CN116702978A - Electric vehicle charging load prediction method and device considering emergency characteristics - Google Patents
Electric vehicle charging load prediction method and device considering emergency characteristics Download PDFInfo
- Publication number
- CN116702978A CN116702978A CN202310673236.0A CN202310673236A CN116702978A CN 116702978 A CN116702978 A CN 116702978A CN 202310673236 A CN202310673236 A CN 202310673236A CN 116702978 A CN116702978 A CN 116702978A
- Authority
- CN
- China
- Prior art keywords
- emergency
- load
- electric vehicle
- optimal
- considering
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 27
- 238000012216 screening Methods 0.000 claims abstract description 21
- 230000008859 change Effects 0.000 claims abstract description 18
- 238000003062 neural network model Methods 0.000 claims abstract description 12
- 239000013598 vector Substances 0.000 claims description 18
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 10
- 230000005611 electricity Effects 0.000 claims description 6
- 238000005316 response function Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 5
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 4
- 230000002441 reversible effect Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 2
- 230000008901 benefit Effects 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000010248 power generation Methods 0.000 abstract description 3
- 241000287127 Passeridae Species 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000010845 search algorithm Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000001744 unit root test Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for predicting the charging load of an electric automobile by considering the characteristics of emergency, wherein the method comprises the following steps: performing feature screening on conventional influencing factors to obtain an optimal conventional feature set; screening the characteristics considering the emergency, and obtaining an optimal characteristic set considering the emergency; and (3) establishing an electric vehicle load prediction model considering the emergency, and predicting the electric vehicle charging load by combining the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data. The electric vehicle load under the emergency can be accurately predicted by constructing a conventional influencing factor feature set and an emergency optimal feature set and predicting through an SSA-BiGRU-CNN neural network model and simultaneously considering the load change caused by the emergency, and the load fluctuation caused by the emergency is positively responded, so that the electric vehicle load prediction method is beneficial for an electric company to reasonably formulate a power generation plan, reduces the power grid cost, improves the satisfaction degree of charging users and improves the economic benefit of a charging station.
Description
Technical Field
The invention belongs to the technical field of electric vehicle charging load prediction methods, relates to an electric vehicle charging load prediction method considering emergency characteristics, and further relates to an electric vehicle charging load prediction device considering the emergency characteristics.
Background
The explosive growth of new energy electric vehicles forms a great test for the stable operation of a power grid. Therefore, developing efficient and accurate electric vehicle charging load prediction is a precondition for safe and stable operation of the power grid.
At present, main research methods for predicting the charging load of an electric automobile are divided into two main categories: model-based and data-based prediction methods. The former establishes a probability model by using mathematical statistics, and a Monte Carlo simulation method is adopted to predict on the basis. Compared with the method, the method has the advantages that the charging load prediction of the electric automobile is more movable by means of the data driving method, and the prediction cost can be reduced. The development of the internet of things promotes the development of a large number of cloud-based electric automobile services, and a data integration platform is established in the province of China. In this context, data-driven based prediction methods have received more attention. Both the above prediction methods only consider some conventional electric vehicle load influencing factors, but do not consider the influence of an emergency on the electric vehicle load. Because the emergency event is aperiodic, it has contingency and persistence to the impact of the electric vehicle load. The sudden event can impact the power grid, and serious electric accidents can be caused when the sudden event is serious.
Disclosure of Invention
The invention aims to provide an electric vehicle charging load prediction method considering the characteristics of an emergency, which solves the problem that the impact of the emergency on the electric vehicle load on a power grid is not considered in the prior art.
The technical scheme adopted by the invention is that the electric vehicle charging load prediction method considering the characteristics of the emergency event comprises the following steps:
step 1, screening the characteristics of conventional influencing factors to obtain an optimal conventional characteristic set;
step 2, screening the characteristics considering the emergency, and obtaining an optimal characteristic set considering the emergency;
and 3, establishing an electric vehicle load prediction model considering the emergency, and predicting the electric vehicle charging load by combining the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data.
The invention is also characterized in that:
conventional influencing factors include weather, electricity prices, date type, historical load data.
The specific process of the step 1 is as follows: firstly carrying out dimensionless treatment on conventional influence factor data, then carrying out feature selection on the treated conventional influence factors by adopting a MIC method, and then carrying out redundancy treatment on the selected features by adopting an mRMR method to obtain an optimal conventional feature set.
The specific process of the step 2 is as follows: and analyzing the importance degree of the emergency social influence, the participated crowd and the traffic control condition on the change of the charging requirement of the electric automobile and the forward and reverse effects on the load data by adopting a vector autoregressive model to obtain an optimal feature set considering the emergency.
The specific process of the step 2 is as follows: firstly, determining the stability of a time sequence formed by social influence, participated crowd and traffic control conditions in an emergency; according to the VAR model, the relationship among the social influence, the participated crowd and the traffic control condition in the emergency is analyzed by using an impulse response function and variance decomposition, and the optimal feature set considering the emergency is obtained.
The specific process of the step 3 is as follows: an SSA-BiGRU-CNN neural network model is built, and the optimal conventional feature set, the optimal feature set considering the emergency and the historical load data are input into the SSA-BiGRU-CNN neural network model for prediction, so that the electric vehicle load is obtained.
The SSA-BiGRU-CNN neural network model comprises the following processing procedures: the BiGRU layer extracts time features of historical load data to obtain two hidden state vectors with past and future information, inputs the hidden state vectors into the CNN layer, captures important local relations through the convolution layer and the pooling layer, and outputs the hidden state vectors through the full connection layer to obtain the electric automobile load.
The specific process of the step 3 is as follows: when an emergency happens, real load data of the emergency is found in the historical load data, then the emergency is assumed to not happen, the optimal conventional feature set and the historical load data are input into an SSA-BiGRU-CNN neural network for prediction, a predicted load sequence when the emergency does not happen is obtained, the predicted load sequence when the emergency does not happen is subtracted by the real load data, and an electric vehicle charging demand change amount historical value caused by the emergency is obtained; when the next emergency is about to happen, inputting an optimal feature set and an electric vehicle charging demand change amount historical value which are considered in the emergency into an SSA-BiGRU-CNN neural network for prediction to obtain a predicted value of the electric vehicle charging demand change amount caused by the emergency; and simultaneously inputting the optimal conventional feature set and the historical load data of the current time into an SSA-BiGRU-CNN neural network for prediction to obtain an electric vehicle load value considering conventional influence factors at the current moment, and adding or subtracting the electric vehicle load value considering the conventional influence factors at the current moment from the predicted value of the electric vehicle charging demand change quantity caused by the emergency to obtain the electric vehicle load.
The VAR model is expressed as follows:
y t ＝C+β 1 y t-1 +β 2 y t-2 +…+β p y t-p +ε t (7)；
in the above, y t As endogenous vector beta p For the matrix to be estimated, C is the model constant, ε t White noise, representing a disturbance vector;
determining an optimal model and hysteresis order based on AIC, HQ:
another object of the present invention is to provide an electric vehicle charging load prediction apparatus considering characteristics of an emergency.
The invention adopts another technical scheme that the electric automobile charging load prediction device considering the characteristics of emergency comprises:
the first feature screening module is used for carrying out feature screening on the conventional influencing factors to obtain an optimal conventional feature set;
the second feature screening module is used for screening the features considering the emergency to obtain an optimal feature set considering the emergency;
the load prediction module is used for establishing an electric vehicle load prediction model considering the emergency, and predicting the electric vehicle charging load by combining the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data.
The beneficial effects of the invention are as follows: the invention relates to an electric vehicle charging load prediction method considering the characteristics of an emergency, which is characterized in that a conventional influencing factor characteristic set and an emergency optimal characteristic set are constructed, the prediction is carried out through an SSA-BiGRU-CNN neural network model, and meanwhile, the load change caused by the emergency is considered, so that the electric vehicle load under the emergency can be accurately predicted, the load fluctuation caused by the emergency is positively responded, the reasonable power generation plan of an electric company is facilitated, the power grid cost is reduced, the satisfaction of a charging user is improved, and the economic benefit of a charging station is improved.
Drawings
Fig. 1 is a flowchart of an electric vehicle charging load prediction method considering an emergency feature of the present invention.
Detailed Description
The invention will be described in detail below with reference to the drawings and the detailed description.
Example 1
The electric vehicle charging load prediction method considering the characteristics of the emergency event comprises the following steps:
step 1, screening the characteristics of conventional influencing factors to obtain an optimal conventional characteristic set;
step 2, screening the characteristics considering the emergency, and obtaining an optimal characteristic set considering the emergency;
and 3, establishing an electric vehicle load prediction model considering the emergency, and predicting the electric vehicle charging load by combining the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data.
Example 2
The electric vehicle charging load prediction method considering the emergency characteristics, as shown in fig. 1, comprises the following steps:
step 1, screening the characteristics of conventional influencing factors to obtain an optimal conventional characteristic set; conventional influencing factors include weather, electricity price, date type, load history load data, as shown in table 1:
TABLE 1 conventional influencing factor feature set
Step 1.1, use P in Table t 、DL t For example, P 0 Indicating the current electricity price, P 1 Representing electricity prices before 1 h; DL (DL) 1 Representing the load value, DL, at the same time before 1 day 2 The load value at the same time before 2 days is shown. And carrying out dimensionless treatment on the conventional influence factor data to ensure that the data have the same specification and accelerate the convergence of the neural network. In this embodiment, the date type sequence defines a workday as 0, a double holiday as 1, and holidaysDay is defined as 2 to distinguish load characteristics at different date types. The electric vehicle load, weather sequence and electricity price data can be normalized to [ -0.5,0.5]Intervals to achieve dimensionless, unlike common [0,1 ]]Since the neural network favors the input of data centered around 0, setting the center of the normalized interval to 0 favors the convergence of the neural network, and the dimensionless formula is:
wherein d max And d min Respectively, the maximum and minimum of data d.
Step 1.2, the invention considers the correlation between the conventional influencing factors and the electric vehicle load, and considers the linear relation and the nonlinear relation at the same time, and can select the MIC method to judge the correlation between the 2 influencing factor sequences, and the method specifically comprises the following steps:
MIC was calculated by the following formula:
wherein d x And d y Values of sequences x and y, respectively, I (·) is a mutual information function, p (·) is a probability density distribution function, and a and b are d, respectively X And d Y Number of discretization in direction, I MIC (x, y) is the MIC of the sequences x and y.
When the sequence is discrete data and the distribution is very uneven, the phenomenon that the MIC is not 1 is likely to occur with the MIC, and the phenomenon belongs to normal conditions and does not influence the conclusion. A subset of features with MIC values greater than 0.6 is selected from high to low according to the MIC method.
And 1.3, because a feature subset selected by the MIC method has a lot of redundant information, selecting the mRMR method on the basis of the MIC to penalize the redundant feature with higher correlation in the selected features. Of all feature sequences, a new feature sequence is incrementally selected, each time a locally optimal feature is selected.
Defining D (S, y) as the correlation between all features and the target variable y, and R (S) as the redundancy of all features, wherein S is the feature set formed by all features together, namely:
wherein d i For the ith feature sequence, m is the number of feature sequences in the final selected feature set, I mRMR mRMR values for the signature sequences. And obtaining a final feature subset by solving the optimization problem formula (6). The invention combines MIC and mRMR, which is more beneficial to selecting conventional factors influencing the load of the electric automobile.
And 2, analyzing the importance degree of the emergency social influence, the participators and the traffic control condition on the change of the charging requirement of the electric automobile and the forward and reverse effects on the load data by adopting a vector autoregressive model, namely, increasing or decreasing the charging requirement of the electric automobile, and constructing an optimal feature set considering the emergency. The social influence of the emergency is obtained through the data volume retrieved by hundred degrees, the number of people participating in the emergency is the number of people actually participating in the emergency, and the traffic control condition is the proportion of the area of the controlled area to the area of the whole area released by authorities. The set of incident influencing factors is shown in table 2:
TABLE 2 Emergency factor feature set
Step 2.1, firstly determining the stability of a time sequence formed by social influence, participated crowd and traffic control conditions in an emergency; specifically, ADF, PP and KPSS unit root tests were performed on the time series, and if the test results met a range of 5% significance, the time series was considered stationary.
Step 2.2, building a VAR model:
in the above, y t For endogenous vectors, C is a model constant, ε t White noise, representing a disturbance vector;
determining an optimal model and hysteresis order based on AIC, HQ:
in the above formula, p is the hysteresis order of the VAR model, n is the sample class, and T is the sample size.
And 2.3, analyzing the relationship among the social influence, the participated crowd and the traffic control condition in the emergency by using an impulse response function and variance decomposition according to the VAR model to obtain an optimal feature set considering the emergency. Specifically, the impulse response function is used for analyzing the dynamic feedback of the reduction of the charging requirement caused by unit impact of the impulse response function, the impulse response function is supplemented through variance decomposition, and finally an optimal feature set considering the emergency is constructed.
And 3, constructing an SSA-BiGRU-CNN neural network model, and inputting an optimal conventional characteristic set, an optimal characteristic set considering an emergency and historical load data into the BiGRU-CNN hybrid neural network for prediction to obtain the electric vehicle load.
Specifically, when an emergency occurs, the real load data of the emergency is found in the historical load data, then the emergency is assumed to not occur, the optimal conventional characteristic set and the historical load data are input into the SSA-BiGRU-CNN neural network for prediction, a predicted load sequence when the emergency does not occur is obtained, the predicted load sequence when the emergency does not occur is subtracted by the real load data, and the historical value of the electric vehicle charging demand change amount caused by the emergency is obtained; when the next emergency is about to happen, inputting an optimal feature set and an electric vehicle charging demand change amount historical value which are considered in the emergency into an SSA-BiGRU-CNN neural network for prediction to obtain a predicted value of the electric vehicle charging demand change amount caused by the emergency; and simultaneously inputting the optimal conventional feature set and the historical load data of the current time into an SSA-BiGRU-CNN neural network for prediction to obtain an electric vehicle load value considering conventional influence factors at the current moment, and adding or subtracting the electric vehicle load value considering the conventional influence factors at the current moment from the predicted value of the electric vehicle charging demand change quantity caused by the emergency to obtain the electric vehicle load. In this embodiment, the SSA-BiGRU-CNN neural network model: and optimizing super parameters such as batch processing, learning rate, hidden layer number, layer neuron number, convolution kernel number, step length and the like in the BiGRU-CNN neural network by using a sparrow search algorithm, and finding out optimal parameters.
The SSA-BiGRU-CNN neural network model has the working principle that: the BiGRU layer extracts time features of the historical load data to obtain two hidden state vectors with past and future information, inputs the hidden state vectors into the CNN layer, captures important local relations through the convolution layer and the pooling layer, and performs local resource integration through the full connection layer to obtain a prediction result.
Furthermore, the CNN is composed of three layers of a convolution layer, a pooling layer and a full connection layer, wherein the convolution layer extracts effective resources in data from input data through a plurality of convolution kernels, the pooling layer reserves strong features and discards weak features, and the full connection layer integrates all local resources together to form a global resource, so that a prediction result is obtained.
Sparrow Search Algorithm (SSA)
The SSA algorithm is a novel group intelligent optimization calculation method which is provided by simulating the behavior of sparrow feeding and enemy avoidance. In the SSA algorithm, the whole sparrow population is divided into discoverers and joiners according to a certain share, and some discoverers and joiners are randomly selected, and the identities of the alerters are simultaneously doubled. The discoverers generally have a high fitness and a wide search range, which is mainly responsible for finding the location of food and providing the direction for the participants to find the food. As the energy of the participants becomes lower, they will follow the discoverer to go to other locations to find food to get more energy. When the whole population is threatened, the alerter can give an alarm to the whole population so as to ensure the safety of the sparrow population.
The basic construction of the biglu network (bi-directional gated recurrent neural network) model is as follows: for a time sequence to be trained, two GRU models are simultaneously arranged in the forward direction and the reverse direction, and hidden layer nodes of the two GRU models are connected to the same output layer. The method may provide complete history and future information for each point in time in the output layer input sequence. The working principle is as follows: the relation between the past load and the future load and the current load is learned, and the time characteristics of the historical load data are extracted to obtain two hidden state vectors with past and future information.
Example 3
An electric vehicle charging load prediction apparatus considering an emergency feature, comprising:
the first feature screening module is used for carrying out feature screening on the conventional influencing factors to obtain an optimal conventional feature set;
the second feature screening module is used for screening the features considering the emergency to obtain an optimal feature set considering the emergency;
the load prediction module is used for establishing an electric vehicle load prediction model considering the emergency, and predicting the electric vehicle charging load by combining the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data.
According to the method for predicting the electric vehicle charging load taking the sudden event characteristics into consideration, the conventional influencing factor characteristic set and the sudden event characteristic set are constructed, the SSA-BiGRU-CNN neural network model is used for predicting, and meanwhile, the load change caused by the sudden event is considered, so that the electric vehicle charging load under the sudden event can be accurately predicted, the load fluctuation caused by the sudden event is positively responded, the reasonable power generation plan of an electric company is facilitated, the power grid cost is reduced, the satisfaction of charging users is improved, and the economic benefit of a charging station is improved.
Claims (10)
1. The electric vehicle charging load prediction method considering the emergency characteristics is characterized by comprising the following steps of:
step 1, screening the characteristics of conventional influencing factors to obtain an optimal conventional characteristic set;
step 2, screening the characteristics considering the emergency, and obtaining an optimal characteristic set considering the emergency;
and step 3, establishing an electric vehicle load prediction model considering the emergency, and predicting the electric vehicle charging load by combining the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data.
2. The method for predicting the charge load of an electric vehicle taking into account the characteristics of an emergency event according to claim 1, wherein the conventional influencing factors include weather, electricity price, date type, and historical load data.
3. The method for predicting the charging load of the electric vehicle taking the emergency characteristics into consideration as set forth in claim 1, wherein the specific process of step 1 is as follows: firstly carrying out dimensionless treatment on the conventional influence factor data, then carrying out feature selection on the treated conventional influence factor by adopting a MIC method, and then carrying out feature selection on the selected feature by adopting an mRMR method to obtain an optimal conventional feature set.
4. The method for predicting the charging load of the electric vehicle taking the emergency characteristics into consideration as set forth in claim 1, wherein the specific process of the step 2 is as follows: and analyzing the importance degree of the emergency social influence, the participated crowd and the traffic control condition on the change of the charging requirement of the electric automobile and the forward and reverse effects on the load data by adopting a vector autoregressive model to obtain an optimal feature set considering the emergency.
5. The method for predicting the charging load of the electric vehicle according to claim 1 or 4, wherein the specific process of step 2 is as follows: firstly, determining the stability of a time sequence formed by social influence, participated crowd and traffic control conditions in an emergency; according to the VAR model, the relationship among the social influence, the participated crowd and the traffic control condition in the emergency is analyzed by using an impulse response function and variance decomposition, and the optimal feature set considering the emergency is obtained.
6. The method for predicting the charging load of the electric vehicle taking the emergency characteristics into consideration as set forth in claim 1, wherein the specific process of the step 3 is as follows: and constructing an SSA-BiGRU-CNN neural network model, and inputting the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data into the SSA-BiGRU-CNN neural network model for prediction to obtain the electric vehicle load.
7. The method for predicting the charging load of the electric vehicle taking the emergency characteristics into consideration as set forth in claim 6, wherein the SSA-biglu-CNN neural network model is processed by: the BiGRU layer extracts time features of historical load data to obtain two hidden state vectors with past and future information, inputs the hidden state vectors into the CNN layer, captures important local relations through the convolution layer and the pooling layer, and outputs the hidden state vectors through the full connection layer to obtain the electric automobile load.
8. The method for predicting the charging load of the electric vehicle taking the emergency characteristics into consideration as set forth in claim 1, wherein the specific process of the step 3 is as follows: when an emergency happens, real load data of the emergency is found in the historical load data, then the emergency is assumed to not happen, the optimal conventional feature set and the historical load data are input into an SSA-BiGRU-CNN neural network for prediction, a predicted load sequence when the emergency does not happen is obtained, the predicted load sequence when the emergency does not happen is subtracted by the real load data, and an electric vehicle charging demand change amount historical value caused by the emergency is obtained; when the next emergency is about to happen, inputting an optimal feature set and an electric vehicle charging demand change amount historical value which are considered in the emergency into an SSA-BiGRU-CNN neural network for prediction to obtain a predicted value of the electric vehicle charging demand change amount caused by the emergency; and simultaneously inputting the optimal conventional feature set and the historical load data of the current time into an SSA-BiGRU-CNN neural network for prediction to obtain an electric vehicle load value considering conventional influence factors at the current moment, and adding or subtracting the electric vehicle load value considering the conventional influence factors at the current moment from the predicted value of the electric vehicle charging demand change quantity caused by the emergency to obtain the electric vehicle load.
9. The method for predicting the charge load of an electric vehicle taking into account the characteristics of an emergency as set forth in claim 5, wherein the VAR model is expressed as follows:
y t ＝C+β 1 y t-1 +β 2 y t-2 +…+β p y t-p +ε t (7)；
in the above, y t As endogenous vector beta p For the matrix to be estimated, C is the model constant, ε t White noise, representing a disturbance vector;
determining an optimal model and hysteresis order based on AIC, HQ:
10. electric automobile charge load prediction device of taking into account incident characteristic, its characterized in that includes:
the first feature screening module is used for carrying out feature screening on the conventional influencing factors to obtain an optimal conventional feature set;
the second feature screening module is used for screening the features considering the emergency to obtain an optimal feature set considering the emergency;
the load prediction module is used for establishing an electric vehicle load prediction model considering the emergency, and predicting the electric vehicle charging load by combining the optimal conventional characteristic set, the optimal characteristic set considering the emergency and the historical load data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310673236.0A CN116702978B (en) | 2023-06-07 | 2023-06-07 | Electric vehicle charging load prediction method and device considering emergency characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310673236.0A CN116702978B (en) | 2023-06-07 | 2023-06-07 | Electric vehicle charging load prediction method and device considering emergency characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116702978A true CN116702978A (en) | 2023-09-05 |
| CN116702978B CN116702978B (en) | 2024-02-13 |
Family
ID=87828732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310673236.0A Active CN116702978B (en) | 2023-06-07 | 2023-06-07 | Electric vehicle charging load prediction method and device considering emergency characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116702978B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140068515A (en) * | 2012-11-28 | 2014-06-09 | 고려대학교 산학협력단 | System and method for electric vehicle charging load forecasting |
| CN108446795A (en) * | 2018-02-28 | 2018-08-24 | 广东电网有限责任公司电力调度控制中心 | Power system load fluction analysis method, apparatus and readable storage medium storing program for executing |
| CN111476441A (en) * | 2020-05-29 | 2020-07-31 | 南方电网科学研究院有限责任公司 | Load prediction method for electric vehicle charging equipment and related device |
| CN112348168A (en) * | 2020-10-27 | 2021-02-09 | 国网四川省电力公司经济技术研究院 | Ultra-short-term load prediction method and system considering data loss and characteristic redundancy |
| CN112488372A (en) * | 2020-11-23 | 2021-03-12 | 国网新疆电力有限公司营销服务中心(资金集约中心、计量中心) | Double-layer optimized scheduling method for electric heating load under multiple time scales |
| WO2021073036A1 (en) * | 2019-10-15 | 2021-04-22 | 江苏大学 | Real-time global optimization intelligent control system and method for fuel cell bus |
| WO2021143075A1 (en) * | 2020-01-17 | 2021-07-22 | 南京东博智慧能源研究院有限公司 | Demand response method taking space-time distribution of electric vehicle charging loads into consideration |
| CN113505534A (en) * | 2021-07-07 | 2021-10-15 | 南京工程学院 | Load prediction method considering demand response |
| CN114418174A (en) * | 2021-12-13 | 2022-04-29 | 国网陕西省电力公司电力科学研究院 | Electric vehicle charging load prediction method |
| CN114925931A (en) * | 2022-06-10 | 2022-08-19 | 北京中恒博瑞数字电力科技有限公司 | Platform area load prediction method and system |
| CN115275977A (en) * | 2022-06-28 | 2022-11-01 | 绿色湾区(广东)能源服务有限公司 | Power load prediction method and device |
| CN115640889A (en) * | 2022-10-19 | 2023-01-24 | 南京邮电大学 | Power load prediction method based on multiple linear regression and improved LSTM |
| CN115660226A (en) * | 2022-12-13 | 2023-01-31 | 国网冀北电力有限公司 | Power load prediction model construction method and construction device based on digital twins |
| US20230051766A1 (en) * | 2021-08-13 | 2023-02-16 | Here Global B.V. | Method, apparatus, and computer program product for predicting electric vehicle charge point utilization |
| US20230074700A1 (en) * | 2021-08-25 | 2023-03-09 | State Grid Shanghai Electric Power Company | Prediction method for charging loads of electric vehicles with consideration of data correlation |
| CN115907122A (en) * | 2022-11-11 | 2023-04-04 | 国网陕西省电力公司电力科学研究院 | Regional electric vehicle charging load prediction method |
-
2023
- 2023-06-07 CN CN202310673236.0A patent/CN116702978B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140068515A (en) * | 2012-11-28 | 2014-06-09 | 고려대학교 산학협력단 | System and method for electric vehicle charging load forecasting |
| CN108446795A (en) * | 2018-02-28 | 2018-08-24 | 广东电网有限责任公司电力调度控制中心 | Power system load fluction analysis method, apparatus and readable storage medium storing program for executing |
| WO2021073036A1 (en) * | 2019-10-15 | 2021-04-22 | 江苏大学 | Real-time global optimization intelligent control system and method for fuel cell bus |
| WO2021143075A1 (en) * | 2020-01-17 | 2021-07-22 | 南京东博智慧能源研究院有限公司 | Demand response method taking space-time distribution of electric vehicle charging loads into consideration |
| CN111476441A (en) * | 2020-05-29 | 2020-07-31 | 南方电网科学研究院有限责任公司 | Load prediction method for electric vehicle charging equipment and related device |
| CN112348168A (en) * | 2020-10-27 | 2021-02-09 | 国网四川省电力公司经济技术研究院 | Ultra-short-term load prediction method and system considering data loss and characteristic redundancy |
| CN112488372A (en) * | 2020-11-23 | 2021-03-12 | 国网新疆电力有限公司营销服务中心(资金集约中心、计量中心) | Double-layer optimized scheduling method for electric heating load under multiple time scales |
| CN113505534A (en) * | 2021-07-07 | 2021-10-15 | 南京工程学院 | Load prediction method considering demand response |
| US20230051766A1 (en) * | 2021-08-13 | 2023-02-16 | Here Global B.V. | Method, apparatus, and computer program product for predicting electric vehicle charge point utilization |
| US20230074700A1 (en) * | 2021-08-25 | 2023-03-09 | State Grid Shanghai Electric Power Company | Prediction method for charging loads of electric vehicles with consideration of data correlation |
| CN114418174A (en) * | 2021-12-13 | 2022-04-29 | 国网陕西省电力公司电力科学研究院 | Electric vehicle charging load prediction method |
| CN114925931A (en) * | 2022-06-10 | 2022-08-19 | 北京中恒博瑞数字电力科技有限公司 | Platform area load prediction method and system |
| CN115275977A (en) * | 2022-06-28 | 2022-11-01 | 绿色湾区(广东)能源服务有限公司 | Power load prediction method and device |
| CN115640889A (en) * | 2022-10-19 | 2023-01-24 | 南京邮电大学 | Power load prediction method based on multiple linear regression and improved LSTM |
| CN115907122A (en) * | 2022-11-11 | 2023-04-04 | 国网陕西省电力公司电力科学研究院 | Regional electric vehicle charging load prediction method |
| CN115660226A (en) * | 2022-12-13 | 2023-01-31 | 国网冀北电力有限公司 | Power load prediction model construction method and construction device based on digital twins |
Non-Patent Citations (2)
| Title |
|---|
| 夏博等: "电力系统短期负荷预测方法研究综述", 电力大数据, vol. 21, no. 7, 31 July 2018 (2018-07-31), pages 22 - 28 * |
| 薛阳等: "基于UTCI-MIC与振幅压缩灰色模型的用户侧微电网短期负荷预测方法", 电网技术, vol. 44, no. 2, 29 February 2020 (2020-02-29), pages 556 - 563 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116702978B (en) | 2024-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Barak et al. | Forecasting energy consumption using ensemble ARIMA–ANFIS hybrid algorithm | |
| Tian et al. | Multi-step short-term wind speed prediction based on integrated multi-model fusion | |
| CN109800875A (en) | Chemical industry fault detection method based on particle group optimizing and noise reduction sparse coding machine | |
| CN110837915B (en) | Low-voltage load point prediction and probability prediction method for power system based on hybrid integrated deep learning | |
| Akgüngör et al. | An artificial intelligent approach to traffic accident estimation: Model development and application | |
| CN111539515A (en) | Complex equipment maintenance decision method based on fault prediction | |
| CN106855957A (en) | Factory's bus load prediction based on similar day and least square method supporting vector machine | |
| Kavousi-Fard et al. | Short term load forecasting of distribution systems by a new hybrid modified FA-backpropagation method | |
| Anbazhagan et al. | A neural network approach to day-ahead deregulated electricity market prices classification | |
| CN108241964A (en) | Capital construction scene management and control mobile solution platform based on BP artificial nerve network model algorithms | |
| CN112884008B (en) | Prediction evaluation method and device for running state of power information acquisition system | |
| Zheng et al. | Early prediction of cooling load in energy-efficient buildings through novel optimizer of shuffled complex evolution | |
| Zhou et al. | Deep learning-based rolling horizon unit commitment under hybrid uncertainties | |
| CN113780684A (en) | Intelligent building user energy consumption behavior prediction method based on LSTM neural network | |
| CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
| Jang et al. | Offline-online reinforcement learning for energy pricing in office demand response: lowering energy and data costs | |
| Tavares et al. | Comparison of PV power generation forecasting in a residential building using ANN and DNN | |
| Poczeta et al. | Application of fuzzy cognitive maps to multi-step ahead prediction of electricity consumption | |
| Li et al. | A multi-factor combination prediction model of carbon emissions based on improved CEEMDAN | |
| CN109408896A (en) | A kind of anerobic sowage processing gas production multi-element intelligent method for real-time monitoring | |
| CN116702978B (en) | Electric vehicle charging load prediction method and device considering emergency characteristics | |
| CN117313795A (en) | Intelligent building energy consumption prediction method based on improved DBO-LSTM | |
| CN116911459A (en) | Multi-input multi-output ultra-short-term power load prediction method suitable for virtual power plant | |
| CN111784019A (en) | Power load processing method and device | |
| CN116523001A (en) | Method, device and computer equipment for constructing weak line identification model of power grid |
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 |