CN116151409A - Urban daily water demand prediction method based on neural network - Google Patents
Urban daily water demand prediction method based on neural network Download PDFInfo
- Publication number
- CN116151409A CN116151409A CN202211249743.3A CN202211249743A CN116151409A CN 116151409 A CN116151409 A CN 116151409A CN 202211249743 A CN202211249743 A CN 202211249743A CN 116151409 A CN116151409 A CN 116151409A
- Authority
- CN
- China
- Prior art keywords
- data
- neural network
- factors
- clustering
- key
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses a city daily water demand prediction method based on a neural network, belongs to the technical field of water supply water quantity prediction methods, and solves the problems of large data quantity and low data processing efficiency among a plurality of cities; secondly, performing dimension reduction treatment on key factors based on data analysis to obtain key influence factors of cities; thirdly, checking the model by taking the obtained key influencing factors and the data of the water used in the past as the input of the Elman neural network algorithm and correcting the model, thereby achieving the effect of improving the data processing efficiency and the data processing reliability.
Description
Technical Field
The invention relates to the field of urban water supply quantity prediction methods, in particular to a method for predicting urban daily water demand based on a neural network.
Background
At present, water resource management is a method for solving the problem of water consumption, and in order to improve the water consumption, the daily water consumption needs to be predicted. The water consumption and the water saving potential can change along with the influence of conditions such as time, weather and the like, and the method has the characteristics of complexity, nonlinearity, timeliness and the like. The mathematical methods currently used for prediction can be divided into three categories: time series method, structure analysis method, and system analysis method.
The neural network is used as an algorithm model, has strong self-organizing, self-learning, induction and fault tolerance capabilities, has a good fitting effect on nonlinear problems, and is increasingly used for solving the prediction problem.
Therefore, based on the above technical premise, the problems to be solved are: how to improve the reliability of the prediction.
Disclosure of Invention
Aiming at the defects of the prior art, the invention at least solves the technical problems in the related art to a certain extent, provides a method for predicting urban daily water demand based on a neural network, and has the advantage of improving the prediction reliability.
In order to solve the technical problems, the technical scheme of the invention is as follows: a city daily water demand prediction method based on a neural network comprises the following steps:
firstly, carrying out fuzzy C clustering on the annual data of a plurality of cities by taking water use variance and annual water use average as feature vectors, and classifying the cities into three types;
secondly, performing dimension reduction treatment on key factors based on data analysis to obtain key influence factors of cities;
thirdly, checking the model by taking the obtained key influencing factors and the annual water consumption data as the input of an Elman neural network algorithm, and correcting the model.
Preferably, the step of fuzzy C-clustering the plurality of city calendar data with the water data as the feature vector includes:
step one: initializing a membership matrix U by using random numbers with values of [0,1 ];
step two: by using
Computing a clustering center C i ，i＝1,2,…,c；
Step three: according to
Calculating a cost function, and stopping the algorithm if the value is smaller than a certain threshold value or the change amount of the value relative to the last cost function value is smaller than a certain threshold value; />
Step four: by using
And (3) calculating a new U matrix, and returning to the step two.
Preferably, the step of correcting the model includes: according to the definition:
a＝|k t+1 -k t |/Y t ″；
b＝sgn(k t+1 -k t )；
the conditional attribute set c= { a, b, C }, the decision attribute set d= { D }, is extracted.
Preferably, the fuzzy C-clustering further comprises the steps of:
step one: preprocessing data;
step two: extracting feature vectors of fuzzy C-means clustering;
step three: setting parameters and carrying out fuzzy C-means clustering;
step four: the main cause analysis synthesizes new components and forms an influence factor group with the original influence factors;
step five: carrying out Pearson correlation analysis on each type of data to extract each type of key influence factors;
step six: normalizing the data; adjusting each type of data and key influencing factors thereof to 0-1;
step seven: elman neural network prediction; taking normalized data as input; setting a training target minimum error, training times, a real frequency and a learning rate; after outputting the result, the number of the neurons can be adjusted to achieve the optimal effect;
step eight: inverse normalization and output of the prediction result;
step nine: constructing a condition attribute set by the output predicted value; constructing a decision attribute set according to the scale factors selected by expert experience; discretizing the set by using a discretization method of equal frequency division to form a decision table; determining a scale factor through a minimum decision rule; and carrying out predicted value correction.
Preferably, the minimum decision rule can be obtained by processing the decision table by adopting an attribute reduction algorithm; thereby determining a scale factor S; and the modification of the rough set to the model is realized.
Compared with the background art, the invention has the technical effects that:
1. the data area is wide, and the data processing is efficient;
2. the urban water prediction model is classified according to the water changes of different cities, and if the number of cities is large, the generalization reduction capability is recognized, so that the urban prediction model can classify the cities before prediction so as to reduce urban differentiation.
Drawings
FIG. 1 is a flow chart of data processing in an embodiment.
Detailed Description
The following detailed description of the invention is provided in connection with the accompanying drawings to facilitate understanding and grasping of the technical scheme of the invention.
Embodiment one:
referring to fig. 1, a method for predicting urban daily water demand based on a neural network is provided: firstly, carrying out fuzzy C clustering on the annual data of a plurality of cities by taking water use variance and annual water use average as feature vectors, and classifying the cities into three types; secondly, performing dimension reduction treatment on key factors based on data analysis to obtain key influence factors of cities; thirdly, checking the model by taking the obtained key influencing factors and the annual water consumption data as the input of an Elman neural network algorithm, and correcting the model.
Fuzzy C-means (FCM) is a soft clustering method; allowing a piece of data to belong to two or more clusters; the data points are classified mainly by membership degrees, and the larger the membership degree is, the closer the distance between the data points and the clustering center is. The fuzzy partitioning is performed by iterative optimization of the objective function described above.
The step of fuzzy C-clustering the plurality of city calendar data with water data as feature vectors comprises:
step one: initializing a membership matrix U by using random numbers with values of [0,1 ];
step two: by using
Computing a clustering center C i ，i＝1,2,…,c；
Step three: according to
Calculating a cost function, and stopping the algorithm if the value is smaller than a certain threshold value or the change amount of the value relative to the last cost function value is smaller than a certain threshold value;
step four: by using
And (3) calculating a new U matrix, and returning to the step two.
The step of modifying the model comprises: according to the definition:
a＝|k t+1 -k t |/Y t ″；
b＝sgn(k t+1 -k t )；
the conditional attribute set c= { a, b, C }, the decision attribute set d= { D }, is extracted.
Therefore, the scheme can process various data, including missing data and data with a plurality of variables, can process inaccuracy and ambiguity of the data, including deterministic and nondeterministic conditions, can obtain the minimum expression of knowledge and various different levels of knowledge, can strip out a mode which is simple in concept and easy to operate from the data, and can generate rules which are accurate and easy to check and verify.
The fuzzy C-clustering further comprises the following steps:
step one: preprocessing data;
step two: extracting feature vectors of fuzzy C-means clustering;
step three: setting parameters and carrying out fuzzy C-means clustering;
step four: the main cause analysis synthesizes new components and forms an influence factor group with the original influence factors;
step five: carrying out Pearson correlation analysis on each type of data to extract each type of key influence factors;
step six: normalizing the data; adjusting each type of data and key influencing factors thereof to 0-1;
step seven: elman neural network prediction; taking normalized data as input; setting a training target minimum error, training times, a real frequency and a learning rate; after outputting the result, the number of the neurons can be adjusted to achieve the optimal effect;
step eight: inverse normalization and output of the prediction result;
step nine: constructing a condition attribute set by the output predicted value; constructing a decision attribute set according to the scale factors selected by expert experience; discretizing the set by using a discretization method of equal frequency division to form a decision table; determining a scale factor through a minimum decision rule; and carrying out predicted value correction.
Processing the decision table by adopting an attribute reduction algorithm to obtain a minimum decision rule; thereby determining a scale factor S; and the modification of the rough set to the model is realized.
The comprehensive influence factors are combined into the original influence factors to form an influence factor group, correlation analysis is carried out on the influence factors and the water consumption data of each type of city respectively, key factors are extracted, and finally the water consumption data and the key influence factors preprocessed by each type of city are used as inputs of an ELman neural network for training and prediction; finally, in order to overcome the defect that the Elman neural network approaches a nonlinear function; the prediction error of the peak value points with overlarge slopes on the two sides is larger, and a rough set is introduced for correction, so that the prediction precision is improved.
Therefore, the scheme has the remarkable effects of wide data collection range and high data processing efficiency. The urban water prediction model is classified according to the water changes of different cities, and if the number of cities is large, the generalization reduction capability is recognized, so that the urban prediction model can classify the cities before prediction so as to reduce urban differentiation.
The words "exemplary" and/or "example" are used herein to mean "serving as an example, instance, or illustration. Any embodiment described herein as "exemplary" and/or "example" is not necessarily to be construed as preferred or advantageous over other embodiments. Likewise, the term "embodiments of the invention" does not require that all embodiments of the invention include the discussed feature, advantage or mode of operation.
Moreover, those of skill in the art will appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.
The methods, sequences and/or algorithms described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. If implemented in software, the functions may be stored on or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one location to another. Storage media may be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Moreover, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Combinations of the above should also be included within the scope of computer-readable media.
Thus, embodiments of the invention may include a computer-readable medium having code embodied thereon for performing the functions, steps, sequences of actions, and/or algorithms disclosed herein.
While the foregoing disclosure shows illustrative embodiments of the invention, it should be noted that various changes and modifications could be made herein without departing from the scope of the invention as defined by the appended claims. The functions, steps and/or actions of the method elements in accordance with the embodiments of the invention described herein need not be performed in any particular order. Furthermore, although elements of the invention may be described or claimed in the singular, the plural is contemplated unless limitation to the singular is explicitly stated.
Of course, the above is only a typical example of the invention, and other embodiments of the invention are also possible, and all technical solutions formed by equivalent substitution or equivalent transformation fall within the scope of the invention claimed.
Claims (5)
1. A city daily water demand prediction method based on a neural network is characterized by comprising the following steps:
firstly, carrying out fuzzy C clustering on the annual data of a plurality of cities by taking water use variance and annual water use average as feature vectors, and classifying the cities into three types;
secondly, performing dimension reduction treatment on key factors based on data analysis to obtain key influence factors of cities;
thirdly, checking the model by taking the obtained key influencing factors and the annual water consumption data as the input of an Elman neural network algorithm, and correcting the model.
2. The neural network-based urban daily water demand prediction method according to claim 1, wherein the method comprises the following steps: the step of fuzzy C-clustering the plurality of city calendar data with water data as feature vectors comprises:
step one: initializing a membership matrix U by using random numbers with values of [0,1 ];
step two: by using
Computing a clustering center C i ，i＝1,2,…,c；
Step three: according to
Calculating a cost function, and stopping the algorithm if the value is smaller than a certain threshold value or the change amount of the value relative to the last cost function value is smaller than a certain threshold value;
step four: by using
And (3) calculating a new U matrix, and returning to the step two.
3. The neural network-based urban daily water demand prediction method according to claim 1, wherein the method comprises the following steps: the step of modifying the model comprises: according to the definition:
a＝|k t+1 -k t |/Y t ″；
b＝sgn(k t+1 -k t )；
the conditional attribute set c= { a, b, C }, the decision attribute set d= { D }, is extracted.
4. The neural network-based urban daily water demand prediction method according to claim 1, wherein the method comprises the following steps: the fuzzy C-clustering further comprises the following steps:
step one: preprocessing data;
step two: extracting feature vectors of fuzzy C-means clustering;
step three: setting parameters and carrying out fuzzy C-means clustering;
step four: the main cause analysis synthesizes new components and forms an influence factor group with the original influence factors;
step five: carrying out Pearson correlation analysis on each type of data to extract each type of key influence factors;
step six: normalizing the data; adjusting each type of data and key influencing factors thereof to 0-1;
step seven: elman neural network prediction; taking normalized data as input; setting a training target minimum error, training times, a real frequency and a learning rate; after outputting the result, the number of the neurons can be adjusted to achieve the optimal effect;
step eight: inverse normalization and output of the prediction result;
step nine: constructing a condition attribute set by the output predicted value; constructing a decision attribute set according to the scale factors selected by expert experience; discretizing the set by using a discretization method of equal frequency division to form a decision table; determining a scale factor through a minimum decision rule; and carrying out predicted value correction.
5. The method for predicting urban daily water demand based on the neural network according to claim 4, wherein the method comprises the following steps: processing the decision table by adopting an attribute reduction algorithm to obtain a minimum decision rule; thereby determining a scale factor S; and the modification of the rough set to the model is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211249743.3A CN116151409A (en) | 2022-10-12 | 2022-10-12 | Urban daily water demand prediction method based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211249743.3A CN116151409A (en) | 2022-10-12 | 2022-10-12 | Urban daily water demand prediction method based on neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116151409A true CN116151409A (en) | 2023-05-23 |
Family
ID=86353228
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211249743.3A Pending CN116151409A (en) | 2022-10-12 | 2022-10-12 | Urban daily water demand prediction method based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116151409A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116451874A (en) * | 2023-06-14 | 2023-07-18 | 埃睿迪信息技术（北京）有限公司 | Urban water consumption prediction method, device and equipment |
-
2022
- 2022-10-12 CN CN202211249743.3A patent/CN116151409A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116451874A (en) * | 2023-06-14 | 2023-07-18 | 埃睿迪信息技术（北京）有限公司 | Urban water consumption prediction method, device and equipment |
| CN116451874B (en) * | 2023-06-14 | 2023-09-05 | 埃睿迪信息技术（北京）有限公司 | Urban water consumption prediction method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109639739B (en) | Abnormal flow detection method based on automatic encoder network | |
| CN111832647A (en) | Abnormal flow detection system and method | |
| CN108875779A (en) | Training method, device and the terminal device of neural network | |
| CN112685504B (en) | Production process-oriented distributed migration chart learning method | |
| CN110708318A (en) | Network abnormal flow prediction method based on improved radial basis function neural network algorithm | |
| CN113268370B (en) | Root cause alarm analysis method, system, equipment and storage medium | |
| CN109657851A (en) | Forecasting of Stock Prices method, apparatus, equipment and the storage medium of event driven | |
| CN110033089A (en) | Deep neural network parameter optimization method and system based on Distributed fusion algorithm | |
| Cao et al. | Prediction of dissolved oxygen content in aquaculture based on clustering and improved ELM | |
| CN114708479B (en) | Self-adaptive defense method based on graph structure and characteristics | |
| CN114255121A (en) | Credit risk prediction model training method and credit risk prediction method | |
| CN116151409A (en) | Urban daily water demand prediction method based on neural network | |
| CN117150416B (en) | Method, system, medium and equipment for detecting abnormal nodes of industrial Internet | |
| CN111914166B (en) | Correction strategy personalized recommendation system applied to community correction personnel | |
| CN111783688B (en) | Remote sensing image scene classification method based on convolutional neural network | |
| CN117113086A (en) | Energy storage unit load prediction method, system, electronic equipment and medium | |
| Shi et al. | Application of deep learning in financial management evaluation | |
| CN116739940A (en) | Point cloud completion method, device, equipment and medium | |
| CN111368168A (en) | Big data-based electricity price obtaining and predicting method, system and computer-readable storage medium | |
| CN116541166A (en) | Super-computing power scheduling server and resource management method | |
| Li | Forecast and simulation of the public opinion on the public policy based on the markov model | |
| CN110991247A (en) | Electronic component identification method based on deep learning and NCA fusion | |
| CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
| CN112581177B (en) | Marketing prediction method combining automatic feature engineering and residual neural network | |
| CN114970357A (en) | Energy-saving effect evaluation method, system, device and storage medium |
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 |