CN113298298A - Charging pile short-term load prediction method and system - Google Patents
Charging pile short-term load prediction method and system Download PDFInfo
- Publication number
- CN113298298A CN113298298A CN202110507429.XA CN202110507429A CN113298298A CN 113298298 A CN113298298 A CN 113298298A CN 202110507429 A CN202110507429 A CN 202110507429A CN 113298298 A CN113298298 A CN 113298298A
- Authority
- CN
- China
- Prior art keywords
- prediction
- charging
- load
- data
- charging pile
- 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 92
- 238000010801 machine learning Methods 0.000 claims description 17
- 238000003062 neural network model Methods 0.000 claims description 17
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 14
- 238000009825 accumulation Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 6
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 239000005431 greenhouse gas Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training 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"
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L53/00—Methods of charging batteries, specially adapted for electric vehicles; Charging stations or on-board charging equipment therefor; Exchange of energy storage elements in electric vehicles
- B60L53/30—Constructional details of charging stations
- B60L53/31—Charging columns specially adapted for electric vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
-
- 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
-
- 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
- Y02T10/7072—Electromobility specific charging systems or methods for batteries, ultracapacitors, supercapacitors or double-layer capacitors
-
- 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
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02T90/10—Technologies relating to charging of electric vehicles
- Y02T90/12—Electric charging stations
-
- 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
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02T90/10—Technologies relating to charging of electric vehicles
- Y02T90/16—Information or communication technologies improving the operation of electric vehicles
- Y02T90/167—Systems integrating technologies related to power network operation and communication or information technologies for supporting the interoperability of electric or hybrid vehicles, i.e. smartgrids as interface for battery charging of electric vehicles [EV] or hybrid vehicles [HEV]
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S30/00—Systems supporting specific end-user applications in the sector of transportation
- Y04S30/10—Systems supporting the interoperability of electric or hybrid vehicles
- Y04S30/12—Remote or cooperative charging
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Power Engineering (AREA)
- Mechanical Engineering (AREA)
- Development Economics (AREA)
- Transportation (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Biophysics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses a method and a system for predicting short-term load of a charging pile, wherein the method for predicting the short-term load of the charging pile comprises the following steps: taking out actual load historical data in a certain time period and prediction data obtained by a plurality of prediction methods from a database; respectively calculating errors E of predicted loads of multiple prediction methods according to actual load historical data and the prediction data; comparing errors E of predicted loads of a plurality of prediction methods: and switching to the prediction method with the minimum error E. By the method and the device, different load prediction methods can be selected in different scenes, and the prediction accuracy is improved.
Description
Technical Field
The invention belongs to the field of charging pile load prediction, and particularly relates to a charging pile short-term load prediction method and a charging pile short-term load prediction system.
Background
In recent years, with the emphasis of the country on natural environment protection, the realization of low-carbon emission has become an important goal of realizing sustainable development in our country. Meanwhile, with the predatory development of nature by human beings, shortage of petroleum resources has become a problem of general worry of countries around the world. Under this background, an electric vehicle, as a new clean energy vehicle, plays an important role in improving energy safety, reducing emissions of greenhouse gases and other pollutants, and alleviating the energy shortage crisis. Therefore, the industrial layout of the electric automobile is accelerated in China, encouragement policies are successively introduced, and the electric automobile is accelerated to be developed by various automobile companies, so that the electric automobile can be expected to become a mainstream travel mode in the near future.
Along with the quantity of electric automobile increases gradually, electric automobile fills electric pile and builds as the most crucial supporting facility along with it, and when these newly-built electric piles were put into use, its electric power network must consider electric automobile when charging the influence to the electric wire netting load. However, when the newly-built charging piles are operated at an initial stage, corresponding charging pile load data are lacked, and a corresponding prediction model cannot be established.
As shown in fig. 1, a method for predicting a charging pile load based on a density peak is proposed in the prior art, and the method includes: classifying historical daily load data of the charging pile to obtain a plurality of historical daily load data clusters; acquiring a historical day attribute cluster corresponding to the historical day load data cluster; establishing a decision tree model based on the historical daily load data cluster and the historical daily attribute; inputting the predicted day attribute into a decision tree model to obtain a target historical day load data cluster; acquiring target historical daily charging related data and target historical daily attributes corresponding to target historical daily load data in a target historical daily load data cluster; training a deep belief network model based on the target historical daily charging related data and the target historical daily attributes; and inputting the predicted day attribute into the deep belief network model to obtain the predicted day charging related data of the charging pile.
However, in view of actual effects, prediction is performed by using a single prediction method in different scenes, and there is a problem that prediction accuracy may be insufficient. The method for forecasting the short-term load of the charging pile is urgently needed at present, various charging pile load forecasting modes are combined, and the forecasting modes are cut according to application scenes, so that the load of the whole operation period of the charging pile can be accurately forecasted.
Disclosure of Invention
Aiming at the problems, the invention discloses a short-term load prediction method of a charging pile, which comprises the following steps:
taking out actual load historical data in a preset time period and prediction data obtained by a plurality of prediction methods from a database;
respectively calculating errors E of predicted loads of multiple prediction methods according to actual load historical data and the prediction data;
comparing errors E of predicted loads of a plurality of prediction methods:
and switching to the prediction method with the minimum error E.
Further, the error E is calculated by accumulating absolute values of differences between the predicted load and the actual load at each time within the preset time period.
Further, the plurality of prediction methods include a monte carlo-based load prediction method and a machine learning-based load prediction method.
Further, the monte carlo-based load prediction method comprises the following steps:
step S1: acquiring basic information of a charging pile;
step S2: randomly extracting the type of each vehicle;
step S3: extracting the initial charging time of each vehicle by a Monte Carlo simulation method, and then sorting the vehicle labels according to the time sequence;
step S4: extracting the initial charging capacity of each vehicle by a Monte Carlo simulation method;
step S5: calculating the charging time of each type of vehicle according to the battery capacity of each type of vehicle, and initializing the charging time of the vehicle to be zero;
step S6: comparing the number of the EVs at the current moment with the number of the available charging piles, if the number of the EVs is larger than the number of the available charging piles, resetting the initial charging time of the EVs at the current moment, placing the initial charging time of the EVs at the next moment in a charging automobile sequence, returning to the step S6, and if the number of the EVs is smaller than the number of the available charging piles, adding all the current EVs into the charging sequence;
step S7: detecting whether the EV in the current charging sequence is the electric vehicle which is completely charged or not, if so, rejecting the EV in the charging EV sequence, and updating the number of the current available charging piles;
step S8: detecting the current EV charging state, and updating the EV charging time of the available charging pile quantity domain;
step S9: and outputting the charging pile load at the current sampling moment.
Further, the basic information of the charging piles comprises the total amount of electric automobiles in the park, the number of the charging piles and vehicle type information;
the vehicle type information includes a vehicle type and a corresponding battery capacity, an initial charge time, and an initial charge capacity probability distribution.
Further, the monte carlo-based load prediction method further includes the step 10: whether the electric vehicle is charged at the present time is detected, and if yes, the process returns to step S6.
Further, the load prediction method based on machine learning specifically includes the following steps:
step S1: inputting a historical data set influencing the load of a charging pile;
step S2: establishing a neural network model of each data through the historical data set;
step S3: respectively acquiring a load prediction data result of a time period of [ t-l, t ] and real load data in the time period of [ t-l, t ] in historical data from prediction data of each single model, respectively calculating the error absolute value accumulation sum of each time of the prediction data of each single model and the real data in the time period of [ t-l, t ], and then arranging the error accumulation sums of each single prediction according to a descending order;
dividing the sequence number of the error accumulation sum of each single prediction by the number sum of all single models to serve as the weight of the error accumulation sum of each single prediction and the corresponding model;
step S4: and respectively multiplying the calculation results of each neural network model by the calculation results of the corresponding weight to obtain the load at the next moment.
Further, step S1 includes step S11:
and the historical data is processed by deletion of missing value data and data normalization.
Further, the method for predicting load based on machine learning further includes step S5: storing the prediction result in the prediction result time series, judging whether the prediction is finished, if not, updating the history database, returning to the step S1, and if so, completing the load prediction.
Further, the charging pile short-term load forecasting system comprises a switching module and a plurality of forecasting modules;
the switching module is used for switching the plurality of prediction modules according to different application scenes;
the plurality of prediction modules are used for predicting the short-term load of the charging pile.
Further, the switching module is configured to obtain actual load historical data within a preset time period and prediction data of the plurality of prediction modules from a database; and respectively calculating load errors E of the plurality of prediction modules, comparing the load errors E of the plurality of prediction modules, and switching the data of the prediction module with the best load error into the current and output prediction data.
Further, the plurality of prediction modules comprise a prediction module A based on a Monte Carlo prediction method, and the prediction module A is used for acquiring system information; randomly extracting the type of each vehicle, and extracting the initial charging time of each vehicle by using a Monte Carlo simulation method; sorting the labels of the vehicles according to the time sequence; extracting the initial charging capacity of each vehicle by a Monte Carlo simulation method; calculating the charging duration of each type of vehicle according to the battery capacity of each type of vehicle, and initializing the vehicle opinion charging duration to be zero; comparing the number of the EVs at the current moment with the number of the available charging piles, if the number of the EVs is larger than the number of the available charging piles, resetting the initial charging time of the EVs at the current moment, placing the initial charging time of the EVs at the next moment in a charging automobile sequence, and re-comparing the number of the EVs at the current moment with the number of the available charging piles, and if the number of the EVs is smaller than the number of the available charging piles, adding all the current EVs into the charging sequence; detecting whether the EV in the current charging sequence is the electric vehicle which is completely charged or not, if so, rejecting the EV in the charging EV sequence, and updating the number of the current available charging piles; detecting the current EV charging state, and updating the EV charging time of the available charging pile quantity domain; and outputting the load of the charging pile at the current sampling moment.
Further, the plurality of prediction modules comprise a prediction module B of a load prediction method based on machine learning, and the prediction module B is used for inputting a historical data set influencing the load of the charging pile; establishing a neural network model of each data through the historical data set; respectively obtaining load prediction data results of the time periods [ t-l, t ] and real load data in the time periods [ t-l, t ] in historical data from the prediction data of each single model, and calculating the weight of each data neural network model; and calculating the load of the next moment according to the calculation result of each neural network model through the weight.
According to the invention, different load prediction methods can be selected under different application scenes, so that the load prediction of the charging pile is more accurate.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 illustrates a density peak based charging pile load prediction method according to the prior art;
fig. 2 illustrates a method of predicting method switching in an embodiment of the present invention;
FIG. 3 illustrates a Monte Carlo-based load prediction method according to an embodiment of the present invention;
fig. 4 illustrates a load prediction method based on machine learning according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention discloses a short-term load forecasting method for a charging pile, which comprises the following steps:
taking out actual load historical data in a preset time period and prediction data obtained by a plurality of prediction methods from a database; respectively calculating errors E of predicted loads of multiple prediction methods; and switching to the prediction method with the minimum error E. Further, in the preset time period, the error E is calculated by accumulating absolute values of differences between the predicted load and the actual load at each time in the time period.
Illustratively, a load prediction method based on Monte Carlo and a load prediction method based on machine learning are adopted by charging piles in a certain park to predict short-term loads of the charging piles in the park, and in order to obtain the most accurate prediction result, corresponding prediction methods are switched according to different application scenes.
Specifically, as shown in fig. 2, actual load history data in the [ t-l, t ] time period, prediction data based on the monte carlo simulation method, and prediction data based on machine learning are taken out from the database; respectively calculating a predicted load error E1 based on a Monte Carlo simulation method and a predicted load error E2 based on machine learning; and judging the numerical values of E1 and E2, if E1 is larger than E2, switching to a prediction mode based on machine learning, and if E1 is smaller than E2, switching to a prediction mode based on Monte Carlo. The method has the advantages that various prediction methods are adopted for switching, so that the accurate prediction of the electric load of the charging pile is realized, the problem that the prediction result is possibly inaccurate due to the fact that a single prediction method cannot deal with the prediction work under multiple scenes is effectively solved, and more accurate guiding reference opinions are provided for the safe operation of the power grid in the park.
The load mentioned above generally refers to the daily load of the charging pile in the park, that is, the superposition of the charging load of the electric vehicle at each moment, and the calculation model is as follows:
wherein, PiRepresents the total power of the charging pile at the ith moment, Pn,iAt the ith moment, the power of the nth vehicle, and N is the total number of the electric vehicles. According to the power calculation model of the charging pile, the initial charging time and the charging capacity of the electric automobile are important factors for load prediction.
Further, as shown in fig. 3, the prediction method using the monte carlo-based prediction method includes the following steps:
step S1: acquiring basic information of a charging pile; specifically, the charging pile basic information includes the total amount of electric vehicles in the park (hereinafter, EV), the number of charging piles and vehicle type information;
the vehicle type information includes a vehicle type and corresponding battery capacity, initial charge time, and initial charge capacity probability distribution.
Step S2: the type of each vehicle is randomly drawn.
Step S3: and extracting the initial charging time of each vehicle by using a Monte Carlo simulation method, and then sorting the vehicle labels according to the time sequence.
Step S4: the initial charge capacity of each vehicle was extracted using a monte carlo simulation.
Step S5: according to the battery capacity of each type of vehicle, the charging time period of each type of vehicle is calculated, and the charging time period of the initialized vehicle is zero.
Step S6: and comparing the number of the EVs at the current moment with the number of the available charging piles, if the number of the EVs is larger than the number of the available charging piles, resetting the initial charging time of the EVs at the current moment, placing the initial charging time of the EVs at the next moment in a charging automobile sequence, returning to the step S6, and if the number of the EVs is smaller than the number of the available charging piles, adding all the current EVs in the charging sequence. Wherein returning to step S6 indicates re-executing step S6.
Step S7: and detecting whether the EV in the current charging sequence is the electric vehicle which is completely charged, if so, rejecting the EV in the charging EV sequence, and updating the number of the current available charging piles.
Step S8: and detecting the current EV charging state, and updating the available charging pile quantity field EV charging time.
Step S9: and outputting the load of the charging pile at the current sampling moment.
Further, the prediction method using the monte carlo-based method further includes step S10: whether the automobile is charged at the current time is detected, and if yes, the process returns to the step S6.
The invention further discloses a load prediction method based on machine learning, as shown in fig. 4, the load prediction method based on machine learning specifically comprises the following steps:
step S1: and inputting a historical data set influencing the load of the charging pile. Further, the historical data is processed, and the processing exemplarily comprises missing value data deletion and data normalization processing.
Step S2: from the historical data set, a neural network model is established for each data.
Step S3: respectively obtaining load prediction data results of the time periods [ t-l, t ] and real load data in the time periods [ t-l, t ] in historical data from the prediction data of each single model, and calculating the weight of each data neural network model;
specifically, the step of calculating the weight of the data neural network model specifically includes:
respectively calculating the error absolute value accumulation sum of each single-term model prediction data and real data at each moment in the [ t-l, t ] time period, and then arranging the error accumulation sums of each single-term prediction according to a descending order;
and dividing the sequence number of the error accumulation sum of each single prediction by the number sum of all the single models to serve as the weight of the error accumulation sum of each single prediction and the corresponding model.
Step S4: and respectively multiplying the calculation results of each neural network model by the calculation results of the corresponding weight to obtain the load at the next moment.
Further, the method for predicting load based on machine learning further includes step S5: storing the prediction result in the prediction result time series, judging whether the prediction is finished, if not, updating the history database, returning to the step S1, if so, finishing the load prediction, and further, the prediction finishing condition is forced finishing of the program.
The invention discloses a charging pile short-term load forecasting system which comprises a switching module and a plurality of forecasting modules. The switching module is used for switching the plurality of prediction modules according to different application scenes, and the plurality of prediction modules are used for predicting the short-term load of the charging pile. The charging pile short-term load forecasting system can forecast the charging pile short-term load by using different forecasting modules in different scenes, so that the forecasting result of the charging pile short-term load forecasting system is more accurate.
Specifically, the switching module is configured to obtain actual load historical data in a preset time period and prediction data of the prediction modules from a database, calculate load errors E of the prediction modules, and compare the load errors E of the prediction modules, where data of the prediction module with the best switching load error is current and output prediction data.
Illustratively, the plurality of prediction modules include a prediction module a based on a monte carlo prediction method, and the prediction module a is used for acquiring system information; randomly extracting the type of each vehicle, and extracting the initial charging time of each vehicle by using a Monte Carlo simulation method; sorting the labels of the vehicles according to the time sequence; extracting the initial charging capacity of each vehicle by a Monte Carlo simulation method; calculating the charging duration of each type of vehicle according to the battery capacity of each type of vehicle, and initializing the vehicle opinion charging duration to be zero; comparing the number of the EVs at the current moment with the number of the available charging piles, if the number of the EVs is larger than the number of the available charging piles, resetting the initial charging time of the EVs at the current moment, placing the initial charging time of the EVs at the next moment in a charging automobile sequence, and re-comparing the number of the EVs at the current moment with the number of the available charging piles, and if the number of the EVs is smaller than the number of the available charging piles, adding all the current EVs into the charging sequence; detecting whether the EV in the current charging sequence is the electric vehicle which is completely charged or not, if so, rejecting the EV in the charging EV sequence, and updating the number of the current available charging piles; detecting the current EV charging state, and updating the EV charging time of the available charging pile quantity domain; and outputting the load of the charging pile at the current sampling moment.
Illustratively, the plurality of prediction modules comprise a prediction module B of a load prediction method based on machine learning, and the prediction module B is used for inputting a historical data set influencing the load of the charging pile; establishing a neural network model of each data through the historical data set; respectively obtaining load prediction data results of the time periods [ t-l, t ] and real load data in the time periods [ t-l, t ] in historical data from the prediction data of each single model, and calculating the weight of each data neural network model; and calculating the load of the next moment according to the calculation result of each neural network model through the weight.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (13)
1. A method for predicting short-term load of a charging pile is characterized in that,
the prediction method comprises the following steps:
taking out actual load historical data in a preset time period and prediction data obtained by a plurality of prediction methods from a database;
respectively calculating errors E of predicted loads of multiple prediction methods according to actual load historical data and the prediction data;
comparing errors E of predicted loads of a plurality of prediction methods:
and switching to the prediction method with the minimum error E.
2. The method of predicting a short-term load of a charging pile according to claim 1,
the error E is calculated by accumulating the absolute value of the difference between the predicted load and the actual load at each moment in the preset time period.
3. The method of predicting a short-term load of a charging pile according to claim 1,
the plurality of prediction methods include a monte carlo-based load prediction method and a machine learning-based load prediction method.
4. The method of predicting a short-term load of a charging pile according to claim 3,
the Monte Carlo-based load prediction method comprises the following steps:
step S1: acquiring basic information of a charging pile;
step S2: randomly extracting the type of each vehicle;
step S3: extracting the initial charging time of each vehicle by a Monte Carlo simulation method, and then sorting the vehicle labels according to the time sequence;
step S4: extracting the initial charging capacity of each vehicle by a Monte Carlo simulation method;
step S5: calculating the charging time of each type of vehicle according to the battery capacity of each type of vehicle, and initializing the charging time of the vehicle to be zero;
step S6: comparing the number of the EVs at the current moment with the number of the available charging piles, if the number of the EVs is larger than the number of the available charging piles, resetting the initial charging time of the EVs at the current moment, placing the initial charging time of the EVs at the next moment in a charging automobile sequence, returning to the step S6, and if the number of the EVs is smaller than the number of the available charging piles, adding all the current EVs into the charging sequence;
step S7: detecting whether the EV in the current charging sequence is the electric vehicle which is completely charged or not, if so, rejecting the EV in the charging EV sequence, and updating the number of the current available charging piles;
step S8: detecting the current EV charging state, and updating the EV charging time of the available charging pile quantity domain;
step S9: and outputting the charging pile load at the current sampling moment.
5. The method of predicting the short-term load of a charging pile according to claim 4,
the charging pile basic information comprises the total quantity of electric automobiles in the park, the number of charging piles and vehicle type information;
the vehicle type information includes a vehicle type and a corresponding battery capacity, an initial charge time, and an initial charge capacity probability distribution.
6. The method of predicting the short-term load of a charging pile according to claim 4,
the Monte Carlo-based load prediction method further comprises the following steps of: whether the electric vehicle is charged at the present time is detected, and if yes, the process returns to step S6.
7. The method of predicting a short-term load of a charging pile according to claim 3,
the load prediction method based on machine learning specifically comprises the following steps:
step S1: inputting a historical data set influencing the load of a charging pile;
step S2: establishing a neural network model of each data through the historical data set;
step S3: respectively acquiring a load prediction data result of a time period of [ t-l, t ] and real load data in the time period of [ t-l, t ] in historical data from prediction data of each single model, respectively calculating the error absolute value accumulation sum of each time of the prediction data of each single model and the real data in the time period of [ t-l, t ], and then arranging the error accumulation sums of each single prediction according to a descending order;
dividing the sequence number of the error accumulation sum of each single prediction by the number sum of all single models to serve as the weight of the error accumulation sum of each single prediction and the corresponding model;
step S4: and respectively multiplying the calculation results of each neural network model by the calculation results of the corresponding weight to obtain the load at the next moment.
8. The method of predicting a short-term load of a charging pile according to claim 7,
the step S1 includes a step S11:
and the historical data is processed by deletion of missing value data and data normalization.
9. The method of predicting a short-term load of a charging pile according to claim 7,
the load prediction method based on machine learning further includes step S5: storing the prediction result in the prediction result time series, judging whether the prediction is finished, if not, updating the history database, returning to the step S1, and if so, completing the load prediction.
10. A short-term load forecasting system of a charging pile is characterized in that,
the charging pile short-term load forecasting system comprises a switching module and a plurality of forecasting modules;
the switching module is used for switching the plurality of prediction modules according to different application scenes;
the plurality of prediction modules are used for predicting the short-term load of the charging pile.
11. The system for predicting the short-term load of a charging pile according to claim 10,
the switching module is used for acquiring actual load historical data in a preset time period and prediction data of the prediction modules from a database; and respectively calculating load errors E of the plurality of prediction modules, comparing the load errors E of the plurality of prediction modules, and switching the data of the prediction module with the best load error into the current and output prediction data.
12. The system for predicting the short-term load of a charging pile according to claim 10,
the plurality of prediction modules comprise a prediction module A based on a Monte Carlo prediction method, and the prediction module A is used for acquiring system information; randomly extracting the type of each vehicle, and extracting the initial charging time of each vehicle by using a Monte Carlo simulation method; sorting the labels of the vehicles according to the time sequence; extracting the initial charging capacity of each vehicle by a Monte Carlo simulation method; calculating the charging duration of each type of vehicle according to the battery capacity of each type of vehicle, and initializing the vehicle opinion charging duration to be zero; comparing the number of the EVs at the current moment with the number of the available charging piles, if the number of the EVs is larger than the number of the available charging piles, resetting the initial charging time of the EVs at the current moment, placing the initial charging time of the EVs at the next moment in a charging automobile sequence, and re-comparing the number of the EVs at the current moment with the number of the available charging piles, and if the number of the EVs is smaller than the number of the available charging piles, adding all the current EVs into the charging sequence; detecting whether the EV in the current charging sequence is the electric vehicle which is completely charged or not, if so, rejecting the EV in the charging EV sequence, and updating the number of the current available charging piles; detecting the current EV charging state, and updating the EV charging time of the available charging pile quantity domain; and outputting the load of the charging pile at the current sampling moment.
13. The system for predicting the short-term load of a charging pile according to claim 10,
the plurality of prediction modules comprise a prediction module B of a load prediction method based on machine learning, and the prediction module B is used for inputting historical data sets influencing charging pile loads; establishing a neural network model of each data through the historical data set; respectively obtaining load prediction data results of the time periods [ t-l, t ] and real load data in the time periods [ t-l, t ] in historical data from the prediction data of each single model, and calculating the weight of each data neural network model; and calculating the load of the next moment according to the calculation result of each neural network model through the weight.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110507429.XA CN113298298B (en) | 2021-05-10 | 2021-05-10 | Short-term load prediction method and system for charging pile |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110507429.XA CN113298298B (en) | 2021-05-10 | 2021-05-10 | Short-term load prediction method and system for charging pile |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113298298A true CN113298298A (en) | 2021-08-24 |
CN113298298B CN113298298B (en) | 2023-12-29 |
Family
ID=77321314
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110507429.XA Active CN113298298B (en) | 2021-05-10 | 2021-05-10 | Short-term load prediction method and system for charging pile |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113298298B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117172390A (en) * | 2023-11-02 | 2023-12-05 | 福建时代星云科技有限公司 | Charging amount prediction method and terminal based on scene division |
Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008052042A2 (en) * | 2006-10-24 | 2008-05-02 | Edsa Micro Corporation | Systems and methods for a real-time synchronized electrical power system simulator for 'what-if' analysis and prediction over electrical power networks |
JP2014042383A (en) * | 2012-08-22 | 2014-03-06 | Hitachi Ltd | Grid load forecast system using position information on electric vehicle |
CN106208388A (en) * | 2016-08-31 | 2016-12-07 | 科大智能电气技术有限公司 | A kind of intelligence charging system and short term basis load prediction implementation method thereof in order |
CN107742038A (en) * | 2017-10-30 | 2018-02-27 | 广东电网有限责任公司惠州供电局 | Charging electric vehicle load forecasting method and device |
CN109146156A (en) * | 2018-08-03 | 2019-01-04 | 大连理工大学 | A method of for predicting charging pile system charge volume |
CN109800537A (en) * | 2019-02-20 | 2019-05-24 | 大连理工大学 | A kind of machine tool thermal error model reliability degree calculation method based on deep neural network and Monte Carlo method |
US20190165580A1 (en) * | 2017-11-27 | 2019-05-30 | Ihi Inc. | System and method for optimal control of energy storage system |
CN109919393A (en) * | 2019-03-22 | 2019-06-21 | 国网上海市电力公司 | A kind of charging load forecasting method of electric taxi |
CN109934403A (en) * | 2019-03-12 | 2019-06-25 | 浙江大学 | Charge load Analysis prediction technique in electric car resident region based on mathematical model |
CN110570042A (en) * | 2019-09-16 | 2019-12-13 | 燕山大学 | Short-term electric vehicle charging load prediction method and system |
CN110570014A (en) * | 2019-08-07 | 2019-12-13 | 浙江大学 | Electric vehicle charging load prediction method based on Monte Carlo and deep learning |
CN110929941A (en) * | 2019-11-26 | 2020-03-27 | 广东电网有限责任公司 | Short-term power load prediction method and system based on multi-load mode |
CN111626514A (en) * | 2020-05-29 | 2020-09-04 | 深圳供电局有限公司 | Electric vehicle charging load prediction method and device |
CN112232559A (en) * | 2020-10-12 | 2021-01-15 | 国网江西省电力有限公司信息通信分公司 | Short-term prediction method and device for load in power area |
CN112508301A (en) * | 2020-12-21 | 2021-03-16 | 北京梧桐车联科技有限责任公司 | Method and device for predicting charging load of electric vehicle, and storage medium |
-
2021
- 2021-05-10 CN CN202110507429.XA patent/CN113298298B/en active Active
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2008052042A2 (en) * | 2006-10-24 | 2008-05-02 | Edsa Micro Corporation | Systems and methods for a real-time synchronized electrical power system simulator for 'what-if' analysis and prediction over electrical power networks |
JP2014042383A (en) * | 2012-08-22 | 2014-03-06 | Hitachi Ltd | Grid load forecast system using position information on electric vehicle |
CN106208388A (en) * | 2016-08-31 | 2016-12-07 | 科大智能电气技术有限公司 | A kind of intelligence charging system and short term basis load prediction implementation method thereof in order |
CN107742038A (en) * | 2017-10-30 | 2018-02-27 | 广东电网有限责任公司惠州供电局 | Charging electric vehicle load forecasting method and device |
US20190165580A1 (en) * | 2017-11-27 | 2019-05-30 | Ihi Inc. | System and method for optimal control of energy storage system |
CN109146156A (en) * | 2018-08-03 | 2019-01-04 | 大连理工大学 | A method of for predicting charging pile system charge volume |
CN109800537A (en) * | 2019-02-20 | 2019-05-24 | 大连理工大学 | A kind of machine tool thermal error model reliability degree calculation method based on deep neural network and Monte Carlo method |
CN109934403A (en) * | 2019-03-12 | 2019-06-25 | 浙江大学 | Charge load Analysis prediction technique in electric car resident region based on mathematical model |
CN109919393A (en) * | 2019-03-22 | 2019-06-21 | 国网上海市电力公司 | A kind of charging load forecasting method of electric taxi |
CN110570014A (en) * | 2019-08-07 | 2019-12-13 | 浙江大学 | Electric vehicle charging load prediction method based on Monte Carlo and deep learning |
CN110570042A (en) * | 2019-09-16 | 2019-12-13 | 燕山大学 | Short-term electric vehicle charging load prediction method and system |
CN110929941A (en) * | 2019-11-26 | 2020-03-27 | 广东电网有限责任公司 | Short-term power load prediction method and system based on multi-load mode |
CN111626514A (en) * | 2020-05-29 | 2020-09-04 | 深圳供电局有限公司 | Electric vehicle charging load prediction method and device |
CN112232559A (en) * | 2020-10-12 | 2021-01-15 | 国网江西省电力有限公司信息通信分公司 | Short-term prediction method and device for load in power area |
CN112508301A (en) * | 2020-12-21 | 2021-03-16 | 北京梧桐车联科技有限责任公司 | Method and device for predicting charging load of electric vehicle, and storage medium |
Non-Patent Citations (7)
Title |
---|
GAO QIANG 等: "Charging Load Forecasting of Electric Vehicle Based on Monte Carlo and Deep Learning", 2019 IEEE SUSTAINABLE POWDER AND ENERGY CONFERENCE(ISPEC), pages 1309 - 1314 * |
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, pages 22 - 28 * |
惠恩: "电动汽车充电负荷的预计及其对小区供电的影响分析", 中国优秀硕士学位论文全文数据库工程科技Ⅱ辑, no. 7, pages 035 - 280 * |
朱慧婷;杨雪;陈友媛;: "电动汽车充电负荷预测方法综述", 电力信息与通信技术, no. 05, pages 44 - 47 * |
李海荣: "电动汽车充电站短期负荷预测研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, no. 1, pages 042 - 1637 * |
沈渊彬;刘庆珍;: "电力系统短期负荷预测研究概述", 电器与能效管理技术, no. 04, pages 28 - 32 * |
龚钢军 等: "基于SAE-ELM 的电动汽车充电站负荷预测模型", 现代电力, vol. 36, no. 6, pages 9 - 15 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117172390A (en) * | 2023-11-02 | 2023-12-05 | 福建时代星云科技有限公司 | Charging amount prediction method and terminal based on scene division |
CN117172390B (en) * | 2023-11-02 | 2024-02-23 | 福建时代星云科技有限公司 | Charging amount prediction method and terminal based on scene division |
Also Published As
Publication number | Publication date |
---|---|
CN113298298B (en) | 2023-12-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111584952B (en) | Method and system for online evaluation of electrochemical cells of energy storage power station | |
CN107516170B (en) | Difference self-healing control method based on equipment failure probability and power grid operation risk | |
US20200161867A1 (en) | Method, system and storage medium for load dispatch optimization for residential microgrid | |
CN103745110B (en) | Method of estimating operational driving range of all-electric buses | |
CN109508826B (en) | Electric vehicle cluster schedulable capacity prediction method based on gradient lifting decision tree | |
CN103488869A (en) | Wind power generation short-term load forecast method of least squares support vector machine | |
CN112884008A (en) | Prediction evaluation method and device for running state of power information acquisition system | |
CN110794308A (en) | Method and device for predicting train battery capacity | |
CN110210670A (en) | A kind of prediction technique based on power-system short-term load | |
CN112766677A (en) | Modeling method and charging system for residential electric vehicle charging demand based on fuzzy comprehensive evaluation method | |
CN115063184A (en) | Electric vehicle charging demand modeling method, system, medium, equipment and terminal | |
CN112036598A (en) | Charging pile use information prediction method based on multi-information coupling | |
CN113298298B (en) | Short-term load prediction method and system for charging pile | |
Wang et al. | Electric vehicle charging load clustering and load forecasting based on long short term memory neural network | |
CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
CN114460481A (en) | Energy storage battery thermal runaway early warning method based on Bi-LSTM and attention mechanism | |
CN109858667A (en) | It is a kind of based on thunder and lightning weather to the short term clustering method of loading effects | |
CN112928766A (en) | Power distribution network electric vehicle accepting capability evaluation method based on multiple influence factors | |
CN113033898A (en) | Electrical load prediction method and system based on K-means clustering and BI-LSTM neural network | |
CN114742118B (en) | Electric automobile cluster charge-discharge load combination prediction method | |
CN114545276A (en) | Power battery service life prediction method based on capacity test and Internet of vehicles big data | |
CN113283638A (en) | Load extreme curve prediction method and system based on fusion model | |
CN113537607A (en) | Power failure prediction method | |
CN110852378A (en) | Road condition kinematics segment extraction method based on navigation system | |
CN112633565A (en) | Photovoltaic power aggregation interval prediction method |
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 |