CN110929950B - Electric automobile load prediction method and system - Google Patents
Electric automobile load prediction method and system Download PDFInfo
- Publication number
- CN110929950B CN110929950B CN201911212935.5A CN201911212935A CN110929950B CN 110929950 B CN110929950 B CN 110929950B CN 201911212935 A CN201911212935 A CN 201911212935A CN 110929950 B CN110929950 B CN 110929950B
- Authority
- CN
- China
- Prior art keywords
- individuals
- state
- electric automobile
- charging
- electric
- 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
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000009826 distribution Methods 0.000 claims abstract description 44
- 230000003993 interaction Effects 0.000 claims abstract description 15
- 230000007704 transition Effects 0.000 claims description 60
- 239000011159 matrix material Substances 0.000 claims description 51
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000003062 neural network model Methods 0.000 claims description 16
- 238000012216 screening Methods 0.000 claims description 10
- 230000008901 benefit Effects 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000004321 preservation Methods 0.000 claims description 6
- 230000002068 genetic effect Effects 0.000 claims description 5
- 230000014759 maintenance of location Effects 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000004088 simulation Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 2
- 238000007599 discharging Methods 0.000 description 5
- 230000002354 daily effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000012546 transfer Methods 0.000 description 4
- 210000000349 chromosome Anatomy 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 108090000623 proteins and genes Proteins 0.000 description 2
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 229910002056 binary alloy Inorganic materials 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000010792 warming Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- 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
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Biomedical Technology (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method and a system for predicting load of an electric automobile. The method comprises the following steps: predicting the electric vehicle holding quantity of the target area; predicting a random path of the electric vehicle in the target area; based on the conservation amount and the random path, predicting the load distribution condition of the electric automobile in one day; and optimizing the load distribution condition based on the electric energy interaction relation between the vehicle and the power grid to obtain the optimized electric vehicle load distribution. The invention can realize the prediction of the electric automobile load aiming at a certain area.
Description
Technical Field
The invention relates to the field of load prediction, in particular to a method and a system for predicting the load of an electric automobile.
Background
Under the conditions of shortage of global resources and global warming, the development of electric vehicles is promoted to relieve the pressure of global resources and climate, so that the electric vehicles are in the spotlight of people. However, with the improvement of the holding amount of the electric vehicle, the electric vehicle is connected to the power grid in a large scale, and the influence on the generation of the power grid is not negligible. The holding quantity of the electric automobile in one area has a larger influence on the charging load of the electric automobile in the area, and the influence factors of the increase of the holding quantity of the electric automobile per year also have a plurality of influence factors. The charge load influence factors on the electric automobile are mainly in space and time. Under the influence of space factors of electric automobile charging, electric automobile's difference of the travel route every day, the randomness is very big. Under the condition of incomplete consideration, the daily load prediction of the electric automobile has great difficulty and inaccuracy, and the running route, weather, time and personal wish of the electric automobile are influenced to a certain extent. The charging time factor of the electric automobile has great influence on the charging load of the electric automobile. The influence of space is mostly reflected in the spatial path of the electric vehicle. And the electric automobile is accessed into the power grid on a large scale in the same time period, so that the peak value of the power grid is greatly affected, and the stability of the power grid is affected to a certain extent. Therefore, load prediction for electric vehicles is a critical technical problem in the current stage.
At present, the load prediction problem of an electric automobile generally comprises the following two prediction methods: (1) Directly using historical data to predict load through a neural network; (2) The data of the daily driving mileage counted abroad is used for prediction. The prediction methods of the former and the latter are too non-representative, and cannot accurately explain the load prediction situation of the electric vehicle in a certain area, so that the electric vehicle cannot be subjected to targeted adjustment based on the influence of the electric vehicle on the power grid.
Disclosure of Invention
The invention aims to provide a method and a system for predicting electric automobile load, which are used for predicting the electric automobile load of a certain area.
In order to achieve the above object, the present invention provides the following solutions:
an electric vehicle load prediction method, comprising:
predicting the electric vehicle holding quantity of the target area;
predicting a random path of the electric vehicle in the target area;
based on the conservation amount and the random path, predicting the load distribution condition of the electric automobile in one day;
and optimizing the load distribution condition based on the electric energy interaction relation between the vehicle and the power grid to obtain the optimized electric vehicle load distribution.
Optionally, the predicting the electric vehicle holding amount in the target area specifically includes:
establishing a neural network model of the relation between the conservation quantity influence factors and the conservation quantity;
acquiring the value of each retention quantity influence factor;
and inputting the value of each conservation quantity influence factor into the neural network model to obtain the conservation quantity of the electric vehicle in the target area.
Optionally, the predicting the random path of the electric vehicle in the target area specifically includes:
establishing a main chain by taking a destination as a first factor affecting a random path;
establishing a main chain and an auxiliary chain by taking weather as a second factor affecting a random path;
establishing a secondary auxiliary chain by taking the date as a third factor affecting the random path; the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment, and the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment; the state of the secondary auxiliary chain at the current moment is influenced by the state of the secondary auxiliary chain at the last moment;
calculating the transition probability among all states of the main chain under the influence of the main chain and the auxiliary chain when the main chain and the auxiliary chain are in any state, and obtaining a first transition probability matrix;
Calculating the transition probability among all states of the main and auxiliary chains under the influence of the auxiliary chains when the auxiliary chains are in any state, and obtaining a second transition probability matrix;
and simulating the random path of the electric automobile by using the first transition probability matrix and the second transition probability matrix to obtain the random path of the electric automobile in the target area.
Optionally, the predicting the load distribution situation of the electric automobile in one day based on the holding amount and the random path specifically includes:
when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is home, determining that the charging mode of the electric automobile is slow charging;
when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is not home, determining that the charging mode of the electric automobile is quick charging;
when the charge state of the electric automobile is higher than the first charge threshold and lower than the second nuclear power threshold, determining that the charging mode of the electric automobile is slow charging;
when the charge state of the electric automobile is higher than the second charge threshold value, determining that the electric automobile is not charged;
and determining the charging power of all the electric vehicles in the charging state at any moment according to the charging modes of all the electric vehicles to obtain a load curve in one day.
Optionally, the optimizing the load distribution condition based on the electric energy interaction relationship between the vehicle and the power grid to obtain the optimized load distribution of the electric automobile specifically includes:
according to the electric energy interaction relation between the vehicle and the power grid, determining that the maximum benefit and the minimum load peak-valley difference are taken as objective functions, optimizing the load distribution condition by taking the limiting condition of the discharge power and the constraint condition that the residual charge of the battery is larger than the minimum residual charge, and initializing the optimized solution as a population individual;
initializing iteration times;
solving objective functions of individuals in each population to obtain fitness values of each solution;
individual selection is carried out according to the fitness value, and the selected individuals are crossed and mutated to obtain offspring individuals;
calculating the crowding distances of the child individuals and the selected individuals, and carrying out non-inferior layering on each crowding distance to obtain non-inferior individuals;
pareto sorting is carried out on the non-inferior individuals to obtain sorted individuals;
dividing the sequenced individuals into four parts according to the sequence, and solving the inverse solution of the individuals in the middle two parts to obtain new individuals in the middle two parts;
calculating the crowding distance of the new individuals in the middle two parts, and carrying out non-inferior layering on the crowding distance of the new individuals in the middle two parts to obtain non-inferior individuals after secondary layering;
Pareto sorting is carried out on the individuals in the first part and the non-inferior individuals after secondary layering, so that the individuals after secondary sorting are obtained;
screening the first N individuals in the individuals after the secondary sequencing, and judging whether the current iteration number is greater than or equal to the maximum iteration number or not to obtain a judging result;
if the judgment result shows that the solution is not the same, adding 1 to the iteration times, updating the population individuals to the first N screened individuals, and returning to the step of solving the objective function of each population individual to obtain the fitness value of each solution;
and if the judgment result shows that the solution is yes, taking the first N screened individuals as the optimal solution.
An electric vehicle load prediction system, comprising:
the holding quantity prediction module is used for predicting the holding quantity of the electric automobile in the target area;
the path prediction module is used for predicting the random path of the electric automobile in the target area;
the load distribution prediction module is used for predicting the load distribution situation of the electric automobile in one day based on the conservation quantity and the random path;
and the optimizing module is used for optimizing the load distribution condition based on the electric energy interaction relation between the vehicle and the power grid to obtain the optimized electric vehicle load distribution.
Optionally, the reserve prediction module includes:
the neural network model building unit is used for building a neural network model of the relation between the preservation quantity influence factors and the preservation quantity;
an input amount acquisition unit configured to acquire values of the respective holding amount influence factors;
the neural network prediction unit is used for inputting the value of each conservation quantity influence factor into the neural network model to obtain the conservation quantity of the electric vehicle in the target area.
Optionally, the path prediction module includes:
a main chain establishing unit for establishing a main chain with the destination as a first factor affecting the random path;
the main and auxiliary chain establishing unit is used for establishing a main and auxiliary chain by taking weather as a second factor affecting a random path;
a secondary auxiliary chain establishing unit for establishing a secondary auxiliary chain by taking the date as a third factor affecting the random path; the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment, and the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment; the state of the secondary auxiliary chain at the current moment is influenced by the state of the secondary auxiliary chain at the last moment;
The first transition probability matrix calculation unit is used for calculating the transition probability among the states of the main chain under the influence of the main chain and the auxiliary chain when the main chain and the auxiliary chain are in any state, so as to obtain a first transition probability matrix;
the second transition probability matrix calculation unit is used for calculating the transition probability among the states of the main and auxiliary chains under the influence of the secondary and auxiliary chains when the secondary and auxiliary chains are in any state, so as to obtain a second transition probability matrix;
and the random path simulation unit is used for simulating the random path of the electric automobile by utilizing the first transition probability matrix and the second transition probability matrix to obtain the random path of the electric automobile in the target area.
Optionally, the load distribution prediction module includes:
the first charging mode determining unit is used for determining that the charging mode of the electric automobile is slow charging when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is home;
the second charging mode determining unit is used for determining that the charging mode of the electric automobile is quick charging when the charging state of the electric automobile is lower than a first charging threshold value and the destination of the electric automobile is not home;
the third charging mode determining unit is used for determining that the charging mode of the electric automobile is slow charging when the charging state of the electric automobile is higher than the first charging threshold and lower than the second nuclear power threshold;
A fourth charging mode determining unit, configured to determine that the electric vehicle is not charged when the state of charge of the electric vehicle is higher than the second charge threshold;
the load curve construction unit is used for determining the charging power of all the electric vehicles in the charging state at any moment according to the charging modes of all the electric vehicles to obtain a load curve in one day.
Optionally, the optimizing module includes:
the optimal solution calculation unit is used for determining that the maximum benefit and the minimum load peak-valley difference are taken as target functions according to the electric energy interaction relation between the vehicle and the power grid, optimizing the load distribution condition by taking the limiting condition of the discharge power and the constraint condition that the residual charge of the battery is larger than the minimum residual charge, and initializing the optimized solution as a population individual;
the iteration initialization unit is used for initializing the iteration times;
the fitness value calculation unit is used for solving the objective function of each population individual to obtain the fitness value of each solution;
the genetic unit is used for selecting individuals according to the fitness value, and intersecting and mutating the selected individuals to obtain offspring individuals;
the first non-inferior layering unit is used for calculating the crowding distances of the child individuals and the selected individuals and performing non-inferior layering on each crowding distance to obtain non-inferior individuals;
The first sorting unit is used for sorting the non-inferior individuals by pareto to obtain sorted individuals;
the reverse solution calculation unit is used for dividing the sequenced individuals into four parts according to the sequence, and solving the reverse solutions of the individuals in the middle two parts to obtain new individuals in the middle two parts;
the second non-inferior layering unit is used for calculating the crowding distance of the new individuals in the middle two parts and performing non-inferior layering on the crowding distance of the new individuals in the middle two parts to obtain non-inferior individuals after secondary layering;
the second sorting unit is used for sorting the individuals in the first part and the non-inferior individuals after secondary layering to obtain individuals after secondary sorting;
the screening and judging unit is used for screening the first N individuals in the secondarily ordered individuals and judging whether the current iteration number is greater than or equal to the maximum iteration number or not to obtain a judging result;
the updating and returning unit is used for adding 1 to the iteration times if the judging result indicates no, updating the population individuals into the first N screened individuals and returning to the fitness value calculating unit;
and the optimal solution determining unit is used for taking the first N screened individuals as optimal solutions if the judging result shows that the judgment result is yes.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: according to the electric vehicle load prediction method and system, the electric vehicle holding quantity and the electric vehicle path of the target area are predicted, so that the electric vehicle load of the target area is predicted. Therefore, the invention can realize the prediction of the electric vehicle load of a certain area.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of a method for predicting load of an electric vehicle according to embodiment 1 of the present invention;
FIG. 2 is a state transition diagram of a multi-chain Markov model;
fig. 3 is a system configuration diagram of an electric vehicle load prediction system according to embodiment 3 of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1:
fig. 1 is a flowchart of a method for predicting load of an electric vehicle according to embodiment 1 of the present invention.
Referring to fig. 1, the electric vehicle load prediction method includes:
step 101: and predicting the electric vehicle holding quantity of the target area.
The step 101 specifically includes:
establishing a neural network model of the relation between the conservation quantity influence factors and the conservation quantity;
acquiring the value of each retention quantity influence factor;
and inputting the value of each conservation quantity influence factor into the neural network model to obtain the conservation quantity of the electric vehicle in the target area.
Step 102: and predicting the random path of the electric automobile in the target area.
The step 102 specifically includes:
establishing a main chain by taking a destination as a first factor affecting a random path;
establishing a main chain and an auxiliary chain by taking weather as a second factor affecting a random path;
establishing a secondary auxiliary chain by taking the date as a third factor affecting the random path; the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment, and the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment; the state of the secondary auxiliary chain at the current moment is influenced by the state of the secondary auxiliary chain at the last moment;
Calculating the transition probability among all states of the main chain under the influence of the main chain and the auxiliary chain when the main chain and the auxiliary chain are in any state, and obtaining a first transition probability matrix;
calculating the transition probability among all states of the main and auxiliary chains under the influence of the auxiliary chains when the auxiliary chains are in any state, and obtaining a second transition probability matrix;
and simulating the random path of the electric automobile by using the first transition probability matrix and the second transition probability matrix to obtain the random path of the electric automobile in the target area.
Step 103: and predicting the load distribution condition of the electric automobile in one day based on the holding quantity and the random path.
The step 103 specifically includes:
when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is home, determining that the charging mode of the electric automobile is slow charging;
when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is not home, determining that the charging mode of the electric automobile is quick charging;
when the charge state of the electric automobile is higher than the first charge threshold and lower than the second nuclear power threshold, determining that the charging mode of the electric automobile is slow charging;
When the charge state of the electric automobile is higher than the second charge threshold value, determining that the electric automobile is not charged;
and determining the charging power of all the electric vehicles in the charging state at any moment according to the charging modes of all the electric vehicles to obtain a load curve in one day.
Step 104: and optimizing the load distribution condition based on the electric energy interaction relation between the vehicle and the power grid to obtain the optimized electric vehicle load distribution.
The step 104 specifically includes:
according to the electric energy interaction relation between the vehicle and the power grid, determining that the maximum benefit and the minimum load peak-valley difference are taken as objective functions, optimizing the load distribution condition by taking the limiting condition of the discharge power and the constraint condition that the residual charge of the battery is larger than the minimum residual charge, and initializing the optimized solution as a population individual;
initializing iteration times;
solving objective functions of individuals in each population to obtain fitness values of each solution;
individual selection is carried out according to the fitness value, and the selected individuals are crossed and mutated to obtain offspring individuals;
calculating the crowding distances of the child individuals and the selected individuals, and carrying out non-inferior layering on each crowding distance to obtain non-inferior individuals;
Pareto sorting is carried out on the non-inferior individuals to obtain sorted individuals;
dividing the sequenced individuals into four parts according to the sequence, and solving the inverse solution of the individuals in the middle two parts to obtain new individuals in the middle two parts;
calculating the crowding distance of the new individuals in the middle two parts, and carrying out non-inferior layering on the crowding distance of the new individuals in the middle two parts to obtain non-inferior individuals after secondary layering;
pareto sorting is carried out on the individuals in the first part and the non-inferior individuals after secondary layering, so that the individuals after secondary sorting are obtained;
screening the first N individuals in the individuals after the secondary sequencing, and judging whether the current iteration number is greater than or equal to the maximum iteration number or not to obtain a judging result;
if the judgment result shows that the solution is not the same, adding 1 to the iteration times, updating the population individuals to the first N screened individuals, and returning to the step of solving the objective function of each population individual to obtain the fitness value of each solution;
and if the judgment result shows that the solution is yes, taking the first N screened individuals as the optimal solution.
Example 2:
this embodiment 2 is a more specific embodiment of embodiment 1.
The basic principle of this embodiment 2 is as follows: and predicting the conservation amount of the electric automobile based on the BP neural network, establishing a Markov model of a multi-chain electric automobile path, and predicting the load distribution of one day by using Monte Carlo. The peak value of the predicted load is optimized using V2G (Vehicle-to-grid).
The procedure of this example 2 is as follows:
(1) Classifying factors influencing the electric vehicle retention, and establishing a BP neural network model to predict the electric vehicle retention.
(2) And establishing a random path model of the electric automobile, dividing factors influencing the random path into main factors and secondary factors, and performing simulation modeling by establishing a multi-chain Markov model.
The method comprises the steps that a main influence factor of a random path of an electric automobile in a region is weather, a secondary influence factor is time, a certain mutual influence exists between the main influence factor and the secondary influence factor, and a model of the random path is built to predict the path of the electric automobile in one day in the region.
(3) And predicting the load distribution condition of the electric automobile in one day by adopting Monte Carlo simulation according to the predicted holding quantity of the target area and the simulation result of the random path of the electric automobile in the target area.
Modeling prediction is realized by determining whether the state of charge (SOC) of the electric vehicle requires charging and selecting a charging mode to calculate the charging power at each moment.
(4) According to the established load distribution condition of the electric automobile in one day, the peak-valley difference of the load distribution of the electric automobile is larger, the peak-valley difference is optimized through V2G, and the peak clipping and valley filling effects are realized.
The maximum benefit obtained by the user is optimized by taking peak clipping and valley filling as objective functions, and the stability of the power grid can be improved while peak clipping and valley filling are realized.
The specific scheme of the step (1) is as follows:
the electric automobile is provided with a plurality of factors, the most important factors are population quantity, economic level and government support force, and the main factors are quite closely related to the electric automobile. Under the influence factors of the holding quantity of the electric automobile comprehensively considered, selecting the population, the distribution of charging piles, the domestic total production value of people, the financial income of cities and the highway passenger traffic as important influence indexes, and selecting the 5 factors as the input indexes of the electric automobile holding quantity prediction model.
The establishment process of the prediction model of the conservation quantity of the electric automobile based on the BP neural network comprises the following steps:
sample data of the input layer is
X＝(x 1 ,x 2 ,…,x I ) (1)
Wherein x is 1 The 1 st relevant influence index for representing the holding quantity of the electric automobile; x is x 2 The 2 nd related influence index for representing the holding quantity of the electric automobile; x is x i An ith relevant influence index for representing the holding quantity of the electric automobile; x is x I The I-th relevant influence index indicating the holding amount of the electric automobile.
Output layer corresponding to input layer in prediction model
Y＝y (2)
The input of each unit of the hidden layer is
Wherein omega ij Represented as input layer to hidden layer connection weights, θ j Expressed as a threshold value of units of an hidden layer, J is expressed as the number of hidden layers, and J is the number of each unit of the hidden layer.
The transfer function adopts a sigmoid function
Then the hidden layer element outputs as
The input of the output layer unit is
L χ Is the input of the output layer unit. χ is the unit of the output layer.
The output of the output layer unit is
Wherein c is the output of the output layer unit, v jχ Represented as implicit layer to output layer link weights, gamma χ Is the threshold of the output layer unit.
The specific scheme of the step (2) is as follows:
definition 1: the Markov process of the random path of the electric automobile means that the state at the time t of the path selection of the electric automobile is only determined by the state at the time t-1. This characteristic, known as markov of random paths of electric vehicles, can be expressed as
P r {X t+1 ＝i t+1 |X 1 ＝i 1 ,X 2 ＝i 2 ,...,X t ＝i t }＝P r {X t+1 ＝i t+1 |X t ＝i t } (8)
In the above formula: p (P) r Is a conditional probability, X t 、X t+1 The state variables i of the electric automobile at the time t and the time t+1 are respectively represented t 、i t+1 Respectively X t 、X t+1 The specific state corresponding to the method.
While this embodiment 2 converts the single-chain markov based on definition 1 into multi-chain markov to simulate the random path and build a model of the random path, the specific procedure is as follows:
X in multiple chain Markov as shown in FIG. 2 1 ,X 2 ,...X t Representing the Markov backbone at 1-tY, state of (2) 1 ,Y 2 ,...,Y t Representing the state of the main and auxiliary chains of Markov at 1-t, Z 1 ,Z 2 ,...,Z t Representing the state of the secondary chain of markov at 1-t. The state of the main chain of the Markov at the time t is influenced by the state of the main chain at the time t-1 and the state of the main chain and the auxiliary chain at the time t-1. Similarly, when the state of the main and auxiliary chains of the Markov at the time t is influenced by the state of the main and auxiliary chains at the time t-1, the state of the auxiliary chain at the time t-1 is also influenced by the state of the auxiliary chain at the time t-1, and the time t of the auxiliary chain is influenced by the state of the auxiliary chain at the time t-1. In the embodiment 2, the multi-chain Markov model is built by taking the selection of the electric automobile path as the main chain, the weather as the main and auxiliary chains and the date as the secondary and auxiliary chains.
S 1 = { home, company, mall, others };
S 2 = { sunny day, cloudy day, rainy day, snowy day, others };
S 3 = { monday, tuesday, wednesday, thursday, friday, sunday };
S 1 ,S 2 ,S 3 respectively representing N state sets of a Markov main chain, M state sets of a Markov main auxiliary chain and K state sets of a Markov secondary auxiliary chain.
Initial state probability matrix pi 1 ,π 2 ,π 3 The probability matrix of the initial time t=1 of the main chain, the auxiliary chain and the auxiliary chain is respectively represented.
The set of state transition probability matrices includes a and B. A comprises A 1 ,A 2 ,...A M Equal M state transition probability matrices, where element A in A m The transition probability among N states of the Markov main chain is represented when the state of the Markov main and auxiliary chains is m. Set B contains B 1 ,B 2 ,...B K Equal K state transition probability matrices, where B k And when the state of the Markov secondary auxiliary chain is k, the transition probability among M states of the Markov primary auxiliary chain is represented.
Historical data of route selection, weather and time of electric vehicle (private car) within 3 months of Qin Royal islandAfter acquisition and processing, the obtained 3 groups of data divide the path selection, weather and time history data into N, M, K respectively, and map the path selection, weather and time sequence into a set S of states 1 ,S 2 ,S 3 。
In the case where the data is sufficient, the probability is replaced with the frequency, and the respective state transition probabilities P are calculated, which can be expressed as:
in which A m Element p of (a) df Representing the probability that the selection of a path transposes state f from state d, B k Element p of (a) gh Representing the probability that the selection of a path transposes state h from state g, 0.ltoreq.p df ≤1，0≤p gh ≤1，
X, Y, Z is used as a route selection, weather and time sequence to obtain a multi-chain Markov main chain X t And main and auxiliary chains Y t And also secondary auxiliary chain Z t Carrying out data statistics on the data to obtain a state transition matrix A m ，B k The m-step state transition matrix and the k-step state transition matrix of weather versus path selection and time versus weather are represented, respectively.
The calculation process of the total state transition matrix with the main chain affected is as follows:
1. and calculating a state transition matrix of the secondary auxiliary chain, and marking the state transition matrix as a matrix L1.
2. And calculating a state transition matrix of the influence of the secondary chain on the primary and secondary chains, and marking the state transition matrix as a matrix L2.
3. And calculating a state transition matrix of the main chain and the auxiliary chain, and marking the state transition matrix as a matrix L3.
4. And carrying out joint operation on the L1, the L2 and the L3 to obtain a main and auxiliary chain overall transfer matrix L4.
5. The main chain self-transfer matrix (denoted as matrix L5) and the state transfer matrix (denoted as matrix L6) of the influence of the main chain and the auxiliary chain on the main chain are calculated.
6. L4, L5 and L6 are multiplied to obtain the total state transition matrix with the main chain affected.
The specific scheme of the step (3) is as follows:
and randomly extracting the initial capacity of the battery and the trip time of the electric automobile, and simulating the path of the electric automobile.
And selectively charging the electric automobile by judging the state of the battery capacity:
SOC t ＜SOC*30％ (11)
SOC*30％＜SOC t ＜SOC*80％ (12)
SOC*80％＜SOC t (13)
SOC t the state of charge of the electric automobile at the time t; the SOC is a state of full charge of the electric vehicle.
When the state of charge of the electric automobile satisfies the formula (11) and the end point of the path is home, the charging mode is selected to be slow charging; when the state of charge of the electric vehicle satisfies the equation (11) and the end point of the route is not home, the charging mode is selected to be quick charging. When the state of charge of the electric vehicle satisfies the equation (12), the charging mode is selected to be slow charging. When the state of charge of the electric vehicle satisfies the expression (13), the charging mode is selected to be non-charging.
Finally, the charging time is calculated, and then the charging power at the moment t can be calculated, so as to obtain a load curve in one day.
The specific scheme of the step (4) is as follows:
the electric automobile has strong flexibility and can be charged through the power grid, and can discharge to the power grid during the peak time of electricity consumption, so that ordered charging and discharging of users are realized, further, the effect of peak clipping and valley filling is achieved, and win-win is realized between the users and the power grid. In this embodiment 2, 45% of electric car users are assumed to have signed a protocol with the power grid to participate in the dispatching of the discharge of the user side. The user participating in the discharging can meet the condition that the user realizes the maximum benefit and simultaneously minimizes the peak-valley difference of the load curve of the electric automobile, and meanwhile, the limitation of the discharging power and the situation that the residual charge of the battery is not lower than 20% are considered.
The objective function and constraint are
In the above formula: p (P) imax Is the maximum value of daily load; p (P) imin Is the minimum of daily load; pd (Pd) dt The power of the discharge; pc (Pc) ct Power for charging; price dt The electricity price is the discharge time; price ct The electricity price is the charging time; efficiency of charge and discharge of η battery; EV (EV) loss The loss of the battery of the electric automobile; EV (EV) soc Residual electric quantity of the electric automobile battery; EV (EV) MAXSOC Total battery capacity of the electric vehicle; pd (Pd) dtmax Is the maximum discharge power; pd (Pd) dt Is the power of the discharge.
The calculation of equation (14) was optimized using NSGA2 (Non-dominant rank genetic algorithm II, non-dominated Sorting Genetic Algorithm II). The chromosome is coded by adopting binary system, the connection between the power grid and the electric automobile has only two states (charging and discharging), and the line-by-line coding method is adopted. When the gene value is 1, the electric automobile is in a charged state; conversely, when the gene value is 0, the electric vehicle is in a discharging state. The length of the chromosome corresponds to the length of time of day when the chromosome is connected into the power grid.
The specific optimization process is as follows:
initializing iteration times;
solving objective functions of individuals in each population to obtain fitness values of each solution;
individual selection is carried out according to the fitness value, and the selected individuals are crossed and mutated to obtain offspring individuals;
Calculating the crowding distances of the child individuals and the selected individuals, and carrying out non-inferior layering on each crowding distance to obtain non-inferior individuals;
pareto sorting is carried out on the non-inferior individuals to obtain sorted individuals;
dividing the sequenced individuals into four parts according to the sequence, and solving the inverse solution of the individuals in the middle two parts to obtain new individuals in the middle two parts;
the calculation formula of the inverse solution is:
wherein X is i,d For an individual to be solved in the negative direction,is a reverse solution.For the best individual after ranking, i.e. the individual with the greatest crowding distance +.>The worst individual after sorting, i.e. the individual with the smallest crowding distance.
μ∈[0,1]Since the search boundary is dynamically moving, when the inverse solution exceeds the search boundary, the method causesrand () is a random valued function. />
Calculating the crowding distance of the new individuals in the middle two parts, and carrying out non-inferior layering on the crowding distance of the new individuals in the middle two parts to obtain non-inferior individuals after secondary layering;
pareto sorting is carried out on the individuals in the first part and the non-inferior individuals after secondary layering, so that the individuals after secondary sorting are obtained;
screening the first N individuals in the individuals after the secondary sequencing, and judging whether the current iteration number is greater than or equal to the maximum iteration number or not to obtain a judging result;
If the judgment result shows that the solution is not the same, adding 1 to the iteration times, updating the population individuals to the first N screened individuals, and returning to the step of solving the objective function of each population individual to obtain the fitness value of each solution;
and if the judgment result shows that the solution is yes, taking the first N screened individuals as the optimal solution.
Example 3:
fig. 3 is a system configuration diagram of an electric vehicle load prediction system according to embodiment 3 of the present invention.
Referring to fig. 3, the electric vehicle load prediction system includes:
the holding amount prediction module 301 is configured to predict an electric vehicle holding amount of a target area;
the path prediction module 302 is configured to predict a random path of the electric vehicle in the target area;
the load distribution prediction module 303 is configured to predict a load distribution situation of the electric vehicle in one day based on the holding amount and the random path;
and the optimizing module 304 is configured to optimize the load distribution condition based on the electric energy interaction relationship between the vehicle and the power grid, so as to obtain the optimized load distribution of the electric vehicle.
Optionally, the holding amount prediction module 301 includes:
the neural network model building unit is used for building a neural network model of the relation between the preservation quantity influence factors and the preservation quantity;
An input amount acquisition unit configured to acquire values of the respective holding amount influence factors;
the neural network prediction unit is used for inputting the value of each conservation quantity influence factor into the neural network model to obtain the conservation quantity of the electric vehicle in the target area.
Optionally, the path prediction module 302 includes:
a main chain establishing unit for establishing a main chain with the destination as a first factor affecting the random path;
the main and auxiliary chain establishing unit is used for establishing a main and auxiliary chain by taking weather as a second factor affecting a random path;
a secondary auxiliary chain establishing unit for establishing a secondary auxiliary chain by taking the date as a third factor affecting the random path; the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment, and the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment; the state of the secondary auxiliary chain at the current moment is influenced by the state of the secondary auxiliary chain at the last moment;
the first transition probability matrix calculation unit is used for calculating the transition probability among the states of the main chain under the influence of the main chain and the auxiliary chain when the main chain and the auxiliary chain are in any state, so as to obtain a first transition probability matrix;
The second transition probability matrix calculation unit is used for calculating the transition probability among the states of the main and auxiliary chains under the influence of the secondary and auxiliary chains when the secondary and auxiliary chains are in any state, so as to obtain a second transition probability matrix;
and the random path simulation unit is used for simulating the random path of the electric automobile by utilizing the first transition probability matrix and the second transition probability matrix to obtain the random path of the electric automobile in the target area.
Optionally, the load distribution prediction module 303 includes:
the first charging mode determining unit is used for determining that the charging mode of the electric automobile is slow charging when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is home;
the second charging mode determining unit is used for determining that the charging mode of the electric automobile is quick charging when the charging state of the electric automobile is lower than a first charging threshold value and the destination of the electric automobile is not home;
the third charging mode determining unit is used for determining that the charging mode of the electric automobile is slow charging when the charging state of the electric automobile is higher than the first charging threshold and lower than the second nuclear power threshold;
a fourth charging mode determining unit, configured to determine that the electric vehicle is not charged when the state of charge of the electric vehicle is higher than the second charge threshold;
The load curve construction unit is used for determining the charging power of all the electric vehicles in the charging state at any moment according to the charging modes of all the electric vehicles to obtain a load curve in one day.
Optionally, the optimizing module 304 includes:
the optimal solution calculation unit is used for determining that the maximum benefit and the minimum load peak-valley difference are taken as target functions according to the electric energy interaction relation between the vehicle and the power grid, optimizing the load distribution condition by taking the limiting condition of the discharge power and the constraint condition that the residual charge of the battery is larger than the minimum residual charge, and initializing the optimized solution as a population individual;
the iteration initialization unit is used for initializing the iteration times;
the fitness value calculation unit is used for solving the objective function of each population individual to obtain the fitness value of each solution;
the genetic unit is used for selecting individuals according to the fitness value, and intersecting and mutating the selected individuals to obtain offspring individuals;
the first non-inferior layering unit is used for calculating the crowding distances of the child individuals and the selected individuals and performing non-inferior layering on each crowding distance to obtain non-inferior individuals;
the first sorting unit is used for sorting the non-inferior individuals by pareto to obtain sorted individuals;
The reverse solution calculation unit is used for dividing the sequenced individuals into four parts according to the sequence, and solving the reverse solutions of the individuals in the middle two parts to obtain new individuals in the middle two parts;
the second non-inferior layering unit is used for calculating the crowding distance of the new individuals in the middle two parts and performing non-inferior layering on the crowding distance of the new individuals in the middle two parts to obtain non-inferior individuals after secondary layering;
the second sorting unit is used for sorting the individuals in the first part and the non-inferior individuals after secondary layering to obtain individuals after secondary sorting;
the screening and judging unit is used for screening the first N individuals in the secondarily ordered individuals and judging whether the current iteration number is greater than or equal to the maximum iteration number or not to obtain a judging result;
the updating and returning unit is used for adding 1 to the iteration times if the judging result indicates no, updating the population individuals into the first N screened individuals and returning to the fitness value calculating unit;
and the optimal solution determining unit is used for taking the first N screened individuals as optimal solutions if the judging result shows that the judgment result is yes.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: according to the electric vehicle load prediction method and system, the electric vehicle holding quantity and the electric vehicle path of the target area are predicted, so that the electric vehicle load of the target area is predicted. Therefore, the invention can realize the prediction of the electric vehicle load of a certain area.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (8)
1. An electric vehicle load prediction method is characterized by comprising the following steps:
predicting the electric vehicle holding quantity of the target area;
predicting a random path of the electric vehicle in the target area;
based on the conservation amount and the random path, predicting the load distribution condition of the electric automobile in one day;
Optimizing the load distribution condition based on the electric energy interaction relation between the vehicle and the power grid to obtain optimized electric vehicle load distribution;
the method for predicting the random path of the electric automobile in the target area specifically comprises the following steps:
establishing a main chain by taking a destination as a first factor affecting a random path;
establishing a main chain and an auxiliary chain by taking weather as a second factor affecting a random path;
establishing a secondary auxiliary chain by taking the date as a third factor affecting the random path; the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment, and the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment; the state of the secondary auxiliary chain at the current moment is influenced by the state of the secondary auxiliary chain at the last moment;
calculating the transition probability among all states of the main chain under the influence of the main chain and the auxiliary chain when the main chain and the auxiliary chain are in any state, and obtaining a first transition probability matrix;
calculating the transition probability among all states of the main and auxiliary chains under the influence of the auxiliary chains when the auxiliary chains are in any state, and obtaining a second transition probability matrix;
And simulating the random path of the electric automobile by using the first transition probability matrix and the second transition probability matrix to obtain the random path of the electric automobile in the target area.
2. The method for predicting the electric vehicle load according to claim 1, wherein predicting the electric vehicle holding amount of the target area specifically includes:
establishing a neural network model of the relation between the conservation quantity influence factors and the conservation quantity;
acquiring the value of each retention quantity influence factor;
and inputting the value of each conservation quantity influence factor into the neural network model to obtain the conservation quantity of the electric vehicle in the target area.
3. The method for predicting load of an electric vehicle according to claim 1, wherein predicting a load distribution of the electric vehicle in one day based on the holding amount and the random path specifically includes:
when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is home, determining that the charging mode of the electric automobile is slow charging;
when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is not home, determining that the charging mode of the electric automobile is quick charging;
When the charge state of the electric automobile is higher than the first charge threshold value and lower than the second charge threshold value, determining that the charging mode of the electric automobile is slow charging;
when the charge state of the electric automobile is higher than the second charge threshold value, determining that the electric automobile is not charged;
and determining the charging power of all the electric vehicles in the charging state at any moment according to the charging modes of all the electric vehicles to obtain a load curve in one day.
4. The electric vehicle load prediction method according to claim 1, wherein the optimizing the load distribution condition based on the electric energy interaction relationship between the vehicle and the power grid, to obtain the optimized electric vehicle load distribution, specifically comprises:
according to the electric energy interaction relation between the vehicle and the power grid, determining that the maximum benefit and the minimum load peak-valley difference are taken as objective functions, optimizing the load distribution condition by taking the limiting condition of the discharge power and the constraint condition that the residual charge of the battery is larger than the minimum residual charge, and initializing the optimized solution as a population individual;
initializing iteration times;
solving objective functions of individuals in each population to obtain fitness values of each solution;
individual selection is carried out according to the fitness value, and the selected individuals are crossed and mutated to obtain offspring individuals;
Calculating the crowding distances of the child individuals and the selected individuals, and carrying out non-inferior layering on each crowding distance to obtain non-inferior individuals;
pareto sorting is carried out on the non-inferior individuals to obtain sorted individuals;
dividing the sequenced individuals into four parts according to the sequence, and solving the inverse solution of the individuals in the middle two parts to obtain new individuals in the middle two parts;
calculating the crowding distance of the new individuals in the middle two parts, and carrying out non-inferior layering on the crowding distance of the new individuals in the middle two parts to obtain non-inferior individuals after secondary layering;
pareto sorting is carried out on the individuals in the first part and the non-inferior individuals after secondary layering, so that the individuals after secondary sorting are obtained;
screening the first N individuals in the individuals after the secondary sequencing, and judging whether the current iteration number is greater than or equal to the maximum iteration number or not to obtain a judging result;
if the judgment result shows that the solution is not the same, adding 1 to the iteration times, updating the population individuals to the first N screened individuals, and returning to the step of solving the objective function of each population individual to obtain the fitness value of each solution;
and if the judgment result shows that the solution is yes, taking the first N screened individuals as the optimal solution.
5. An electric vehicle load prediction system, comprising:
the holding quantity prediction module is used for predicting the holding quantity of the electric automobile in the target area;
the path prediction module is used for predicting the random path of the electric automobile in the target area;
the load distribution prediction module is used for predicting the load distribution situation of the electric automobile in one day based on the conservation quantity and the random path;
the optimizing module is used for optimizing the load distribution condition based on the electric energy interaction relation between the vehicle and the power grid to obtain optimized electric vehicle load distribution;
wherein the path prediction module comprises:
a main chain establishing unit for establishing a main chain with the destination as a first factor affecting the random path;
the main and auxiliary chain establishing unit is used for establishing a main and auxiliary chain by taking weather as a second factor affecting a random path;
a secondary auxiliary chain establishing unit for establishing a secondary auxiliary chain by taking the date as a third factor affecting the random path; the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment, and the state of the main chain at the current moment is influenced by the state of the main chain at the moment and the state of the auxiliary chain at the moment; the state of the secondary auxiliary chain at the current moment is influenced by the state of the secondary auxiliary chain at the last moment;
The first transition probability matrix calculation unit is used for calculating the transition probability among the states of the main chain under the influence of the main chain and the auxiliary chain when the main chain and the auxiliary chain are in any state, so as to obtain a first transition probability matrix;
the second transition probability matrix calculation unit is used for calculating the transition probability among the states of the main and auxiliary chains under the influence of the secondary and auxiliary chains when the secondary and auxiliary chains are in any state, so as to obtain a second transition probability matrix;
and the random path simulation unit is used for simulating the random path of the electric automobile by utilizing the first transition probability matrix and the second transition probability matrix to obtain the random path of the electric automobile in the target area.
6. The electric vehicle load prediction system according to claim 5, wherein the holding amount prediction module includes:
the neural network model building unit is used for building a neural network model of the relation between the preservation quantity influence factors and the preservation quantity;
an input amount acquisition unit configured to acquire values of the respective holding amount influence factors;
the neural network prediction unit is used for inputting the value of each conservation quantity influence factor into the neural network model to obtain the conservation quantity of the electric vehicle in the target area.
7. The electric vehicle load prediction system of claim 5, wherein the load distribution prediction module comprises:
the first charging mode determining unit is used for determining that the charging mode of the electric automobile is slow charging when the charge state of the electric automobile is lower than a first charge threshold value and the destination of the electric automobile is home;
the second charging mode determining unit is used for determining that the charging mode of the electric automobile is quick charging when the charging state of the electric automobile is lower than a first charging threshold value and the destination of the electric automobile is not home;
the third charging mode determining unit is used for determining that the charging mode of the electric automobile is slow charging when the charging state of the electric automobile is higher than the first charging threshold value and lower than the second charging threshold value;
a fourth charging mode determining unit, configured to determine that the electric vehicle is not charged when the state of charge of the electric vehicle is higher than the second charge threshold;
the load curve construction unit is used for determining the charging power of all the electric vehicles in the charging state at any moment according to the charging modes of all the electric vehicles to obtain a load curve in one day.
8. The electric vehicle load prediction system of claim 5, wherein the optimization module comprises:
The optimal solution calculation unit is used for determining that the maximum benefit and the minimum load peak-valley difference are taken as target functions according to the electric energy interaction relation between the vehicle and the power grid, optimizing the load distribution condition by taking the limiting condition of the discharge power and the constraint condition that the residual charge of the battery is larger than the minimum residual charge, and initializing the optimized solution as a population individual;
the iteration initialization unit is used for initializing the iteration times;
the fitness value calculation unit is used for solving the objective function of each population individual to obtain the fitness value of each solution;
the genetic unit is used for selecting individuals according to the fitness value, and intersecting and mutating the selected individuals to obtain offspring individuals;
the first non-inferior layering unit is used for calculating the crowding distances of the child individuals and the selected individuals and performing non-inferior layering on each crowding distance to obtain non-inferior individuals;
the first sorting unit is used for sorting the non-inferior individuals by pareto to obtain sorted individuals;
the reverse solution calculation unit is used for dividing the sequenced individuals into four parts according to the sequence, and solving the reverse solutions of the individuals in the middle two parts to obtain new individuals in the middle two parts;
The second non-inferior layering unit is used for calculating the crowding distance of the new individuals in the middle two parts and performing non-inferior layering on the crowding distance of the new individuals in the middle two parts to obtain non-inferior individuals after secondary layering;
the second sorting unit is used for sorting the individuals in the first part and the non-inferior individuals after secondary layering to obtain individuals after secondary sorting;
the screening and judging unit is used for screening the first N individuals in the secondarily ordered individuals and judging whether the current iteration number is greater than or equal to the maximum iteration number or not to obtain a judging result;
the updating and returning unit is used for adding 1 to the iteration times if the judging result indicates no, updating the population individuals into the first N screened individuals and returning to the fitness value calculating unit;
and the optimal solution determining unit is used for taking the first N screened individuals as optimal solutions if the judging result shows that the judgment result is yes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911212935.5A CN110929950B (en) | 2019-12-02 | 2019-12-02 | Electric automobile load prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911212935.5A CN110929950B (en) | 2019-12-02 | 2019-12-02 | Electric automobile load prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110929950A CN110929950A (en) | 2020-03-27 |
| CN110929950B true CN110929950B (en) | 2023-09-26 |
Family
ID=69848136
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911212935.5A Active CN110929950B (en) | 2019-12-02 | 2019-12-02 | Electric automobile load prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110929950B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111983480B (en) * | 2020-08-19 | 2023-02-28 | 华晟（青岛）智能装备科技有限公司 | AGV electric quantity state prediction method and system based on Mahalanobis process |
| CN114611811B (en) * | 2022-03-21 | 2022-10-25 | 特斯联科技集团有限公司 | Low-carbon park optimal scheduling method and system based on EV load participation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105591433A (en) * | 2016-01-15 | 2016-05-18 | 国家电网公司 | Electric automobile charging load optimization method based on electric automobile charging power dynamic distribution |
| CN108944531A (en) * | 2018-07-24 | 2018-12-07 | 河海大学常州校区 | A kind of orderly charge control method of electric car |
| CN110363332A (en) * | 2019-06-21 | 2019-10-22 | 国网天津市电力公司电力科学研究院 | A kind of electric car charging load spatial and temporal distributions prediction technique based on individual behavior characteristic |
| CN115456180A (en) * | 2022-08-17 | 2022-12-09 | 上海电机学院 | Electric vehicle quantity prediction method based on three-chain Markov model |
-
2019
- 2019-12-02 CN CN201911212935.5A patent/CN110929950B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105591433A (en) * | 2016-01-15 | 2016-05-18 | 国家电网公司 | Electric automobile charging load optimization method based on electric automobile charging power dynamic distribution |
| CN108944531A (en) * | 2018-07-24 | 2018-12-07 | 河海大学常州校区 | A kind of orderly charge control method of electric car |
| CN110363332A (en) * | 2019-06-21 | 2019-10-22 | 国网天津市电力公司电力科学研究院 | A kind of electric car charging load spatial and temporal distributions prediction technique based on individual behavior characteristic |
| CN115456180A (en) * | 2022-08-17 | 2022-12-09 | 上海电机学院 | Electric vehicle quantity prediction method based on three-chain Markov model |
Non-Patent Citations (2)
| Title |
|---|
| Fenglei Li.Optimal Scheduling Strategy of Distribution Network Based on Electric Vehicle Forecasting.electronics.2019,全文. * |
| 康树祝.基于改进NSGA2 的临近空间通信网络优化设计.《计算机工程与应用》.2018,第203-210页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110929950A (en) | 2020-03-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111178619B (en) | Multi-objective optimization method considering distributed power supply and charging station joint planning | |
| CN111311122B (en) | Charging station construction method for smart city | |
| CN108288112A (en) | Region electric automobile charging station load forecasting method based on user's trip simulation | |
| CN110929950B (en) | Electric automobile load prediction method and system | |
| CN113780665B (en) | Private car stay position prediction method and system based on enhanced recurrent neural network | |
| CN103745110A (en) | Method of estimating operational driving range of all-electric buses | |
| CN107194605B (en) | Energy consumption estimation system and method | |
| CN103078152B (en) | Intelligent charging method for centralized charging station | |
| CN114139794A (en) | Regional public parking lot charging demand prediction method based on parking big data | |
| CN114444802A (en) | Electric vehicle charging guide optimization method based on graph neural network reinforcement learning | |
| CN115239032A (en) | Highway service area microgrid planning method and system considering energy self-consistency rate | |
| Tadayon‐Roody et al. | Multi‐objective locating of electric vehicle charging stations considering travel comfort in urban transportation system | |
| CN111626492B (en) | Fuzzy multi-target opportunity constraint planning method for electric vehicle charging network | |
| CN105262167A (en) | Intra-regional electric vehicle ordered charging control method | |
| CN115239004A (en) | Charging parking lot location and volume optimization method based on charging demand prediction | |
| CN118410902A (en) | Electric vehicle charging demand space-time distribution simulation and scheduling optimization method considering vehicle owner travel rule | |
| Pourvaziri et al. | Planning of electric vehicle charging stations: An integrated deep learning and queueing theory approach | |
| CN113836735A (en) | Method for establishing two-stage model of electric bus cluster battery charging and replacing strategy in battery replacing mode | |
| CN114118724A (en) | Electric vehicle charging scheduling method considering requirement matching degree | |
| CN113672857A (en) | Method for carrying out hydrogenation station site selection based on hydrogen energy automobile and hydrogen production plant data | |
| Erdoğan et al. | Finding an energy efficient path for plug-in electric vehicles with speed optimization and travel time restrictions | |
| Zhao et al. | Robust Optimization of Mixed-Load School Bus Route Based on Multi-Objective Genetic Algorithm. | |
| CN110674988A (en) | Urban charging station planning method based on electric vehicle travel big data | |
| Ma et al. | An agent-based training system for optimizing the layout of AFVs' initial filling stations | |
| CN114298379B (en) | Automatic bus-substituting parking lot layout optimization method based on simulation optimization |
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 |