CN115860192A - Water supply network optimal scheduling method based on fuzzy neural network and genetic algorithm - Google Patents
Water supply network optimal scheduling method based on fuzzy neural network and genetic algorithm Download PDFInfo
- Publication number
- CN115860192A CN115860192A CN202211447118.XA CN202211447118A CN115860192A CN 115860192 A CN115860192 A CN 115860192A CN 202211447118 A CN202211447118 A CN 202211447118A CN 115860192 A CN115860192 A CN 115860192A
- Authority
- CN
- China
- Prior art keywords
- water supply
- water
- flow
- network
- pressure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention relates to a water supply network optimal scheduling method based on a fuzzy neural network and a genetic algorithm. The technical scheme of the invention is as follows: s1, randomly generating a plurality of initial individuals to form an initial population; s2, inputting water supply flow and pressure parameters of water supply pump stations of each water plant in the individual into the trained neural network model, and predicting flow and pressure values of each measuring point in the water supply network; s3, calculating a fitness function value of the individual based on the water supply flow and pressure parameters of the water supply pump stations of the water plants in the individual and the flow and pressure values of the measuring points obtained through prediction; s4, selecting a population, crossing the population, varying the population and optimizing parameters in a genetic algorithm based on fitness function values of all individuals in the population to form a new generation of population; and S5, repeating the steps S2 to S4 until the genetic algebra reaches a set algebra, and obtaining water supply flow and pressure parameters of the water supply pump stations of the water plants optimized by the genetic algorithm. The method is suitable for the technical field of water supply pipe network optimization scheduling.
Description
Technical Field
The invention relates to a water supply network optimal scheduling method based on a fuzzy neural network and a genetic algorithm. The method is suitable for the technical field of water supply network optimization scheduling.
Background
Along with the continuous development of national economy, the living standard of people is continuously improved, the scale of cities is continuously enlarged, the population of the cities is greatly increased, and along with the increasing demand of people on high-quality water, the demand on a water supply system is also higher and higher. Because urban water supply network system has the topological relation complicacy, characteristics such as operating condition are changeable, and now along with user's demand for water consumption is constantly improving, the scale of water supply network system is constantly expanding, if still rely on artificial experience's management mode to be difficult to satisfy modernized water supply management requirement.
At present, several key problems which are relatively concerned by each water supply enterprise are the problems of operation energy consumption of a water pump station in a water supply plant, water production cost of the water supply plant and reliability of a water supply network system, and the like, for a first-line large city, the scale of water supply is large, the topology system of a pipe network is complicated, the requirement on energy consumption is high, the energy consumption of water supply of each ton of water is about 0.2-0.3 degrees, according to statistics, the total water consumption in the whole country in 2020 is nearly 6000 billion tons, the energy consumption of water supply of one year exceeds 1200 billion degrees, and the water supply enterprise becomes the most main electricity consumption unit in each city; for water supply enterprises, the energy consumption of water plants is one of the important factors influencing the production cost of the enterprises, and the energy consumption cost accounts for more than 35-40% of the production cost. For example, a design scale of 10 4 m 3 In the second-stage pumping station of the/d, the total efficiency of the water pump unit is assumed to be 80%, if the pump station lift is increased by 1m, the annual consumed electric energy of the pumping station is up to 30 ten thousand kWh, and if the electric charge of each 1kWh is 0.85 yuan, the RMB is reduced to 25.5 yuan. Therefore, more reasonable water supply has great significance for reducing energy consumption cost of water supply enterprisesAnd (5) defining.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the existing problems, the water supply network optimal scheduling method based on the fuzzy neural network and the genetic algorithm is provided.
The technical scheme adopted by the invention is as follows: a water supply network optimal scheduling method based on a fuzzy neural network and a genetic algorithm is characterized in that:
s1, randomly generating a plurality of initial individuals to form an initial population, wherein the chromosome codes of the individuals comprise water supply flow and pressure parameters of water supply pump stations of water plants in the water supply pipe network;
s2, inputting water supply flow and pressure parameters of water supply pump stations of each water plant in an individual into the trained neural network model, and predicting flow and pressure values of each measuring point in the water supply network;
the training data set of the neural network model comprises a large number of water supply flow and pressure parameters of water supply pump stations of each water plant in the water supply network, and flow and pressure measured values of measurement points in the water supply network corresponding to the flow and pressure parameters of the water supply pump stations;
s3, calculating a fitness function value of the individual based on the water supply flow and pressure parameters of the water supply pump stations of the water plants in the individual and the flow and pressure values of the measuring points obtained through prediction;
s4, selecting a population, crossing the population, mutating the population and optimizing parameters in a genetic algorithm based on fitness function values of individuals in the population to form a new generation of population;
and S5, repeating the steps S2 to S4 until the genetic algebra reaches a set algebra, and obtaining water supply flow and pressure parameters of the water supply pump stations of the water plants optimized by the genetic algorithm.
The constraint conditions of the water supply flow and pressure parameters of the water supply pump station of each water plant and the flow and pressure values of each measuring point comprise: daily water supply capacity constraint of a water supply pump station, lifting lift constraint of the water supply pump station, pressure constraint of a measuring point and water supply and demand balance constraint.
The fitness function value of the individual is calculated based on the water supply flow and pressure parameters of the water supply pump stations of the water plants in the individual and the flow and pressure values of the measuring points obtained through prediction, and the fitness function value comprises the following steps:
the method comprises the steps of calculating an energy consumption objective function value F1 based on water supply flow and pressure parameters of water supply pump stations of water plants in individuals, calculating a pipe network reliability objective function value F2 based on flow and pressure values of various measuring points obtained through prediction, and calculating a fitness function value of the individual based on the energy consumption objective function value of the individual and a corresponding pipe network reliability objective function.
The energy consumption objective function value F1 includes:
in the formula: rho- -represents the density of water, kg/m 3 ；
G- -represents gravitational acceleration, G/s 2 ；
Q ij -representing the flow (m) of j pumping stations during the i period 3 /h)；
H ij -representing the lift (m) of the j pumping stations during the time i, based on the supply flow and pressure of the supply pumping stations
Calculating force parameters;
m represents the time period number (generally 24 hours) for supplying water to the pipe network by the plant;
n-represents the total number of pumping stations in the water supply system.
The pipe network reliability objective function value F2 comprises:
in the formula: q. q.s i -node traffic, m 3 /s；
h i -node i actual measurement water pressure, KPa;
n-represents the number of nodes in the water supply system.
The calculating of the individual fitness function value based on the individual energy consumption objective function value and the corresponding pipe network reliability objective function comprises the following steps:
wherein f is a fitness function; c is a constant; f is the objective function, F = α F1+ β F2,
0≤β≤1，
a storage medium having stored thereon a computer program executable by a processor, the computer program comprising: the computer program when executed implements the steps of the water supply network optimized scheduling method based on the fuzzy neural network and the genetic algorithm.
A computer device having a memory and a processor, the memory having stored thereon a computer program executable by the processor, the computer program comprising: the computer program when executed implements the steps of the water supply network optimal scheduling method based on the fuzzy neural network and the genetic algorithm.
The invention has the beneficial effects that: the method utilizes a genetic algorithm to carry out scheduling optimization of the water supply network, uses the water supply flow and pressure parameters of the water supply pump stations of each water plant in the water supply network as chromosome codes of individuals in the genetic algorithm, inputs the water supply flow and pressure parameters of the water supply pump stations in the individuals into a neural network model trained by a large amount of actual data to obtain the flow and pressure values corresponding to the water supply flow and pressure parameters of each measuring point in the water supply network and each water supply pump station in the individuals, calculates the fitness function value of the individuals by utilizing the water supply flow and pressure parameters of the water supply pump stations and the flow and pressure values of the measuring points, and obtains a scheduling optimization scheme of the water supply network through a plurality of genetic iterations.
The invention obtains the water supply flow and pressure parameters of each water supply pump station and the internal relation between each measuring point through the neural network model, thereby predicting the parameter value of each measuring point based on the parameters in the genetic algorithm individual, and further calculating the fitness function value of the individual by using the parameter value of each measuring point to carry out parameter optimization.
Drawings
FIG. 1 is a flow chart of an embodiment.
Fig. 2 is a diagram of a pipe network system according to an embodiment.
Detailed Description
As shown in fig. 1, the embodiment is a water supply network optimal scheduling method based on a fuzzy neural network and a genetic algorithm, and specifically includes the following steps:
s1, randomly generating a plurality of initial individuals to form an initial population, wherein chromosome codes of the individuals comprise water supply flow and pressure parameters of water supply pump stations of various water plants in the water supply pipe network, and binary codes of the water supply flow and pressure parameters of the water supply pump stations are provided.
And S2, inputting the water supply flow and pressure parameters of the water supply pump stations of the water plants in the individual into the trained neural network model, and predicting to obtain the flow and pressure values of the measuring points in the corresponding time period of the individual in the water supply network.
In this embodiment, the training data set of the neural network model includes actual data of a large number of water supply networks, including water supply flow and pressure parameters of water supply pump stations of water plants, and actual flow and pressure values corresponding to the flow and pressure parameters of the water supply pump stations at each measuring point in the water supply network.
And S3, calculating a fitness function value of the individual based on the water supply flow and pressure parameters of the water supply pump stations of the water plants in the individual and the flow and pressure values of the measuring points obtained through prediction.
The method comprises the steps of calculating an energy consumption objective function value F1 based on water supply flow and pressure parameters of water supply pump stations of water plants in individuals, calculating a pipe network reliability objective function value F2 based on flow and pressure values of various measuring points obtained through prediction, and calculating a fitness function value of the individual based on the energy consumption objective function value of the individual and a corresponding pipe network reliability objective function.
A secondary pump room in the water supply pipe network system lifts water to a water distribution pipe network and can meet the water supply requirement of the worst point, and the electric energy consumed in the process is used as an energy consumption target function F1 of primary optimization scheduling of the pipe network;
in the formula: rho- -represents the density of water, kg/m 3 ；
G- -represents gravitational acceleration, G/s 2 ；
Q ij -representing the water supply (m) of j pumping stations during the i period 3 /h)；
H ij -representing the lift (m) of the j pump station during the i time;
m represents the time period number (generally 24 hours) for supplying water to the pipe network by the plant;
n-represents the total number of pump stations in the water supply system;
the embodiment is used for measuring the actual measurement water pressure P when each node operates during the normal operation of the water supply network 1 And its minimum required water pressure P 0 The difference value of (2) is called as node surplus water head, the lower the weighted average value of the surplus water heads of the nodes is, the higher the reliability of the pipe network system is, namely the reliability of the pipe network system is taken as a pipe network reliability target function F2, then
In the formula: q. q.s i -node traffic, m 3 /s；
h i -node i actual measurement water pressure, KPa;
n-represents the number of nodes in the water supply system.
In this embodiment, calculating the fitness function value of the individual based on the individual energy consumption objective function value and the corresponding pipe network reliability objective function includes:
wherein f is a fitness function; c is a constant; f is the objective function, F = α F1+ β F2,
0≤β≤1，
and S4, selecting, crossing and mutating in a genetic algorithm based on the fitness function values of all individuals in the population to form a new generation of population.
Selecting population in genetic algorithm, endowing individual with fitness f with selection probability Pi, then
The generation with the largest adaptation is inherited to the next generation population.
And S5, repeating the steps S2 to S4 until the genetic algebra reaches a set algebra, and obtaining water supply flow and pressure parameters of the water supply pump stations of the water plants optimized by the genetic algorithm.
In this embodiment, the water supply flow and pressure parameters of each water supply pump station of the water plant and the flow and pressure values of each measuring point have constraint conditions, including: daily water supply capacity constraint of a water supply pump station, lifting lift constraint of the water supply pump station, pressure constraint of a measuring point and water supply and demand balance constraint.
(a) The constraint conditions of the daily water supply capacity of the water supply pump station are as follows:
in the formula: q ij Express j pumpWater supply amount (m) at time i 3 /d)
Q maxj Maximum daily water supply (m) for jth pumping station 3 /d)
j＝1、2、3、4………n。
(b) The water supply pump station lifting lift constraint condition is as follows:
H mini，j ≤H i，j ≤H maxi，j
in the formula: h mini，j 、H maxi，j Respectively representing the design minimum value and the design maximum value of the lifting head of the pump station j in the water supply period i.
(c) Pressure constraint conditions of the pressure monitoring points:
h jimin ≤h ji ≤h jimax
in the formula:
h j、imin -represents the lowest water pressure, m, of the jth water pressure monitoring point during the period i;
h j、imax and m represents the highest bearable water pressure of the pipe section in the period i at the j water pressure monitoring point.
(d) Water supply and demand balance constraint conditions:
in the formula: q i -indicating the water demand (m) of the pipe network during the period i 3 /h)。
Fig. 2 is a diagram of a pipe network system according to an embodiment, where constraint conditions for performing water supply network optimal scheduling on the pipe network system include:
TABLE 1. Water supply interval value (unit: m) for two water plants 3 /h)
TABLE 2 head of water pump station of each water plant (unit: MPa)
TABLE 3 Upper and lower values of pressure at monitoring point in water supply pipe network system (unit: MPa)
The scheduling scheme optimized by the embodiment is as follows:
TABLE 4 optimized and empirical scheduling schemes
The present embodiment also provides a storage medium having stored thereon a computer program executable by a processor, the computer program when executed performing the steps of the water supply network optimized dispatch method in this example based on a fuzzy neural network and a genetic algorithm.
The present embodiment also provides a computer device having a memory and a processor, the memory having stored thereon a computer program executable by the processor, the computer program when executed performing the steps of the water supply network optimization scheduling method based on the fuzzy neural network and the genetic algorithm in this example.
Claims (8)
1. A water supply network optimal scheduling method based on a fuzzy neural network and a genetic algorithm is characterized in that:
s1, randomly generating a plurality of initial individuals to form an initial population, wherein chromosome codes of the individuals comprise water supply flow and pressure parameters of water supply pump stations of water plants in the water supply pipe network;
s2, inputting water supply flow and pressure parameters of water supply pump stations of each water plant in the individual into the trained neural network model, and predicting flow and pressure values of each measuring point in the water supply network;
the training data set of the neural network model comprises a large number of water supply flow and pressure parameters of water supply pump stations of each water plant in the water supply network, and flow and pressure measured values of measurement points in the water supply network corresponding to the flow and pressure parameters of the water supply pump stations;
s3, calculating a fitness function value of the individual based on the water supply flow and pressure parameters of the water supply pump stations of the water plants in the individual and the flow and pressure values of the measuring points obtained through prediction;
s4, selecting, crossing and mutating in a genetic algorithm based on fitness function values of all individuals in the population to form a new generation of population;
and S5, turning to the step S2 until the genetic algebra reaches a set algebra to obtain water supply flow and pressure parameters of the water supply pump stations of the water plants optimized by the genetic algorithm.
2. The water supply network optimal scheduling method based on the fuzzy neural network and the genetic algorithm as claimed in claim 1, wherein: the water supply flow and pressure parameters of each water supply pump station of each water plant and the flow and pressure of each measuring point have constraint conditions, and the method comprises the following steps: the method comprises the following steps of daily water supply capacity constraint of a water supply pump station, lifting lift constraint of the water supply pump station, pressure constraint of a measuring point and water supply and demand balance constraint.
3. The water supply network optimal scheduling method based on the fuzzy neural network and the genetic algorithm as claimed in claim 1, wherein: the fitness function value of the individual is calculated based on the water supply flow and pressure parameters of the water supply pump stations of the water plants in the individual and the flow and pressure values of the measuring points obtained through prediction, and the fitness function value comprises the following steps:
the method comprises the steps of calculating an energy consumption objective function value F1 based on water supply flow and pressure parameters of water supply pump stations of water plants in individuals, calculating a pipe network reliability objective function value F2 based on flow and pressure values of various measuring points obtained through prediction, and calculating a fitness function value of the individual based on the energy consumption objective function value of the individual and a corresponding pipe network reliability objective function.
4. The water supply network optimization scheduling method based on the fuzzy neural network and the genetic algorithm as claimed in claim 3, wherein the energy consumption objective function value F1 comprises:
in the formula: rho- -represents the density of water, kg/m 3 ；
G- -represents gravitational acceleration, G/s 2 ；
Q ij -representing the flow (m) of j pumping stations during the i period 3 /h)；
H ij -representing the lift (m) of the j pump station within the time i, calculated on the basis of the water supply flow and pressure parameters of the water supply pump station;
m represents the number of time sections (generally 24 hours) for supplying water to the pipe network by the plant station;
n-represents the total number of pumping stations in the water supply system.
5. The water supply network optimization scheduling method based on the fuzzy neural network and the genetic algorithm as claimed in claim 3, wherein the pipe network reliability objective function value F2 comprises:
in the formula: q. q.s i -node traffic, m 3 /s；
h i -node i actual measurement water pressure, KPa;
n-represents the number of nodes in the water supply system.
6. The water supply network optimization scheduling method based on the fuzzy neural network and the genetic algorithm as claimed in claim 3, wherein the calculating the fitness function value of the individual based on the energy consumption objective function value of the individual and the corresponding network reliability objective function comprises:
wherein f is a fitness function; c is a constant; f is the objective function, F = α F1+ β F2,
0≤β≤1，
7. a storage medium having stored thereon a computer program executable by a processor, the computer program comprising: the computer program when executed implements the steps of the water supply network optimized dispatching method based on the fuzzy neural network and the genetic algorithm of any one of claims 1 to 6.
8. A computer device having a memory and a processor, the memory having stored thereon a computer program executable by the processor, the computer program comprising: the computer program when executed implements the steps of the water supply network optimized dispatching method based on the fuzzy neural network and the genetic algorithm of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211447118.XA CN115860192A (en) | 2022-11-18 | 2022-11-18 | Water supply network optimal scheduling method based on fuzzy neural network and genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211447118.XA CN115860192A (en) | 2022-11-18 | 2022-11-18 | Water supply network optimal scheduling method based on fuzzy neural network and genetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115860192A true CN115860192A (en) | 2023-03-28 |
Family
ID=85664160
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211447118.XA Pending CN115860192A (en) | 2022-11-18 | 2022-11-18 | Water supply network optimal scheduling method based on fuzzy neural network and genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115860192A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116466591A (en) * | 2023-06-13 | 2023-07-21 | 埃睿迪信息技术（北京）有限公司 | Method and device for determining water supply strategy of water supply system |
| CN116596280A (en) * | 2023-07-17 | 2023-08-15 | 青岛国源中创电气自动化工程有限公司 | Cooperative scheduling method for water pump set of sewage treatment plant |
-
2022
- 2022-11-18 CN CN202211447118.XA patent/CN115860192A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116466591A (en) * | 2023-06-13 | 2023-07-21 | 埃睿迪信息技术（北京）有限公司 | Method and device for determining water supply strategy of water supply system |
| CN116466591B (en) * | 2023-06-13 | 2023-08-29 | 埃睿迪信息技术（北京）有限公司 | Method and device for determining water supply strategy of water supply system |
| CN116596280A (en) * | 2023-07-17 | 2023-08-15 | 青岛国源中创电气自动化工程有限公司 | Cooperative scheduling method for water pump set of sewage treatment plant |
| CN116596280B (en) * | 2023-07-17 | 2023-10-03 | 青岛国源中创电气自动化工程有限公司 | Cooperative scheduling method for water pump set of sewage treatment plant |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115860192A (en) | Water supply network optimal scheduling method based on fuzzy neural network and genetic algorithm | |
| CN111401604B (en) | Power system load power prediction method and energy storage power station power distribution method | |
| CN110071502B (en) | Calculation method for short-term power load prediction | |
| CN113972694B (en) | Investment decision-making method for distributed photovoltaic and energy storage power station of power distribution network | |
| CN110854891B (en) | Power distribution network pre-disaster resource allocation method and system | |
| CN110867852B (en) | Microgrid energy storage optimization configuration method and device considering whole life cycle cost | |
| CN110165698B (en) | Wind power plant smooth grid-connection method for realizing prospective error asset conversion | |
| CN116681312B (en) | Ecological-oriented multi-objective reservoir optimal scheduling decision method and system | |
| CN112884270A (en) | Multi-scene power distribution network planning method and system considering uncertainty factors | |
| CN113162090A (en) | Energy storage system capacity configuration optimization method considering battery module capacity | |
| CN115983710A (en) | High-proportion new energy access electric power system infrastructure project decision method and system | |
| CN116961052A (en) | Method and system for locating and sizing distributed energy storage system | |
| CN112651177B (en) | Flexible resource allocation method and system for power distribution network considering flexible service cost | |
| CN109214610A (en) | A kind of saturation Methods of electric load forecasting based on shot and long term Memory Neural Networks | |
| CN111914900B (en) | User electricity utilization mode classification method | |
| CN110610264B (en) | Single-target optimization method for water supply pipe network regulation and control under uncertain situation | |
| CN116796990A (en) | Control method for water level fluctuation of cascade reservoir before flood under uncertain electricity price condition | |
| CN116683530A (en) | Wind-light-containing hybrid type pumping and storing station cascade reservoir random optimization scheduling method | |
| CN114123175B (en) | Distributed control method for power distribution network considering generalized demand side resource alliance characteristics | |
| CN115239117A (en) | Classification and feature selection method based on multi-objective optimization | |
| CN115186882A (en) | Clustering-based controllable load spatial density prediction method | |
| CN112785022B (en) | Method and system for excavating electric energy substitution potential | |
| CN114742285B (en) | Construction method and application of resident power consumption mode prediction model | |
| CN113642760B (en) | Regional pump station optimization site selection method for large-scale reconstruction and expansion water supply network | |
| CN111932115B (en) | Power system interval expansion planning method considering source-load interaction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |