Search

CN-121981475-A - Pump station energy consumption optimal scheduling method and system for intelligent water affairs

CN121981475ACN 121981475 ACN121981475 ACN 121981475ACN-121981475-A

Abstract

The invention discloses a pump station energy consumption optimal scheduling method and system for intelligent water affairs, and relates to the technical field of pump station energy consumption optimal scheduling, wherein the method comprises the steps of collecting pump station water pump operation parameters and external multi-source data, and forming a structured input data set after time alignment and cleaning treatment; the method comprises the steps of constructing a nonlinear efficiency curved surface model based on water pump performance characteristics, generating a preliminary pump set start-stop and rotating speed configuration scheme meeting flow constraint by utilizing a structured input data set in combination with current net lift and water supply requirements, inputting the structured input data set into a water demand prediction model, outputting a time-by-time water consumption prediction result and a corresponding prediction uncertainty index in a future scheduling period, dynamically calculating minimum necessary safety margin based on the prediction uncertainty index and the current water storage capacity of a clean water basin, superposing the water consumption prediction result with the minimum necessary safety margin, and introducing the superposed water consumption prediction result serving as a feedforward constraint into a scheduling target to generate a corrected scheduling intention of fused time-of-use electricity price information.

Inventors

  • LI JIE
  • YUAN YA
  • MA ZHONGRUI
  • Ma Chaorong
  • HU BIN

Assignees

  • 聚创(广东)智能装备有限公司

Dates

Publication Date
20260505
Application Date
20260126

Claims (10)

  1. 1. An intelligent water service oriented pump station energy consumption optimization scheduling method is characterized by comprising the following steps: Collecting pump station water pump operation parameters and external multi-source data, and forming a structured input data set after time alignment and cleaning treatment; constructing a nonlinear efficiency curved surface model based on water pump performance characteristics, and generating a preliminary pump set start-stop and rotation speed configuration scheme meeting flow constraint by utilizing a structured input data set in combination with current net lift and water supply requirements; Inputting the structured input data set into a water demand prediction model, outputting a time-by-time-interval water consumption prediction result and a corresponding prediction uncertainty index in a future scheduling period, dynamically calculating a minimum necessary safety margin based on the prediction uncertainty index and the current water storage capacity of the clean water basin, overlapping the water consumption prediction result with the minimum necessary safety margin, and introducing the overlapped water consumption prediction result serving as a feedforward constraint into a scheduling target to generate a corrected scheduling intention fused with time-of-use electricity price information; Establishing a digital twin hydraulic simulation model comprising a pump station, a clean water tank and a water distribution pipe network, using the corrected scheduling intention as a drive, and simulating the dynamic response of the water level, the pressure and the water age of the system; According to the simulation result of the digital twin hydraulic simulation model, under the condition that the water level safety, the pipe network service pressure and the water age limit are met, the corrected scheduling intention is subjected to rolling optimization by combining the energy efficiency inertial cost of water pump combination switching, and a final pump station scheduling instruction is output; And issuing and executing the final pump station dispatching instruction, judging whether to trigger a rescheduling condition or not based on the real-time feedback operation data, and if so, returning to execute the steps again by using the updated data to realize closed-loop self-adaptive dispatching.
  2. 2. The intelligent water service oriented pump station energy consumption optimization scheduling method of claim 1, wherein the method is characterized in that the pump station operation parameters and external multi-source data are collected, and a structured input data set is formed after time alignment and cleaning treatment, and comprises the following specific steps: The method comprises the steps of collecting instantaneous flow, lift, current, voltage and clean water tank water level of each water pump in real time through the sensors of the Internet of things, which are arranged at the outlet of the water pump, the motor terminal and the clean water tank, wherein the sampling period is 15 minutes; synchronously acquiring the temperature and rainfall of 72 hours from the weather service platform every hour in the future, acquiring holiday identifiers from the municipal calendar system, and acquiring a history of 30-day-near water use record from the regional intelligent water meter platform; Performing linear interpolation alignment on the heterogeneous data according to the uniform 15-minute time granularity, and removing abnormal points which are continuously missing for more than two time periods or deviate from the sliding window mean value by three times of standard deviation; And (3) normalizing the cleaned data to a [0,1] interval by using Min-Max to generate a structured input data set.
  3. 3. The intelligent water service oriented pump station energy consumption optimization scheduling method of claim 2, wherein the method is characterized in that a nonlinear efficiency curved surface model is built based on water pump performance characteristics, and a primary pump set start-stop and rotation speed configuration scheme meeting flow constraint is generated by utilizing a structured input data set in combination with current net lift and water supply requirements, and comprises the following specific steps: For the first in the pump station The platform water pump is used for constructing a continuous and micro efficiency curved surface by utilizing discrete working condition points measured in a factory performance test and adopting radial basis function interpolation Wherein The flow rate is indicated by the flow rate, Representing the lift; In the current scheduling period, calculating the net lift according to the water level of the clean water tank and the pressure of the least adverse control point of the pipe network ; Setting the total water supply requirement of the next scheduling period as Introducing binary variables Representing the start-stop state of the water pump and continuous variable The output frequency of the frequency converter is represented, and the following optimization problem is established by taking the total input power minimization as a target: ; Wherein, the , Is water density; Gravitational acceleration; And (3) with The rated flow rate and the rated frequency of the water pump are respectively defined, and the constraint conditions comprise And is also provided with Solving the mixed integer nonlinear programming problem to obtain a preliminary pump set start-stop and rotating speed configuration scheme.
  4. 4. The intelligent water service oriented pump station energy consumption optimization scheduling method of claim 3, wherein the structured input data set is input into a water demand prediction model, a time-by-time water consumption prediction result and a corresponding prediction uncertainty index in a future scheduling period are output, a minimum necessary safety margin is dynamically calculated based on the prediction uncertainty index and the current water storage capacity of a clean water basin, the water consumption prediction result is overlapped with the minimum necessary safety margin and is used as a feedforward constraint to be introduced into a scheduling target, and a corrected scheduling intention integrating time-of-use electricity price information is generated, and the method comprises the following specific steps: inputting the structured input data set into a pre-trained long-period memory neural network, and outputting the predicted value of water consumption per hour in the future 24 hours Standard deviation of prediction ; Define the effective regulating volume of clean water pool as Wherein Is the cross-sectional area of the clean water tank, For the current water level, Is the lowest operating water level; Based on And (3) with Construction of minimum necessary safety margin The calculation formula is as follows: ; Wherein, the >0、 And >0 is a preset adjustment coefficient, The total effective volume of the clean water tank; Equation representation when predicting uncertainty Larger or clean water basin buffer capacity The smaller the safety margin is, the automatically increased, and the risk of undersupply is avoided; Updating the total scheduling requirement to ; Time-of-use electricity price introduced into power grid Constructing an economic-energy efficiency combined objective function, wherein the expression is as follows: ; Wherein, the For a period of time The total power consumption is that the total power consumption, Is an energy efficiency weight coefficient; And solving an economic-energy efficiency combined objective function to generate a corrected scheduling intention.
  5. 5. The intelligent water service oriented pump station energy consumption optimization scheduling method of claim 4, wherein the digital twin hydraulic simulation model comprising a pump station, a clean water basin and a water distribution network is established to correct the scheduling intention as driving, simulate the dynamic response of the water level, the pressure and the age of water of the system, and comprises the following specific steps: Constructing a pipe network topology model covering a pump station water outlet to key user nodes based on an EPANET hydraulic modeling engine; the time-interval total flow in the scheduling intention will be corrected And net lift As a boundary condition for the water source; setting an initial water level The initial water age of each node is zero; Performing transient hydraulic power and water quality joint simulation by taking 15 minutes as a step length, and solving a clean water tank water level dynamic equation: ; Wherein, the In order to schedule the step size, For a period of time A pipe network total outflow; synchronously tracking the water age evolution, and outputting a clear water tank water level sequence, a most adverse point pressure sequence and a maximum node water age sequence for subsequent constraint verification.
  6. 6. The intelligent water service oriented pump station energy consumption optimization scheduling method of claim 5, wherein the method is characterized in that the method is used for carrying out rolling optimization on the corrected scheduling intention by combining the energy efficiency inertial cost of water pump combination switching under the condition of meeting the water level safety, pipe network service pressure and water age limit according to the simulation result of the digital twin hydraulic simulation model and outputting a final pump station scheduling instruction, and comprises the following specific steps: setting hard constraint that the water level of the clean water tank is not lower than And must not be higher than The pressure at the most unfavorable point of the pipe network is not lower than the lower limit of the service pressure The maximum water age must not exceed the upper allowable limit ; Defining energy efficiency inertial costs The method comprises the following steps: ; Wherein, the For a period of time The start-stop state vector of the pump group, For the time interval from the last water pump combination switch, For the variance of the most unfavorable point pressure of the network in the last three periods, , , Is a non-negative weight coefficient; the cost item quantifies equipment abrasion, hydraulic impact and unstable control caused by frequent switching; Constructing a rolling optimization target, wherein the expression is as follows: ; and under the condition that the rigid constraint and the physical limit value of the water pump are met, adopting sequence quadratic programming to carry out iterative adjustment on flow distribution of each pump in the corrected scheduling intention until the change of the objective function is smaller than a preset convergence threshold value, and outputting a final pump station scheduling instruction.
  7. 7. The intelligent water service oriented pump station energy consumption optimization scheduling method of claim 6, wherein the final pump station scheduling instruction is issued and executed, whether a rescheduling condition is triggered is judged based on real-time return operation data, if yes, the updated data is returned to be used for re-executing the steps, and closed-loop self-adaptive scheduling is realized, and the specific steps are as follows: The water pump start-stop state and the frequency setting value of the frequency converter in the final pump station dispatching instruction are issued to the programmable logic controller through an industrial communication protocol; In the dispatching execution process, continuously receiving actual total flow returned by the sensor of the Internet of things, the water level of the clean water tank and the pressure of the most adverse point of the pipe network; The rescheduling is triggered when any of the following situations occur: the absolute value of the deviation between the actual total water consumption and the predicted value of two continuous scheduling periods exceeds 10%; Any water pump fails to stop; receiving a new time-sharing electricity price signal; The water level of the clean water tank enters a warning zone within 10% of the lowest or highest limit value; Immediately after triggering, the current scheduling period is terminated to regenerate the structured input data set from the latest acquired operating parameters and external data.
  8. 8. An intelligent water service oriented pump station energy consumption optimal scheduling system is based on the intelligent water service oriented pump station energy consumption optimal scheduling method according to any one of claims 1-7, and is characterized by comprising the following steps: the system comprises a multi-source data fusion module, a high-efficiency combination decision module, a dynamic margin scheduling module, a digital twin simulation module, an inertial rolling optimization module and a closed-loop self-adaptive execution module; the multi-source data fusion module is used for collecting and processing pump station operation and external environment data and generating a structured input data set; the efficient combination decision module is used for generating a preliminary pump set start-stop and rotation speed configuration scheme meeting flow requirements based on a nonlinear efficiency curved surface of the water pump and the current working condition; The dynamic margin scheduling module is used for dynamically calculating safety margin by combining water use prediction and uncertainty thereof and the water storage capacity of the clean water tank to generate a corrected scheduling intention fused with time-of-use electricity price; The digital twin simulation module is used for constructing a pump station-clean water tank-pipe network joint simulation model and simulating water level, pressure and water age response; the inertial rolling optimization module is used for introducing energy efficiency inertial cost to conduct rolling optimization on the corrected scheduling intention on the premise of meeting the constraint of water power and water quality, and outputting a final scheduling instruction; the closed-loop self-adaptive execution module is used for issuing a scheduling instruction and monitoring the running state, and when a rescheduling condition is triggered, the driving system re-executes the whole flow.
  9. 9. The intelligent water service oriented pump station energy consumption optimization scheduling method is characterized in that the intelligent water service oriented pump station energy consumption optimization scheduling method is realized by the processor when the computer program is executed.
  10. 10. A computer readable storage medium is provided, which is characterized in that the computer program is executed by a processor to realize the steps of the intelligent water service oriented pump station energy consumption optimizing and scheduling method in any one of claims 1-7.

Description

Pump station energy consumption optimal scheduling method and system for intelligent water affairs Technical Field The invention relates to the technical field of pump station energy consumption optimal scheduling, in particular to a pump station energy consumption optimal scheduling method and system for intelligent water affairs. Background The pump station energy consumption optimizing and scheduling technology is one intelligent control method for making the pump station consume minimum electric energy (or unit water supply energy consumption) in completing water supply task through scientifically determining the control parameters of the water pump, such as start-stop combination, running number, rotation speed regulation, etc. on the premise of meeting the service requirements of water supply amount, water pressure, water quality, etc. The key objective is to achieve the optimal balance between ensuring safe and reliable water supply and reducing the energy consumption cost of operation, and is a key technical link for realizing green low-carbon operation in intelligent water service. The existing scheduling is mainly based on deterministic water consumption prediction, a predicted value is directly used as a demand, uncertainty of prediction is not considered, once actual water consumption deviates from the prediction, clear water tank water level out-of-limit, pipe network low pressure or insufficient water supply are easily caused, and the traditional method is usually started and stopped by taking a single water pump high-efficiency area as a basis, or an average distribution flow strategy is adopted, cooperative potential among heterogeneous pump groups is not considered, and downstream pipe network pressure demand and water age constraint are not included in decisions. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. The invention provides an intelligent water service oriented pump station energy consumption optimization scheduling method, which solves the problems that the existing scheduling is based on deterministic water prediction, a predicted value is directly used as a demand input, uncertainty of prediction is not considered, clear water tank water level out-of-limit, pipe network low pressure or insufficient water supply are easily caused once actual water deviates from the prediction, a traditional method usually starts and stops based on a single water pump high-efficiency area, or an average distribution flow strategy is adopted, the cooperative potential among heterogeneous pump groups is not considered, and the pressure demand and water age constraint of a downstream pipe network are not included in a decision. In order to solve the technical problems, the invention provides the following technical scheme: in a first aspect, the invention provides an intelligent water service oriented pump station energy consumption optimization scheduling method, which comprises the following steps: Collecting pump station water pump operation parameters and external multi-source data, and forming a structured input data set after time alignment and cleaning treatment; constructing a nonlinear efficiency curved surface model based on water pump performance characteristics, and generating a preliminary pump set start-stop and rotation speed configuration scheme meeting flow constraint by utilizing a structured input data set in combination with current net lift and water supply requirements; Inputting the structured input data set into a water demand prediction model, outputting a time-by-time-interval water consumption prediction result and a corresponding prediction uncertainty index in a future scheduling period, dynamically calculating a minimum necessary safety margin based on the prediction uncertainty index and the current water storage capacity of the clean water basin, overlapping the water consumption prediction result with the minimum necessary safety margin, and introducing the overlapped water consumption prediction result serving as a feedforward constraint into a scheduling target to generate a corrected scheduling intention fused with time-of-use electricity price information; Establishing a digital twin hydraulic simulation model comprising a pump station, a clean water tank and a water distribution pipe network, using the corrected scheduling intention as a drive, and simulating the dynamic response of the water level, the pressure and the water age of the system; According to the simulation result of the digital twin hydraulic simulation model, under the condition that the water level safety, the pipe network service pressure and the water age limit are met, the corrected scheduling intention is subjected to rolling optimization by combining the energy efficiency inertial cost of water pump combination switching, and a final pump station scheduling instruction is output; Issuing and executing a final p