CN117057533A - Method and device for optimizing subdivision period ordered charging strategy based on load evaluation - Google Patents
Method and device for optimizing subdivision period ordered charging strategy based on load evaluation Download PDFInfo
- Publication number
- CN117057533A CN117057533A CN202310854084.4A CN202310854084A CN117057533A CN 117057533 A CN117057533 A CN 117057533A CN 202310854084 A CN202310854084 A CN 202310854084A CN 117057533 A CN117057533 A CN 117057533A
- Authority
- CN
- China
- Prior art keywords
- charging
- data
- electric automobile
- load
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000011156 evaluation Methods 0.000 title claims abstract description 23
- 230000005611 electricity Effects 0.000 claims abstract description 99
- 239000002245 particle Substances 0.000 claims abstract description 28
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 23
- 241000282461 Canis lupus Species 0.000 claims abstract description 13
- 238000005457 optimization Methods 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims description 19
- 230000007704 transition Effects 0.000 claims description 18
- 238000003860 storage Methods 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 8
- 230000006399 behavior Effects 0.000 claims description 6
- 230000001186 cumulative effect Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 239000002956 ash Substances 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 5
- 238000010845 search algorithm Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 4
- 230000003313 weakening effect Effects 0.000 claims description 4
- 235000002918 Fraxinus excelsior Nutrition 0.000 claims description 3
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 125000004122 cyclic group Chemical group 0.000 claims description 3
- 238000011161 development Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 230000004069 differentiation Effects 0.000 claims description 2
- 238000003032 molecular docking Methods 0.000 claims description 2
- 238000012546 transfer Methods 0.000 claims description 2
- 230000007547 defect Effects 0.000 abstract description 4
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 10
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 4
- 238000009472 formulation Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000001360 synchronised effect Effects 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
- G06Q10/06—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
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses a subdivision time period ordered charging strategy optimization method and device based on load evaluation, which are used for acquiring the data of the holding quantity and the charging demand of an electric vehicle in a target residential area; obtaining the charging load of the electric automobile in the target residential area according to the data of the holding quantity and the charging demand of the electric automobile in the target residential area; evaluating and grading the charging load of the target residential area; based on the evaluation grading, determining initial electricity prices of the corresponding grades, obtaining ordered charging weighted time-of-use electricity prices according to the weighted subdivision time-of-use electricity prices, and determining an ordered charging strategy according to an ordered charging theory of the ordered charging weighted time-of-use electricity prices; and constructing a master-slave game model based on the ordered charging strategy and the game theory, and solving the master-slave game model by utilizing an improved particle swarm algorithm combined with wolf swarm search. The advantages are that: the defect that in the prior art, a power selling party and an electric vehicle user win-win situation are difficult to achieve in the ordered charging of the electric vehicles due to the fact that the number of the electric vehicles is increased by the existing prediction mode is overcome.
Description
Technical Field
The invention relates to a subdivision period ordered charging strategy optimization method and device based on load evaluation, and belongs to the technical field of electric automobile charging.
Background
With the surge of the global electric automobile scale, a large number of electric automobiles are connected into a power grid, so that the inherent randomness of the charging behavior of the electric automobiles with high power load is brought, the load characteristic and economic operation of a power distribution system are affected, especially in urban residential areas with high permeability of the electric automobiles, the charging behavior of the electric automobiles in the overlapping or electricity utilization peak period of charging time is influenced by the rule of the users, larger load peaks are brought to the power grid, the problems of peak-valley difference increase, power loss increase, transformer overload and the like are caused, the burden of a power distribution network is increased, and meanwhile, the running economy of the power grid is affected.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a subdivision period ordered charging strategy optimization method based on load evaluation, which predicts charging loads of electric vehicles in residential areas, adds a reasonable ordered charging strategy and optimizes the ordered charging strategy to achieve ideal benefits of ordered charging peak clipping and valley filling.
In order to solve the technical problems, the invention provides a subdivision period ordered charging strategy optimization method based on load evaluation, which comprises the following steps:
acquiring the data of the electric automobile storage quantity of a target residential area;
acquiring charging demand data of an electric automobile in a target residential area;
obtaining the charging load of the electric vehicle in the target residential area according to the data of the storage quantity of the electric vehicle in the target residential area and the charging demand data of the electric vehicle in the target residential area;
evaluating and grading the charging load of the target residential area;
based on the evaluation grading, determining initial electricity prices of the corresponding grades, acquiring weighted subdivision time-of-day electricity prices, obtaining ordered charging weighted subdivision time-of-day electricity prices according to the weighted subdivision time-of-day electricity prices, and determining an ordered charging strategy according to an ordered charging theory of the ordered charging weighted subdivision time-of-day electricity prices;
based on the ordered charging strategy, a master-slave game model is built by combining a game theory, and the master-slave game model is solved by utilizing an improved particle swarm algorithm combined with wolf swarm search, so that load peak-valley difference, load fluctuation rate, maximum load rate and user charging cost data of power grid operation are obtained.
Further, the obtaining the data of the electric vehicle in the target residential area includes:
and acquiring historical single-day holding quantity data of the electric vehicles in the target residential area, and predicting the electric vehicle holding quantity data of the target day of the target residential area by using a gray-radial basis function neural network model according to the historical single-day holding quantity data of the electric vehicles in the target residential area.
Further, the predicting the electric vehicle holding amount data of the target residential area on the target day by using the gray-radial basis function neural network model according to the electric vehicle history single day holding amount data of the target residential area includes:
step 11: the historical single-day conservation quantity data sequence of the electric automobile in the target residential area is as follows:
n is the number of the original data;
step 12: introducing a first-order weakening operator D into the original data sequence to obtain a first-order weakened data sequence
X (0) D, expressed as:
in the method, in the process of the invention,each specific data after the weakening is represented,
k=1,2,…,n;
step 13: according toCreating a cumulative generation sequence X (1) Expressed as:
representing each specific data of the accumulated sequence;
step 14: for a pair ofFitting the sequence by adopting a univariate first-order differential equation to obtain a first-order gray model GM (1, 1), which is expressed as follows:
wherein a is the number of developed ashes and represents x (0) And x (1) B is the relation of controlling ash number and reaction data,representing a data sequence x (1) Differentiation of time t;
step 15: estimating the development gray number a and the control gray number b by using a least square method to obtain the time response of a first-order gray model GM (1, 1)Expressed as:
in the method, in the process of the invention,representing the raw data when k is 1, e representing an exponential constant;
step 16: the prediction residual e (n+l) of the first-order gray model GM (1, 1) is expressed as:
e(n+l)=C(n+l)-C y (n+l)
wherein l is a predicted step length, C is the electric vehicle holding quantity, C (n+l) is a true value, C y (n+l) is the predicted value of the first-order gray model GM (1, 1);
step 17: acquiring initial data of the electric vehicle, inputting the initial data into an RBF neural network, giving a target error of the RBF neural network, completing the cyclic training of the RBF neural network, and outputting a predicted residual value of a first-order gray model GM (1, 1);
step 18: inputting a first-order gray model GM (1, 1) predicted value with step length of l into the RBF neural network after trainingRBF neural network outputs +.A predicted residual value for a first-order gray model GM (1, 1)>Will predict the residual value +.>And predictive value->And summing to obtain predicted electric vehicle storage quantity data of the target residence target day.
Further, the obtaining the charging demand data of the electric automobile in the target residential area includes:
and acquiring charging start and end time and charging electric quantity data of the historical electric automobile of the target residential area, and predicting and acquiring charging demand data of the electric automobile of the residential area based on a Markov chain Monte Carlo algorithm according to the charging start and end time and the charging electric quantity data of the historical electric automobile of the target residential area.
Further, the charging start time, the charging end time and the charging electric quantity data of the target residential area historical electric automobile are predicted based on a Markov chain Monte Carlo algorithm to obtain residential area electric automobile charging demand data, and the method comprises the following steps:
step 21: electric vehicle state P (S) according to a multi-dimensional state space markov chain i →S j ) The description is as follows:
P(S i →S j )=P ij
wherein S is i S is the state of the electric automobile at the current moment of the electric automobile j For the state of the electric automobile at the next moment, P ij Is the transition probability;
step 22:based on the difference of user traveling habits and the randomness of user charging demands, taking the position state, the charge state and the battery charge state of the electric automobile as 3 elements of a state vector of the electric automobile, and establishing a three-dimensional state vector S= [ S ] of the electric automobile 1 ,s 2 ,s 3 ]Wherein s is 1 Is the position state of the electric automobile, s 2 Is a state of charge, which is used for representing the current residual electric quantity of the electric automobile, s 3 Is the battery state of charge;
step 23: acquiring actual travel data of the electric automobile, solving state transition probability based on the three-dimensional state vector of the electric automobile and the actual travel data of the electric automobile, and obtaining a state transition matrix H of the electric automobile as follows:
in the formula, h mn For the cumulative number of times that the destination originating at location M is at location n, M is the total number of key locations;
step 24: simulating and generating an initial state of the electric automobile by using a Monte Carlo method according to state transfer behaviors of the electric automobile in three dimensions of a position state, a charge state and a battery charge state;
step 25: obtaining S= [ S ] from the state transition matrix and the initial state 1 ,s 2 ,s 3 ]The probability of occurrence of the state is:
step 26: determining transition probability P ij The calculation formula is as follows:
step 27: based on transition probability P ij Generating charging requirement of electric vehicle in target residential area by combining current time state of electric vehicle userAnd (5) calculating data.
Further, based on the evaluation grading, determining an initial electricity price of the corresponding grade, obtaining a weighted subdivision time-of-day electricity price, and obtaining an ordered charging weighted time-of-day electricity price according to the weighted subdivision time-of-day electricity price, including:
according to the predicted data of the charging demand of the electric automobile, dividing the load into three levels by calculating the load fluctuation rate of the platform region, wherein the working day is level I, the holiday is level II, the holiday is level III, and the initial electricity prices of the level I, the level II and the level III are determined as d Ⅰ 、d Ⅱ And d Ⅲ Obtaining a grading initial time-sharing electricity price matrix d f =[d Ⅰ ,d Ⅱ ,d Ⅲ ]。
The subdivision time-sharing electricity price adopts a calculation mode of a piecewise function, and the calculation formula is as follows:
wherein: t is the time of orderly charging participation, the unit is min, a 1 、a 2 、a 3 And b 1 、b 2 、b 3 Obtaining an ordered charging electricity price matrix d by taking the electricity price matrix d as a constant n =[d 1 ,d 2 ,d 3 ];
The actual electricity price is obtained by weighting the classified initial electricity price and the subdivision time-sharing electricity price, and the weighted calculation formula is as follows:
wherein: d is the actual electricity price after participating in ordered charging, w represents a weighting coefficient, and w is E [0,1].
Further, determining a corresponding ordered charging strategy according to an ordered charging theory of ordered charging weighted time-of-use electricity price comprises:
step 31: leading residential electric vehicle holding quantity prediction data and target residential electric vehicle charging demand prediction data into a charging station, and combining the number of electric vehiclesQuantity constraint, charging power constraint, total charging quantity constraint and transformer capacity constraint, and a time-period charging power limit value matrix P is generated lim =[P ph ,P p ,P 1 ],P ph Recorded as peak period power limit value, P p Is recorded as a normal power limit value, P 1 Recording as a valley period power limit;
step 32: according to the number N of electric vehicles in a target residential area and charging demand data, optimizing charging power in each period through a variable power ordered charging theory to generate P c =[P′ ph ,P′p,P′ 1 ],P′ ph The peak charge power, P' p The charge power is recorded as the normal charge power, P' 1 The charging power is recorded as valley period charging power;
step 33: and (3) docking the charging piles of the electric automobile, uploading SOC data, battery capacity, user arrival time and user-set departure time of the electric automobile, counting actual charging load, judging a load prediction error coefficient epsilon, and optimizing the charging time in the peak period by an optimization controller if the error coefficient epsilon is greater than a preset precision requirement, so as to limit the charging time of the electric automobile in the power utilization peak period.
Further, based on the ordered charging strategy, a master-slave game model is constructed in combination with a game theory, including:
based on master-slave games, an electricity seller is taken as a leader of the games, a leader policy is charging electricity price, an electric automobile user is a follower, the follower policy is charging plan, and a corresponding policy establishment is established as follows:
the profit of the electricity seller is as follows:
wherein maxA represents the maximum daily profit of the charging station, P ev (t) represents the charging load at time t of the electric vehicle, d t Represents the real-time charging electricity price, d b,t Represents electricity purchasing price from electricity seller, c p Unit penalty cost, Q, representing load peak-valley difference max Indicating maximum loadValue, Q min Representing a minimum value of the load;
the charging expenditure per unit time of the electric automobile user is as follows:
wherein minB represents the minimum charge expenditure per unit time of the electric vehicle, C ev Representing the number of electric vehicles, P ev And (t) represents a charging load of the electric automobile at a certain time t.
Further, the method for solving the game model by utilizing the improved particle swarm algorithm combined with the wolf swarm search comprises the following steps:
step 41: predicting the basic electricity load of the current day, and initializing the weighted time-sharing electricity price;
step 42: importing the arrival time and the required charging electric quantity information of the electric automobile according to the living rule of the residential area;
step 43: initializing population information, generating initial population of electric automobile users and electricity sellers, initializing positions and speeds of particles, limiting the charging quantity of the electric automobile according to the capacity constraint condition of a transformer, and limiting the charging price according to the constraint condition of an electricity selling station;
step 44: the population diversity and optimizing effect of the particle swarm algorithm are improved based on the wolf swarm search algorithm, and the position and speed of the particles are iteratively updated;
step 45: calculating an electric automobile user objective function, solving an optimal value, and solving particles meeting constraint conditions and maximizing user satisfaction according to the power price per hour;
step 46: calculating a charging station objective function based on a game model, solving an optimal value, and solving the income maximization particles meeting constraint conditions according to a user charging strategy;
step 47: judging whether the iteration termination condition is met, if not, turning to the step 44 and the step 45 to recalculate, and finally obtaining the optimal charging scheme.
A subdivision period ordered charging strategy optimization device based on load assessment, comprising:
the acquisition module is used for acquiring the storage quantity data of the electric automobile in the target residential area and the charging demand data of the electric automobile in the target residential area;
the first determining module is used for determining the charging load of the electric vehicle in the target residential area according to the storage quantity data of the electric vehicle in the target residential area and the charging demand data of the electric vehicle in the target residential area;
the grading module is used for evaluating and grading the charging load of the target residential area;
the second determining module is used for determining initial electricity prices of the corresponding grades based on the evaluation grades, acquiring weighted subdivision time-of-day electricity prices, obtaining ordered charging weighted time-of-day electricity prices according to the weighted subdivision time-of-day electricity prices, and determining corresponding ordered charging strategies according to an ordered charging theory of the ordered charging weighted time-of-day electricity prices;
and the model construction and solving module is used for constructing a master-slave game model based on the ordered charging strategy and combining with a game theory, solving the master-slave game model by utilizing an improved particle swarm algorithm combined with wolf swarm search, and obtaining load peak-valley difference, load fluctuation rate, maximum load rate and user charging cost data of the power grid operation.
The invention has the beneficial effects that:
the invention predicts the charging load of the electric vehicle in a multi-level manner based on the prediction of the holding quantity and the charging demand of the electric vehicle in the residential area scene, and overcomes the defect of the existing prediction mode that the number of the electric vehicles is increased. The ordered charging strategy formulation based on the load evaluation is relatively more reasonable, the defect that the prior art is difficult to achieve win-win of an electricity seller and an electric vehicle user in ordered charging of the electric vehicle is overcome by combining the improvement of a game theory and a model solving algorithm, and a foundation is laid for popularization and development of the ordered charging technology of the electric vehicle in the future.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
FIG. 2 is a schematic diagram of a GM-RBF model prediction flow provided by an embodiment of the present invention;
fig. 3 is a schematic diagram of an electric vehicle charging demand prediction flow based on an MCMC algorithm according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an orderly charging strategy formulation flow provided by an embodiment of the present invention;
fig. 5 is a schematic diagram of a WPSO algorithm solving game model provided by an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
As shown in fig. 1, a subdivision period ordered charging strategy optimization method based on load evaluation includes:
s1, predicting residential electric automobile maintenance data based on a gray-radial basis function neural network model (Grey Models-Radial Basis Function, GM-RBF);
s2, predicting and obtaining charging demand data of the electric automobile in the residential area based on a Markov Chain Monte Carlo algorithm (MCMC);
s3, combining the data of the storage quantity with the charging probability of the typical daily electric vehicle in the residential area to obtain the charging load of the electric vehicle in the residential area;
s4, evaluating and grading charging loads of residential areas;
s5, determining initial electricity prices of the corresponding grades based on the evaluation grades, formulating corresponding ordered charging strategies, and weighting by combining ordered charging participation time to obtain ordered charging actual electricity prices;
s6, a master-slave game model is constructed based on a game theory, a improved particle swarm algorithm (Wolf Colony Algorithm-Particle Swarm Optimization, WPSO) combined with wolf swarm search is utilized to solve the game model, load peak-valley difference, load fluctuation rate, maximum load rate and user charging cost data of power grid operation are obtained, the reduction of the load peak-valley difference, load fluctuation rate and maximum load rate of power grid operation is verified, and the reduction of peak clipping and valley filling and user charging cost of power grid operation is realized.
As shown in fig. 2, the factors of purchasing the electric automobile by residents are analyzed based on the GM-RBF prediction model, and the long-term keeping amount of the electric automobile in the residential area is predicted, wherein the prediction process is as follows:
step 11: assume that the original data sequence is:n is the number of the original data;
step 12: introducing a first-order weakening operator D into the original data sequence to obtain
Wherein the method comprises the steps of
Step 13: according toCreating a cumulative sequence
n is the number of the original data;
step 14: fitting the sequence by adopting a univariate first-order differential equation to obtain a first-order gray model GM (1, 1):wherein a is the number of developed ashes and represents X (0) And X is (1) B is the relation of the mutual change of the reaction data, and the gray number is controlled;
step 15: estimating the parameter a and the parameter b by using a least square method to obtain the GM (1, 1) model time response as follows:
step 16: GM (1, 1) model prediction residual is e (n+l) =C (n+l) -C y (n+l) in the formulan is the number of original data, l is the prediction step length, C is the electric car holding quantity, e (n+l) is the prediction residual error, C (n+l) is the true value, C y (n+l) is a GM (1, 1) model predictive value;
step 17: the method comprises the steps that initial data of the electric automobile conservation quantity are input into an RBF neural network, target errors of the RBF neural network are given, cyclic training of the RBF neural network is completed, and a prediction residual value of a GM (1, 1) model is output;
step 18: inputting GM (1, 1) model predictive value with step length of l into RBF network after trainingRBF network output is GM (1, 1) model prediction residual +.>Summing the predicted residual error and the predicted value to obtain an actual predicted value of the electric vehicle holding capacity +.>The calculation formula is as follows: />
As shown in fig. 3, the electric vehicle charging demand prediction is performed based on MCMC, and the prediction process is as follows:
step 21: the state description of the electric automobile according to the multidimensional state space Markov chain is as follows:
P(S i →S j )=P(S i |S j )=P ij
wherein: s is S i S is the state of the electric automobile at the current moment of the electric automobile j For the state of the electric automobile at the next moment, P ij Is the transition probability;
step 22: based on the difference of user traveling habits and the randomness of user charging demands, the position, the state of charge and the battery charging state of the electric automobile are taken as 3 elements of the state vector of the electric automobile, and a three-dimensional state vector S= [ S ] of the electric automobile is established 1 ,s 2 ,s 3 ]Wherein s is 1 For use inStatus of household position s 2 Is the state of charge, i.e. the current residual electric quantity of the electric automobile, s 3 The battery behavior state of the electric automobile;
step 23: based on the three-dimensional state vector of the electric automobile, solving the state transition probability according to actual vehicle travel data, and obtaining the state transition matrix H of the electric automobile as follows:
h in ij For the cumulative number of times that the destination originating at location i is at location j, M is the total number of key locations;
step 24: according to state transition behaviors of the electric automobile in three-dimensional state statistics, simulating and generating an initial state of the electric automobile by using a Monte Carlo method;
step 25: obtaining S= [ S ] from the state transition matrix and the initial state 1 ,s 2 ,s 3 ]The probability of occurrence of the state is:
step 26: determining transition probability P ij The calculation formula is as follows:
step 27: and combining the current time state of the electric automobile user and the charging demand state of the user to generate electric automobile charging demand data in a residential area scene.
And combining the residential electric vehicle storage quantity prediction data and the charging demand prediction data to obtain residential electric vehicle charging load prediction data.
And (3) carrying out evaluation and grading according to the load quantity, dividing the load grade into three grades, wherein the working day is grade I, the double holidays are grade II, and the holidays are grade III.
As shown in fig. 4, the ordered charging power is adjusted based on the evaluation and classification, and an ordered charging strategy is formulated, wherein the ordered charging strategy formulation process is as follows:
step 31: the load prediction result is led into a charging station intelligent control system, and a time-period charging power limit value matrix P is generated by combining constraint conditions such as transformer capacity limit, user charging demand, electric vehicle residence time and the like lim =[P ph ,P p ,P 1 ],P ph Recorded as peak period power limit value, P p Is recorded as a normal power limit value, P 1 Recording as a valley period power limit;
step 32: according to the number N of electric vehicles and the charging load, the charging power of each period is optimized by an optimization controller to generate P c =[P′ ph ,P′p,P′ 1 ],P′ ph The peak charge power, P' p The charge power is recorded as the normal charge power, P' 1 The charging power is recorded as valley period charging power;
step 33: the electric automobile charging pile is in butt joint, the battery energy management system uploads the automobile SOC data, the battery capacity, the user arrival time and the user-set departure time, actual charging load is counted, the load prediction error coefficient epsilon is judged, if the error coefficient epsilon is larger than the precision requirement, the optimization controller optimizes the charging time in the peak period, and the charging time of the electric automobile in the electricity utilization peak period is limited.
According to the load evaluation grade, determining the initial electricity prices of the grade I, the grade II and the grade III as d, d and d to obtain d f =[d,d,d]The participation duration of ordered charging can be accurate to each minute, and a user can freely select continuous ordered charging participation time, and the ordered charging electricity price adopts a calculation mode of a piecewise function, and the calculation formula is as follows:
wherein: t is the time of orderly charging participation, a 1 、a 2 、a 3 And b 1 、b 2 、b 3 As a constant, obtain the ordered charge electricity priceMatrix d n =[d 1 ,d 2 ,d 3 ]The initial electricity price and the ordered charging electricity price are weighted to obtain the actual electricity price, and the weighted calculation formula is as follows:
wherein: d is the actual electricity price amount after participation in ordered charging, w is [0,1], and the ordered charging is combined with a load evaluation grading mode, so that the synchronous rising of the participation rate and the participation time of ordered charging can be realized.
And establishing a charging model by using three elements of the game, namely participants, strategies and benefits, and establishing a game strategy by taking both sides of an electricity selling party and an electric automobile user as game participants and considering the targets of the two parties.
Based on master-slave gaming, taking an electricity seller as a leader of the gaming, taking a strategy as charging electricity price, taking an electric automobile user as a follower, taking the strategy as a charging plan, namely charging starting time, and establishing a model for the corresponding strategy as follows:
the profit of the electricity seller is as follows:
wherein: maxA represents the maximum daily profit of the charging station, P ev (t) represents the charging load of the electric vehicle at a certain time t, d t Represents the real-time charging electricity price, d b,t Representing electricity purchase price from electric network company, c p Unit penalty cost, Q, representing load peak-valley difference max Represents the maximum load, Q min Representing the minimum value of the load.
The charging expenditure per unit time of the electric automobile user is as follows:
wherein: minB represents the minimum charge expenditure per unit time of the electric automobile, C ev Representation ofNumber of electric vehicles, P ev (t) represents the charging load of the electric vehicle at a certain time t, d t Indicating the real-time charge electricity price.
The strategies of both game sides are abstracted into particles in an improved particle swarm algorithm combined with a wolf swarm search algorithm, a master-slave game strategy is solved by utilizing the improved algorithm, game balance is achieved, a period of time which is close to charging access time and has relatively low electricity price is simulated and selected according to current load, a charging strategy is formulated, a new load prediction curve is generated, population diversity and optimizing effects of the particle swarm algorithm are improved based on the wolf swarm search algorithm, an electricity seller adjusts electricity price according to the load, the previous round of strategy simulation is repeatedly carried out until the economical efficiency of a selling power station and the satisfaction degree of users of an electric automobile reach comprehensive optimal, and the method comprises the following specific steps of:
step 41: predicting the basic electricity load of the current day, and initializing the time-sharing electricity price;
step 42: importing the arrival time and the required charging electric quantity information of the electric automobile according to the living rule of the residential area;
step 43: initializing population information, generating initial population of electric automobile users and electricity sellers, initializing positions and speeds of particles, limiting the charging quantity of the electric automobile according to the capacity constraint condition of a transformer, and limiting the charging price according to the constraint condition of an electricity selling station;
step 44: the population diversity and optimizing effect of the particle swarm algorithm are improved based on the wolf swarm search algorithm, and the position and speed of the particles are iteratively updated;
step 45: calculating an electric automobile user objective function, solving an optimal value, and solving particles meeting constraint conditions and maximizing user satisfaction according to the power price per hour;
step 46: calculating a charging station objective function based on a game model, solving an optimal value, and solving the income maximization particles meeting constraint conditions according to a user charging strategy;
step 47: judging whether the iteration termination condition is met, if not, turning to the step 4 and the step 5 to recalculate, and finally obtaining the optimal charging scheme.
The invention also provides a subdivision period ordered charging strategy optimizing device based on load evaluation, which comprises the following components:
the acquisition module is used for acquiring the storage quantity data of the electric automobile in the target residential area and the charging demand data of the electric automobile in the target residential area;
the first determining module is used for determining the charging load of the electric vehicle in the target residential area according to the storage quantity data of the electric vehicle in the target residential area and the charging demand data of the electric vehicle in the target residential area;
the grading module is used for evaluating and grading the charging load of the target residential area;
the second determining module is used for determining initial electricity prices of the corresponding grades based on the evaluation grades, acquiring weighted subdivision time-of-day electricity prices, obtaining ordered charging weighted time-of-day electricity prices according to the weighted subdivision time-of-day electricity prices, and determining corresponding ordered charging strategies according to an ordered charging theory of the ordered charging weighted time-of-day electricity prices;
and the model construction and solving module is used for constructing a master-slave game model based on the ordered charging strategy and combining with a game theory, solving the master-slave game model by utilizing an improved particle swarm algorithm combined with wolf swarm search, and obtaining load peak-valley difference, load fluctuation rate, maximum load rate and user charging cost data of the power grid operation.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (10)
1. The subdivision period ordered charging strategy optimization method based on load evaluation is characterized by comprising the following steps of:
acquiring the data of the electric automobile storage quantity of a target residential area;
acquiring charging demand data of an electric automobile in a target residential area;
obtaining the charging load of the electric vehicle in the target residential area according to the data of the storage quantity of the electric vehicle in the target residential area and the charging demand data of the electric vehicle in the target residential area;
evaluating and grading the charging load of the target residential area;
based on the evaluation grading, determining initial electricity prices of the corresponding grades, acquiring weighted subdivision time-of-day electricity prices, obtaining ordered charging weighted subdivision time-of-day electricity prices according to the weighted subdivision time-of-day electricity prices, and determining an ordered charging strategy according to an ordered charging theory of the ordered charging weighted subdivision time-of-day electricity prices;
based on the ordered charging strategy, a master-slave game model is built by combining a game theory, and the master-slave game model is solved by utilizing an improved particle swarm algorithm combined with wolf swarm search, so that load peak-valley difference, load fluctuation rate, maximum load rate and user charging cost data of power grid operation are obtained.
2. The method for optimizing a sub-divided period ordered charging strategy based on load assessment according to claim 1, wherein the obtaining the target residential electric vehicle holding amount data comprises:
and acquiring historical single-day holding quantity data of the electric vehicles in the target residential area, and predicting the electric vehicle holding quantity data of the target day of the target residential area by using a gray-radial basis function neural network model according to the historical single-day holding quantity data of the electric vehicles in the target residential area.
3. The method for optimizing a sub-divided period ordered charging strategy based on load assessment according to claim 2, wherein the predicting the electric vehicle holding amount data of the target residential area on the target day by using the gray-radial basis function neural network model according to the electric vehicle historical single day holding amount data of the target residential area comprises:
step 11: the historical single-day conservation quantity data sequence of the electric automobile in the target residential area is as follows:
n is the number of the original data;
step 12: introducing first-order reduction to original data sequencesub-D, obtaining a first-order weakened data sequence X (0) D, expressed as:
in the method, in the process of the invention,representing specific data after weakening +_>
Step 13: according toCreating a cumulative generation sequence X (1) Expressed as:
representing each specific data of the accumulated sequence;
step 14: for a pair ofFitting the sequence by adopting a univariate first-order differential equation to obtain a first-order gray model GM (1, 1), which is expressed as follows:
wherein a is the number of developed ashes and representsx (0) And x (1) B is the relation of controlling ash number and reaction data,representing a data sequence x (1) Differentiation of time t;
step 15: estimating the development gray number a and the control gray number b by using a least square method to obtain the time response of a first-order gray model GM (1, 1)Expressed as:
in the method, in the process of the invention,representing the raw data when k is 1, e representing an exponential constant;
step 16: the prediction residual e (n+l) of the first-order gray model GM (1, 1) is expressed as:
e(n+l)=C(n+l)-C y (n+l)
wherein l is a predicted step length, C is the electric vehicle holding quantity, C (n+l) is a true value, C y (n+l) is the predicted value of the first-order gray model GM (1, 1);
step 17: acquiring initial data of the electric vehicle, inputting the initial data into an RBF neural network, giving a target error of the RBF neural network, completing the cyclic training of the RBF neural network, and outputting a predicted residual value of a first-order gray model GM (1, 1);
step 18: inputting a first-order gray model GM (1, 1) predicted value with step length of l into the RBF neural network after trainingRBF neural network outputs +.A predicted residual value for a first-order gray model GM (1, 1)>Will predict the residual value +.>And predictive value->And summing to obtain predicted electric vehicle storage quantity data of the target residence target day.
4. The method for optimizing a sub-divided period ordered charging strategy based on load assessment according to claim 1, wherein the obtaining the charging demand data of the electric vehicle in the target residential area comprises:
and acquiring charging start and end time and charging electric quantity data of the historical electric automobile of the target residential area, and predicting and acquiring charging demand data of the electric automobile of the residential area based on a Markov chain Monte Carlo algorithm according to the charging start and end time and the charging electric quantity data of the historical electric automobile of the target residential area.
5. The method according to claim 4, wherein the predicting charging load demand data of the target residential area historical electric vehicle based on the markov chain monte carlo algorithm includes:
step 21: electric vehicle state P (S) according to a multi-dimensional state space markov chain i →S j ) The description is as follows:
P(S i →S j )=P ij
wherein S is i S is the state of the electric automobile at the current moment of the electric automobile j For the state of the electric automobile at the next moment, P ij Is the transition probability;
step 22: based on the difference of the travel habits of users and the randomness of the charging demands of users, the electric automobile is charged in the position state, the charge state and the batteryThe electric state is used as 3 elements of a state vector of the electric automobile, and a three-dimensional state vector S= [ S ] of the electric automobile is established 1 ,s 2 ,s 3 ]Wherein s is 1 Is the position state of the electric automobile, s 2 Is a state of charge, which is used for representing the current residual electric quantity of the electric automobile, s 3 Is the battery state of charge;
step 23: acquiring actual travel data of the electric automobile, solving state transition probability based on the three-dimensional state vector of the electric automobile and the actual travel data of the electric automobile, and obtaining a state transition matrix H of the electric automobile as follows:
in the formula, h mn For the cumulative number of times that the destination originating at location M is at location n, M is the total number of key locations;
step 24: simulating and generating an initial state of the electric automobile by using a Monte Carlo method according to state transfer behaviors of the electric automobile in three dimensions of a position state, a charge state and a battery charge state;
step 25: obtaining S= [ S ] from the state transition matrix and the initial state 1 ,s 2 ,s 3 ]The probability of occurrence of the state is:
step 26: determining transition probability P ij The calculation formula is as follows:
step 27: based on transition probability P ij And generating the electric vehicle charging demand data of the target residential area according to the current state of the electric vehicle user.
6. The method for optimizing a sub-division period ordered charging strategy based on a weighted sub-division period time-of-use power price for load assessment according to claim 1, wherein the determining initial power price of a corresponding grade based on assessment grading, obtaining the weighted sub-division period time-of-use power price, obtaining the ordered charging weighted time-of-use power price according to the weighted sub-division period time-of-use power price, comprises:
according to the predicted data of the charging demand of the electric automobile, dividing the load into three levels by calculating the load fluctuation rate of the platform region, wherein the working day is level I, the holiday is level II, the holiday is level III, and the initial electricity prices of the level I, the level II and the level III are determined as d Ⅰ、 d Ⅱ And d Ⅲ Obtaining a grading initial time-sharing electricity price matrix d f =[d Ⅰ ,d Ⅱ ,d Ⅲ ]。
The subdivision time-sharing electricity price adopts a calculation mode of a piecewise function, and the calculation formula is as follows:
wherein: t is the time of orderly charging participation, the unit is min, a 1 、a 2 、a 3 And b 1 、b 2 、b 3 Obtaining an ordered charging electricity price matrix d by taking the electricity price matrix d as a constant n =[d 1 ,d 2 ,d 3 ];
The actual electricity price is obtained by weighting the classified initial electricity price and the subdivision time-sharing electricity price, and the weighted calculation formula is as follows:
wherein: d is the actual electricity price after participating in ordered charging, w represents a weighting coefficient, and w is E [0,1].
7. The method of optimizing a sub-segment ordered charging strategy based on load assessment according to claim 7, wherein determining a corresponding ordered charging strategy according to an ordered charging theory of ordered charging weighted time-of-use electricity prices comprises:
step 31: leading residential electric vehicle holding quantity prediction data and target residential electric vehicle charging demand prediction data into a charging station, and generating a time-period charging power limit value matrix P by combining electric vehicle quantity constraint, charging power constraint, charging total quantity constraint and transformer capacity constraint lim =[P ph ,P p ,P 1 ],P ph Recorded as peak period power limit value, P p Is recorded as a normal power limit value, P 1 Recording as a valley period power limit;
step 32: according to the number N of electric vehicles in a target residential area and charging demand data, optimizing charging power in each period through a variable power ordered charging theory to generate P c =[P′ ph ,P′ p ,P′ 1 ],P′ ph The peak charge power, P' p The charge power is recorded as the normal charge power, P' 1 The charging power is recorded as valley period charging power;
step 33: and (3) docking the charging piles of the electric automobile, uploading SOC data, battery capacity, user arrival time and user-set departure time of the electric automobile, counting actual charging load, judging a load prediction error coefficient epsilon, and optimizing the charging time in the peak period by an optimization controller if the error coefficient epsilon is greater than a preset precision requirement, so as to limit the charging time of the electric automobile in the power utilization peak period.
8. The method for optimizing the sub-segment ordered charging strategy based on the load assessment according to claim 1, wherein the building of the master-slave gaming model based on the ordered charging strategy in combination with the gaming theory comprises the following steps:
based on master-slave games, an electricity seller is taken as a leader of the games, a leader policy is charging electricity price, an electric automobile user is a follower, the follower policy is charging plan, and a corresponding policy establishment is established as follows:
the profit of the electricity seller is as follows:
wherein maxA represents the maximum daily profit of the charging station, P ev (t) represents the charging load at time t of the electric vehicle, d t Represents the real-time charging electricity price, d b,t Represents electricity purchasing price from electricity seller, c p Unit penalty cost, Q, representing load peak-valley difference max Represents the maximum load, Q min Representing a minimum value of the load;
the charging expenditure per unit time of the electric automobile user is as follows:
wherein minB represents the minimum charge expenditure per unit time of the electric vehicle, C ev Representing the number of electric vehicles, P ev And (t) represents a charging load of the electric automobile at a certain time t.
9. The method for optimizing a sub-segment ordered charging strategy based on load assessment according to claim 9, wherein solving a betting model using an improved particle swarm algorithm combined with a wolf-swarm search comprises:
step 41: predicting the basic electricity load of the current day, and initializing the weighted time-sharing electricity price;
step 42: importing the arrival time and the required charging electric quantity information of the electric automobile according to the living rule of the residential area;
step 43: initializing population information, generating initial population of electric automobile users and electricity sellers, initializing positions and speeds of particles, limiting the charging quantity of the electric automobile according to the capacity constraint condition of a transformer, and limiting the charging price according to the constraint condition of an electricity selling station;
step 44: the population diversity and optimizing effect of the particle swarm algorithm are improved based on the wolf swarm search algorithm, and the position and speed of the particles are iteratively updated;
step 45: calculating an electric automobile user objective function, solving an optimal value, and solving particles meeting constraint conditions and maximizing user satisfaction according to the power price per hour;
step 46: calculating a charging station objective function based on a game model, solving an optimal value, and solving the income maximization particles meeting constraint conditions according to a user charging strategy;
step 47: judging whether the iteration termination condition is met, if not, turning to the step 44 and the step 45 to recalculate, and finally obtaining the optimal charging scheme.
10. A subdivision period ordered charging strategy optimization device based on load evaluation, characterized by comprising:
the acquisition module is used for acquiring the storage quantity data of the electric automobile in the target residential area and the charging demand data of the electric automobile in the target residential area;
the first determining module is used for determining the charging load of the electric vehicle in the target residential area according to the storage quantity data of the electric vehicle in the target residential area and the charging demand data of the electric vehicle in the target residential area;
the grading module is used for evaluating and grading the charging load of the target residential area;
the second determining module is used for determining initial electricity prices of the corresponding grades based on the evaluation grades, acquiring weighted subdivision time-of-day electricity prices, obtaining ordered charging weighted time-of-day electricity prices according to the weighted subdivision time-of-day electricity prices, and determining corresponding ordered charging strategies according to an ordered charging theory of the ordered charging weighted time-of-day electricity prices;
and the model construction and solving module is used for constructing a master-slave game model based on the ordered charging strategy and combining with a game theory, solving the master-slave game model by utilizing an improved particle swarm algorithm combined with wolf swarm search, and obtaining load peak-valley difference, load fluctuation rate, maximum load rate and user charging cost data of the power grid operation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310854084.4A CN117057533A (en) | 2023-07-12 | 2023-07-12 | Method and device for optimizing subdivision period ordered charging strategy based on load evaluation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310854084.4A CN117057533A (en) | 2023-07-12 | 2023-07-12 | Method and device for optimizing subdivision period ordered charging strategy based on load evaluation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117057533A true CN117057533A (en) | 2023-11-14 |
Family
ID=88659757
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310854084.4A Pending CN117057533A (en) | 2023-07-12 | 2023-07-12 | Method and device for optimizing subdivision period ordered charging strategy based on load evaluation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117057533A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117374975A (en) * | 2023-12-06 | 2024-01-09 | 国网湖北省电力有限公司电力科学研究院 | Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming |
CN117996754A (en) * | 2024-04-01 | 2024-05-07 | 南京邮电大学 | Electric automobile ordered charge and discharge control method based on improved DBO algorithm |
-
2023
- 2023-07-12 CN CN202310854084.4A patent/CN117057533A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117374975A (en) * | 2023-12-06 | 2024-01-09 | 国网湖北省电力有限公司电力科学研究院 | Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming |
CN117374975B (en) * | 2023-12-06 | 2024-02-27 | 国网湖北省电力有限公司电力科学研究院 | Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming |
CN117996754A (en) * | 2024-04-01 | 2024-05-07 | 南京邮电大学 | Electric automobile ordered charge and discharge control method based on improved DBO algorithm |
CN117996754B (en) * | 2024-04-01 | 2024-06-04 | 南京邮电大学 | Electric automobile ordered charge and discharge control method based on improved DBO algorithm |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zheng et al. | Day-ahead bidding strategy for electric vehicle aggregator enabling multiple agent modes in uncertain electricity markets | |
Ahmadian et al. | A review on plug-in electric vehicles: Introduction, current status, and load modeling techniques | |
Li et al. | Electric vehicle charging management based on deep reinforcement learning | |
Bi et al. | Estimating remaining driving range of battery electric vehicles based on real-world data: A case study of Beijing, China | |
Yang et al. | Computational scheduling methods for integrating plug-in electric vehicles with power systems: A review | |
CN117057533A (en) | Method and device for optimizing subdivision period ordered charging strategy based on load evaluation | |
Yang et al. | Agent-based modeling for scale evolution of plug-in electric vehicles and charging demand | |
Qiu et al. | Reinforcement learning for electric vehicle applications in power systems: A critical review | |
CN113103905B (en) | Intelligent charging distribution adjusting method, device, equipment and medium for electric automobile | |
Zhang et al. | Multistep multiagent reinforcement learning for optimal energy schedule strategy of charging stations in smart grid | |
CN114004450A (en) | Ordered charging model guided by electric vehicle charging load interactive real-time pricing strategy | |
CN112238781B (en) | Electric automobile ordered charging control method based on layered architecture | |
CN109800917A (en) | A kind of planing method in electric car parking lot, device and calculate equipment | |
CN111553750A (en) | Energy storage bidding strategy method considering power price uncertainty and loss cost | |
Nweye et al. | MERLIN: Multi-agent offline and transfer learning for occupant-centric operation of grid-interactive communities | |
Dai et al. | Robust energy management for a corporate energy system with shift-working V2G | |
CN111313449B (en) | Cluster electric vehicle power optimization management method based on machine learning | |
Wu et al. | Electric vehicle charging scheduling considering infrastructure constraints | |
CN115663804A (en) | Electric power system regulation and control method based on deep reinforcement learning | |
Zhang et al. | A safe reinforcement learning-based charging strategy for electric vehicles in residential microgrid | |
Tan et al. | Bi-level optimization of charging scheduling of a battery swap station based on deep reinforcement learning | |
CN112001564A (en) | User leaving time prediction method based on long-term and short-term memory network | |
CN114498632A (en) | Power distribution station load prediction method and device based on new energy and charging facility | |
CN114692965A (en) | Electric vehicle cluster schedulable potential prediction method and device based on digital-analog hybrid driving | |
CN117543581A (en) | Virtual power plant optimal scheduling method considering electric automobile demand response and application thereof |
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 |