CN116090675B - Short-time charging scheduling method based on combination of block chain and neural network - Google Patents
Short-time charging scheduling method based on combination of block chain and neural network Download PDFInfo
- Publication number
- CN116090675B CN116090675B CN202310371117.XA CN202310371117A CN116090675B CN 116090675 B CN116090675 B CN 116090675B CN 202310371117 A CN202310371117 A CN 202310371117A CN 116090675 B CN116090675 B CN 116090675B
- Authority
- CN
- China
- Prior art keywords
- charging
- model
- data
- scheduling
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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
- B60L53/60—Monitoring or controlling charging stations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
-
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Computer Security & Cryptography (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Educational Administration (AREA)
- Molecular Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Power Engineering (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Bioethics (AREA)
- Computer Hardware Design (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses a short-time charging scheduling method based on combination of a block chain and a neural network, which comprises the following steps: building a distributed data communication model; acquiring the initial data of charging of a plurality of groups of electric vehicles, and building a BP neural network model; training by using a BP neural network model to obtain a first prediction scheduling model; presetting a scheduling power threshold; collecting charging data of any current charging pile by using a distributed data communication model; updating the first predictive scheduling model to obtain a second predictive scheduling model; the idle charging pile is accessed and judged, the second prediction scheduling model is updated, and the updated second prediction scheduling model is stored in a distributed mode; and continuously acquiring the charging data of any charging pile if the corresponding power data in the charging scheduling data output by the second predictive scheduling model is equal to or greater than the scheduling power threshold. Through the scheme, the invention has the advantages of simple logic, reliable dispatching and the like, and has high practical value and popularization value in the technical field of charging dispatching.
Description
Technical Field
The invention relates to the technical field of charge scheduling, in particular to a short-time charge scheduling method based on combination of a block chain and a neural network.
Background
The Charging pile (Charging pile) refers to Charging equipment for providing Charging service for an electric automobile, and is mainly divided into a floor type Charging pile and a wall-mounted Charging pile, wherein the Charging pile mainly adopts a Charging mode of timing, electricity metering and money metering. Along with the high-speed development of the charging pile (station) and the electric automobile, the charging voltage, the charging current and the charging efficiency are also obviously improved; for example: some vehicle enterprises propose super charging stations (piles). One ac transformer device will be equipped with several charging piles (stations), for example, a 10kV transformer is equipped with several charging piles (stations) on which 1 dc or ac gun is arranged.
In design, a coefficient exists between the total power of the alternating-current transformer equipment and the rated power sum of all charging piles (stations), and the coefficient is smaller than 1; typically, the coefficient is between 0.75 and 0.95. The reason for this is: first, in general, the possibility of full-load maximum power charging of a charging pile (station) in the same ac transformer device in the same time period is extremely low, except in the theoretical limit test case (i.e. all the charging piles (stations) are at the same time and are charged with maximum power, which only stays in the test phase and is not applicable to the actual situation). Secondly, in order to prolong the service life of the battery of the electric vehicle and avoid the problems of overcharging and the like of the battery, the battery is higher in charging power of 30% -80%, and after the battery is charged to 80%, the charging power is reduced, and slow charging is performed. Third, the highest charging power of not all electric vehicles is equal to the rated power of the charging pile (station), and the charging power is related to the electric vehicle itself.
Therefore, the ac transformer and the charging pile (station) face complicated and variable charging conditions, and how to ensure efficient and reliable charging distribution has become an important research topic. Currently, numerous methods of charge schedule management also occur in the prior art, such as "patent publication No.: CN114638440a, name: a Chinese patent invention of a charging load ultra-short-term prediction method based on the usage degree of a charging pile comprises the following steps: s1, acquiring charging load transaction data of all charging piles; s2, eliminating the charging load of the regular abnormal day by using a density clustering algorithm, and calculating to obtain the daily average load of the charging pile; s3, calculating to obtain the utilization degree of the charging pile according to the daily average load of the charging pile; s4, fusing the charging load historical data and the charging pile usage degree data to form a two-dimensional input data set; s5, inputting the two-dimensional input data set into the constructed long-short-period memory neural network model, performing a large number of supervised learning training, and performing ultra-short-period prediction of the charging load by using the trained model. The technology is to evaluate daily average load, the load curve is still wider, for example, the quick charge time of an electric vehicle is about half an hour to about 1 hour, the technology cannot be accurate to the number of minutes or seconds, and only long-period prediction can be realized. Typically, 9:00 to 11 am: 00. 13:00 to 17:00 pm, 19:00 to 21:00 pm are relative peak periods of charge, and 23:00 to 7:00 are valley periods of charge. During the peak period of charging, the work of a charging gun can be switched in or out within one minute or less, and the complex change state of the charging gun can put more flexible scheduling requirements on the charging equipment. In addition, not all electric vehicles are charged from low power to full load, which may require high power charging, possibly low power charging. Therefore, this technique cannot be applied to a charging schedule for an extremely short time.
And the patent publication number is as follows: CN114648171a, name: a Chinese invention patent of electric car charging pile load prediction method, system and storage medium, which comprises: inputting a pre-constructed first neural network model into a charging pile to be analyzed, and adjusting the first neural network model according to target charging information of the charging pile to be analyzed to obtain a second neural network model; the first neural network model is constructed through the first quantity of charging information, and the difference value between the quantity of target charging information and the first quantity is larger than a preset difference value; and acquiring real-time charging information in the charging pile to be analyzed, inputting the real-time charging information into the second neural network model, and acquiring load state information of the charging pile to be analyzed, which is output by the second neural network model. And the patent publication number is as follows: CN115034507a, name: a china patent for electric load prediction method and related components of a charging pile, comprising: acquiring a historical power load of a target charging pile to obtain an input sequence, and inputting the input sequence into a prediction model constructed by a convolutional neural network and a long-term and short-term memory network; performing first processing on the input sequence by using the convolutional neural network to obtain an output characteristic corresponding to the input sequence, and performing second processing on the output characteristic by using the long-term and short-term memory network to obtain an output power load; and predicting the electric load of the target charging pile according to an output sequence of the prediction model, wherein the output sequence consists of the output electric loads. Both the above techniques are to adjust a certain charging pile (station), and are not applicable to dynamically adjusting a plurality of charging piles (stations) as a whole. In CN114648171a, for example, the charging schedule is dynamically changed for different vehicle types and battery conditions, and it cannot be applied to joint scheduling of multiple charging piles (stations).
With the development of technology, blockchains are promoted in various industries, and are no exception in the electric automobile charging technology, however, the prior art stays in control among charging piles, a user side and a management server, and scheduling among a plurality of charging piles of the same site cannot be realized. For example, "patent publication No.: CN115545486a, name: the invention relates to a block chain-based electric automobile charging scheduling method, which comprises the following steps: (1) user: the method comprises the steps of forming an electric automobile to be scheduled; when an electric automobile needs to be charged, basic information of the current automobile needs to be uploaded to a blockchain, wherein the basic information comprises automobile position, required electric quantity, current electric quantity, reputation score and power consumption per kilometer at normal running speed; after the dispatching result is obtained, the vehicle obtains updated credit score according to whether the user performs according to the dispatching result so as to be used for next dispatching; (2) charging station: is responsible for providing power resources for users; before starting scheduling, the charging station needs to upload basic information to the blockchain, including position and time-of-use electricity price, and provide service when charging is performed before a user; (3) blockchain: the method is responsible for recording basic information, transaction information, decision for completing scheduling, information issuing and updating user reputation scores; after receiving the related information of the charging station and the user, the blockchain utilizes the intelligent contract on the blockchain to carry out charging scheduling on the vehicle; and updating the credit of the electric automobile according to the performance condition.
And the patent publication number is as follows: CN115063927a, name: the invention relates to a block chain-based shared charging system and running equipment, which comprises a charging pile, a user end and a management server end, wherein the charging pile, the user end and the management server end are connected with each other after being uplink to form a block chain network, and the charging pile is as follows: charging amount management is carried out, and owners of charging piles sign charging pile information through equipment and upload the charging pile information to a blockchain; the user terminal: each user terminal at least corresponds to one device to be charged, the device signs the information of the device to be charged through the device, and then uploads the information to the block chain, the reservation charging is successful, and charging cost is paid to the management server terminal after charging; the management server side: and acquiring information of the charging pile and information of equipment to be charged through the blockchain so as to carry out resource scheduling according to requirements, supervising the completion of the charging transaction and uploading charging transaction data to the blockchain.
Therefore, it is highly desirable to provide a short-time charging scheduling method based on the combination of the blockchain and the neural network, which is simple in logic and reliable in scheduling.
Disclosure of Invention
Aiming at the problems, the invention aims to provide a short-time charging scheduling method based on the combination of a block chain and a neural network, and the technical scheme adopted by the invention is as follows:
a short-time charging scheduling method based on the combination of a block chain and a neural network comprises the following steps:
constructing a distributed data communication model of a plurality of charging piles under the same alternating-current transformer equipment by adopting consensus communication;
acquiring the initial data of charging of a plurality of groups of electric vehicles, and building a BP neural network model; training by using a BP neural network model to obtain a first predictive scheduling model corresponding to the original data; the original data comprise the current residual capacity of the first electric vehicle battery, a first charging voltage and a first charging current;
the first predictive scheduling model is stored in any charging pile in a distributed mode by adopting consensus communication, and a scheduling power threshold value is preset;
collecting charging data of any current charging pile by using a distributed data communication model; inputting the charging data into a first predictive scheduling model, and updating the first predictive scheduling model to obtain a second predictive scheduling model; outputting current charging schedule data by using the second predictive schedule model;
if the corresponding power data in the charging schedule data output by the second predictive schedule model is smaller than the schedule power threshold and a charging pile in a non-charging state exists, acquiring an electric vehicle access charging pile signal by adopting trusted identity verification, acquiring third current battery working condition data and third charging data of the electric vehicle, and carrying out distributed storage on the third current battery working condition data and the third charging data;
inputting third current battery working condition data and third charging data into a second predictive scheduling model, obtaining a second error value, updating the second predictive scheduling model by adopting counter propagation updating weight, and performing distributed storage on the updated second predictive scheduling model;
and if the corresponding power data in the charging schedule data output by the second predictive schedule model is equal to or greater than the schedule power threshold, continuously collecting the charging data of any charging pile, and updating the second predictive schedule model.
Further, training by using the BP neural network model to obtain a first predictive scheduling model corresponding to the original data, including:
selecting a certain charging pile under the same alternating-current transformer equipment, charging according to original data, and keeping the rest charging pile under the same alternating-current transformer equipment in a non-charging state;
and inputting the original data into the BP neural network model for training, and obtaining a first predictive scheduling model corresponding to the original data.
Further, collecting charging data of any current charging pile by using a distributed data communication model; inputting the charging data into the first predictive scheduling model, updating the first predictive scheduling model, and obtaining a second predictive scheduling model, wherein the method comprises the following steps:
sequentially acquiring charging data of a charging pile in a charging use state under the same alternating-current transformer equipment by using a distributed data communication model; the charging data comprise the current residual capacity of the second electric vehicle battery, a second charging voltage and a second charging current;
inputting the current residual capacity, the second charging voltage and the second charging current of the second electric vehicle battery into a first prediction scheduling model, performing prediction training, obtaining a first error value, and updating the weight by adopting back propagation;
and updating the first predictive scheduling model and obtaining a second predictive scheduling model.
Preferably, a BP neural network model is built, and a loss function is set between a hidden layer and an output layer of the BP neural network model; the loss function is a mean square error loss function.
Further, obtaining the original data of charging of the plurality of groups of electric vehicles and the charging data of any charging pile, and adopting linear fitting to respectively remove abnormal data points in the original data and the charging data.
Further, training of the BP neural network model includes forward propagation and backward propagation; the back propagation takes as input the second error value and transmits to the hidden layer of the BP neural network model.
Preferably, the scheduling power threshold
The expression of (2) is:
wherein k represents the ratio coefficient of the rated power sum of the charging piles connected to the same alternating-current transformer equipment and the total power of the alternating-current transformer equipment; n represents the number of charging piles;
indicating the rated power of the charging pile.
Further, the short-time charging scheduling method based on the combination of the blockchain and the neural network further comprises the following steps: updating the first predictive scheduling model to obtain a second predictive scheduling model, and carrying out distributed storage on the second predictive scheduling model.
Further, the short-time charging scheduling method based on the combination of the blockchain and the neural network further comprises the following steps: and if the corresponding power data in the charging schedule data output by the second predictive schedule model is smaller than the schedule power threshold and any charging pile is in a charging state, acquiring the charging data of any current charging pile by using the distributed data communication model, and updating the first predictive schedule model.
Further, the distributed data communication model collects charging data of any charging pile, performs distributed storage, and marks a current time stamp.
Compared with the prior art, the invention has the following beneficial effects:
according to the invention, the original data of the charging of the plurality of groups of electric vehicles are skillfully acquired, a certain charging pile is used as a research object for theoretical training, a first prediction scheduling model in an ideal state is obtained, and any charging pile receives the theoretical first prediction scheduling model in an initial state, so that a foundation is provided for actual correction and adjustment in a later period.
The invention collects the charging data of any current charging pile through the sharing mechanism of the block chain, corrects the charging data with the first prediction scheduling model to obtain the current real-time prediction scheduling model, judges whether the charging pile in the non-charging state can be accessed according to the current charging scheduling data, has strong dynamic reliability, and utilizes the first error value to carry out back propagation and update the weight so as to ensure the real-time dynamization of the current scheduling model.
The invention skillfully builds a distributed data communication model of a plurality of charging piles under the same alternating current transformer equipment, and the data of the distributed data communication model is shared in real time and can be traced, so that the dynamic linkage scheduling of the plurality of charging piles under the same alternating current transformer equipment is facilitated.
According to the invention, the block chain distributed storage is carried out on the third current battery working condition data and the third charging data of the connected electric vehicle, and the second predictive scheduling model training is carried out, so that the second predictive scheduling model is updated, and the scheduling dynamics is ensured. Similarly, after one or a plurality of charging piles exit from charging, the redistribution type data communication model collects charging data of any current charging pile, and the first prediction scheduling model can be updated.
The invention skillfully adopts the BP neural network model, and utilizes the error value to carry out back propagation so as to update the model and ensure that the model is closer to the charging schedule data required at present.
The invention skillfully sets the dispatching power threshold, and has a margin for the access of the charging pile so as to judge whether the current charging state can be accessed to the electric vehicle for charging or not, thereby ensuring the reliable operation of the alternating-current transformer equipment.
In conclusion, the invention has the advantages of simple logic, reliable dispatching and the like, and has high practical value and popularization value in the technical field of charging dispatching.
Drawings
For a clearer description of the technical solutions of the embodiments of the present invention, the drawings to be used in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and should not be considered as limiting the scope of protection, and other related drawings may be obtained according to these drawings without the need of inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of a logic structure of the present invention.
Description of the embodiments
For the purposes, technical solutions and advantages of the present application, the present invention will be further described with reference to the accompanying drawings and examples, and embodiments of the present invention include, but are not limited to, the following examples. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present application based on the embodiments herein.
In this embodiment, the term "and/or" is merely an association relationship describing the association object, and indicates that three relationships may exist, for example, a and/or B may indicate: a exists alone, A and B exist together, and B exists alone.
The terms first and second and the like in the description and in the claims of the present embodiment are used for distinguishing between different objects and not for describing a particular sequential order of objects. For example, the first target object and the second target object, etc., are used to distinguish between different target objects, and are not used to describe a particular order of target objects.
In the embodiments of the present application, words such as "exemplary" or "such as" are used to mean serving as examples, illustrations, or descriptions. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.
In the description of the embodiments of the present application, unless otherwise indicated, the meaning of "a plurality" means two or more. For example, the plurality of processing units refers to two or more processing units; the plurality of systems means two or more systems.
Blockchain (Blockchain) is a term in the field of information technology, and is essentially a shared database, in which data or information is stored, and has the characteristics of 'non-counterfeitable', 'whole-course trace', 'traceable', 'transparent disclosure', 'collective maintenance', and the like. Therefore, the data acquisition and sharing of any charging pile are carried out through the block chain, and reliable adjustment is facilitated. In addition, the implementation adopts the trusted identity verification of the blockchain, when the electric vehicle needs to be charged, the electric vehicle is connected into the charging pile, and the trusted identity verification is carried out, so that the reliable sharing of information is ensured.
As shown in fig. 1, the present embodiment provides a short-time charging scheduling method based on the combination of a blockchain and a neural network, which implements charging scheduling of a plurality of charging piles under the same ac transformer device, and specifically includes the following steps:
firstly, constructing a distributed data communication model of a plurality of charging piles under the same alternating-current transformer equipment by adopting the consensus communication of a block chain. The data sharing of a plurality of charging piles in the same distributed data communication model is accessed, and the real-time performance is realized. For example, 5 charging piles are disposed under the same ac transformer apparatus. The information among the plurality of charging piles of the same distributed data communication model has sharing property, namely the disclosure is transparent, so that the later data processing and sharing are facilitated.
Secondly, an initial scheduling model is built according to charging requirements, initial model training can be conducted by adopting initial data of charging of the plurality of groups of electric vehicles of the same type of charging stations, and the initial data comprise the current residual electric quantity of the first electric vehicle battery, first charging voltage and first charging current. The more comprehensive and more numerous the original data are collected, the more accurate and reliable initial model is trained, and the later correction amount is relatively reduced. In this embodiment, in order to obtain reverse correction and accurately close to the scheduling requirement, a loss function is set between a hidden layer and an output layer of the BP neural network model through the BP neural network model, where the loss function may be a mean square error loss function.
In this embodiment, training is performed by using a BP neural network model to obtain a first prediction scheduling model corresponding to original data, and the first prediction scheduling model is used as an initial scheduling model of the charging station (i.e., ac transformer equipment, 5 charging piles). The initial scheduling model belongs to theory, provides a basis for later actual correction adjustment, and shared data of the block chain is used as the later actual correction adjustment.
Specifically, training is performed by using a BP neural network model to obtain a first predictive scheduling model corresponding to original data, and the method comprises the following steps:
(1) Selecting a certain charging pile under the same alternating-current transformer equipment, charging according to original data, and keeping the rest charging pile under the same alternating-current transformer equipment in a non-charging state;
(2) And inputting the original data into the BP neural network model for training, and obtaining a first predictive scheduling model corresponding to the original data.
Third, the first predictive scheduling model is stored in any charging pile in a distributed manner by adopting the consensus communication of the block chain, and a scheduling power threshold is preset, and in the embodiment, the scheduling power threshold is
The expression of (2) is:
wherein k represents the ratio coefficient of the rated power sum of the charging piles connected to the same alternating-current transformer equipment and the total power of the alternating-current transformer equipment; n represents the number of charging piles;
indicating the rated power of the charging pile.
And judging whether the current charging state can be connected with the electric vehicle for charging or not by using the dispatching power threshold value, and judging whether the current charging state has charging power allowance or not, so that the transformer equipment can work at rated power or below.
And fourthly, updating the first prediction scheduling model according to the current actual situation, and acquiring the charging data of any current charging pile, namely the sharing characteristic of the block chain by using the distributed data communication model. Inputting the charging data into a first predictive scheduling model, and updating the first predictive scheduling model to obtain a second predictive scheduling model; and outputting the current charging schedule data by using the second predictive schedule model. For example, when the electric vehicle is just put into operation, no electric vehicle is connected into the charging pile for charging, and then the electric vehicle is still the original first prediction scheduling model; with charging investment, the original predictive scheduling model is necessarily revised and updated. For example: at present, two electric vehicles are connected into a charging pile for charging, so that charging data of the electric vehicles are collected, and correction of a first prediction scheduling model is carried out, specifically:
(1) Sequentially acquiring charging data of a charging pile in a charging use state under the same alternating-current transformer equipment by using a distributed data communication model; the charging data comprises the current residual capacity of the second electric vehicle battery, a second charging voltage and a second charging current. In the embodiment, the characteristics of sharing, disclosure transparency, illegal tampering prevention and the like of the block chain are utilized to ensure the real-time sharing of the data of any charging pile, and provide a guarantee for the later scheduling adjustment.
(2) And inputting the current residual quantity of the second electric vehicle battery, the second charging voltage and the second charging current into the first prediction scheduling model, performing prediction training, obtaining a first error value (namely an error value between the output of the original data input into the first prediction scheduling model and the output of the charging data input into the first prediction scheduling model), and updating the first prediction scheduling model and obtaining the second prediction scheduling model by adopting counter propagation updating weights. The first error is the difference between the theoretical output and the predicted output, and the difference is propagated in the opposite direction so as to obtain more accurate scheduling data.
In this embodiment, if the corresponding power data in the charging schedule data output by the second prediction schedule model is smaller than the schedule power threshold and there is a charging pile in a non-charging state, then a new and idle charging pile may be connected to the electric vehicle for charging. In the embodiment, the block chain trusted identity verification is adopted to acquire the electric vehicle access charging pile signal, acquire the third current battery working condition data and the third charging data of the electric vehicle, and perform distributed storage on the third current battery working condition data and the third charging data. The third current battery working condition data and the third charging data are data such as the battery residual capacity, the charging voltage, the charging current and the like of the electric vehicle. Because a new electric vehicle is charged and connected, and scheduling correction is needed, third current battery working condition data and third charging data are input into a second prediction scheduling model to obtain a second error value, a counter-propagation updating weight is adopted to update the second prediction scheduling model, the updated second prediction scheduling model is stored in a distributed mode, and the scheduling model is updated by analogy.
In addition, if the corresponding power data in the charging schedule data output by the second predictive schedule model is smaller than the schedule power threshold, and any charging pile is in a charging state; for example, one or a plurality of charging piles are in high-power charging, the rest of charging piles are in a slow charging process, at this time, the charging data of any current charging pile is collected by using a distributed data communication model, and a second prediction scheduling model is updated. Or when all the charging piles are in a charging state, and corresponding power data in the charging scheduling data output by the second prediction scheduling model is larger than or equal to a scheduling power threshold value and smaller than the maximum rated power of the alternating-current transformer equipment, collecting the charging data of any current charging pile, and updating the second prediction scheduling model.
Under special conditions, the corresponding power data in the charging schedule data output by the second predictive schedule model is equal to or greater than the schedule power threshold, one charging pile is in an idle state, and a plurality of charging piles in a charging working state are indicated to be close to the rated power of the charging pile (station), at the moment, the charging pile which is idle is temporarily not allowed to be accessed for charging, the charging data of any charging pile in the current charging state is still acquired, and the second predictive schedule model is updated.
The above embodiments are only preferred embodiments of the present invention and are not intended to limit the scope of the present invention, but all changes made by adopting the design principle of the present invention and performing non-creative work on the basis thereof shall fall within the scope of the present invention.
Claims (10)
1. The short-time charging scheduling method based on the combination of the block chain and the neural network is characterized by comprising the following steps of:
constructing a distributed data communication model of a plurality of charging piles under the same alternating-current transformer equipment by adopting consensus communication;
acquiring the initial data of charging of a plurality of groups of electric vehicles, and building a BP neural network model; training by using a BP neural network model to obtain a first predictive scheduling model corresponding to the original data; the original data comprise the current residual capacity of the first electric vehicle battery, a first charging voltage and a first charging current;
the first predictive scheduling model is stored in any charging pile in a distributed mode by adopting consensus communication, and a scheduling power threshold value is preset;
collecting charging data of any current charging pile by using a distributed data communication model; inputting the charging data into a first predictive scheduling model, and updating the first predictive scheduling model to obtain a second predictive scheduling model; outputting current charging schedule data by using the second predictive schedule model;
if the corresponding power data in the charging schedule data output by the second predictive schedule model is smaller than the schedule power threshold and a charging pile in a non-charging state exists, acquiring an electric vehicle access charging pile signal by adopting trusted identity verification, acquiring third current battery working condition data and third charging data of the electric vehicle, and carrying out distributed storage on the third current battery working condition data and the third charging data;
inputting third current battery working condition data and third charging data into a second predictive scheduling model, obtaining a second error value, updating the second predictive scheduling model by adopting counter propagation updating weight, and performing distributed storage on the updated second predictive scheduling model;
and if the corresponding power data in the charging schedule data output by the second predictive schedule model is equal to or greater than the schedule power threshold, continuously collecting the charging data of any charging pile, and updating the second predictive schedule model.
2. The short-time charge scheduling method based on the combination of blockchain and neural network according to claim 1, wherein training by using a BP neural network model to obtain a first predictive scheduling model corresponding to original data comprises:
selecting a certain charging pile under the same alternating-current transformer equipment, charging according to original data, and keeping the rest charging pile under the same alternating-current transformer equipment in a non-charging state;
and inputting the original data into the BP neural network model for training, and obtaining a first predictive scheduling model corresponding to the original data.
3. The short-time charging scheduling method based on the combination of the blockchain and the neural network according to claim 1, wherein charging data of any current charging pile is collected by using a distributed data communication model; inputting the charging data into the first predictive scheduling model, updating the first predictive scheduling model, and obtaining a second predictive scheduling model, wherein the method comprises the following steps:
sequentially acquiring charging data of a charging pile in a charging use state under the same alternating-current transformer equipment by using a distributed data communication model; the charging data comprise the current residual capacity of the second electric vehicle battery, a second charging voltage and a second charging current;
inputting the current residual capacity, the second charging voltage and the second charging current of the second electric vehicle battery into a first prediction scheduling model, performing prediction training, obtaining a first error value, and updating the weight by adopting back propagation;
and updating the first predictive scheduling model and obtaining a second predictive scheduling model.
4. The short-time charge scheduling method based on the combination of the blockchain and the neural network according to claim 1, 2 or 3, wherein a BP neural network model is built, and a loss function is set between a hidden layer and an output layer of the BP neural network model; the loss function is a mean square error loss function.
5. The short-time charging scheduling method based on the combination of the blockchain and the neural network according to claim 1 or 2, wherein the method is characterized in that the original data of the charging of the plurality of groups of electric vehicles and the charging data of any charging pile are obtained, and abnormal data points in the original data and the charging data are removed respectively by adopting linear fitting.
6. The blockchain and neural network joint-based short-time charge scheduling method of claim 4, wherein training of the BP neural network model includes forward propagation and backward propagation; the back propagation takes as input the second error value and transmits to the hidden layer of the BP neural network model.
7. The blockchain and neural network joint-based short-time charge scheduling method of claim 1, wherein the scheduling power threshold value
The expression of (2) is:
wherein k represents the ratio coefficient of the rated power sum of the charging piles connected to the same alternating-current transformer equipment and the total power of the alternating-current transformer equipment; n represents the number of charging piles;
indicating the rated power of the charging pile.
8. The blockchain and neural network joint-based short-time charge scheduling method of claim 3, further comprising: updating the first predictive scheduling model to obtain a second predictive scheduling model, and carrying out distributed storage on the second predictive scheduling model.
9. The blockchain and neural network joint-based short-time charge scheduling method of claim 1, further comprising: and if the corresponding power data in the charging schedule data output by the second predictive schedule model is smaller than the schedule power threshold and any charging pile is in a charging state, acquiring the charging data of any current charging pile by using the distributed data communication model, and updating the first predictive schedule model.
10. The short-time charging scheduling method based on the combination of the blockchain and the neural network according to claim 1, wherein the distributed data communication model collects charging data of any charging pile, performs distributed storage, and is stamped with a current time stamp.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310371117.XA CN116090675B (en) | 2023-04-10 | 2023-04-10 | Short-time charging scheduling method based on combination of block chain and neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310371117.XA CN116090675B (en) | 2023-04-10 | 2023-04-10 | Short-time charging scheduling method based on combination of block chain and neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116090675A CN116090675A (en) | 2023-05-09 |
| CN116090675B true CN116090675B (en) | 2023-06-09 |
Family
ID=86208629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310371117.XA Active CN116090675B (en) | 2023-04-10 | 2023-04-10 | Short-time charging scheduling method based on combination of block chain and neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116090675B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111993944A (en) * | 2020-07-27 | 2020-11-27 | 中国能源建设集团天津电力设计院有限公司 | Electric automobile that possesses marginal calculation function fills electric pile |
| CN112019338A (en) * | 2019-05-31 | 2020-12-01 | 浙江工商大学 | Lightweight safety smart power grid communication method and system based on block chain |
| CN113103905A (en) * | 2021-04-02 | 2021-07-13 | 山东大卫国际建筑设计有限公司 | Intelligent charging distribution adjusting method, device, equipment and medium for electric automobile |
| EP3960530A1 (en) * | 2020-08-31 | 2022-03-02 | Bayerische Motoren Werke Aktiengesellschaft | Apparatus, method and computer program for determining a plurality of information of a location of a charging station |
| CN114648171A (en) * | 2022-04-08 | 2022-06-21 | 国网湖南省电力有限公司 | Electric vehicle charging pile load prediction method, system and storage medium |
| CN115545486A (en) * | 2022-10-11 | 2022-12-30 | 北京工业大学 | Electric vehicle charging scheduling method based on block chain |
| CN115837855A (en) * | 2023-02-15 | 2023-03-24 | 南方电网产业投资集团有限责任公司 | Charging pile output power control method and device, charging pile and storage medium |
| CN115907076A (en) * | 2022-09-30 | 2023-04-04 | 国网重庆市电力公司营销服务中心 | Mountain city charging load probability prediction system and method |
-
2023
- 2023-04-10 CN CN202310371117.XA patent/CN116090675B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112019338A (en) * | 2019-05-31 | 2020-12-01 | 浙江工商大学 | Lightweight safety smart power grid communication method and system based on block chain |
| CN111993944A (en) * | 2020-07-27 | 2020-11-27 | 中国能源建设集团天津电力设计院有限公司 | Electric automobile that possesses marginal calculation function fills electric pile |
| EP3960530A1 (en) * | 2020-08-31 | 2022-03-02 | Bayerische Motoren Werke Aktiengesellschaft | Apparatus, method and computer program for determining a plurality of information of a location of a charging station |
| CN113103905A (en) * | 2021-04-02 | 2021-07-13 | 山东大卫国际建筑设计有限公司 | Intelligent charging distribution adjusting method, device, equipment and medium for electric automobile |
| CN114648171A (en) * | 2022-04-08 | 2022-06-21 | 国网湖南省电力有限公司 | Electric vehicle charging pile load prediction method, system and storage medium |
| CN115907076A (en) * | 2022-09-30 | 2023-04-04 | 国网重庆市电力公司营销服务中心 | Mountain city charging load probability prediction system and method |
| CN115545486A (en) * | 2022-10-11 | 2022-12-30 | 北京工业大学 | Electric vehicle charging scheduling method based on block chain |
| CN115837855A (en) * | 2023-02-15 | 2023-03-24 | 南方电网产业投资集团有限责任公司 | Charging pile output power control method and device, charging pile and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| 基于大数据分析的充电桩负荷预测研究与数据平台设计;郭沫;《中国优秀硕士学位论文全文数据库工程科技II辑》(第2期);C034-1834 * |
| 基于时空信息的充电站负荷预测及V2G定价策略;罗清松;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》(第1期);C034-1100 * |
| 基于联盟区块链的电动汽车可信充电模型;穆蕾 等;《计算机系统应用》;第32卷(第2期);119-127页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116090675A (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112134300B (en) | Reservation-based rolling optimization operation method and system for electric vehicle light storage charging station | |
| Wang et al. | Aggregated electric vehicle load modeling in large-scale electric power systems | |
| He et al. | Optimal scheduling for charging and discharging of electric vehicles | |
| Li et al. | Emission-concerned wind-EV coordination on the transmission grid side with network constraints: Concept and case study | |
| CN107491867B (en) | Safety checking and evaluating method for multi-cycle transmission and transformation maintenance plan | |
| CN104376385A (en) | Microgrid power price optimizing method | |
| Erden et al. | Adaptive V2G peak shaving and smart charging control for grid integration of PEVs | |
| Haque et al. | Exploration of dispatch model integrating wind generators and electric vehicles | |
| Sharifi et al. | Leveraging owners’ flexibility in smart charge/discharge scheduling of electric vehicles to support renewable energy integration | |
| Zheng et al. | Optimal short-term power dispatch scheduling for a wind farm with battery energy storage system | |
| CN113971530A (en) | Novel power system source network and storage cooperation oriented power balancing method | |
| Nguyen et al. | Smart charging and discharging of electric vehicles to support grid with high penetration of renewable energy | |
| Li et al. | Energy management model of charging station micro-grid considering random arrival of electric vehicles | |
| Hayati et al. | A Two-Stage Stochastic Optimization Scheduling Approach for Integrating Renewable Energy Sources and Deferrable Demand in the Spinning Reserve Market | |
| Chang et al. | Model predictive control based energy collaborative optimization management for energy storage system of virtual power plant | |
| Dong et al. | Research on day-ahead optimal dispatching of virtual power plants considering the coordinated operation of diverse flexible loads and new energy | |
| CN116090675B (en) | Short-time charging scheduling method based on combination of block chain and neural network | |
| Diaz-Londono et al. | Coordination strategies for electric vehicle chargers integration in electrical grids | |
| Zhang et al. | Transfer deep reinforcement learning-based large-scale V2G continuous charging coordination with renewable energy sources | |
| CN116307437A (en) | Virtual power plant optimal scheduling method and system based on carbon emission right exchange mechanism | |
| CN112417048B (en) | Block chain-based intelligent micro-grid system scheduling method, storage medium and equipment | |
| Yao et al. | Determination of a dispatch strategy to maximize income for a wind turbine-BESS power station | |
| CN114400653A (en) | Electric vehicle charging optimization method based on day-ahead load predicted value | |
| CN113887902A (en) | Wind power cluster electric quantity distribution method based on scheduling electric quantity proportion | |
| Zhou et al. | A decentralized access control algorithm for PHEV charging in smart 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 |