CN110322063B - Power consumption simulation prediction method and storage medium - Google Patents
Power consumption simulation prediction method and storage medium Download PDFInfo
- Publication number
- CN110322063B CN110322063B CN201910569997.5A CN201910569997A CN110322063B CN 110322063 B CN110322063 B CN 110322063B CN 201910569997 A CN201910569997 A CN 201910569997A CN 110322063 B CN110322063 B CN 110322063B
- Authority
- CN
- China
- Prior art keywords
- power consumption
- time
- prediction
- data
- simulation
- 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.)
- Active
Links
- 238000004088 simulation Methods 0.000 title claims abstract description 73
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000003860 storage Methods 0.000 title claims abstract description 10
- 238000013135 deep learning Methods 0.000 claims abstract description 21
- 230000008569 process Effects 0.000 claims abstract description 14
- 230000005611 electricity Effects 0.000 claims description 86
- 238000012360 testing method Methods 0.000 claims description 26
- 238000012549 training Methods 0.000 claims description 25
- 238000004364 calculation method Methods 0.000 claims description 23
- 238000013136 deep learning model Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 14
- 238000010276 construction Methods 0.000 claims description 12
- 238000004220 aggregation Methods 0.000 claims description 11
- 230000002776 aggregation Effects 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000007476 Maximum Likelihood Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000006116 polymerization reaction Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000009826 distribution Methods 0.000 abstract description 19
- 230000006399 behavior Effects 0.000 abstract description 4
- 230000008859 change Effects 0.000 abstract description 4
- 238000005259 measurement Methods 0.000 abstract description 4
- 230000002354 daily effect Effects 0.000 description 11
- 238000005265 energy consumption Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010248 power generation Methods 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
- G06Q10/04—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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Abstract
The invention provides a power consumption simulation prediction method and a storage medium, wherein the power consumption simulation prediction method comprises the steps of performing poisson fitting on instantaneous power consumption of a preset history section twice to obtain fluctuation rate distribution of a prediction time section and single unit quantity distribution of the prediction time section; and calculating the instantaneous power consumption of the prediction time period according to the obtained fluctuation rate distribution of the prediction time period and the single unit quantity distribution of the prediction time period, and improving the simulation prediction accuracy of the invention by accurate measurement of each minute. The invention also adopts a deep learning mode to predict the average daily power consumption, and the data adopted in the deep learning process not only comprises instantaneous power consumption, but also comprises meteorological data and air quality data, and the weather condition and the air quality can influence the behavior of the power utilization unit so as to influence the power consumption of the power utilization entity and finally cause the power consumption to change, so that the prediction accuracy can be further improved by considering the meteorological data and the air quality data.
Description
Technical Field
The invention relates to the technical field of electricity consumption, in particular to a power consumption simulation prediction method and a storage medium.
Background
With the development of power supply technology and the diversification of power generation modes, electric energy gradually takes the dominant role in various energy sources. As the world's largest energy consumption country, the energy consumption in china has been an important and urgent task for both the supply side and the consumption side of energy, as the industrial skill level and the living standard of people have been improved. According to the demand side response theory, the power consumption can be reduced by 4% -15% by directly feeding back the refined information of the real-time electric energy consumption of the user, and the demand side response theory is beneficial to maintaining the supply and demand balance of the electric energy and the safety and stability of the electric power system. In order to reasonably arrange energy production and management, and solve the problem of uneven energy consumption in electricity consumption peak time, effective simulation and prediction of electric energy data become an urgent task.
Most of the existing energy prediction methods are based on the general trend of judging energy consumption, lack of reliable prediction and simulation based on historical data of specific electricity utilization units, and do not consider the instantaneous fluctuation rate of electricity utilization power, so that the judging result is not accurate enough. In recent years, with the continuous development of big data and artificial intelligence technology, a deep learning model is widely applied to the prediction field, and effective practice is obtained in the energy consumption prediction field. However, since the power consumption data of a specific power consumption unit also has the characteristics of high noise and instability, effective simulation of power consumption is still a technical problem to be solved.
Therefore, it is urgently needed to provide a power consumption simulation prediction method and system, wherein the instantaneous fluctuation rate is used in a simulation model, so that the problems of inaccurate prediction result and the like can be solved.
Disclosure of Invention
The invention aims to provide a power consumption simulation prediction method and a storage medium, which can effectively solve the problems of large traditional prediction time range and the like by combining the prediction of the fluctuation rate and the number of user units to further obtain the instantaneous power consumption prediction of a certain time period or time point in a certain time range in the future; and the meteorological data and the air quality data are introduced into the prediction model through deep learning, so that the prediction accuracy can be improved.
In order to solve the technical problems, the invention provides a power consumption simulation prediction method, which comprises the following steps: a database establishing step of inputting electricity utilization data of at least one electricity utilization entity into a database, wherein each electricity utilization entity comprises at least one electricity utilization unit; the electricity consumption data of each electricity entity comprises the instantaneous electricity consumption power of the electricity entity in a preset historical time period and the number of electricity consumption units in the electricity entity, wherein the preset historical time period comprises a plurality of first time sequences; a first sample collection step of collecting first samples from the database, each first sample including a first time series of instantaneous power consumption and the number of power consumption units; a first simulation step of constructing a first simulation model by using the first sample; and outputting a second time series of average instantaneous power consumption per day, the second time series being later than the preset historical period.
Further, the electricity utilization data of each electricity utilization entity further comprises air quality data and air temperature data, and the preset historical time period further comprises a plurality of third time sequences; the power consumption simulation prediction method further comprises the following steps: a second sample collection step of collecting a second sample from the database, the second sample including a third time series of instantaneous power consumption, air quality data, and air temperature data; a second sample classification step of randomly classifying the second sample into two types, namely a second training sample and a second test sample; a second prediction model construction step of training and constructing a prediction model by using more than two second training samples; and a second prediction step of inputting the second test sample into the second prediction model to obtain a fourth time sequence of power consumption, wherein the fourth time sequence is included in the third time sequence.
Further, the time interval of the instantaneous power consumption of the preset historical time period is 0.5-1.5 minutes; and/or the time range of the first time sequence is 1-28 days; and/or, the time range of the third time sequence is 28 days-84 days.
Further, after the database building step, a step of processing the data missing value of the preset historical time period is further included, and specifically includes the following steps: inquiring a missing time point in the historical time period, wherein the database does not contain instantaneous power consumption data corresponding to the missing time point; and an interpolation step of calculating instantaneous power consumption data corresponding to the missing time point and interpolating the instantaneous power consumption data corresponding to the missing time point to the database.
Further, in the interpolation step, according to the instantaneous power consumption data corresponding to the time points before and after the missing time point and the instantaneous power consumption data corresponding to the missing time point in other first time sequences, calculating and interpolating the instantaneous power consumption data corresponding to the missing time point by adopting a mean value interpolation method, a homogeneous mean value interpolation method, a maximum likelihood estimation method or a multiple interpolation method.
Further, the first simulation step includes the steps of: a data preprocessing step, namely normalizing the instantaneous power consumption of the first time sequence and calculating the power consumption ratio of power consumption per minute to the whole first time sequence to obtain the first time sequenceInstantaneous power consumption of column X j The calculation formula comprises:
wherein ,pij For instantaneous power consumption at the j-th minute on the i-th day within the first time series range; x is x ij -an instantaneous power consumption duty cycle of the powered entity at the j-th minute of the i-th day, for the first time series range; a first poisson simulation step by, for each X j Poisson simulation is performed to obtain a power consumption fluctuation rate sigma= (sigma) of the second time sequence 1 ,σ 2 ,...,σ n ) Wherein n=1440, the first time interval is included in the first time sequence, and the first time interval includes 10 to 50 time intervals; a second poisson simulation step, wherein the speed of the electricity utilization unit entering and exiting the electricity utilization entity is calculated by using a poisson process to obtain the number num= (Num) of the electricity utilization units in the second time sequence 1 ,Num 2 ,...,Num 1440 ) The method comprises the steps of carrying out a first treatment on the surface of the And a second time series power consumption calculation step of calculating the instantaneous power consumption of the second time series by using the number of power consumption units in the second time series and the instantaneous power consumption fluctuation rate of the second time series to obtain Pow, wherein a calculation formula comprises: />
Wherein the electricity utilization unit comprises a first electricity utilization sheet D 1 ～N(b 1 ,c 1 ) Element and second electricity utilization unit D 1 ～N(b 2 ,c 2 ) Delta is the number of first electric units in the first time sequence, mu i Average power consumption ratio on the i-th day of the first time sequence of the power utilization unit; the second time-series low-fluctuation-rate power utilization unit is included in the second time-series power utilization unit, and the instantaneous power consumption Pow of the second time series is subjected to Gaussian distribution.
Further, the second poisson simulation step specifically includes the following steps: a first data aggregation step, for X j The instantaneous power consumption duty ratio of the fifth time sequence is obtained by aggregation at 10-50 time intervals
The polymerization mode is realized by calculating X in 10-50 time intervals j Average value of (2); the average power consumption of the power utilization unit is calculated by X j The average instantaneous power consumption ratio mu of the power utilization unit is calculated, and the calculation formula is as follows: />
Wherein m=num, X max For the maximum value of the power consumption duty ratio of the first time sequence, X min A minimum value of the power consumption duty ratio for the first time series; calculating the speed of the electricity utilization unit entering and exiting the electricity utilization entity, and calculating the number of the electricity utilization units consuming the electricity utilization power in unit time, wherein a calculation formula comprises: />
Poisson fitting step according to parameter lambda 2 And carrying out poisson process fitting to obtain the number of the power utilization units in the second time sequence.
Further, the first poisson simulation step specifically includes the following steps: a first time series segmentation step of dividing X j Performing cutting at a first time interval, wherein the first time interval comprises 10-50 time intervals; an output step of, for each X in the first time interval j Performing poisson process fitting to obtain poisson fitting values poi of a second time interval; a step of calculating the power consumption fluctuation rate by using X in the first time interval j And poi calculates the fluctuation rate of each time interval, wherein the calculation formula comprises:
where k is the length of the first time interval.
Further, the fluctuation rate of the first electricity unit is smaller than the fluctuation rate of the second electricity unit; in the second time-series power consumption calculation step, Δ, c 1 、c 2 Obtained by solving the following equation set:
Further, the second sample collection step further includes the steps of: a second data aggregation step of obtaining an average daily instantaneous power consumption of the third time series by average aggregation of the instantaneous power consumption of the third time series at 1 day as a time interval; and normalizing the average daily power consumption of the third time sequence to obtain the second sample.
Further, the second prediction model construction step includes the steps of: a second prediction model learning step, wherein the second prediction model performs deep learning according to a first formula, and the first formula comprises:
The present invention also provides a computer-readable storage medium having a computer program stored thereon, which when executed by a processor, implements any of the above-described power consumption simulation prediction methods.
The beneficial effects of the invention are as follows: the invention provides a power consumption simulation prediction method and a storage medium, wherein the power consumption simulation prediction method comprises the steps of performing poisson fitting on instantaneous power consumption of a preset history section twice to obtain fluctuation rate distribution of a predicted time section and power consumption unit quantity distribution of the predicted time section; and calculating the instantaneous power consumption of the prediction time period according to the fluctuation rate distribution of the obtained prediction time period and the quantity distribution of the power consumption units of the prediction time period, and accurately obtaining each minute, thereby improving the prediction accuracy of the invention. The invention also adopts a deep learning mode to predict the daily power consumption, the deep learning is used as an artificial intelligent method, compared with the traditional time sequence method and machine learning method, the prediction accuracy is higher, the data adopted in the deep learning process not only has instantaneous power consumption, but also has weather data and air quality data, and the weather condition and the air quality can influence the behavior of the power consumption unit, thereby influencing the power consumption of the power consumption entity and finally causing the power consumption to change, so the prediction accuracy can be further improved by considering the weather data and the air quality data.
Drawings
The invention is further described below with reference to the drawings and examples.
FIG. 1 is a flow chart of a power consumption simulation prediction method provided by the invention;
FIG. 2 is a flowchart showing the steps of processing the data missing values in the preset historical time period according to the present invention;
FIG. 3 is a flowchart of a first simulation step provided by the present invention;
FIG. 4 is a flowchart of a first poisson simulation procedure according to the present invention;
FIG. 5 is a flowchart of a second Poisson simulation step provided by the present invention;
FIG. 6 is a flowchart of a second sample collection step provided by the present invention;
FIG. 7 is a flowchart of a second predictive model construction step provided by the present invention;
FIG. 8 is a functional block diagram of a power consumption simulation prediction system according to the present invention;
FIG. 9 is a functional block diagram of a processing unit for data missing values in a preset historical time period according to the present invention;
FIG. 10 is a functional block diagram of a second prediction model building unit according to the present invention;
a power consumption simulation prediction system 100;
a database 110; a data processing system 200; a preset history period data missing value processing unit 21;
a first sample acquisition unit 22; a first simulation unit 23; a second sample acquisition unit 24;
a second sample classification unit 25; a second prediction model construction unit 26; a second prediction unit 27;
a query unit 211; an interpolation unit 212; an initialization unit 261;
a deep learning unit 262; a capability measuring unit 263; load prediction unit 264.
Detailed Description
The following description of the embodiments refers to the accompanying drawings, which illustrate specific embodiments of the invention that may be wet. The names of the elements, such as first, second, etc., mentioned in the present invention are merely distinguishing different elements, and may be better expressed. In the drawings, like elements are referred to by like reference numerals and adjacent or similar elements are referred to by similar reference numerals.
Embodiments of the present invention will be described in detail herein with reference to the accompanying drawings. This invention may take many different forms and should not be construed as limited to the particular embodiments set forth herein. These embodiments are provided so that this disclosure will be thorough and complete, and will fully enable others skilled in the art to understand the various embodiments of the invention and with various modifications as are suited to the particular use contemplated.
As shown in FIG. 1, the invention provides a computer software method, which realizes a power consumption simulation prediction method by a computer and comprises the following steps S1 to S8.
S1) a database establishing step, namely inputting electricity utilization data of at least one electricity utilization entity into a database, wherein each electricity utilization entity comprises at least one electricity utilization unit; the electricity consumption data of each electricity consumption entity comprises instantaneous electricity consumption power of the electricity consumption entity in a preset historical time period, the number of electricity consumption units in the electricity consumption entity, air quality data and air temperature data, wherein the preset historical time period comprises a plurality of first time sequences and a plurality of third time sequences.
The time interval of the instantaneous power consumption of the preset historical time period is 0.5-1.5 minutes, and the optimal time interval is 1 minute; the time range of the first time sequence is 1-28 days.
S2) a data missing value processing step of a preset historical time period, which is used for interpolating the missing instantaneous power consumption data of the preset historical time period; as shown in fig. 2, the method specifically includes the following steps S21 to S22.
S21) inquiring a missing time point in the historical time period, wherein the database does not have instantaneous power consumption data corresponding to the missing time point.
S22) calculating instantaneous power consumption data corresponding to the missing time point, and interpolating the instantaneous power consumption data corresponding to the missing time point into the database.
In the interpolation step, the instantaneous power consumption data corresponding to the missing time point is calculated and interpolated by adopting a mean value interpolation method, a homogeneous mean value interpolation method, a maximum likelihood estimation method or a multiple interpolation method according to the instantaneous power consumption data corresponding to the time points before and after the missing time point and the instantaneous power consumption data corresponding to the missing time point in other first time sequences.
S3) a first sample acquisition step, namely acquiring first samples from the database, wherein each first sample comprises a first time sequence of instantaneous power consumption and the number of power consumption units.
S4) a first simulation step, namely constructing a first simulation model by using the first sample; and outputting a second time series of instantaneous power consumption, the second time series being later than the preset historical period.
As shown in fig. 3, the first simulation step specifically includes the following steps S41 to S44.
S41) a data preprocessing step, namely normalizing the instantaneous power consumption of the first time sequence and calculating the power consumption ratio of power consumption per minute to the whole first time sequence to obtain the instantaneous power consumption ratio X of the first time sequence j The calculation formula comprises:
wherein ,pij For instantaneous power consumption at the j-th minute on the i-th day within the first time series range; x is x ij In order to be in the first time sequence range, the instantaneous power consumption duty ratio of the electricity utilization entity in the j-th minute of the i-th day.
The first time series is preferably every minute of the day, i.e. the first time series has 1440 data.
S42) a first Poisson simulation step by, for each X j Poisson simulation is performed to obtain a power consumption fluctuation rate sigma= (sigma) of the second time sequence 1 ,σ 2 ,...,σ n ) Wherein n=1440, the first time interval is comprised in theAnd a first time sequence, wherein the first time interval comprises 10-50 time intervals.
As shown in fig. 4, the first poisson simulation step specifically includes the following steps S421 to S423.
S421) a first time series segmentation step, X is defined as j Performing cutting at first time intervals, wherein the first time intervals comprise 10-50 time intervals, preferably 15 time intervals, namely 15 minutes;
s422) outputting step, for each X in the first time interval j (X 1 ～X 15 、X 16 ～X 30 The term "poisson process" refers to a poisson process fit to obtain poisson fit values poi (poi) for a first time interval 1 ～poi 15 、poi 16 ～poi 30 、...)；
S423) a power consumption fluctuation ratio calculation step of calculating a power consumption fluctuation ratio by using X in a plurality of first time intervals j And poi calculates the fluctuation rate of each time interval, wherein the calculation formula comprises:
According to said first time sequence, preferably a time of day, then sigma i ＝(σ 1 ,σ 2 ,...,σ 96 )。
S43) a second Poisson simulation step, namely, simulating and calculating the speed of the electricity utilization unit entering and exiting the electricity utilization entity by using a Poisson process to obtain the number of electricity utilization units Num= (Num) in the second time sequence 1 ,Num 2 ,...,Num 1440 )。
As shown in fig. 5, the second poisson simulation step specifically includes the following steps S431 to S434.
S431) a first data aggregation step, for X j The instantaneous power consumption duty ratio of the fifth time sequence is obtained by aggregation at 10-50 time intervals
The polymerization mode is realized by calculating X in 10-50 time intervals j Average value of (2); the fifth time sequence is 15 time intervals, i.e. 15 minutes, & gt>
S432) calculating average power consumption of the power utilization unit by X j The average instantaneous power consumption ratio of the power utilization unit is calculated, and the calculation formula is as follows:
wherein m=num, which is the number of electricity units in the second time sequence, X max For the maximum value of the power consumption duty ratio of the first time sequence, X min Is the minimum value of the power consumption duty cycle of the first time series.
S433), calculating the speed of the electricity utilization units in and out of the electricity utilization entity, and calculating the number of the electricity utilization units consuming the electricity utilization power in unit time, namely calculating how many electricity utilization units consume the electricity utilization power in unit time, wherein a calculation formula comprises:
as can be seen from step S431, i ε [1,96]。
S434) poisson fitting step according to the parameter λ 2 Fitting a poisson process to obtain the number of power utilization units in the second time sequence; each lambda at the time of output 2 Including 15 time intervals, i.e. Num of output 1440 time intervals.
S44) a second time-series power consumption calculation step of calculating an instantaneous power consumption of the second time series by using the number of power consumption units in the second time series and the instantaneous power consumption fluctuation rate of the second time series to obtain Pow, wherein a calculation formula includes:
wherein the electricity utilization unit comprises a first electricity utilization sheet D 1 ～N(b 1 ,c 1 ) Element and second electricity utilization unit D 1 ～N(b 2 ,c 2 ) The fluctuation rate of the first electricity unit is smaller than that of the second electricity unit; the total predicted number of power usage units can be regarded as
Is a gaussian distribution of (c).
Delta is the number of first electric units in the first time sequence, mu i Average power consumption ratio on the i-th day of the first time sequence of the power utilization unit; the second time-series low-fluctuation-rate power utilization unit is included in the second time-series power utilization unit, and the instantaneous power consumption Pow of the second time series is subjected to Gaussian distribution.
In the second time-series power consumption calculation step, Δ, c 1 、c 2 Obtained by solving the following equation set:
wherein ,
sampling data of a first electric unit in Num and the corresponding power consumption fluctuation rate of the second time sequence; />
And the power consumption fluctuation rate of the second time sequence corresponding to the data sampling of the second power consumption unit in Num.
S5) a second sample acquisition step, namely acquiring data from the database and performing data processing to obtain a second sample, wherein the second sample comprises a third time sequence of instantaneous power consumption, air quality data and air temperature data.
As shown in fig. 6, the second sample collection step includes the following steps S51 to S52.
S51) a second data aggregation step of obtaining an average power consumption per day of the third time series by averaging the instantaneous power consumption of the third time series at 1 day intervals.
S52) normalizing the average power consumption per day of the third time series to obtain the second sample.
S6) a second sample classification step, namely randomly classifying the second samples into two types, namely a second training sample and a second test sample.
S7) a second prediction model construction step, namely training and constructing a prediction model by using more than two second training samples.
S8) a second prediction step, namely inputting the second test sample into the second prediction model to obtain a fourth time sequence of daily average power consumption, wherein the fourth time sequence is later than the preset historical time period.
As shown in fig. 7, the second prediction model construction step includes the following steps S81 to S82.
S81) a second prediction model learning step of performing deep learning according to a first formula including:
wherein ,xi For the ith training sample in the training set, n is the number of training samples in the training set, L () is a mean square loss function, lambda is a regularization coefficient, J (w) is a regularization term, w is a weight parameter, y i For the ith test sample in the test set, v is the power consumption prediction data of the user per day in the fourth time sequence, f (x) i ) A second current prediction model;
s82) a first prediction error calculation step of determining a first prediction error of the current deep learning model using a second formula comprising:
wherein M is the first prediction error, n is the number of power load prediction data of the user entity in the fourth time sequence, and X t ' forecast data for the average power consumption of the user per day, X, the t th time series in the fourth time series t Is the true value of the daily power consumption in the t second test sample in the test set.
If the first prediction error is lower than a preset first error lower limit, the prediction capacity meets the preset requirement; and if the prediction capability does not meet the preset requirement, optimizing the current second prediction model until a preset first error lower limit is met.
The invention provides a power consumption simulation prediction method, which is characterized in that the instantaneous power consumption of a preset history period is subjected to poisson fitting twice to obtain the fluctuation rate distribution of a predicted period and the quantity distribution of power consumption units of the predicted period; and calculating the instantaneous power consumption of the prediction time period according to the fluctuation rate distribution of the obtained prediction time period and the quantity distribution of the power consumption units of the prediction time period, and accurate to each minute, thereby improving the prediction accuracy of the invention.
The invention also adopts a deep learning mode to predict the daily power consumption, the deep learning is used as an artificial intelligent method, compared with the traditional time sequence method and machine learning method, the prediction accuracy is higher, the data adopted in the deep learning process not only has the daily average power consumption, but also has weather data and air quality data, and the weather condition and the air quality can influence the behavior of the power consumption unit, thereby influencing the power consumption of the power consumption entity and finally causing the power consumption to change, so the prediction accuracy can be improved by considering the weather data and the air quality data.
The invention also provides an electronic device comprising a memory and a processor, the memory being for storing executable program code; the processor executes a program corresponding to the executable program code by reading the executable program code to perform the steps in the power consumption prediction method described above.
As shown in fig. 8, the electronic device includes a power consumption simulation prediction system 100, which includes a database 110 and a data processing system 200, wherein the database 110 is stored in the memory, and the data processing system 200 is the processor described above.
The database 110 includes electricity consumption data of at least one electricity consumption entity, where the electricity consumption data of each electricity consumption entity includes instantaneous power consumption of the electricity consumption entity in a preset historical time period, the number of electricity consumption units in the electricity consumption entity, air quality data and air temperature data, and the preset historical time period includes a plurality of first time sequences and a plurality of third time sequences.
The data processing system 200 includes: the data missing value processing unit 21, the first sample acquisition unit 22, the first simulation unit 23, the second sample acquisition unit 24, the second sample classification unit 25, the second prediction model construction unit 26, and the second prediction unit 27 of the preset history period.
As shown in fig. 9, the preset history period data loss value processing unit 21 specifically includes a query unit 211 and an interpolation unit 212.
The query unit 211 queries the missing time point in the historical time period, and no instantaneous power consumption data corresponding to the missing time point exists in the database 110.
The interpolation unit 212 is configured to calculate instantaneous power consumption data corresponding to the missing time point, and interpolate the instantaneous power consumption data corresponding to the missing time point to the database 110.
The first sample collection unit 22 collects first samples from the database 110, each first sample including a first time series of instantaneous power consumption and a number of power usage units.
The first simulation unit 23 constructs a first simulation model using the first sample; and outputting a second time series of instantaneous power consumption, the second time series being later than the preset historical period.
The second sample collection unit 24 collects a second sample from the database 110, the second sample including a third time series of instantaneous power consumption, air quality data, and air temperature data.
The second sample classification unit 25 randomly classifies the second samples into two classes, a second training sample and a second test sample.
The second prediction model construction unit 26 trains and constructs a prediction model using two or more second training samples.
The second prediction unit 27 inputs the second test sample to the second prediction model to obtain a fourth time sequence of power consumption, where the fourth time sequence is later than the preset historical period.
As shown in fig. 10, the second prediction model construction unit 26 is configured to determine a deep learning model for predicting the instantaneous power consumption per day of the sixth time series, and the second prediction model construction unit 26 includes an initialization unit 261, a deep learning unit 262, a capacity measurement unit 263, and a load prediction unit 264.
The initialization unit 261 is configured to perform initialization setting on a network structure and model parameters of the deep learning model;
the deep learning unit 262 is configured to input the second training sample into a current deep learning model, and perform deep learning through the current deep learning model, so as to obtain daily instantaneous power consumption data of the user entity in the fourth time sequence.
The capacity measurement unit 263 is configured to measure a prediction capacity of a current deep learning model according to the average daily power consumption measurement data of the user entity in the fourth time interval and the second test sample, and if the prediction capacity does not meet a predetermined requirement, adjust a network structure and/or model parameters of the current deep learning model, and return to the deep learning unit 262; otherwise, the current deep learning model is used as the deep learning model for carrying out power consumption simulation prediction, and the load prediction module is transferred to.
The load prediction unit 264 is configured to input the second test sample into the deep learning model for performing power consumption simulation prediction, so as to obtain an average power consumption of the user per day in the sixth time sequence.
The feature extraction unit is used for extracting features affecting the prediction result of the deep learning model for carrying out power consumption simulation prediction and the influence degree value corresponding to each feature.
The electronic device provided by the invention is provided with the power consumption simulation prediction system 100, the first simulation unit 23 outputs the instantaneous power consumption every minute every day in the future according to the data of the historical stage, and the output time precision is smaller. And initializing a model structure and/or model parameters by the second prediction model to obtain an initial deep learning model, then performing deep learning by the initial deep learning model by using a training set to obtain a prediction result, determining the prediction capability of the model according to the prediction result and a test set, and if the prediction capability does not meet the requirement, adjusting the model structure and/or parameters until the prediction capability meets the requirement to obtain a final deep learning model, and performing testing by using the test set to obtain a desired test result.
The invention also provides a computer readable storage medium, which stores a computer program, and the power consumption simulation prediction method can be realized when the computer program is executed in a processor of the electronic equipment.
The invention provides a power consumption simulation prediction method and a storage medium, wherein the power consumption simulation prediction method comprises the steps of performing poisson fitting on instantaneous power consumption of a preset history section twice to obtain fluctuation rate distribution of a predicted time section and power consumption unit quantity distribution of the predicted time section; and calculating the instantaneous power consumption of the prediction time period according to the fluctuation rate distribution of the obtained prediction time period and the quantity distribution of the power consumption units of the prediction time period, and accurately obtaining each minute, thereby improving the prediction accuracy of the invention.
The invention also adopts a deep learning mode to predict the daily power consumption, the deep learning is used as an artificial intelligent method, compared with the traditional time sequence method and machine learning method, the prediction accuracy is higher, the data adopted in the deep learning process not only has instantaneous power consumption, but also has weather data and air quality data, and the weather condition and the air quality can influence the behavior of the power consumption unit, thereby influencing the power consumption of the power consumption entity and finally causing the power consumption to change, so the prediction accuracy can be further improved by considering the weather data and the air quality data.
It should be noted that numerous variations and modifications are possible in light of the fully described invention, and are not limited to the specific examples of implementation described above. The above-described embodiments are merely illustrative of the present invention and are not intended to be limiting. In general, the scope of the present invention should include those variations or alternatives and modifications apparent to those skilled in the art.
Claims (9)
1. The power consumption simulation prediction method is characterized by comprising the following steps of:
a database establishing step of inputting electricity utilization data of at least one electricity utilization entity into a database, wherein each electricity utilization entity comprises at least one electricity utilization unit; the electricity consumption data of each electricity entity comprises the instantaneous electricity consumption power of the electricity entity in a preset historical time period and the number of electricity consumption units in the electricity entity, wherein the preset historical time period comprises a plurality of first time sequences;
a first sample collection step of collecting first samples from the database, wherein each first sample comprises instantaneous power consumption of the electricity utilization entity in a first time sequence and the number of electricity utilization units in the electricity utilization entity; and
a first simulation step of constructing a first simulation model by using the first sample and outputting a second time series of instantaneous power consumption, wherein the second time series is later than the preset historical time period;
the first simulation step includes the steps of:
a data preprocessing step, which is to normalize the instantaneous power consumption of the electricity utilization entity in the first time sequence and calculate the ratio of power consumption per minute to the whole power consumption of the first time sequence,obtaining the instantaneous power consumption ratio X of the first time sequence j The calculation formula is as follows:
wherein ,pij For instantaneous power consumption by the powered entity at the j-th minute of the i-th day within the first time series range; x is x ij -an instantaneous power consumption duty cycle of the powered entity at the j-th minute of the i-th day, for the first time series range;
a first poisson simulation step by, for each X j Poisson simulation processing is performed to obtain a power consumption fluctuation rate sigma= (sigma) of the second time series 1 ,σ 2 ,...,σ 1440 ) The first time interval is included in the first time sequence, and the first time interval comprises 10-50 time intervals;
a second poisson simulation step, wherein the speed of the electricity utilization unit entering and exiting the electricity utilization entity is simulated by using a poisson process to obtain the number num= (Num) of the electricity utilization units in the second time sequence 1 ,Num 2 ,...,Num 1440 ) The method comprises the steps of carrying out a first treatment on the surface of the The second poisson simulation step specifically comprises the following steps:
a first data aggregation step, for X j The instantaneous power consumption duty ratio of the fifth time sequence is obtained by aggregation at 10-50 time intervals
The polymerization mode is realized by calculating X in 10-50 time intervals j Average value of (2);
the average power consumption of the power utilization unit is calculated by X j The average instantaneous power consumption ratio mu of the power utilization unit is calculated, and the calculation formula is as follows:
wherein m=num, X max For the maximum value of the power consumption duty ratio of the first time sequence, X min Is the firstA time series of minimum values of power consumption duty cycle;
calculating the speed of the electricity utilization units entering and exiting the electricity utilization entity, and calculating the number of the electricity utilization units consuming the electricity utilization power in unit time, wherein a calculation formula comprises:
and
Poisson fitting step according to poisson parameter lambda 2 Fitting a poisson process to obtain the number of power utilization units in the second time sequence; and
a second time series power consumption calculation step of calculating a second time series instantaneous power consumption Pow according to the number of power consumption units in the second time series and the second time series instantaneous power consumption fluctuation rate, wherein a calculation formula comprises:
wherein the electricity utilization unit comprises a first electricity utilization unit D 1 ～N(b 1 ,c 1 ) Second electricity utilization unit D 1 ～N(b 2 ,c 2 )，
Delta is the number of first electrical units in the first time sequence;
μ i average power consumption ratio on the i-th day of the first time sequence of the power utilization unit;
wherein ,Δ、c1 、c 2 Obtained by solving the following equation set:
wherein ,
sampling data of a first electric unit in Num and the corresponding power consumption fluctuation rate of the second time sequence;
2. The method for power consumption simulation prediction according to claim 1, wherein,
the electricity usage data for each electrical entity also includes air quality data and air temperature data,
the preset historical time period also comprises a plurality of third time sequences;
the power consumption simulation prediction method further comprises the following steps:
a second sample collection step of collecting a second sample from the database, the second sample including a third time series of instantaneous power consumption, air quality data, and air temperature data;
a second sample classification step of randomly classifying the second sample into two types, namely a second training sample and a second test sample;
a second prediction model construction step of training and constructing a prediction model by using more than two second training samples;
and a second prediction step of inputting the second test sample into the second prediction model to obtain a fourth time sequence of average power consumption per day, wherein the fourth time sequence is included in the third time sequence.
3. The method for power consumption simulation prediction according to claim 1, wherein,
the time interval of the instantaneous power consumption of the preset historical time period is 0.5-1.5 minutes; and/or the number of the groups of groups,
the time range of the first time sequence is 1-28 days.
4. The method for power consumption simulation prediction according to claim 1, wherein,
after the database establishing step, a step of processing the data missing value of the preset historical time period is further included, and the method specifically includes the following steps:
inquiring a missing time point in the historical time period, wherein the database does not contain instantaneous power consumption data corresponding to the missing time point; and
and an interpolation step, calculating the instantaneous power consumption data corresponding to the missing time point, and interpolating the instantaneous power consumption data corresponding to the missing time point into the database.
5. The method for power consumption simulation prediction according to claim 4, wherein,
in the step of the interpolation, the interpolation is performed,
according to the instantaneous power consumption data corresponding to the time points before and after the missing time point, and
other instantaneous power consumption data corresponding to the missing time point in the first time sequence,
and calculating and interpolating the instantaneous power consumption data corresponding to the missing time point by adopting a mean value interpolation method, a homogeneous mean value interpolation method, a maximum likelihood estimation method or a multiple interpolation method.
6. The method for power consumption simulation prediction according to claim 1, wherein,
the first poisson simulation step specifically comprises the following steps:
a first time series segmentation step of dividing X j Dividing the first time interval length, wherein the first time interval comprises 10-50 time intervals;
an output step of, for lambda 1 I.e. X in each first time interval j Fitting in a poisson process to obtain a poisson fitting value poi of a first time interval;
a power consumption fluctuation rate calculation step of calculating a power consumption fluctuation rate by using X in a plurality of first time intervals j And poi calculates the fluctuation rate of each time interval, wherein the calculation formula comprises:
where k is the length of the first time interval.
7. The method for power consumption simulation prediction according to claim 2, wherein,
the second sample collection step further comprises the steps of:
a second data aggregation step of obtaining an average daily instantaneous power consumption of the third time series by average aggregation of the instantaneous power consumption of the third time series at 1 day as a time interval;
and normalizing the average daily power consumption of the third time sequence to obtain the second sample.
8. The method for power consumption simulation prediction according to claim 2, wherein,
the second prediction model construction step includes the steps of:
a second prediction model learning step, wherein the second prediction model performs deep learning according to a first formula, and the first formula comprises:
wherein ,xi For the ith training sample in the training set, n is the number of training samples in the training set, L () is a mean square loss function, lambda is a regularization coefficient, J (w) is a regularization term, w is a weight parameter, y i For the ith test sample in the test set, v is the power consumption simulation prediction data of the user entity in the fourth time sequence, f (x) i ) A second current prediction model;
a first prediction error calculating step, wherein a second formula is adopted to determine the first prediction error of the current deep learning model, and the second formula comprises:
wherein M is the first prediction error, n is the number of power load prediction data of the user entity in the fourth time sequence, and X t ' forecast data for the power consumption of the user at the t-th day in the fourth time sequence, X t A true value for the average daily power consumption in a t second test sample in the test set;
if the first prediction error is lower than a preset first error lower limit, the prediction capacity meets the preset requirement; and if the prediction capability does not meet the preset requirement, optimizing the current second prediction model until the preset first error lower limit is met.
9. A storage medium having a computer program stored thereon, wherein the method according to any of claims 1-8 is implemented when the computer program is executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910569997.5A CN110322063B (en) | 2019-06-27 | 2019-06-27 | Power consumption simulation prediction method and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910569997.5A CN110322063B (en) | 2019-06-27 | 2019-06-27 | Power consumption simulation prediction method and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110322063A CN110322063A (en) | 2019-10-11 |
| CN110322063B true CN110322063B (en) | 2023-05-30 |
Family
ID=68121293
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910569997.5A Active CN110322063B (en) | 2019-06-27 | 2019-06-27 | Power consumption simulation prediction method and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110322063B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126780B (en) * | 2019-10-31 | 2023-04-07 | 内蒙古电力(集团)有限责任公司包头供电局 | Non-invasive load monitoring method and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014110679A (en) * | 2012-11-30 | 2014-06-12 | Mitsubishi Electric Corp | Power consumption estimation system, power consumption estimation method, and program |
| CN107239852A (en) * | 2017-05-05 | 2017-10-10 | 南京邮电大学 | A kind of electric quantity consumption Forecasting Methodology based on deep learning |
| CN107769235A (en) * | 2017-09-29 | 2018-03-06 | 国网上海市电力公司 | A kind of microgrid energy management method based on hybrid energy-storing and electric automobile |
| CN108416695A (en) * | 2018-02-24 | 2018-08-17 | 合肥工业大学 | Electric load probability density prediction technique based on deep learning and system, medium |
-
2019
- 2019-06-27 CN CN201910569997.5A patent/CN110322063B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2014110679A (en) * | 2012-11-30 | 2014-06-12 | Mitsubishi Electric Corp | Power consumption estimation system, power consumption estimation method, and program |
| CN107239852A (en) * | 2017-05-05 | 2017-10-10 | 南京邮电大学 | A kind of electric quantity consumption Forecasting Methodology based on deep learning |
| CN107769235A (en) * | 2017-09-29 | 2018-03-06 | 国网上海市电力公司 | A kind of microgrid energy management method based on hybrid energy-storing and electric automobile |
| CN108416695A (en) * | 2018-02-24 | 2018-08-17 | 合肥工业大学 | Electric load probability density prediction technique based on deep learning and system, medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110322063A (en) | 2019-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108593990B (en) | Electricity stealing detection method based on electricity consumption behavior mode of electric energy user and application | |
| CN108280552B (en) | Power load prediction method and system based on deep learning and storage medium | |
| CN102999791A (en) | Power load forecasting method based on customer segmentation in power industry | |
| CN106600037B (en) | Multi-parameter auxiliary load prediction method based on principal component analysis | |
| CN113554466B (en) | Short-term electricity consumption prediction model construction method, prediction method and device | |
| CN112990500B (en) | Transformer area line loss analysis method and system based on improved weighted gray correlation analysis | |
| CN109359665B (en) | Household appliance load identification method and device based on support vector machine | |
| CN111027775A (en) | Step hydropower station generating capacity prediction method based on long-term and short-term memory network | |
| CN112330077B (en) | Power load prediction method, power load prediction device, computer equipment and storage medium | |
| CN110796307A (en) | Distributed load prediction method and system for comprehensive energy system | |
| CN113126019A (en) | Intelligent ammeter error remote estimation method, system, terminal and storage medium | |
| CN112365056A (en) | Electrical load joint prediction method and device, terminal and storage medium | |
| CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
| CN115222106A (en) | User day-ahead load prediction method of self-adaptive model | |
| CN114119273A (en) | Park comprehensive energy system non-invasive load decomposition method and system | |
| CN111179108A (en) | Method and device for predicting power consumption | |
| CN106295877B (en) | Method for predicting electric energy consumption of smart power grid | |
| CN110322063B (en) | Power consumption simulation prediction method and storage medium | |
| CN112307675B (en) | Neural network-based temperature-sensitive load separation identification method and system | |
| CN117273195A (en) | Steam heating system demand load prediction method | |
| CN116470491A (en) | Photovoltaic power probability prediction method and system based on copula function | |
| CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
| Ming-guang et al. | Short-term load combined forecasting method based on BPNN and LS-SVM | |
| CN115358441A (en) | New energy cluster consumption intelligent control method and device based on federal learning | |
| CN115130764A (en) | Power distribution network situation prediction method and system based on state evaluation |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231130 Address after: 201108 3 / F and 4 / F, building 2, 598 Guanghua Road, Minhang District, Shanghai Patentee after: Shanghai Jientropy Data Technology Co.,Ltd. Address before: 201108 room D62, 1st floor, building 6, 4299 Jindu Road, Minhang District, Shanghai Patentee before: SHANGHAI MAXTROPY DATA TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |