CN110751416A - Method, device and equipment for predicting water consumption - Google Patents
Method, device and equipment for predicting water consumption Download PDFInfo
- Publication number
- CN110751416A CN110751416A CN201911037989.2A CN201911037989A CN110751416A CN 110751416 A CN110751416 A CN 110751416A CN 201911037989 A CN201911037989 A CN 201911037989A CN 110751416 A CN110751416 A CN 110751416A
- Authority
- CN
- China
- Prior art keywords
- water
- data
- neural network
- network model
- domestic
- 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
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 359
- 238000000034 method Methods 0.000 title claims abstract description 39
- 239000008235 industrial water Substances 0.000 claims abstract description 134
- 238000003062 neural network model Methods 0.000 claims abstract description 115
- 230000008859 change Effects 0.000 claims abstract description 14
- 238000010606 normalization Methods 0.000 claims description 13
- 238000007621 cluster analysis Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 11
- 238000001556 precipitation Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 abstract 1
- 238000012549 training Methods 0.000 description 12
- 239000011159 matrix material Substances 0.000 description 10
- 238000003860 storage Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000008399 tap water Substances 0.000 description 1
- 235000020679 tap water Nutrition 0.000 description 1
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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- 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
-
- 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
- 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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Educational Administration (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method, a device and equipment for predicting water consumption. The water consumption prediction method comprises the following steps: acquiring historical water consumption data of a target area; dividing historical water data into industrial water data and domestic water data according to the change rule of the historical water data along with time; respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data; and predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model. According to the technical scheme of the embodiment of the invention, according to the historical water use data and the change rule thereof, the historical water use data is firstly divided into industrial water and domestic water, and then different convolutional neural network models are respectively established for predicting the water use amount, so that the automatic prediction of the water use amount is realized, and the prediction result has strong adaptability and high precision.
Description
Technical Field
The embodiment of the invention relates to the technical field of data processing, in particular to a method, a device and equipment for predicting water consumption.
Background
The urban and rural water consumption prediction can analyze the water consumption in a period of time in the future, and plays an increasingly important guiding role in aspects of water resource planning, water consumption management, scientific scheduling of a water supply system and the like. Therefore, research on urban water consumption prediction technology and method is an important work for realizing sustainable utilization of water resources.
The scale prediction of the existing urban and rural tap water consumption is mostly determined according to urban water supply engineering planning specification GB50282-98 and rural water supply design specification CECS82:96 by multiplying the average water consumption index by the planned annual predicted population number.
The water demand calculated by the calculation method is a determined index, has no characteristics of changing along with time and external conditions, has a certain difference with the actual condition, is used for guiding a large error possibly generated during the production of a water plant, and often needs to be finely adjusted by water plant workers according to experience in the actual operation. The prediction result has large error and poor adaptivity.
Disclosure of Invention
The invention provides a method, a device and equipment for predicting water consumption, which are used for realizing automatic and accurate prediction of the water consumption.
In a first aspect, an embodiment of the present invention provides a method for predicting water consumption, including:
acquiring historical water consumption data of a target area;
dividing historical water data into industrial water data and domestic water data according to the change rule of the historical water data along with time;
respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data;
and predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
In a second aspect, embodiments of the present invention further provide a device for predicting water consumption, including:
the historical information acquisition module is used for acquiring historical water consumption data of the target area;
the information dividing module is used for dividing the historical water use data into industrial water use data and domestic water use data according to the change rule of the historical water use data along with time;
the neural network model establishing module is used for respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data;
and the water consumption prediction module is used for predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
In a third aspect, an embodiment of the present invention further provides an apparatus, where the apparatus includes:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method for predicting water usage provided by any embodiment of the present invention.
According to the technical scheme of the embodiment of the invention, the historical water consumption data is obtained and classified according to the change rule of the historical water consumption data into two types of industrial water and domestic water, different neural network models are respectively established for the two types of data, the water consumption is predicted according to the established neural network models, the local historical data is taken as the basic data for prediction, the prediction of the water consumption is more consistent with the actual condition of water consumption, and the different neural network models are established for prediction by classifying the data, so that the prediction adaptability is strong, and the accuracy is high.
Drawings
FIG. 1 is a flow chart of a method for predicting water usage in accordance with a first embodiment of the present invention;
FIG. 2 is a flow chart of a method for predicting water consumption in a second embodiment of the present invention;
FIG. 3 is a schematic view of a water consumption predicting apparatus according to a third embodiment of the present invention;
fig. 4 is a schematic diagram of an apparatus according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flow chart of a method for predicting water consumption according to an embodiment of the present invention, which can be applied to a case of predicting water consumption, and the method can be executed by a device or system for predicting water consumption, which can be implemented by software and/or hardware, as shown in fig. 1, and specifically includes the following steps:
and step 110, acquiring historical water consumption data of the target area.
The target area refers to an area for predicting water consumption, and may be a city or other designated area. The historical water usage data includes water usage data or water usage for the target area some time prior to the current time that needs to be predicted. The historical water usage data may also include date and time of day, etc.
Further, the historical water usage data of the set time period of the target area may be acquired. Wherein the set time period may be 1 year, 2 years, 3 years or other time length before the current time.
Specifically, historical water usage data for the target area may be obtained from a water utility company. The historical water use data of the target area can also be obtained from other related units.
And 120, dividing the historical water use data into industrial water use data and domestic water use data according to the change rule of the historical water use data along with time.
The industrial water data comprises water use data used in an industrial production process and water use data of staff life in a factory area, and the life book data comprises water use data used in daily life of human beings in a non-industrial area. The domestic water data can also comprise water data with high randomness, such as municipal water, fire water and the like.
Since the water characteristics of industrial water and domestic water are greatly different, the prediction error of the model is large when the industrial water and the domestic water are mixed for neural network modeling, and therefore, the historical water data is required to be classified to separate the domestic water data from the historical water data. The characteristic water consumption of the industrial water is relatively uniform, and the industrial water has the characteristic of small change along with seasons and weather; the uncertainty of the domestic water is larger, and more factors need to be considered. Therefore, the historical water data is divided into industrial water data and domestic water data, and models are respectively established for calculation.
Optionally, the historical water data can be divided into industrial water data and domestic water data according to the charging standard of the historical water data.
And step 130, respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data.
Optionally, each neural network model mentioned in any embodiment of the present invention includes an input layer, two convolutional layers, a fully-connected layer, and an output layer.
For example, assuming that historical water usage data of the previous two years, or data of the previous 730 days, is obtained, the data of the previous 730-31 days may be selected as a training set for neural network model training, and the data of the previous 30-1 days may be selected as a test set for neural network model testing to determine whether the neural network model is qualified.
The specific training process of the neural network model related to any embodiment of the application is as follows: firstly, initializing values of all parameters of a neural network model, inputting data in a training set into an input layer, transmitting the data to an output layer after passing through a convolutional layer and a full-connection layer to obtain an output value, calculating an error between the output value and a target value (which can be water consumption at the moment of prediction in the training set or a test set), when the error is greater than an expected value, transmitting the error back to the network, sequentially obtaining the error of each layer, updating the parameter values according to the error until the error of all data in the training set is less than or equal to the expected value, finishing training, and determining the neural network model. An upper limit of the training times can also be set, and when the upper limit is larger than the upper limit, the training is ended.
Optionally, respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data, including: respectively determining influence factors of the industrial water data and the domestic water data, and acquiring values of the influence factors; establishing an industrial water prediction neural network model according to the industrial water data and the values of the influence factors thereof; and establishing a domestic water prediction neural network model according to the domestic water data and the values of the influence factors thereof.
Optionally, the influence factors of the industrial water data include season, holidays, time, historical average water consumption and the like; the influence factors of the domestic water data comprise: season, holiday, week, time, temperature, precipitation, historical average water consumption and the like.
Specifically, the influence factors of the industrial water data and the domestic water data can be respectively determined according to the water use characteristics. Taking the domestic water data as an example, the obtained value of the influence factor can be a value of the date and time of the influence factor determined according to the domestic water data; determining the values of the influence factors of the season, the week and the holidays corresponding to the date according to the date; determining the value of the influence factor air temperature according to the values of the influence factor date and time and the information of a meteorological department, wherein the value of the influence factor air temperature can be the highest air temperature or the average air temperature; and the value of the historical average water consumption may be an average of the water consumption of the domestic water data 28 days away from the current time. The industrial water data is similar to the above and is not described in detail herein.
For example, the data of industrial water in 8 months and 8 days in 2018 are shown in table 1, 0-1 in the time in table 1 represents a time period from 0 point to 1 point, and the average water consumption correspondingly represents the average water consumption of the current time period, and the unit is ton. The industrial water data includes dates (8.8.2018), and information such as seasons, holidays, weeks and the like can be obtained according to the dates, specifically summer, wednesday and non-holidays. And then, according to the temperature and weather conditions of 8 months and 8 days in 2018 of the region published by the weather station, the highest temperature, the average temperature, the lowest temperature, rainfall information and the like can be obtained.
Data for industrial water on 8 months and 8 days in 12018 years
Specifically, the above-mentioned respective influence factors may be quantified as input parameters of the neural network model. Illustratively, the date may be a positive integer between 1 and 366, such as a value of 1 for 1/2018 and a value of 32 for 1/2/2018, which refers to the day being the day of the year. The value of holiday may be 0 or 1, with 0 representing no holiday and 1 representing holiday. The week is a positive integer between 1 and 7, the numerical value corresponds to the week represented by the week, the numerical value corresponding to the wednesday is 3, and the numerical value corresponding to the sunday or the sunday is 7. The time is represented by positive integers from 1 to 24 or from 0 to 23, and respectively corresponds to each integer point, if the corresponding value of 7 is 7, the corresponding value of 15 is 15, and the corresponding value of 24 can be 24 or 0. The maximum air temperature and the average air temperature are expressed by decimal numbers, or can be classified according to the temperature range, the grade is expressed, for example, 5 grades can be divided, the integers 1-5 respectively represent the grades 1-5, the temperature range corresponding to the grade 1 with the lowest temperature is the grade 5 represents the temperature range with the highest temperature, specifically, when the temperature is the maximum temperature, 1 represents that the maximum air temperature is lower than 10 ℃, 2 represents that the maximum air temperature is within the range of 10-30 ℃, 3 represents that the maximum air temperature is within the range of 30-35 ℃, 4 represents that the maximum air temperature is within the range of 35-40 ℃, and 5 represents that the maximum air temperature is higher than 40 ℃, of course, the grade can be divided by other numerical values, and the divided grades can be other numbers. The rainfall can be the average rainfall per hour, and is consistent with the calculation interval of the water consumption, of course, in order to reduce the calculation amount, the total rainfall in 24 hours can be used for representing, the decimal number can be used for representing, the grade can be used for representing, for example, when the rainfall is less than 10mm, the rainfall is light rain, and the grade value is 1; if the rainfall is between 10 and 25mm, the rain is medium rain, and the grade value is 2; if the rainfall is between 25 and 50mm, the rain is heavy rain, and the grade value is 3; if the rainfall is 50-100 mm, the rainstorm is the case, and the grade value is 4; if the rainfall is between 100 and 200mm, the rainstorm is heavy, and the grade value is 5; the rainfall is greater than 200mm, the rainstorm is extremely heavy, and the grade value is 6.
Specifically, in building the industrial water prediction neural network model, the values of the various influencing factors of the industrial water prediction neural network model can be determined according to the industrial water data, and then the various influencing factors of the industrial water data of the past 730 days to 31 days, such as the date DIHoliday HITime TIAnd average water consumption WIjWhere j denotes the predicted current time, WIjThe average value of the water consumption 28 days before the current time is shown, the lower corner mark I is shown as an industry related parameter, the industrial water prediction neural network model is input to obtain the predicted water consumption of the past 30 days, and then the test is carried out according to the actual average water consumption of each hour of the past 30 days so as to correct the parameter of the network model. Similarly, in establishing the domestic water prediction neural network model, the values of the respective influence factors can be determined according to the domestic water data, and then the respective influence factors of the domestic water data of the past 730 days to 31 days, such as the date DDHoliday HDTime TDWeek XDMaximum air temperature CDPrecipitation RDAnd average water consumption WDjWhere j denotes the predicted current time, WDjThe average value of the water consumption 28 days before the current time is shown, the lower corner mark D is a life related parameter, the life water prediction neural network model is input to obtain the predicted water consumption of the past 30 days, and then the test is carried out according to the actual average water consumption of each hour of the past 30 days so as to correct the parameter of the network model.
And 140, predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
Specifically, the industrial water consumption of the target area can be predicted according to the industrial water prediction neural network model, the domestic water consumption of the target area can be predicted according to the domestic water prediction neural network model, and the water consumption of the target area can be determined according to the industrial water consumption and the domestic water consumption. Wherein the water consumption can be the sum of domestic water consumption and industrial water consumption.
According to the technical scheme of the embodiment of the invention, the historical water consumption data is obtained and classified according to the change rule of the historical water consumption data into two types of industrial water and domestic water, different neural network models are respectively established for the two types of data, the water consumption is predicted according to the established neural network models, the local historical data is taken as the basic data for prediction, the prediction of the water consumption is more consistent with the actual condition of water consumption, and the different neural network models are established for prediction by classifying the data, so that the prediction adaptability is strong, and the accuracy is high.
Example two
Fig. 2 is a flow chart of a method for predicting water consumption according to a second embodiment of the present invention, which is a further refinement of the previous embodiment, as shown in fig. 2, the method includes the following steps:
and step 210, acquiring historical water consumption data of the target area.
And step 220, dividing the historical water use data into industrial water use data and domestic water use data according to the change rule of the historical water use data along with time.
And step 230, determining influence factors of the industrial water data and the domestic water data respectively, and acquiring values of the influence factors.
And 240, carrying out normalization processing on the industrial water data and the domestic water data to obtain normalized industrial water data and normalized domestic water data.
Wherein, normalization refers to normalizing the values of the water consumption data to a set value range. Normalized industrial water data refers to normalized industrial water data, and normalized domestic water data refers to normalized domestic water data.
Further, the industrial water data and the domestic water data can be normalized according to a preset range to obtain normalized industrial water data and normalized domestic water data.
Because the value ranges of the water consumption data (water consumption) at different times are different, the proportion of the water consumption data in the prediction is different, and the data with smaller value is submerged, so that the prediction result is inaccurate. In order to improve the accuracy of prediction, each influence factor needs to be normalized.
Specifically, in order to ensure the accuracy of the model and prevent the data from being submerged, the value of each water use data (including industrial water use data and domestic water use data) may be normalized to set a value range, such as [0.2,0.8] and [0,1 ]. The specific normalization treatment is as follows:
wherein, a and b are respectively normalized to obtain upper and lower limits of value range, DxThe data of the raw water consumption is the data of the raw water consumption,
for normalized water use data, DminAnd DmaxThe data of the raw water are respectively the minimum value and the maximum value, wherein the data of the raw water can be industrial water data or domestic water data.
And 250, establishing an industrial water prediction neural network model according to the normalized industrial water data and the influence factors of the industrial water data.
Specifically, the neural network model related to any embodiment of the present application includes an industrial water prediction neural network model, including an input layer, two convolution layers, a full connection layer and an output layer, the characteristic size of the input layer is 3 × 3, and the characteristic channel is 1; the feature size of the convolution layer 1 is 3 multiplied by 3, the feature channel is 3, and the convolution kernel is 2 multiplied by 2; the characteristic size of the convolution layer 2 is 3 multiplied by 3, the characteristic channel is 9, and the convolution kernel is 2 multiplied by 2; the characteristic size of the full connection layer is 1 multiplied by 27, the characteristic channel is 1, and the convolution kernel is 54 multiplied by 1; the output layer characteristic size is 1.
And establishing an industrial water prediction neural network model according to the normalized industrial water data and the influence factors of the industrial water data, wherein the steps are basically consistent with the steps for establishing the industrial water prediction neural network model according to the industrial water data and the influence factors of the industrial water data, and only the input industrial water data which is not subjected to normalization processing is replaced by the normalized industrial water data.
Optionally, establishing an industrial water prediction neural network model according to the normalized industrial water data and the influence factor of the industrial water data, including:
performing cluster analysis on the normalized industrial water data according to water use characteristics to obtain each industrial water subset; and establishing a sub-industrial water prediction neural network model corresponding to each industrial water subset according to the industrial water subsets and the influence factors of the industrial water data, wherein the industrial water prediction neural network model consists of each sub-industrial water prediction neural network model.
And step 260, performing cluster analysis on the normalized domestic water data according to water characteristics to obtain each domestic water subset.
Wherein, the water consumption characteristics comprise the distribution rule of the water consumption of users. Clustering analysis is an analytical process that classifies data.
Due to the fact that water consumption of users with domestic water data is different greatly, such as single-living office workers, large families in three-generation classrooms and the like, water consumption of different users is different. Industrial water also varies with the size and nature of the industry. Therefore, the water use data needs to be classified to classify the domestic water data of users with similar or similar water use characteristics into one class, so as to obtain each domestic water subset, to refine the classification of the data, and to improve the accuracy of prediction. The steps related to clustering performed on the domestic water data are also applied to the industrial water data, and the domestic water is described as an example below.
Optionally, the performing cluster analysis on the domestic water data according to water characteristics includes:
and performing cluster analysis on the domestic water data according to the domestic water data and the correlation coefficient of the influence factors thereof.
Optionally, the expression of the correlation coefficient is:
where ρ represents the correlation coefficient, QtThe water consumption at the moment t represents the current moment, the initial value of t is 1, the upper limit value is n, n represents the upper limit time for calculating the correlation coefficient, QavgAverage value of water consumption for a set time period before time t, etIs the value of the impact factor of the domestic water data at time t, eavgAnd averaging the values of the influence factors of the domestic water data in a set time period before the time t.
And 270, establishing a sub domestic water prediction neural network model corresponding to each domestic water subset according to the domestic water subsets and the influence factors of the domestic water data.
The domestic water prediction neural network model is composed of sub domestic water prediction neural network models.
Step 260 and step 270 may be executed first, and then step 250 may be executed.
Specifically, establishing a sub domestic water prediction neural network model corresponding to each domestic water subset according to the domestic water subsets and the influence factors of the domestic water data includes:
generating an initial input matrix according to the domestic water subset and the influence factors thereof, wherein the number of columns of the initial input matrix is the same as the number of the influence factors of the domestic water data; performing zero padding operation on the initial input matrix according to the column number to obtain an intermediate input matrix, wherein the column number of the intermediate input matrix is the square of a set numerical value; and establishing a sub domestic water prediction neural network model corresponding to each domestic water subset according to the intermediate input matrix.
For example, assuming that the domestic water data is the average water consumption of the residents in the target area per hour through step 270, the domestic water data may be classified into Y categories, each including D groups of data, wherein the number of data corresponding to different categories may be different or the same. Then, the size of the domestic water data corresponding to the sub-domestic water prediction neural network model is (730 × 24 × D) × 7, where 730 denotes that water consumption of the past 730 days is recorded, 24 denotes that average water consumption per hour per day needs to be recorded, D denotes that domestic water data of D users are included in the group, and 7 denotes that 7 influence factors (season, holiday, week, time, air temperature, precipitation, and historical average water consumption) correspond to each water consumption. In order to facilitate calculation of the neural network model, row-column zero padding operation is performed on the data, and 2 columns of zero vectors are padded at the end of the original 7 columns of data, so that the size of the data is (730 × 24 × D) × 9. Each row vector of the data is matrixed, that is, the original 1 × 9 row vector is converted into a matrix form of 3 × 3, thereby obtaining a sample matrix (size (730 × 24 × D) × 3 × 3). And (3) performing dimensionality improvement on the data to serve as an input of the sub-domestic water prediction neural network model: the sample matrix (730 × 24 × D) × 3 × 3 liters is dimensioned in a four-dimensional matrix form having a size of (730 × 24 × D) × 3 × 3 × 1, where 1 represents a group to which the current data belongs, i.e., a value of Y. And for the data after the dimensionality is increased, selecting data which is 730-31 days away from the current forecast day as a training set, and selecting data which is 30-1 days away from the current forecast day as a test set. And determining parameters of the sub-domestic water prediction neural network model by the neural network model training, testing and other methods.
And step 280, predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
Specifically, the predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model comprises the following steps:
predicting the normalized industrial water consumption of the target area according to the industrial water prediction neural network model; predicting the normalized domestic water consumption of the target area according to a domestic water prediction neural network model; and predicting the water consumption of the target area according to the normalized industrial water consumption and the normalized domestic water consumption.
Optionally, predicting the water usage of the target area based on the normalized industrial water usage and the normalized domestic water usage comprises:
determining the industrial water consumption according to the normalized industrial water consumption and the inverse process of the normalized processing of the industrial water data; determining the domestic water consumption according to the normalized domestic water consumption and the inverse process of the normalization processing of the domestic water data; and determining the sum of the industrial water consumption and the domestic water consumption as the predicted water consumption of the target area.
Specifically, since the water consumption data input to the model is normalized, the data output from the neural network model needs to be subjected to inverse normalization processing to restore the real value of the water consumption.
From the normalization of the processed content in step 240, it can be known that the formula used for normalization is:
correspondingly, the normalization inverse process is as follows:
wherein, a and b are respectively normalized to obtain upper and lower limits of value range, DminAnd DmaxRespectively the minimum value and the maximum value of raw water data, wherein the raw water data can be industrial water data or domestic water data,
use of predictions for neural network modelsAmount of water, DyAnd (5) normalizing the water use data after the reverse treatment.
Further, the predicted water consumption can be displayed in a set form according to the predicted water consumption of the target area.
Specifically, the prediction result of the water consumption of a single user can be displayed, the prediction result of the water consumption of a set area can also be displayed, and the prediction trend graph of the water consumption of the area every day can also be displayed by longitudinal superposition. And a rolling mode can be adopted, and the water consumption trend in the set time in the future is predicted by using historical data.
Further, still include: and updating the historical water consumption data, and repeatedly executing the steps 220-280 to update the neural network model so as to improve the accuracy and the real-time performance of the model.
According to the technical scheme of the embodiment of the invention, the data is prevented from being submerged by carrying out normalization processing on the data, and the prediction precision is effectively ensured; the domestic water data are further divided, data classification is refined, and a plurality of sub-domestic water prediction neural network models are generated, so that the training of the models is more in line with the characteristics of users of the type, and the prediction accuracy of the models is improved.
EXAMPLE III
Fig. 3 is a schematic diagram of a water consumption predicting apparatus according to a third embodiment of the present invention, as shown in fig. 3, the apparatus includes: a historical information acquisition module 310, an information partitioning module 320, a neural network model building module 330, and a water usage prediction module 340.
The historical information acquiring module 310 is configured to acquire historical water consumption data of a target area; the information dividing module 320 is used for dividing the historical water data into industrial water data and domestic water data according to the change rule of the historical water data along with time; the neural network model establishing module 330 is configured to respectively establish an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data; and the water consumption prediction module 340 is used for predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
According to the technical scheme of the embodiment of the invention, the historical water consumption data is obtained and classified according to the change rule of the historical water consumption data into two types of industrial water and domestic water, different neural network models are respectively established for the two types of data, the water consumption is predicted according to the established neural network models, the local historical data is taken as the basic data for prediction, the prediction of the water consumption is more consistent with the actual condition of water consumption, and the different neural network models are established for prediction by classifying the data, so that the prediction adaptability is strong, and the accuracy is high.
Optionally, the neural network model building module 330 includes:
the influence factor determining unit is used for respectively determining the influence factors of the industrial water data and the domestic water data and acquiring the values of the influence factors; the industrial model establishing unit is used for establishing an industrial water prediction neural network model according to the industrial water data and the values of the influence factors thereof; and the life model establishing unit is used for establishing a life water prediction neural network model according to the life water data and the value of the influence factor thereof.
Optionally, the device for predicting water consumption further comprises:
and the normalization processing module is used for performing normalization processing on the industrial water data and the domestic water data before respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data.
Optionally, the living model building unit includes:
the cluster analysis subunit is used for carrying out cluster analysis on the domestic water data according to the water characteristics so as to obtain each domestic water subset; and the life model establishing subunit is used for establishing a sub-life water prediction neural network model corresponding to each life water subset according to the life water subsets and the influence factors of the life water data, wherein the life water prediction neural network model consists of each sub-life water prediction neural network model.
Optionally, the cluster analysis subunit is specifically configured to:
and performing cluster analysis on the domestic water data according to the domestic water data and the correlation coefficient of the influence factors thereof.
Wherein, the expression of the correlation coefficient is:
where ρ represents the correlation coefficient, QtThe water consumption at the moment t represents the current moment, the initial value of t is 1, the upper limit value is n, n represents the upper limit time for calculating the correlation coefficient, QavgAverage value of water consumption for a set time period before time t, etIs the value of the impact factor of the domestic water data at time t, eavgAnd averaging the values of the influence factors of the domestic water data in a set time period before the time t.
The water consumption prediction device provided by the embodiment of the invention can execute the water consumption prediction method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 4 is a schematic structural diagram of a water consumption prediction apparatus according to a fourth embodiment of the present invention, as shown in fig. 4, the apparatus includes a processor 410, a memory 420, an input device 430, and an output device 440; the number of the device processors 410 may be one or more, and one processor 410 is taken as an example in fig. 4; the processor 410, the memory 420, the input device 430 and the output device 440 in the apparatus may be connected by a bus or other means, for example, in fig. 4.
The memory 420 serves as a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the method for predicting water consumption according to the embodiment of the present invention (for example, the historical information acquisition module 410, the information division module 420, the neural network model creation module 430, and the water consumption prediction module 440 in the water consumption prediction apparatus). The processor 410 executes software programs, instructions and modules stored in the memory 420 to perform various functional applications of the device and data processing, i.e., to implement the above-described water consumption prediction method.
The memory 420 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 420 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 420 may further include memory located remotely from the processor 410, which may be connected to the device/terminal/server via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input means 430 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function control of the apparatus. The output device 440 may include a display device such as a display screen.
EXAMPLE five
Further provided in accordance with an embodiment of the present invention is a storage medium containing computer-executable instructions which, when executed by a computer processor, perform a method of predicting water usage, the method comprising:
acquiring historical water consumption data of a target area;
dividing historical water data into industrial water data and domestic water data according to the change rule of the historical water data along with time;
respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data;
and predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
Of course, the embodiments of the present invention provide a storage medium containing computer-executable instructions, which are not limited to the operations of the method described above, but can also perform related operations in the method for predicting water consumption provided by any embodiments of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the water consumption prediction device, the units and modules included in the embodiment are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method for predicting water usage, comprising:
acquiring historical water consumption data of a target area;
dividing historical water data into industrial water data and domestic water data according to the change rule of the historical water data along with time;
respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data;
and predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
2. The method of claim 1, wherein building an industrial water prediction neural network model and a domestic water prediction neural network model from the industrial water data and the domestic water data, respectively, comprises:
respectively determining influence factors of the industrial water data and the domestic water data, and acquiring values of the influence factors;
establishing an industrial water prediction neural network model according to the industrial water data and the values of the influence factors thereof;
and establishing a domestic water prediction neural network model according to the domestic water data and the values of the influence factors thereof.
3. The method of claim 2, wherein the impact factors of the industrial water data include at least one of season, holidays, time of day, and historical average water usage; the influence factors of the domestic water data comprise: at least one of season, holiday, week, time of day, air temperature, precipitation and historical average water usage.
4. The method of claim 1, further comprising, prior to building an industrial water prediction neural network model and a domestic water prediction neural network model from the industrial water data and the domestic water data, respectively:
and carrying out normalization processing on the industrial water data and the domestic water data.
5. The method of claim 2, wherein the building of the domestic water prediction neural network model from the domestic water data and the values of its impact factors comprises:
performing cluster analysis on the domestic water data according to water characteristics to obtain each domestic water subset;
and establishing a sub domestic water prediction neural network model corresponding to each domestic water subset according to the domestic water subsets and the influence factors of the domestic water data, wherein the domestic water prediction neural network model consists of the sub domestic water prediction neural network models.
6. The method of claim 5, wherein the performing cluster analysis on the domestic water data according to water usage characteristics comprises:
and performing cluster analysis on the domestic water data according to the domestic water data and the correlation coefficient of the influence factors thereof.
7. The method of claim 6, wherein the correlation coefficient is expressed by:
where ρ represents the correlation coefficient, QtThe water consumption at the moment t represents the current moment, the initial value of t is 1, the upper limit value is n, n represents the upper limit time for calculating the correlation coefficient, QavgAverage value of water consumption for a set time period before time t, etIs the value of the impact factor of the domestic water data at time t, eavgAnd averaging the values of the influence factors of the domestic water data in a set time period before the time t.
8. The method of claim 2, wherein building an industrial water prediction neural network model from the industrial water data and values of its impact factors comprises:
performing cluster analysis on the industrial water data according to water use characteristics to obtain each industrial water subset;
and establishing a sub-industrial water prediction neural network model corresponding to each industrial water subset according to the industrial water subsets and the influence factors of the industrial water data, wherein the industrial water prediction neural network model consists of each sub-industrial water prediction neural network model.
9. An apparatus for predicting water consumption, comprising:
the historical information acquisition module is used for acquiring historical water consumption data of the target area;
the information dividing module is used for dividing the historical water use data into industrial water use data and domestic water use data according to the change rule of the historical water use data along with time;
the neural network model establishing module is used for respectively establishing an industrial water prediction neural network model and a domestic water prediction neural network model according to the industrial water data and the domestic water data;
and the water consumption prediction module is used for predicting the water consumption of the target area according to the industrial water prediction neural network model and the domestic water prediction neural network model.
10. A water usage prediction apparatus, the apparatus comprising:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of predicting water usage according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911037989.2A CN110751416A (en) | 2019-10-29 | 2019-10-29 | Method, device and equipment for predicting water consumption |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911037989.2A CN110751416A (en) | 2019-10-29 | 2019-10-29 | Method, device and equipment for predicting water consumption |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110751416A true CN110751416A (en) | 2020-02-04 |
Family
ID=69280823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911037989.2A Pending CN110751416A (en) | 2019-10-29 | 2019-10-29 | Method, device and equipment for predicting water consumption |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110751416A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111210093A (en) * | 2020-03-05 | 2020-05-29 | 重庆森鑫炬科技有限公司 | Daily water consumption prediction method based on big data |
| CN111415192A (en) * | 2020-02-27 | 2020-07-14 | 重庆森鑫炬科技有限公司 | Water quality prediction method for user based on big data |
| CN111476626A (en) * | 2020-03-04 | 2020-07-31 | 珠海市百岛科技有限公司 | Intelligent water ordering method and intelligent water station system |
| CN112696344A (en) * | 2020-12-23 | 2021-04-23 | 上海熊猫机械(集团)有限公司 | Intelligent control method for water supply booster pump station |
| CN112799156A (en) * | 2021-03-30 | 2021-05-14 | 中数通信息有限公司 | Meteorological integrated emergency early warning issuing method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104715292A (en) * | 2015-03-27 | 2015-06-17 | 上海交通大学 | City short-term water consumption prediction method based on least square support vector machine model |
| CN107977735A (en) * | 2017-11-15 | 2018-05-01 | 河海大学 | A kind of municipal daily water consumption Forecasting Methodology based on deep learning |
| CN108764473A (en) * | 2018-05-23 | 2018-11-06 | 河北工程大学 | A kind of BP neural network water demands forecasting method based on correlation analysis |
-
2019
- 2019-10-29 CN CN201911037989.2A patent/CN110751416A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104715292A (en) * | 2015-03-27 | 2015-06-17 | 上海交通大学 | City short-term water consumption prediction method based on least square support vector machine model |
| CN107977735A (en) * | 2017-11-15 | 2018-05-01 | 河海大学 | A kind of municipal daily water consumption Forecasting Methodology based on deep learning |
| CN108764473A (en) * | 2018-05-23 | 2018-11-06 | 河北工程大学 | A kind of BP neural network water demands forecasting method based on correlation analysis |
Non-Patent Citations (4)
| Title |
|---|
| 张喜才: "《互联网时代农村流通全产业链整合发展》", 30 September 2017, 中国商业出版社 * |
| 张志果等: "基于恩格尔系数与霍夫曼系数的城市需水量预测", 《水利学报》 * |
| 李志敏: "《垃圾邮件识别与处理技术研究》", 31 December 2015, 北京理工大学出版社 * |
| 荀港益: "基于聚类分析与随机森林的短期负荷滚动预测", 《智能城市》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111415192A (en) * | 2020-02-27 | 2020-07-14 | 重庆森鑫炬科技有限公司 | Water quality prediction method for user based on big data |
| CN111476626A (en) * | 2020-03-04 | 2020-07-31 | 珠海市百岛科技有限公司 | Intelligent water ordering method and intelligent water station system |
| CN111210093A (en) * | 2020-03-05 | 2020-05-29 | 重庆森鑫炬科技有限公司 | Daily water consumption prediction method based on big data |
| CN112696344A (en) * | 2020-12-23 | 2021-04-23 | 上海熊猫机械(集团)有限公司 | Intelligent control method for water supply booster pump station |
| CN112696344B (en) * | 2020-12-23 | 2022-09-02 | 上海熊猫机械(集团)有限公司 | Intelligent control method for water supply booster pump station |
| CN112799156A (en) * | 2021-03-30 | 2021-05-14 | 中数通信息有限公司 | Meteorological integrated emergency early warning issuing method and device |
| CN112799156B (en) * | 2021-03-30 | 2021-07-20 | 中数通信息有限公司 | Meteorological integrated emergency early warning issuing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110751416A (en) | Method, device and equipment for predicting water consumption | |
| CN111552923B (en) | Load prediction method and load prediction system based on universal distribution | |
| Antuchevičiene et al. | Multiple criteria construction management decisions considering relations between criteria | |
| CN110751326B (en) | Photovoltaic day-ahead power prediction method and device and storage medium | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| CN112734135B (en) | Power load prediction method, intelligent terminal and computer readable storage medium | |
| CN112330077B (en) | Power load prediction method, power load prediction device, computer equipment and storage medium | |
| CN114330935B (en) | New energy power prediction method and system based on multiple combination strategies integrated learning | |
| CN114493052B (en) | Multi-model fusion self-adaptive new energy power prediction method and system | |
| CN110796307A (en) | Distributed load prediction method and system for comprehensive energy system | |
| CN114372360A (en) | Method, terminal and storage medium for power load prediction | |
| CN114648155A (en) | Source analysis method and emergency response system based on weather typing and weather forecast | |
| CN106600029A (en) | Macro-economy predictive quantization correction method based on electric power data | |
| CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
| CN116957356A (en) | Scenic spot carbon neutralization management method and system based on big data | |
| CN114676931B (en) | Electric quantity prediction system based on data center technology | |
| CN115187134A (en) | Grid-based power distribution network planning method and device and terminal equipment | |
| CN116050636A (en) | Output prediction method, device, equipment and medium of photovoltaic power station | |
| CN115238854A (en) | Short-term load prediction method based on TCN-LSTM-AM | |
| CN111311025B (en) | Load prediction method based on meteorological similar days | |
| Hong et al. | The study on affecting factors of regional marine industrial structure upgrading | |
| CN110929219A (en) | Distribution network equipment group state evaluation method, device and system and computer storage medium | |
| CN110991083A (en) | Photovoltaic power station model determination method, device, equipment and storage medium | |
| Gaing et al. | Optimal grey topological predicting approach to short-term load forecasting in power system | |
| Jyothsna et al. | Analysis and Prediction of Electricity Load Forecasting |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 206-002, 2 / F, building 8, Xixi bafangcheng, Wuchang Street, Yuhang District, Hangzhou City, Zhejiang Province Applicant after: HANGZHOU RUHR TECHNOLOGY Co.,Ltd. Address before: A4-4-201, No. 643, Shuangliu, Zhuantang street, Xihu District, Hangzhou City, Zhejiang Province Applicant before: HANGZHOU RUHR TECHNOLOGY Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200204 |