to a cooperative adaptive scheduling method for pumping stations and gate stations based on multi-source data. Background Technology

[0002] Pumping stations, as core facilities of water conservancy projects, undertake key functions such as flood control and drainage, urban and rural water supply, and ecological water replenishment. The scientific nature and real-time performance of their scheduling and control directly affect the operational safety and effectiveness of water conservancy projects. Existing pumping station scheduling relies heavily on manual experience or single threshold triggering logic, which is no longer suitable for the operational needs of modern smart water conservancy. A core issue is the problem of delayed scheduling response. Traditional control relies solely on real-time monitored field data such as water level and flow rate for passive triggering operations, without incorporating meteorological warnings or upstream water inflow forecasts for proactive scheduling. In the face of sudden rainfall or a surge in upstream water inflow, untimely scheduling can easily lead to flood risks, demonstrating a lack of overall predictive ability regarding hydrological changes.

[0003] Meanwhile, traditional scheduling methods also suffer from the dual defects of isolated scenarios and poor coordination between pumps and gates. On the one hand, there is no flexible switching mechanism between different scheduling objectives such as flood control, water supply, and ecological water replenishment. The control system often sets parameters for a single objective, which can easily lead to problems such as excessive drainage for flood control, neglecting ecological flow requirements for water supply, and failing to consider energy consumption costs when replenishing ecological water, resulting in energy waste or scheduling risks. On the other hand, the start-up and shutdown of pump stations, frequency conversion regulation, and gate opening regulation lack deeply coupled control logic. The two operations are independent of each other, which can easily lead to violent fluctuations in the water flow state in the area, causing hydraulic problems such as water hammer effect and river scouring damage. This not only affects the service life of water conservancy facilities but also creates hidden dangers for the operation of local projects.

[0004] There are currently no effective solutions to the problems in the relevant technologies. Summary of the Invention

[0005] To address the problems in related technologies, this invention proposes a cooperative adaptive scheduling method for pumping stations based on multi-source data, in order to overcome the aforementioned technical problems existing in the existing related technologies.

[0006] The technical solution of this invention is implemented as follows:

[0007] A cooperative adaptive scheduling method for pumping stations based on multi-source data includes the following steps:

[0008] Pre-extraction of contextual feature parameters, including real-time acquisition of meteorological and rainfall forecast data. Upstream water level Downstream water level Inflow into reservoirs and rivers Equipment operating status data and downstream water demand instructions ;

[0009] A comprehensive evaluation function S is constructed. Based on the extracted context feature parameters, a quantitative value is calculated through the comprehensive evaluation function to realize the automatic identification and triggering of scheduling modes.

[0010] Based on the identified scheduling mode, a Model Predictive Control (MPC) pump gate control model with a time span of T is constructed. In each control sampling period, the gate opening degree a and the pump group operating frequency ω are used as control variables to solve a constrained optimization problem to achieve coordinated regulation of the pump gate.

[0011] For the constructed constrained nonlinear optimization problem, a rolling time-domain solution is performed to obtain the optimal control vector sequence in the prediction time domain within each control sampling period. ,in, This is the optimal gate opening sequence. This is the optimal operating frequency sequence for the pump set.

[0012] Wherein, the optimal control vector sequence in the prediction time domain is obtained. This includes: constructing a digital twin model of the pump station to simulate the changes in hydraulic state and equipment operating status after the execution of the control command sequence, and to verify whether there are any abnormal situations.

[0013] The comprehensive evaluation function S is expressed as:

[0014] ;

[0015] In the formula, These are the weighting coefficients; This is the upper limit of the safe water level upstream; The safe threshold for inflow into the reservoir and river; This is the quantification value for the critical level of rainfall forecasting;

[0016] The automatic identification and triggering of the scheduling modes include: flood control scheduling mode, water supply economic scheduling mode, and ecological water replenishment mode;

[0017] Among them, the flood control scheduling mode, that is, when the comprehensive evaluation function calculates the value... At that time, the flood control dispatch mode is triggered, in which, The threshold for triggering the flood control mode; the objective function under this mode is... This means maximizing the sum of the gate's flow rate and the pumping station's discharge rate to ensure the upstream water level in the pumping station's control area. , This is the upstream flood control warning level;

[0018] Among them, the water supply economic dispatch mode, that is, when the comprehensive evaluation function calculates the value... At that time, the water supply economic dispatch mode is triggered, in which, The lower limit threshold for pattern recognition; the objective function for this pattern is... That is, to minimize the time integral of the system's overall energy consumption, and to achieve energy-saving and economical operation of the pump station while ensuring the downstream water demand command D;

[0019] Among them, the ecological water replenishment mode is triggered directly when the monitored dissolved oxygen (DO) content is lower than the preset ecological threshold, without being limited by the comprehensive evaluation function S value; the objective function of this mode is to maintain the minimum ecological flow downstream of the control area. This meets the flow requirements of the river ecological water network.

[0020] The objective function J of the optimization problem is expressed as:

[0021] ;

[0022] In the formula, T represents the time domain of model prediction; These are the weight coefficients for each optimization term; To predict the upstream water level at time t within the time domain; This represents the target upstream water level under the current dispatching mode. For the pump station unit at the operating frequency Yangcheng Real-time power; This represents the change in gate opening between adjacent sampling periods; This represents the change in pump unit operating frequency between adjacent sampling periods; For the pump station head, This is the real-time water level downstream.

[0023] The rolling time-domain solution includes: the dynamic changes in water level within the calibration control area follow the water balance equation, expressed as:

[0024] ;

[0025] In the formula, To control the area of ​​hydraulic storage, For the gate flow rate, This refers to the pumping flow rate of the pumping station.

[0026] Among them, the gate flow rate , is represented as:

[0027] ;

[0028] In the formula, For flow coefficient, The width of the gate opening. This is the acceleration due to gravity.

[0029] The beneficial effects of this invention are:

[0030] This invention extracts contextual features and utilizes a multi-scenario decision engine built upon a comprehensive evaluation function to automatically identify and trigger flood control, economic water supply, and ecological water replenishment scheduling modes. It then combines this with a model predictive control algorithm to construct a pump and gate control model, solving a constrained optimization problem using gate opening and pump operating frequency as joint control variables. Nonlinear solutions and rolling execution are achieved through particle swarm optimization or interior-point methods, realizing full-process automation and intelligence in pump and gate scheduling. This not only introduces meteorological rainfall forecasting and inflow prediction, transforming the response from passive water level to proactive inflow prediction, significantly improving the predictability and timeliness of flood control scheduling and effectively mitigating flood risks from sudden hydrological events, but also achieves gradual adjustment of gates and pump stations by setting a smoothness penalty term and actuator stability constraints in the objective function, completely resolving the water hammer effect caused by excessively rapid actions in traditional control methods. This method addresses issues such as river scouring and damage, ensuring the stability of water flow and the lifespan of hydraulic facilities. Simultaneously, by utilizing dynamically adjusted weighting coefficients, the system focuses on core objectives under different scheduling modes, achieving a dynamic balance between multiple objectives: flood control, water supply security, ecological water replenishment, and energy conservation. This solves the problems of isolation and difficulty in coordinating multiple objectives inherent in traditional scheduling modes, improving the overall operational efficiency of pumping stations. Furthermore, this method requires no manual intervention throughout the process, significantly reducing the labor intensity and experience dependence of manual operations. Each weighting coefficient, trigger threshold, and constraint condition can be flexibly calibrated and adjusted according to the hydraulic characteristics of different regions and types of pumping stations. It is also compatible with newly added monitoring parameters and scheduling objectives, possessing good versatility, scalability, and engineering feasibility. This effectively improves the precision of pumping station scheduling and control, achieving deep coupling and coordination of pumping stations and optimizing overall operational efficiency. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a flowchart illustrating a multi-source data-driven collaborative adaptive scheduling method for pumping stations according to an embodiment of the present invention. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0034] Example 1: According to an embodiment of the present invention, a cooperative adaptive scheduling method for pumping stations based on multi-source data is provided.

[0035] like Figure 1 As shown, the pump station cooperative adaptive scheduling method based on multi-source data driven according to an embodiment of the present invention includes the following steps:

[0036] Step S1: Real-time acquisition of meteorological and rainfall forecast data Hydrological monitoring data, i.e., upstream water level Downstream water level Inflow into reservoirs and rivers Equipment operating status data and downstream water demand instructions And perform data preprocessing and context feature parameter extraction;

[0037] This technical solution involves a system that acquires data in real time through a meteorological monitoring module, a hydrological sensor module, and a water conservancy dispatching instruction module, and performs data preprocessing and contextual feature extraction. Contextual feature extraction includes data acquisition based on meteorological rainfall forecasts. Forecast inflow to reservoirs and rivers calculated using a rainfall-runoff inference model Real-time water level upstream of the pump station control area The system collects downstream real-time water demand commands (D); simultaneously, it collects dissolved oxygen (DO), pump station unit operating status, and gate opening status as auxiliary parameters.

[0038] Step S2: Construct a comprehensive evaluation function S. Based on the situational feature parameters extracted in step S1, calculate the quantitative value through the comprehensive evaluation function to realize the automatic identification and triggering of three scheduling modes: flood control, water supply economy, and ecological water replenishment.

[0039] The comprehensive evaluation function S is expressed as:

[0040] ;

[0041] In the formula, The weighting coefficients are pre-calibrated based on the characteristics of regional water conservancy projects and scheduling priorities, and can be dynamically adjusted. This is the upper limit of the safe water level upstream; The safe threshold for inflow into the reservoir and river; The critical level of rainfall forecast is quantified by classifying it into different levels and assigning corresponding quantified values ​​based on rainfall intensity, duration, and coverage.

[0042] Specifically, the triggering conditions and objective functions for the three scheduling modes are as follows:

[0043] Among them, the flood control scheduling mode, that is, when the comprehensive evaluation function calculates the value... At that time, the flood control dispatch mode is triggered, in which, The threshold for triggering the flood control mode; the objective function under this mode is... This means maximizing the sum of the gate's flow rate and the pumping station's discharge rate to ensure the upstream water level in the pumping station's control area. , This is the upstream flood control warning level;

[0044] Water supply economic dispatch mode, that is, when the comprehensive evaluation function calculates the value At that time, the water supply economic dispatch mode is triggered, in which, The lower limit threshold for pattern recognition; the objective function for this pattern is... That is, to minimize the time integral of the system's overall energy consumption, and to achieve energy-saving and economical operation of the pump station while ensuring the downstream water demand command D;

[0045] The ecological water replenishment mode is triggered directly when the monitored dissolved oxygen (DO) content falls below a preset ecological threshold, without being limited by the comprehensive evaluation function S-value. The objective function under this mode is to maintain the minimum ecological flow downstream of the control area. This meets the flow requirements of the river ecological water network.

[0046] Step S3: Under the scheduling mode identified in step S2, construct a model predictive control (MPC) pump gate control model with a time span of T. In each control sampling period, use the gate opening degree a and the pump group operating frequency ω as control variables to solve the constrained optimization problem and realize the coordinated regulation of the pump gate.

[0047] The objective function J of the optimization problem is expressed as:

[0048] ;

[0049] In the formula, T is the time domain for model prediction, which is set according to the hydraulic response characteristics of the pumping station; The weight coefficients for each optimization item are dynamically adjusted based on the current scheduling mode; To predict the upstream water level at time t within the time domain; This represents the target upstream water level under the current dispatching mode. For the pump station unit at the operating frequency Yangcheng Real-time power; This represents the change in gate opening between adjacent sampling periods; This represents the change in pump unit operating frequency between adjacent sampling periods; For the pump station head, The downstream real-time water level is used as the reference. Meanwhile, to ensure the safe operation of the pumping station and the physical constraints of the equipment, the following constraints are set:

[0050] Water level safety constraints, i.e. ,in, This is the lower limit of the safe water level upstream. This is the upper limit of the safe water level upstream;

[0051] Physical constraints of the gate, i.e. ,in, This is the minimum opening of the gate. This represents the maximum opening of the gate;

[0052] Pump frequency conversion constraint, i.e. ,in, This is the minimum operating frequency of the pump set. This is the maximum operating frequency of the pump set;

[0053] Stability constraints of the implementing agency, namely ,in, This represents the maximum change in gate opening in a single operation. This is the maximum single change in the pump unit's operating frequency. This constraint is used to prevent problems such as water hammer and river scouring caused by excessively rapid gate and pump station movements.

[0054] Step S4: For the constrained nonlinear optimization problem constructed in step S3, a rolling time-domain solution is performed using the particle swarm optimization algorithm or the interior point method. Within each control sampling period, the optimal control vector sequence in the prediction time domain is obtained. ,in, This is the optimal gate opening sequence. This is the optimal operating frequency sequence for the pump set;

[0055] Specifically, to ensure the real-time performance and stability of the control, a rolling execution strategy is adopted. Only the first control value in the optimal control vector sequence is sent to the pump station PLC to adjust the gate opening and pump operating frequency once. In the next control sampling cycle, multi-source data is collected again, model parameters are updated, and the optimization problem is solved again to realize the dynamic rolling update of control commands.

[0056] The dynamic changes in water level within the control area follow the water balance equation, expressed as:

[0057] ;

[0058] In the formula, To control the area of ​​hydraulic storage, For the gate flow rate, This refers to the pumping flow rate of the pumping station.

[0059] Among them, the gate overflow rate , is represented as:

[0060] ;

[0061] In the formula, For flow coefficient, The width of the gate opening. This is the acceleration due to gravity.

[0062] Example 2, according to an embodiment of the present invention, further includes, based on Example 1: pre-constructing a digital twin model of the pump station to simulate the changes in hydraulic state and equipment operating state after the execution of the control command sequence, and to verify whether there are abnormal situations such as water backflow, sudden increase in water pressure, or equipment overload;

[0063] Specifically, to further enhance the safety of scheduling and control, before sending the control commands obtained in step S4 to the PLC, a digital twin model of the pump station built in the cloud is used to simulate the changes in hydraulic state and equipment operating status after the execution of the control command sequence, and to verify whether there are any abnormal situations such as water backflow, sudden increase in water pressure, or equipment overload.

[0064] If the simulation results are normal, the control command will be formally issued.

[0065] If the simulation results are abnormal, return to step S3 to adjust the weight coefficients or constraints and solve the optimization problem again.

[0066] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Those skilled in the art, upon considering the disclosure in the specification and embodiments, will readily conceive of other embodiments of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0067] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A cooperative adaptive scheduling method for pumping stations based on multi-source data, characterized in that, Includes the following steps: Pre-extraction of contextual feature parameters, including real-time acquisition of meteorological and rainfall forecast data. Upstream water level Downstream water level Inflow into reservoirs and rivers Equipment operating status data and downstream water demand instructions ; A comprehensive evaluation function S is constructed. Based on the extracted context feature parameters, a quantitative value is calculated through the comprehensive evaluation function to realize the automatic identification and triggering of scheduling modes. Based on the identified scheduling mode, a Model Predictive Control (MPC) pump gate control model with a time span of T is constructed. In each control sampling period, the gate opening degree a and the pump group operating frequency ω are used as control variables to solve a constrained optimization problem to achieve coordinated regulation of the pump gate. For the constructed constrained nonlinear optimization problem, a rolling time-domain solution is performed to obtain the optimal control vector sequence in the prediction time domain within each control sampling period. ,in, This is the optimal gate opening sequence. This is the optimal operating frequency sequence for the pump set.

2. The pump station cooperative adaptive scheduling method based on multi-source data driving according to claim 1, characterized in that, The optimal control vector sequence in the prediction time domain is obtained. This includes: constructing a digital twin model of the pump station to simulate the changes in hydraulic state and equipment operating status after the execution of the control command sequence, and to verify whether there are any abnormal situations.

3. The pump station cooperative adaptive scheduling method based on multi-source data driving according to claim 1, characterized in that, The comprehensive evaluation function S is expressed as: ； In the formula, These are the weighting coefficients; This is the upper limit of the safe water level upstream; The safe threshold for inflow into the reservoir and river; This is the quantification value for the critical level of rainfall forecasting.

4. The pump station cooperative adaptive scheduling method based on multi-source data driving according to claim 3, characterized in that, The automatic identification and triggering of the scheduling modes include: flood control scheduling mode, water supply economic scheduling mode, and ecological water replenishment mode; Among them, the flood control scheduling mode, that is, when the comprehensive evaluation function calculates the value... At that time, the flood control dispatch mode is triggered, in which, The threshold for triggering the flood control mode; the objective function under this mode is... This means maximizing the sum of the gate's flow rate and the pumping station's discharge rate to ensure the upstream water level in the pumping station's control area. , This is the upstream flood control warning level; Among them, the water supply economic dispatch mode, that is, when the comprehensive evaluation function calculates the value... At that time, the water supply economic dispatch mode is triggered, in which, The lower limit threshold for pattern recognition; the objective function for this pattern is... That is, to minimize the time integral of the system's overall energy consumption, and to achieve energy-saving and economical operation of the pump station while ensuring the downstream water demand command D; Among them, the ecological water replenishment mode is triggered directly when the monitored dissolved oxygen (DO) content is lower than the preset ecological threshold, without being limited by the comprehensive evaluation function S value; the objective function of this mode is to maintain the minimum ecological flow downstream of the control area. This meets the flow requirements of the river ecological water network.

5. The pump station cooperative adaptive scheduling method based on multi-source data driving according to claim 1, characterized in that, The objective function J of the optimization problem is expressed as: ； In the formula, T represents the time domain of model prediction; These are the weight coefficients for each optimization term; To predict the upstream water level at time t within the time domain; This represents the target upstream water level under the current dispatching mode. For the pump station unit at the operating frequency Yangcheng Real-time power; This represents the change in gate opening between adjacent sampling periods; This represents the change in pump unit operating frequency between adjacent sampling periods; For the pump station head, This is the real-time water level downstream.

6. The pump station cooperative adaptive scheduling method based on multi-source data driving according to claim 1, characterized in that, The rolling time-domain solution includes: the dynamic changes in water level within the calibration control area follow the water balance equation, expressed as: ； In the formula, To control the area of ​​hydraulic storage, For the gate flow rate, This refers to the pumping flow rate of the pumping station.

7. The pump station cooperative adaptive scheduling method based on multi-source data driving according to claim 6, characterized in that, The gate overflow , is represented as: ； In the formula, For flow coefficient, The width of the gate opening. This is the acceleration due to gravity.

Images