CN112686469A - Family population number prediction method based on electric power big data - Google Patents
Family population number prediction method based on electric power big data Download PDFInfo
- Publication number
- CN112686469A CN112686469A CN202110049155.4A CN202110049155A CN112686469A CN 112686469 A CN112686469 A CN 112686469A CN 202110049155 A CN202110049155 A CN 202110049155A CN 112686469 A CN112686469 A CN 112686469A
- Authority
- CN
- China
- Prior art keywords
- family
- users
- population
- electricity consumption
- family users
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A family population number prediction method based on electric power big data belongs to the technical field of big data analysis. The invention aims to solve the problem that the existing method for predicting the family population depends on the average power consumption of people, and the average power consumption of people is difficult to obtain reliably. The method comprises the following steps: drawing a normal distribution curve; determining a power consumption interval according to the user proportion of the household users with different population numbers in all the household users; then acquiring population probability density and power consumption curves of the family users with different population; the method comprises the steps of collecting daily electricity consumption in a household target period to be predicted, obtaining a current mean value and a current variance, determining power consumption intervals corresponding to curves of different population quantity probability densities and power consumption by taking a value obtained by subtracting the current variance from the current mean value as a minimum value and a value obtained by adding the current mean value and the current variance as a maximum value, and taking the population number of a household user corresponding to the curve when the area of the electricity consumption interval is maximum as a prediction result. The method and the device realize the prediction of the number of the family population under the condition of unknown electricity consumption per capita.
Description
Technical Field
The invention relates to a family population number prediction method based on electric power big data, and belongs to the technical field of big data analysis.
Background
At present, the prediction of family population based on electric power data mainly depends on the average electricity consumption of people in an area, and the number of the family population is predicted through the total household electricity consumption and the average electricity consumption of people. However, the average power consumption of people in the area is extremely difficult to obtain, the result is unreliable, and in the case of no average power consumption of people in the area, the prediction of family population cannot be realized, so that the method is not suitable for practical situations.
At present, household electricity consumption data in an area obtained by a machine learning method does not have description on the population number, so that training requirements are not met, for example, if the family population number does not correspond to the accuracy, algorithm training cannot be carried out, and prediction of the population number cannot be achieved.
Disclosure of Invention
The invention provides a family population prediction method based on large electric power data, aiming at the problem that the existing family population prediction method depends on the average power consumption of people, and the average power consumption of people is difficult to obtain reliably.
The invention relates to a family population number prediction method based on electric power big data, which comprises the following steps,
collecting daily electricity consumption of all home users in a monitoring area in a preset period; drawing a normal distribution curve by taking the electricity consumption as a horizontal coordinate and the frequency of the electricity consumption corresponding to the appearance as a vertical coordinate;
based on population survey data, acquiring the user proportion of the family users with different population numbers in all the family users;
determining power consumption intervals of household users with different population numbers corresponding to the normal distribution curve according to the user proportion;
obtaining the mean value and the variance of a corresponding section of a normal distribution curve according to the electricity consumption interval, and then calculating by a normal distribution formula to obtain population number probability density and electricity consumption curves of household users with different population numbers;
the method comprises the steps of collecting daily electricity consumption in a target period of a family to be predicted, calculating to obtain a current mean value and a current variance of the daily electricity consumption of the family to be predicted, using the value obtained by subtracting the current variance from the current mean value as a minimum value, using the value obtained by adding the current variance to the current mean value as a maximum value, determining a corresponding electricity consumption interval of the family to be predicted in a family population probability density and electricity consumption curve of family users with different population numbers, and taking the population number probability density of the family with the largest electricity consumption interval and the population number of the family users corresponding to the electricity consumption curve as a prediction result of the population number of the family to be.
According to the household population prediction method based on the electric power big data, the abscissa of the electric power consumption corresponding to the normal distribution curve takes 0.5 degrees as an interval.
According to the household population prediction method based on the electric power big data, the household users with different population numbers comprise one household user, two household users, three household users, four household users, five household users and six or more household users;
the user proportion comprises the user proportion alpha of one family user to all the family users1The user ratio alpha of two-person family users to all family users2The user ratio alpha of three family users to all the family users3The user ratio alpha of four-person family users to all family users4The user ratio alpha of five family users to all the family users5The user ratio alpha of six or more family users to all the family users6。
According to the household population prediction method based on the electric power big data, the step of determining the power consumption intervals of the household users with different population numbers corresponding to the normal distribution curve according to the user proportion comprises the following steps:
in the projection area of the normal distribution curve and the coordinate axis, the proportion of the total consumption to the total power consumption is alpha1、α2、α3、α4、α5And alpha6Determining the power consumption intervals of one-person family users, two-person family users, three-person family users, four-person family users, five-person family users and six or more family users.
The invention has the beneficial effects that: the invention is based on a mathematical statistics method, and the family population number is predicted according to the family power consumption, thereby realizing the prediction of the family population number under the condition of unknown average power consumption; the invention does not need to adopt a machine learning mode, and has no early sample training process, so the requirement on the data quality is lower; the process of manually screening the training samples is avoided, the labor cost is reduced, and the objectivity of the prediction result is improved.
When population prediction is carried out, the electricity consumption data in the time period closest to the current query time can be selected for prediction, and the iteration of the prediction result is quick and timely.
Drawings
FIG. 1 is a normal distribution curve obtained by the present invention by using the amount of electricity and the frequency of occurrence of electricity in correspondence with the amount of electricity used;
FIG. 2 is a power consumption interval of a normal distribution curve determined according to a user ratio of household users with different population numbers to all household users;
FIG. 3 is a graph of population probability density versus power usage for a two-person family user;
FIG. 4 is a graph of population probability density versus power usage for three home users;
FIG. 5 is a graph of population probability density versus power usage for four family users;
FIG. 6 is a power consumption interval corresponding to the graph shown in FIG. 3 for the household power consumption to be predicted;
fig. 7 is a power consumption interval corresponding to the graph shown in fig. 4 for the household power consumption to be predicted;
fig. 8 is a power consumption interval corresponding to the graph shown in fig. 5 for the household power consumption to be predicted;
FIG. 9 is a graph of the population trend of a certain cell over time versus the total electricity consumption obtained by population prediction using the present invention; the number of population on the abscissa and the amount of electricity consumed on the ordinate are shown in the figure.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
First embodiment, as shown in fig. 1 to 8, the present invention provides a method for predicting a household population based on big power data, including,
collecting daily electricity consumption of all home users in a monitoring area in a preset period; drawing a normal distribution curve by taking the electricity consumption as a horizontal coordinate and the frequency of the electricity consumption corresponding to the appearance as a vertical coordinate;
based on population survey data, acquiring the user proportion of the family users with different population numbers in all the family users;
determining power consumption intervals of household users with different population numbers corresponding to the normal distribution curve according to the user proportion;
obtaining the mean value and the variance of a corresponding section of a normal distribution curve according to the electricity consumption interval, and then calculating by a normal distribution formula to obtain population number probability density and electricity consumption curves of household users with different population numbers;
the method comprises the steps of collecting daily electricity consumption in a target period of a family to be predicted, calculating to obtain a current mean value and a current variance of the daily electricity consumption of the family to be predicted, using the value obtained by subtracting the current variance from the current mean value as a minimum value, using the value obtained by adding the current variance to the current mean value as a maximum value, determining a corresponding electricity consumption interval of the family to be predicted in a family population probability density and electricity consumption curve of family users with different population numbers, and taking the population number probability density of the family with the largest electricity consumption interval and the population number of the family users corresponding to the electricity consumption curve as a prediction result of the population number of the family to be.
As an example, the normal distribution curve corresponds to the abscissa of the power consumption at intervals of 0.5 degrees.
The embodiment first performs frequency-based household electricity consumption statistics:
the preset period may be selected to be one month. Counting the number of the daily electricity consumption of all the family users in the monitoring area in the last month, for example, the daily electricity consumption of 1000 families in the monitoring area in one month is counted to obtain 30000 shares of electricity consumption data. In 30000 parts of electricity consumption data, the minimum value is greater than or equal to 0 degree, the maximum value is not limited, and the number of occurrences of 0 degree, the number of occurrences of 0.5 degree, … …, the number of occurrences of the maximum value and the like are counted. The number of occurrences is used as the ordinate and the electricity consumption is used as the abscissa, and a curve similar to normal distribution is generated.
As an example, the family users with different population numbers include one-family users, two-family users, three-family users, four-family users, five-family users, and six or more family users;
the user proportion comprises the user proportion alpha of one family user to all the family users1The user ratio alpha of two-person family users to all family users2Three-person family user accountUser ratio alpha of all home users3The user ratio alpha of four-person family users to all family users4The user ratio alpha of five family users to all the family users5The user ratio alpha of six or more family users to all the family users6。
According to the population survey data, the proportion of the family users with different population numbers in the monitored area to all the family users can be obtained. For example, from the results of a survey of family population in a certain area, the structural distribution of family population shown in table 1 can be obtained:
TABLE 1
| Number of family members | In proportion to all families | Cumulative percentage of |
| 1 | 12.88％ | 12.88％ |
| 2 | 28.86％ | 41.74％ |
| 3 | 36.5％ | 78.24％ |
| 4 | 12.58％ | 90.82％ |
| 5 | 7.44％ | 98.26％ |
| >＝6 | 3.12％ | 100％ |
Further, as shown in fig. 2 to 8, the determining, according to the user ratio, the power consumption intervals of the normal distribution curves corresponding to the household users with different population numbers includes:
in the projection area of the normal distribution curve and the coordinate axis, the proportion of the total consumption to the total power consumption is alpha1、α2、α3、α4、α5And alpha6Determining the power consumption intervals of one-person family users, two-person family users, three-person family users, four-person family users, five-person family users and six or more family users.
Taking the area obtained by projection of a normal distribution curve and a coordinate axis as the total power consumption, starting from the original point, and intercepting alpha of the total power consumption1The area of the proportion is used as the electricity consumption interval of one family user, and alpha is obtained in sequence2、α3、α4、α5And alpha6Area of proportion. Thereby determining the electricity consumption intervals of different population household users.
After obtaining the mean value and the variance of the corresponding section of the normal distribution curve, substituting the mean value and the variance into a normal distribution formula; population probability density and power consumption curves corresponding to one family user, two family users, three family users, four family users, five family users and six or more family users can be obtained respectively.
Finally, for the household to be predicted, the target period may be selected to be one week. And collecting daily electric quantity of the family to be predicted in weeks. The collection of the power consumption can be data collection at a fixed time every day. And calculating the mean value and the variance of the power consumption in the period of time according to the acquisition result to obtain a power consumption interval of [ mean-variance, mean + variance ]. And comparing the probability density of the number of the six populations with the area of the electricity consumption interval on the electricity consumption curve to obtain a prediction result of the number of the family population to be predicted. For example, if the curve having the largest curve projection area of the family to be predicted is a curve having a population of 2, the population of the family is predicted to be 2.
The following further details the implementation of the present invention:
1) obtaining a power consumption statistical curve, namely the normal distribution curve, as shown in fig. 1;
2) according to the family population survey data of the monitored area, the family structure distribution is shown in table 2:
TABLE 2
| Number of family members | In proportion to the total family | Cumulative percentage of |
| 1 | α1 | α1 |
| 2 | α2 | α1+α2 |
| 3 | α3 | α1+α2+α3 |
| 4 | α4 | α1+α2+α3+α4 |
| 5 | α5 | α1+α2+α3+α4+α5 |
| >＝6 | α6 | α1+α2+α3+α4+α5+α6 |
Wherein alpha isi∈[0,1]，
According to the proportional data shown in table 2, the electricity consumption intervals corresponding to the household users with different population numbers are obtained, as shown in fig. 2. Corresponding to X1、X2、X3、X4、X5The boundary value of the electricity consumption corresponding to each electricity consumption interval;
3) calculating the power consumption average value mu corresponding to the family users with obviously different population numbers according to the power consumption interval1And variance σ2The probability density-power consumption curve of the corresponding household population is obtained by the mean and the variance, and is a household power consumption statistical curve with the household population numbers of 2, 3 and 4 as shown in fig. 3 to 5.
4) In the process of predicting the population of the family, the family to be predicted is counted by the daily electricity consumption of the last month or a week to obtain the average value mu of the daily electricity consumption2And variance θ2Will [ mu ] of2-θ2，μ2+θ2]Carry over to fig. 3 to 5In the illustrated probability density-power consumption curve, the projected area of the curve on the coordinate axis is the largest, and the probability representing the number of family population is the largest, thereby determining the prediction result, as shown in fig. 6 to 8. Comparing the shaded areas of fig. 6 to fig. 8, it is determined that the shaded area in fig. 6 is the largest, and thus the prediction result is that the number of the family population to be predicted is 2.
And (3) experimental verification: and constructing an incidence relation between the residential electricity consumption and the population number of the residential based on the historical data, and analyzing the historical electricity consumption trend. The number of household population of residents in a certain cell 3457 is determined according to the electricity consumption, and a relation curve of population change trend of the whole cell along with time and the total electricity consumption is obtained firstly, as shown in fig. 9.
The results of the family population prediction using the method of the present invention are shown in table 3:
TABLE 3
The family demographics obtained from census data are shown in table 4:
TABLE 4
The results of tables 3 and 4 were calculated to obtain the following accuracy of the present invention:
TABLE 5
| Prediction | Practice of | Rate of accuracy | Actual ratio of occupation | Actual weight | ||
| Vacant | 331 | 350 | 94.57％ | 10.1 | 0.101 | 9.55％ |
| 1 mouth | 546 | 520 | 95.00％ | 15.1 | 0.151 | 14.35％ |
| 2 mouth | 1378 | 1330 | 96.39％ | 38.5 | 0.385 | 37.11％ |
| 3 mouth | 982 | 1003 | 97.91％ | 29 | 0.29 | 28.39％ |
| 4-port | 181 | 209 | 86.60％ | 6 | 0.06 | 5.20％ |
| Social residence | 39 | 45 | 86.67％ | 1.3 | 0.013 | 1.13％ |
As can be seen from Table 5, the data prediction accuracy of the invention is 94.57% for the unoccupied residents, 95% for the number of households in one mouthful, 96.39% for the number of households in 2 mouths, 97.91% for the number of households in 3 mouths, 86.60% for the number of households in 4 mouths and 86.67% for the number of the grouped residents; the total accuracy rate can reach 95.72%. Therefore, the invention has great practical value.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that features described in different dependent claims and herein may be combined in ways different from those described in the original claims. It is also to be understood that features described in connection with individual embodiments may be used in other described embodiments.
Claims (4)
1. A family population number prediction method based on electric power big data is characterized by comprising the following steps,
collecting daily electricity consumption of all home users in a monitoring area in a preset period; drawing a normal distribution curve by taking the electricity consumption as a horizontal coordinate and the frequency of the electricity consumption corresponding to the appearance as a vertical coordinate;
based on population survey data, acquiring the user proportion of the family users with different population numbers in all the family users;
determining power consumption intervals of household users with different population numbers corresponding to the normal distribution curve according to the user proportion;
obtaining the mean value and the variance of a corresponding section of a normal distribution curve according to the electricity consumption interval, and then calculating by a normal distribution formula to obtain population number probability density and electricity consumption curves of household users with different population numbers;
the method comprises the steps of collecting daily electricity consumption in a target period of a family to be predicted, calculating to obtain a current mean value and a current variance of the daily electricity consumption of the family to be predicted, using the value obtained by subtracting the current variance from the current mean value as a minimum value, using the value obtained by adding the current variance to the current mean value as a maximum value, determining a corresponding electricity consumption interval of the family to be predicted in a family population probability density and electricity consumption curve of family users with different population numbers, and taking the population number probability density of the family with the largest electricity consumption interval and the population number of the family users corresponding to the electricity consumption curve as a prediction result of the population number of the family to be.
2. The method of claim 1, wherein the power big data-based family population prediction method,
and the abscissa of the power consumption corresponding to the normal distribution curve takes 0.5 degrees as an interval.
3. The method for predicting the population of households based on the power big data according to claim 1 or 2,
the family users with different population numbers comprise one family user, two family users, three family users, four family users, five family users and six or more family users;
the user proportion comprises the user proportion alpha of one family user to all the family users1The user ratio alpha of two-person family users to all family users2The user ratio alpha of three family users to all the family users3The user ratio alpha of four-person family users to all family users4The user ratio alpha of five family users to all the family users5The user ratio alpha of six or more family users to all the family users6。
4. The method of claim 3, wherein the power big data-based family population prediction method,
the step of determining the power consumption intervals of the family users with different population numbers corresponding to the normal distribution curve according to the user proportion comprises the following steps:
in the projection area of the normal distribution curve and the coordinate axis, the proportion of the total consumption to the total power consumption is alpha1、α2、α3、α4、α5And alpha6Determining the power consumption intervals of one-person family users, two-person family users, three-person family users, four-person family users, five-person family users and six or more family users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110049155.4A CN112686469A (en) | 2021-01-14 | 2021-01-14 | Family population number prediction method based on electric power big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110049155.4A CN112686469A (en) | 2021-01-14 | 2021-01-14 | Family population number prediction method based on electric power big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112686469A true CN112686469A (en) | 2021-04-20 |
Family
ID=75457949
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110049155.4A Pending CN112686469A (en) | 2021-01-14 | 2021-01-14 | Family population number prediction method based on electric power big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112686469A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116029480A (en) * | 2023-03-28 | 2023-04-28 | 广东电网有限责任公司 | Proxy purchase electricity measuring method and system thereof |
-
2021
- 2021-01-14 CN CN202110049155.4A patent/CN112686469A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116029480A (en) * | 2023-03-28 | 2023-04-28 | 广东电网有限责任公司 | Proxy purchase electricity measuring method and system thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106952159B (en) | Real estate collateral risk control method, system and storage medium | |
| CN108776939A (en) | The analysis method and system of user power utilization behavior | |
| CN106096805A (en) | A kind of residential electricity consumption load classification method based on entropy assessment feature selection | |
| CN112633924B (en) | District electric energy substitution demand analysis method based on load decomposition | |
| CN109190950B (en) | Energy consumption abnormity identification method for communicating electric meter and water meter files | |
| CN103577883A (en) | Grid-load intelligent interaction method and device | |
| CN107426026B (en) | Cloud computing server load short-term prediction method based on EEMD-ARIMA | |
| CN107103387B (en) | Load prediction method based on total power and occupancy coefficient of household average equipment | |
| US20230153767A1 (en) | Method and internet of things system for waste cleaning volume prediction in smart city | |
| CN104244314B (en) | A kind of potential group customer recognition methods based on Mc interface signaling | |
| CN111553444A (en) | Load identification method based on non-invasive load terminal data | |
| CN112286063A (en) | Non-intrusive measurement-based regional energy consumption monitoring system and method | |
| CN112686469A (en) | Family population number prediction method based on electric power big data | |
| CN109118128B (en) | Local area industrial enterprise power demand response potential evaluation method | |
| CN116797404A (en) | Intelligent building operation and maintenance supervision system based on big data and data processing | |
| CN112488738A (en) | Method and equipment for identifying resident vacant residents based on electric power big data | |
| CN110807546A (en) | Community grid population change early warning method and system | |
| CN112862172A (en) | State network 95598 power failure complaint prediction method and device, computer equipment and storage medium | |
| CN107145986B (en) | Charging electric quantity prediction method and device | |
| CN114398941A (en) | User payment behavior classification method, device, equipment and readable storage medium | |
| Fitzmaurice | Model selection with overdispersed data | |
| CN113112177A (en) | Transformer area line loss processing method and system based on mixed indexes | |
| CN110489893B (en) | Variable weight-based bus load prediction method and system | |
| Ziegler et al. | A Probabilistic Modelling Approach for Residential Load Profiles | |
| CN116757402A (en) | Service policy determining method and device, storage medium and electronic device |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210420 |
|
| WD01 | Invention patent application deemed withdrawn after publication |