CN117458475A - Charging pile electricity load prediction method and system based on electricity time sequence data - Google Patents
Charging pile electricity load prediction method and system based on electricity time sequence data Download PDFInfo
- Publication number
- CN117458475A CN117458475A CN202311488052.3A CN202311488052A CN117458475A CN 117458475 A CN117458475 A CN 117458475A CN 202311488052 A CN202311488052 A CN 202311488052A CN 117458475 A CN117458475 A CN 117458475A
- Authority
- CN
- China
- Prior art keywords
- data
- charging pile
- model
- load prediction
- electricity
- 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
- 230000005611 electricity Effects 0.000 title claims abstract description 69
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012549 training Methods 0.000 claims abstract description 47
- 238000013507 mapping Methods 0.000 claims abstract description 18
- 239000013598 vector Substances 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000010276 construction Methods 0.000 claims description 12
- 230000000737 periodic effect Effects 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 4
- 230000002547 anomalous effect Effects 0.000 claims description 3
- 238000013210 evaluation model Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
Classifications
-
- 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
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses a charging pile electricity load prediction method based on electricity time sequence data, which comprises the steps of obtaining charging pile data information of a target area; mapping the data information into feature vectors and clustering to obtain a charging pile feature class label; acquiring historical electricity load data of a charging pile of a target area and processing the historical electricity load data to obtain a training data set; constructing an initial charging pile electric load prediction model, training and evaluating to obtain a final charging pile electric load prediction model; and carrying out charging pile electricity load prediction based on the electricity time sequence data by adopting a charging pile electricity load prediction model. The invention also discloses a system for realizing the charging pile electricity load prediction method based on the electricity time sequence data. The method can not only realize the prediction of the power consumption load of the charging pile based on the power consumption time sequence data, but also has higher reliability, better accuracy and more objectivity and science.
Description
Technical Field
The invention belongs to the field of electric automation, and particularly relates to a charging pile electricity load prediction method and system based on electricity time sequence data.
Background
Along with the development of economic technology and the improvement of living standard of people, electric energy becomes an indispensable secondary energy source in the production and living of people, and brings endless convenience to the production and living of people. Therefore, ensuring stable and reliable supply of electric energy becomes one of the most important tasks of the electric power system.
At present, with the rapid growth of new energy automobile users, a charging pile is used as a supporting infrastructure of the new energy automobile, and large-scale access to a power grid is started. The distribution range of the charging piles is extremely wide, such as being installed in residential areas, expressway service areas, industrial parks and the like. In addition, because the charging piles belong to intermittent energy supply equipment, electric energy consumption can be generated in the charging process of the electric vehicle, and charging load peaks of the charging piles in different areas are different, the distribution characteristics and the power utilization time sequence of the charging piles are combined to predict the power utilization load of the charging piles, the prediction of the power utilization condition of each class of charging piles is facilitated, the power utilization load distribution and the power utilization load prediction of a power system are optimized, and further the balance of the power grid load and the safe and stable operation of the power grid are realized.
At present, conventional schemes for electricity load prediction are generally based on gray prediction models, ARIMA models and the like. Such models are typically used to fit the power usage using a linear time series model to achieve load prediction of the power system. However, the charging pile load data is subjected to a plurality of influencing factors, and the conventional load prediction model cannot process the plurality of influencing factors. Therefore, in the conventional load prediction scheme, when load prediction is performed on the charging pile, the problems of poor prediction accuracy and poor reliability of a prediction structure often exist.
Disclosure of Invention
The invention aims to provide a charging pile electricity load prediction method based on electricity time sequence data, which is high in reliability, good in accuracy and objective and scientific.
The second object of the present invention is to provide a system for implementing the method for predicting the power consumption load of the charging pile based on the power consumption time sequence data.
The invention provides a charging pile electricity load prediction method based on electricity time sequence data, which comprises the following steps:
s1, acquiring charging pile data information of a target area;
s2, mapping the data information obtained in the step S1 into feature vectors by adopting a data dictionary mapping mode and clustering, so as to obtain a charging pile feature type label;
s3, acquiring historical electricity load data of the charging piles of the target area according to the characteristic class labels of the charging piles obtained in the step S2, and processing the historical electricity load data to obtain a training data set;
s4, constructing an initial charge pile electricity load prediction model according to influence factors of charge pile loads;
s5, training and evaluating the initial model of the electric load prediction for the charging pile constructed in the step S4 by adopting the training data set obtained in the step S3 to obtain a final model of the electric load prediction for the charging pile;
s6, adopting the charging pile electricity load prediction model obtained in the step S5 to conduct charging pile electricity load prediction based on electricity time sequence data.
The step S1 of acquiring the charging pile data information of the target area specifically comprises the following steps:
acquiring distribution and attribute information of charging piles; the distribution and attribute information comprises a charging pile scale, a service vehicle type and a station type;
acquiring characteristic data of an area where the charging pile is located; the characteristic data of the region comprises a service area, a residential area and a business district;
acquiring characteristic data of a radio station area where the charging pile is located; the characteristic data of the station area comprises the type of the station area, the number of users in the station area and the power supply voltage;
and carrying out association processing on the acquired distribution and attribute information, the characteristic data of the region and the characteristic data of the station area to obtain a charging pile distribution characteristic sample label set.
The method of mapping the data dictionary in the step S2 maps the data information obtained in the step S1 into feature vectors and clusters the feature vectors to obtain feature class labels of the charging piles, and specifically includes the following steps:
according to the charging pile distribution characteristic sample label set obtained in the step S1, adopting a dictionary mapping mode to form characteristic vectors from label data with different dimensions;
adopting a K-means algorithm to perform clustering division on the charging piles:
the number k of the clustered categories is calculated by adopting the following formula:
wherein argmax is the maximum value;is a classification index, and->tr () is the trace of the matrix, B k Is covariance matrix among categories, W k A covariance matrix in the category, num is the number of samples, and k is the number of categories;
and forming a final charging pile characteristic category label according to the obtained category number k.
And step S3, acquiring historical electricity load data of the charging piles of the target area according to the charging pile characteristic type label obtained in the step S2, and processing the historical electricity load data to obtain a training data set, wherein the method specifically comprises the following steps of:
acquiring historical electricity load data of the charging piles under different categories according to the characteristic category labels of the charging piles obtained in the step S2, and generating historical period electricity time sequence data; the historical time period power utilization time sequence data comprise historical time date, holiday time, charging pile power utilization data and meteorological data;
and preprocessing abnormal data and missing data in the obtained historical period power utilization time sequence data:
for anomalous data: searching abnormal data by adopting a 3sigma principle, and filling the abnormal data by adopting a mean value filling method;
for missing data: filling is carried out by adopting a mean value filling method.
And step S4, constructing an initial model for predicting the electric load of the charging pile according to the influence factors of the load of the charging pile, and specifically comprising the following steps:
the following formula is adopted as an initial charge pile electricity load prediction model y (t):
y(t)=g(t)+s(t)+p·h(t)+ε(t)
wherein g (t) is a trend function, andc is a bearing capacity parameter of the model, alpha is an increase rate of the model, and beta is an offset of the model; s (t) is a periodic function, and +.>N is the Fourier order, A n Is the first intermediate parameter and +.>P is a period value, B n Is a second intermediate parameter and(a 1 ,...,a N ) And (b) 1 ,...,b N ) The model parameters to be predicted are; p is the factor value of the holiday effect on the charging pile, h (t) is the holiday function and +.>L is the number of holidays, τ i For the factor value of the influence of the ith holiday on the load, < >>Days for the ith holiday; epsilon (t) is the error term.
The evaluation in step S5 specifically includes the following steps:
and calculating a root mean square error index and an average absolute error index of the trained charging pile power load prediction initial model by adopting evaluation data:
the root mean square error index RMSE is calculated using the following equation:
the average absolute error index MAE is calculated using the following equation:
wherein h is the number of charging piles; y is g Is the actual value at time g;is the predicted value of the g time;
when evaluating, the method comprises the following steps:
if the root mean square error index RMSE and the average absolute error index MAE both meet the set requirements, the trained electric load prediction initial model for the charging pile is considered to meet the set requirements, and the evaluation model is used as a final electric load prediction model for the charging pile;
if the root mean square error index (RMSE) or the average absolute error index (MAE) does not meet the set requirement, the trained initial model of the electric load prediction of the charging pile is determined to not meet the set requirement, and the initial model of the electric load prediction of the charging pile is trained again.
The invention also provides a system for realizing the charging pile electricity load prediction method based on the electricity time sequence data, which comprises a data acquisition module, a label acquisition module, a training set acquisition module, a model construction module, a model training module and a load prediction module; the data acquisition module, the label acquisition module, the training set acquisition module, the model construction module, the model training module and the load prediction module are sequentially connected in series; the data acquisition module is used for acquiring the data information of the charging pile in the target area and uploading the data to the tag acquisition module; the tag acquisition module is used for mapping the acquired data information into feature vectors and clustering the feature vectors in a data dictionary mapping mode according to the received data information, so as to obtain a charging pile feature type tag, and uploading the data to the training set acquisition module; the training set acquisition module is used for acquiring historical electricity load data of the charging piles of the target area according to the received data information and the characteristic class labels of the charging piles, processing the historical electricity load data to obtain a training data set, and uploading the data to the model construction module; the model construction module is used for constructing an initial model of the charging pile power load prediction according to the received data information and the influence factors of the charging pile load, and uploading the data to the model training module; the model training module is used for training and evaluating the constructed initial model of the electric load prediction of the charging pile by adopting a training data set according to the received data information to obtain a final model of the electric load prediction of the charging pile, and uploading the data to the load prediction module; and the load prediction module is used for predicting the power consumption load of the charging pile based on the power consumption time sequence data by adopting a power consumption load prediction model of the charging pile according to the received data information.
According to the charging pile electricity load prediction method and system based on the electricity time sequence data, the charging pile attribute is analyzed to obtain the charging pile characteristic label, then the charging piles are clustered, a prediction model is generated based on the training of the charging pile electricity time sequence data, and finally the classification prediction of the charging pile electricity load is realized by adopting the prediction model; therefore, the method can not only realize the prediction of the power consumption load of the charging pile based on the power consumption time sequence data, but also has higher reliability, better accuracy and more objectivity and science.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
FIG. 2 is a schematic diagram of functional modules of the system of the present invention.
Detailed Description
A schematic process flow diagram of the method of the present invention is shown in fig. 1: the invention discloses a charging pile electricity load prediction method based on electricity time sequence data, which comprises the following steps:
s1, acquiring charging pile data information of a target area; the method specifically comprises the following steps:
acquiring distribution and attribute information of charging piles; the distribution and attribute information comprises a charging pile scale, a service vehicle type and a station type;
acquiring characteristic data of an area where the charging pile is located; the characteristic data of the region comprises a service area, a residential area and a business district;
acquiring characteristic data of a radio station area where the charging pile is located; the characteristic data of the station area comprises the type of the station area, the number of users in the station area and the power supply voltage;
carrying out association processing on the acquired distribution and attribute information, the characteristic data of the region and the characteristic data of the station area to obtain a charging pile distribution characteristic sample label set;
s2, mapping the data information obtained in the step S1 into feature vectors by adopting a data dictionary mapping mode and clustering, so as to obtain a charging pile feature type label; the method specifically comprises the following steps:
according to the charging pile distribution characteristic sample label set obtained in the step S1, adopting a dictionary mapping mode to form characteristic vectors from label data with different dimensions;
adopting a K-means algorithm to perform clustering division on the charging piles:
the number k of the clustered categories is calculated by adopting the following formula:
wherein argmax is the maximum value;is a classification index, and->tr () is the trace of the matrix, B k Is covariance matrix among categories, W k A covariance matrix in the category, num is the number of samples, and k is the number of categories;
forming final charging pile characteristic category labels, such as residential areas/slow charging, expressway servers/fast charging, business circles/slow charging and the like, according to the obtained category number k;
s3, acquiring historical electricity load data of the charging piles of the target area according to the characteristic class labels of the charging piles obtained in the step S2, and processing the historical electricity load data to obtain a training data set; the method specifically comprises the following steps:
acquiring historical electricity load data of the charging piles under different categories according to the characteristic category labels of the charging piles obtained in the step S2, and generating historical period electricity time sequence data; the historical time period power utilization time sequence data comprise historical time date, holiday time, charging pile power utilization data and meteorological data;
and preprocessing abnormal data and missing data in the obtained historical period power utilization time sequence data:
for anomalous data: searching abnormal data by adopting a 3sigma principle, and filling the abnormal data by adopting a mean value filling method;
for missing data: filling by adopting a mean value filling method;
s4, constructing an initial charge pile electricity load prediction model according to influence factors of charge pile loads; the method specifically comprises the following steps:
the following formula is adopted as an initial charge pile electricity load prediction model y (t):
y(t)=g(t)+s(t)+p·h(t)+ε(t)
wherein g (t) is a trend function for representing the non-periodic trend in the power consumption time sequence, such as the rising or falling of the power consumptionC is a bearing capacity parameter of the model, alpha is an increase rate of the model, and beta is an offset of the model; s (t) is a periodic function and is used for representing the periodic transformation or seasonal variation trend of the electricity consumption, for example, the periodic transformation is carried out according to the year and month, the electricity consumption time sequence data comprises periodic fluctuation of periodic types such as the year, month and the like, a Fourier sequence is adopted to establish a periodic function model, and the periodic function model is constructed>N is the fourier order, the larger N is the more variable the periodic pattern is, n=10 denotes the period in yearsSex variation, n=3 represents a periodic variation in units of weeks, a n Is the first intermediate parameter and +.>P is a period value (p=365.25 when the period is 1 year, p=7 when the period is 1 week), B n Is the second intermediate parameter and->(a 1 ,...,a N ) And (b) 1 ,...,b N ) The model parameters to be predicted are; p is the factor value of the holiday effect on the charging pile, h (t) is the holiday function for representing the irregular time node effect of holiday, burst time and the like in the power utilization time sequence data, and->L is the number of holidays, τ i For the factor value of the influence of the ith holiday on the load, < >>Days for the ith holiday; epsilon (t) is the error term;
s5, training and evaluating the initial model of the electric load prediction for the charging pile constructed in the step S4 by adopting the training data set obtained in the step S3 to obtain a final model of the electric load prediction for the charging pile;
the method specifically comprises the following steps of:
and calculating a root mean square error index and an average absolute error index of the trained charging pile power load prediction initial model by adopting evaluation data:
the root mean square error index RMSE is calculated using the following equation:
the average absolute error index MAE is calculated using the following equation:
wherein h is the number of charging piles; y is g Is the actual value at time g;is the predicted value of the g time;
when evaluating, the method comprises the following steps:
if the root mean square error index RMSE and the average absolute error index MAE both meet the set requirements, the trained electric load prediction initial model for the charging pile is considered to meet the set requirements, and the evaluation model is used as a final electric load prediction model for the charging pile;
if the root mean square error index (RMSE) or the average absolute error index (MAE) does not meet the set requirement, the trained charging pile electric load prediction initial model is determined to not meet the set requirement, and the charging pile electric load prediction initial model is trained again;
s6, adopting the charging pile electricity load prediction model obtained in the step S5 to conduct charging pile electricity load prediction based on electricity time sequence data.
FIG. 2 is a schematic diagram of functional modules of the system of the present invention: the system for realizing the charging pile electricity load prediction method based on the electricity time sequence data comprises a data acquisition module, a label acquisition module, a training set acquisition module, a model construction module, a model training module and a load prediction module; the data acquisition module, the label acquisition module, the training set acquisition module, the model construction module, the model training module and the load prediction module are sequentially connected in series; the data acquisition module is used for acquiring the data information of the charging pile in the target area and uploading the data to the tag acquisition module; the tag acquisition module is used for mapping the acquired data information into feature vectors and clustering the feature vectors in a data dictionary mapping mode according to the received data information, so as to obtain a charging pile feature type tag, and uploading the data to the training set acquisition module; the training set acquisition module is used for acquiring historical electricity load data of the charging piles of the target area according to the received data information and the characteristic class labels of the charging piles, processing the historical electricity load data to obtain a training data set, and uploading the data to the model construction module; the model construction module is used for constructing an initial model of the charging pile power load prediction according to the received data information and the influence factors of the charging pile load, and uploading the data to the model training module; the model training module is used for training and evaluating the constructed initial model of the electric load prediction of the charging pile by adopting a training data set according to the received data information to obtain a final model of the electric load prediction of the charging pile, and uploading the data to the load prediction module; and the load prediction module is used for predicting the power consumption load of the charging pile based on the power consumption time sequence data by adopting a power consumption load prediction model of the charging pile according to the received data information.
Claims (7)
1. A charging pile electricity load prediction method based on electricity time sequence data comprises the following steps:
s1, acquiring charging pile data information of a target area;
s2, mapping the data information obtained in the step S1 into feature vectors by adopting a data dictionary mapping mode and clustering, so as to obtain a charging pile feature type label;
s3, acquiring historical electricity load data of the charging piles of the target area according to the characteristic class labels of the charging piles obtained in the step S2, and processing the historical electricity load data to obtain a training data set;
s4, constructing an initial charge pile electricity load prediction model according to influence factors of charge pile loads;
s5, training and evaluating the initial model of the electric load prediction for the charging pile constructed in the step S4 by adopting the training data set obtained in the step S3 to obtain a final model of the electric load prediction for the charging pile;
s6, adopting the charging pile electricity load prediction model obtained in the step S5 to conduct charging pile electricity load prediction based on electricity time sequence data.
2. The method for predicting the power consumption load of the charging pile based on the power consumption time sequence data according to claim 1, wherein the step S1 of obtaining the charging pile data information of the target area specifically comprises the following steps:
acquiring distribution and attribute information of charging piles; the distribution and attribute information comprises a charging pile scale, a service vehicle type and a station type;
acquiring characteristic data of an area where the charging pile is located; the characteristic data of the region comprises a service area, a residential area and a business district;
acquiring characteristic data of a radio station area where the charging pile is located; the characteristic data of the station area comprises the type of the station area, the number of users in the station area and the power supply voltage;
and carrying out association processing on the acquired distribution and attribute information, the characteristic data of the region and the characteristic data of the station area to obtain a charging pile distribution characteristic sample label set.
3. The method for predicting the power consumption load of the charging pile based on the power consumption time sequence data according to claim 2, wherein the method for mapping the data dictionary in the step S2 is characterized in that the data information obtained in the step S1 is mapped into feature vectors and clustered, so as to obtain the charging pile feature class label, and specifically comprises the following steps:
according to the charging pile distribution characteristic sample label set obtained in the step S1, adopting a dictionary mapping mode to form characteristic vectors from label data with different dimensions;
adopting a K-means algorithm to perform clustering division on the charging piles:
the number k of the clustered categories is calculated by adopting the following formula:
wherein argmax is the maximum value;is a classification index, and->tr () is the trace of the matrix, B k Is covariance matrix among categories, W k A covariance matrix in the category, num is the number of samples, and k is the number of categories;
and forming a final charging pile characteristic category label according to the obtained category number k.
4. The method for predicting the electric load of the charging pile based on the electric time sequence data according to claim 3, wherein the step S3 is characterized in that the charging pile feature class label obtained according to the step S2 obtains the historical electric load data of the charging pile of the target area, and processes the historical electric load data to obtain the training data set, and specifically comprises the following steps:
acquiring historical electricity load data of the charging piles under different categories according to the characteristic category labels of the charging piles obtained in the step S2, and generating historical period electricity time sequence data; the historical time period power utilization time sequence data comprise historical time date, holiday time, charging pile power utilization data and meteorological data;
and preprocessing abnormal data and missing data in the obtained historical period power utilization time sequence data:
for anomalous data: searching abnormal data by adopting a 3sigma principle, and filling the abnormal data by adopting a mean value filling method;
for missing data: filling is carried out by adopting a mean value filling method.
5. The method for predicting the electric load of the charging pile based on the electric time sequence data according to claim 4 is characterized in that in the step S4, an initial model for predicting the electric load of the charging pile is constructed according to the influence factors of the load of the charging pile, and the method specifically comprises the following steps:
the following formula is adopted as an initial charge pile electricity load prediction model y (t):
y(t)=g(t)+s(t)+p·h(t)+ε(t)
wherein g (t) is a trend function, andc is a bearing capacity parameter of the model, alpha is an increase rate of the model, and beta is an offset of the model; s (t) is a periodic function, and +.>N is the Fourier order, A n Is the first intermediate parameter and +.>P is a period value, B n Is a second intermediate parameter and(a 1 ,...,a N ) And (b) 1 ,...,b N ) The model parameters to be predicted are; p is the factor value of the holiday effect on the charging pile, h (t) is the holiday function and +.>L is the number of holidays, τ i For the factor value of the influence of the ith holiday on the load, < >>Days for the ith holiday; epsilon (t) is the error term.
6. The method for predicting the electric load of the charging pile based on the electric time sequence data as set forth in claim 5, wherein the evaluation in step S5 specifically includes the steps of:
and calculating a root mean square error index and an average absolute error index of the trained charging pile power load prediction initial model by adopting evaluation data:
the root mean square error index RMSE is calculated using the following equation:
the average absolute error index MAE is calculated using the following equation:
wherein h is the number of charging piles; y is g Is the actual value at time g;is the predicted value of the g time;
when evaluating, the method comprises the following steps:
if the root mean square error index RMSE and the average absolute error index MAE both meet the set requirements, the trained electric load prediction initial model for the charging pile is considered to meet the set requirements, and the evaluation model is used as a final electric load prediction model for the charging pile;
if the root mean square error index (RMSE) or the average absolute error index (MAE) does not meet the set requirement, the trained initial model of the electric load prediction of the charging pile is determined to not meet the set requirement, and the initial model of the electric load prediction of the charging pile is trained again.
7. A system for realizing the electric load prediction method of the charging pile based on the electric time sequence data according to one of claims 1 to 6, which is characterized by comprising a data acquisition module, a label acquisition module, a training set acquisition module, a model construction module, a model training module and a load prediction module; the data acquisition module, the label acquisition module, the training set acquisition module, the model construction module, the model training module and the load prediction module are sequentially connected in series; the data acquisition module is used for acquiring the data information of the charging pile in the target area and uploading the data to the tag acquisition module; the tag acquisition module is used for mapping the acquired data information into feature vectors and clustering the feature vectors in a data dictionary mapping mode according to the received data information, so as to obtain a charging pile feature type tag, and uploading the data to the training set acquisition module; the training set acquisition module is used for acquiring historical electricity load data of the charging piles of the target area according to the received data information and the characteristic class labels of the charging piles, processing the historical electricity load data to obtain a training data set, and uploading the data to the model construction module; the model construction module is used for constructing an initial model of the charging pile power load prediction according to the received data information and the influence factors of the charging pile load, and uploading the data to the model training module; the model training module is used for training and evaluating the constructed initial model of the electric load prediction of the charging pile by adopting a training data set according to the received data information to obtain a final model of the electric load prediction of the charging pile, and uploading the data to the load prediction module; and the load prediction module is used for predicting the power consumption load of the charging pile based on the power consumption time sequence data by adopting a power consumption load prediction model of the charging pile according to the received data information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311488052.3A CN117458475A (en) | 2023-11-09 | 2023-11-09 | Charging pile electricity load prediction method and system based on electricity time sequence data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311488052.3A CN117458475A (en) | 2023-11-09 | 2023-11-09 | Charging pile electricity load prediction method and system based on electricity time sequence data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117458475A true CN117458475A (en) | 2024-01-26 |
Family
ID=89588983
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311488052.3A Pending CN117458475A (en) | 2023-11-09 | 2023-11-09 | Charging pile electricity load prediction method and system based on electricity time sequence data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117458475A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118194055A (en) * | 2024-05-14 | 2024-06-14 | 国网江西省电力有限公司信息通信分公司 | Charging pile power curve matching method |
-
2023
- 2023-11-09 CN CN202311488052.3A patent/CN117458475A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118194055A (en) * | 2024-05-14 | 2024-06-14 | 国网江西省电力有限公司信息通信分公司 | Charging pile power curve matching method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | Research and application of a hybrid forecasting framework based on multi-objective optimization for electrical power system | |
CN113962364B (en) | Multi-factor power load prediction method based on deep learning | |
CN109919353B (en) | Distributed photovoltaic prediction method of ARIMA model based on spatial correlation | |
CN110909912B (en) | Park electric power system net load combination prediction method based on self-adaptive error feedback | |
CN110443413B (en) | Electric power material demand prediction system and construction method of electric power material demand model | |
CN114792156B (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
CN109978242A (en) | The photovoltaic power generation cluster power forecasting method and device of scale are risen based on statistics | |
CN117458475A (en) | Charging pile electricity load prediction method and system based on electricity time sequence data | |
CN113554466A (en) | Short-term power consumption prediction model construction method, prediction method and device | |
CN112149890A (en) | Comprehensive energy load prediction method and system based on user energy label | |
CN113255900A (en) | Impulse load prediction method considering improved spectral clustering and Bi-LSTM neural network | |
CN114021848A (en) | Generating capacity demand prediction method based on LSTM deep learning | |
CN116307076A (en) | Industrial park energy efficiency management and control method based on Internet of things | |
CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
CN115481918A (en) | Active sensing and predictive analysis system for unit state based on source network load storage | |
CN116826710A (en) | Peak clipping strategy recommendation method and device based on load prediction and storage medium | |
CN105279582B (en) | Super short-period wind power prediction technique based on dynamic correlation feature | |
CN108694475A (en) | Short-term time scale photovoltaic cell capable of generating power amount prediction technique based on mixed model | |
CN116070458A (en) | New wind power plant scene generation method based on RAC-GAN | |
CN114372835B (en) | Comprehensive energy service potential customer identification method, system and computer equipment | |
Ananthu et al. | A study of 100kWp PV Plant Output Power Forecasting: A case study | |
CN115907228A (en) | Short-term power load prediction analysis method based on PSO-LSSVM | |
CN112348235B (en) | Wind-solar bus load self-adaptive prediction method, device and computer equipment | |
CN115759469A (en) | Method and device for predicting peak load and occurrence time thereof | |
Kuang et al. | Short-term power load forecasting method in rural areas based on cnn-lstm |
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 |