CN115036952A - Real-time power control method for electric vehicle participating in load stabilization based on MPC - Google Patents

Real-time power control method for electric vehicle participating in load stabilization based on MPC Download PDF

Info

Publication number
CN115036952A
CN115036952A CN202210673083.5A CN202210673083A CN115036952A CN 115036952 A CN115036952 A CN 115036952A CN 202210673083 A CN202210673083 A CN 202210673083A CN 115036952 A CN115036952 A CN 115036952A
Authority
CN
China
Prior art keywords
load
power
time
charge
unit
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
Application number
CN202210673083.5A
Other languages
Chinese (zh)
Inventor
赖信辉
胡俊杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
North China Electric Power University
Original Assignee
North China Electric Power University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by North China Electric Power University filed Critical North China Electric Power University
Priority to CN202210673083.5A priority Critical patent/CN115036952A/en
Publication of CN115036952A publication Critical patent/CN115036952A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/28—Arrangements for balancing of the load in a network by storage of energy
    • H02J3/32—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means
    • H02J3/322—Arrangements for balancing of the load in a network by storage of energy using batteries with converting means the battery being on-board an electric or hybrid vehicle, e.g. vehicle to grid arrangements [V2G], power aggregation, use of the battery for network load balancing, coordinated or cooperative battery charging
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/008—Circuit arrangements for ac mains or ac distribution networks involving trading of energy or energy transmission rights
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/0047—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries with monitoring or indicating devices or circuits
    • H02J7/0048—Detection of remaining charge capacity or state of charge [SOC]
    • 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
    • Y04S20/00—Management or operation of end-user stationary applications or the last stages of power distribution; Controlling, monitoring or operating thereof
    • Y04S20/20—End-user application control systems
    • Y04S20/222—Demand response systems, e.g. load shedding, peak shaving

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)

Abstract

The invention belongs to the field of electric vehicle participation vehicle network interaction, and particularly relates to a real-time power control method for electric vehicle participation load stabilization based on MPC, which comprises the following steps: collecting travel, load power, operation of a thermal power generating unit and time-of-use electricity price information of the electric automobile; predicting the load by adopting a neural network algorithm according to various collected historical information data to obtain a load predicted value of the next day; establishing a day-to-day regulation and control model considering load stabilization, system power balance constraint, thermal power unit constraint, EV charge and discharge cost, battery loss cost and thermal power unit operation cost; carrying out multi-objective solution on the model in an optimization period by adopting a weight multi-objective method; and (3) performing rolling optimization real-time updating on the load information based on a model predictive control theory to obtain a scheduling power and load stabilizing result in one day. The method has universality, can optimize the system operation cost and improve the unit and EV operation economy while effectively realizing load stabilization, and has practical significance in the aspect of electric automobile participating in vehicle network interaction.

Description

Real-time power control method for electric vehicle participating in load stabilization based on MPC
Technical Field
The invention belongs to the field of electric vehicle participation vehicle network interaction, and particularly relates to a real-time power control method for load stabilization of electric vehicles based on MPC.
Background
Electric vehicles have been rapidly developed in recent years as an important part of achieving the "dual carbon" goal. With the increasing of the number of electric vehicles, a large number of electric vehicles are connected into a power grid and are in an unordered charging state, so that the problems of increasing power grid loss, reducing power quality, increasing the difficulty of optimizing and controlling the operation of the power grid and the like are caused, and under the condition that the current new energy gives a serious problem, the electric vehicle is a huge threat to the safe and stable operation of the power grid. Therefore, the charging of the electric automobile needs to be optimized and guided, so that the electric automobile can be charged and discharged orderly under the condition of meeting the travel requirements of users, the peak-valley difference of the load of the power distribution network is reduced, the load fluctuation is inhibited, the influence of the disordered charging of the electric automobile on the power grid is effectively limited, and the economic benefit of the electric automobile users is improved. The existing power control method for the electric automobile to participate in load stabilization considers the optimization of the system less, but considers the load side cost or the power generation cost more singly, and does not relate to the real-time control level of the electric automobile. Therefore, based on the problem, an effective, fast and accurate real-time power control method for load stabilization of the electric vehicle is needed.
Disclosure of Invention
The invention discloses a real-time power control method for an electric automobile to participate in load stabilization based on MPC, which comprises the following steps:
step A, collecting travel, load power, operation of a thermal power generating unit and time-of-use electricity price information of the electric automobile;
b, predicting the load by adopting a neural network algorithm according to various collected historical information data to obtain a load predicted value of the next day;
c, establishing an intra-day regulation and control model considering load stabilization, system power balance constraint, thermal power unit constraint and various costs;
and D, solving by adopting a weight multi-objective method based on the model established in the step C, and obtaining a scheduling result.
And E, performing rolling optimization on the load information and updating the scheduling result in real time.
Further, the basic information of the cluster electric vehicle in the step A comprises EV arrival time, predicted departure time, battery capacity, the charge state of a vehicle-mounted battery when arriving at the station and an expected SOC value required to be reached when leaving the station, the load power information is load power data of a plurality of historical days in a certain area, the operation information of the thermal power unit comprises information such as maximum output, minimum output, climbing rate constraint and operation cost coefficient of the thermal power unit, and the time-sharing electricity price information comprises charging electricity prices and discharging electricity prices of different time periods in a certain area day.
Further, in step B, according to the collected historical information, the load is predicted by using a long short term memory network (LSTM) algorithm to obtain the load power of the next day:
Figure BDA0003695405270000021
S(t),S(t-1),...,S(t-96+1),
C(t),C(t-1),...,C(t-96+1))
in the formula: r p (t) represents the charge and discharge power vectors of all EVs in the t-th scheduling period, and the magnitude of the charge and discharge power vectors is N x 1; s (t) represents the presence status vectors of all EVs in the t-th scheduling period; and C (t) represents a charge and discharge electricity price vector of the t scheduling period, and the magnitude of the charge and discharge electricity price vector is 2 x 1.
Further, the daily model objective function established in step C is as follows, and the load stabilizing objective is:
Figure BDA0003695405270000022
min f 1 ＝K var,day
t∈[t,t+3]
the cost optimal target is:
Figure BDA0003695405270000023
Figure BDA0003695405270000024
Figure BDA0003695405270000025
L ET ＝L c E s DoD
Figure BDA0003695405270000031
min f 2 ＝C ev +C B +C plant +C wind
wherein, P ev (t) represents the total net charge-discharge power of all electric vehicles at time t;
Figure BDA0003695405270000032
respectively representing the charging electric quantity and the discharging electric quantity of the nth EV in a time period t; k var,day Respectively representing the variance of the optimization and the reference value in the day;
Figure BDA0003695405270000033
predicting power for the load;
Figure BDA0003695405270000034
the representation of the optimized reference power is sent by a dispatching center; lambda [ alpha ] tou，t 、λ feed，t Represents a charging power rate and a discharging power rate at a time period t; b is Cost Representing the depreciation cost of the battery generated by the unit discharge capacity of the EV; c bat Represents the base cost of the battery; l is ET Represents battery life in kWh; l is c Battery life in cycles; e s Represents the total stored energy of the battery; DoD denotes EV depth of discharge; a. b and c are quadratic fitting coefficients of the operation cost of the thermal power generating unit; s c Is the unit coal price.
Furthermore, in the step C, various physical constraints and EV travel constraints of each unit of the system are fully considered, and the constraint conditions include:
Figure BDA0003695405270000035
P plant,min <P plant (t)<P plant,max
-ε 1 P plant,max <P plant (t)-P plant (t-1)<ε 2 P plant,max
Figure BDA0003695405270000036
Figure BDA0003695405270000037
δ dischg ·δ chg ＝0
Figure BDA0003695405270000038
Figure BDA0003695405270000039
wherein, P plant (t) representing the output of the thermal power generating unit at the moment t; p plant,min 、P plant,max Respectively representing theoretical minimum output power and theoretical maximum output power of the thermal power generating unit; epsilon 1 、ε 2 Respectively representing the downward and upward climbing rates of the unit, limiting the power adjustment amount of the unit in unit time,
Figure BDA0003695405270000041
respectively representing the maximum charge and discharge electric quantity of the nth EV in a t period; delta chg 、δ dischg Representing EV charge-discharge state 0-1 variable to ensure that charge and discharge are not performed simultaneously, SOC n,min Represents the SOC minimum value of the nth EV; SOC n,arr 、SOC n,dep Respectively representing an on-site SOC value and an off-site SOC value of the nth EV; eta chg,n 、η dischg,n The charge and discharge efficiencies of the nth EV are shown, respectively.
Preferably, in step D, for the two objective functions, a weighted multi-objective method is used to solve the multi-objective problem.
Preferably, in step E, the load information is predicted and updated in real time based on a model predictive control theory (MPC) theory, so as to update the scheduling result in real time.
Drawings
FIG. 1 is a control framework diagram of the present invention;
FIG. 2 is a flow chart of the method of the present invention;
FIG. 3 is a MPC based load leveling real-time control schematic;
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings.
Figure 1 shows the control framework of the present invention. Information transmission exists between a power grid and a aggregator, the aggregator acquires travel information data of electric vehicles under the aggregator and controls charging and discharging power of the electric vehicles, and the power grid acquires operation information of a thermal power generating unit and adjusts the operation power of the thermal power generating unit according to an overall optimization result of the system.
Fig. 2 is a flow chart of real-time power control of an electric vehicle participating in load leveling. Comprises the following steps:
and step A, collecting the travel, load power, operation of the thermal power generating unit and time-of-use electricity price information of the electric automobile.
On the aggregator level, the information required to be collected comprises electric vehicle travel information, and the basic information of the cluster electric vehicle comprises EV arrival time, predicted departure time, battery capacity, on-board battery charge state when arriving and expected SOC value required to be reached when departing; on the power grid level, the information required to be collected comprises load power and thermal power unit operation information, the load power information is load power data of a plurality of historical days and is used for reference of future load power prediction, and the thermal power unit operation information comprises information such as maximum output, minimum output, climbing rate constraint and operation cost coefficient (related to coal consumption cost) of the thermal power unit; in the system operation aspect, the collected time-of-use electricity price information comprises charging electricity prices and discharging electricity prices of different time periods in a certain area day, and is used for optimizing the charging and discharging cost of the electric automobile.
And B, according to the collected historical information, considering that a long-short term memory network algorithm is adopted to predict the load so as to obtain the load power of the next day, wherein the output relation of the LSTM unit input is as follows:
Figure BDA0003695405270000051
S(t),S(t-1),...,S(t-96+1),
C(t),C(t-1),...,C(t-96+1))
in the formula: r p (t) represents the charge and discharge power vectors of all EVs in the t-th scheduling period, and the magnitude of the charge and discharge power vectors is N x 1; s (t) represents the presence status vectors of all EVs in the t-th scheduling period; and C (t) represents a charge and discharge electricity price vector of the t scheduling period, and the magnitude of the charge and discharge electricity price vector is 2 x 1.
The LSTM unit carries out accurate prediction through a door mechanism (a forgetting door, an input door and an output door), and each door has the following functions:
1) forget the door: selecting weak related information in data such as load power, EV field state, EV charge state, charge and discharge price and the like
f t ＝σ(W f ·[h t-1 ,x t ]+b f )
In the formula: h is a total of t-1 The output (information such as storage load power, EV in-field state, EV charge state, electricity price and the like) of the hidden layer at the time t-1 is represented; x is the number of t The input load power value at the time t is represented; sigma represents a sigmoid activation function; f. of t Indicating a forgotten gate output; w f Representing a forgetting gate weight matrix; b f A matrix representing forgotten gate offset values;
2) an input gate: updating the state of the memory unit together with the sigmoid layer and the hidden layer
i t ＝σ(W i ·[h t-1 ,x t ]+b i )
Figure BDA0003695405270000052
Figure BDA0003695405270000053
In the formula: i.e. i t Representing the input gate output; w i Representing an input gate weight matrix; b i Representing a matrix of input gate offset values;
Figure BDA0003695405270000061
the updated value of the information of the state of the unit at the time t, such as load power, the on-site state of the EV, the state of charge of the EV, the price of electricity and the like; w C 、b C Respectively representing a neuron weight matrix and an offset value matrix; c t And important information representing data such as load power, EV presence state, EV state of charge, price and the like which are reserved in the unit state at the time t.
3) An output gate: and outputting the load information prediction result.
C, establishing an intra-day regulation and control model considering load stabilization, system power balance constraint, thermal power unit constraint and various costs;
and step C1, according to the various load power information collected in the step A, according to the power balance constraint, the thermal power output constraint and the EV charge-discharge constraint, establishing constraint conditions of an objective function as follows:
Figure BDA0003695405270000062
P plant,min <P plant (t)<P plant,max
-ε 1 P plant,max <P plant (t)-P plant (t-1)<ε 2 P plant,max
Figure BDA0003695405270000063
Figure BDA0003695405270000064
δ dischg +δ chg ＝0
Figure BDA0003695405270000065
Figure BDA0003695405270000066
wherein, P plant (t) representing the power of the thermal power generating unit at the moment t; p plant,min 、P plant,max Respectively representing the theoretical minimum output power and the theoretical maximum output power of the thermal power generating unit; epsilon 1 、ε 2 Respectively representing the downward and upward climbing rates of the unit and limiting the power adjustment amount of the unit in unit time;
Figure BDA0003695405270000067
respectively representing the maximum charge and discharge electric quantity of the nth EV in the t period; delta chg 、δ dischg Representing EV charge-discharge state 0-1 variable to ensure that charge and discharge are not performed simultaneously, SOC n,min Represents the SOC minimum value of the nth EV; SOC n,arr 、SOC n,dep Respectively representing an on-site SOC value and an off-site SOC value of the nth EV; eta chg,n 、η dischg,n The charge and discharge efficiencies of the nth EV are shown, respectively.
Step C2: and establishing a day-to-day regulation and control model considering EV charge-discharge cost, battery loss cost, thermal power unit operation cost and wind abandoning cost.
And C, controlling the electric automobile and the thermal power generating unit in real time based on a Model Predictive Control (MPC) theory in the scheduling process in the day according to the load reference value obtained by the long-term and short-term memory neural network prediction in the step B.
The target function of the built in-day regulation and control model is as follows, and the load stabilizing target is as follows:
Figure BDA0003695405270000071
min f 1 ＝K var,day
t∈[t,t+3]
the cost optimal target is:
Figure BDA0003695405270000072
Figure BDA0003695405270000073
Figure BDA0003695405270000074
L ET ＝L c E s DoD
Figure BDA0003695405270000075
min f 2 ＝C ev +C B +C plant +C wind
wherein, P ev (t) represents the total net charge-discharge power of all electric vehicles at time t;
Figure BDA0003695405270000076
respectively representing the charging electric quantity and the discharging electric quantity of the nth EV in a time period t; k var,day Respectively representing the variance of the optimization and the reference value in the day;
Figure BDA0003695405270000077
predicting power for the load;
Figure BDA0003695405270000078
representing the optimized reference power, and transmitted by a scheduling center; lambda [ alpha ] tou，t 、λ feed，t Is shown at time period tCharging electricity prices and discharging electricity prices of (c); b is Cost Representing the depreciation cost of the battery generated by the discharge capacity of the EV unit; c bat Represents the base cost of the battery; l is ET Represents battery life in kWh; l is c Battery life in cycles; e s Represents the total stored energy of the battery; DoD denotes EV depth of discharge; a. b and c are quadratic fitting coefficients of the operation cost of the thermal power generating unit; s c Is the unit coal price.
And D, solving by adopting a weight multi-objective method based on the real-time control model established in the step C, and obtaining a scheduling result.
The two objective functions are respectively a load translation objective function and a cost optimal function, different weight values are respectively set for the two objective functions when multi-objective solving is carried out, so that objective algorithm solving with different weight deviation is carried out, and the total objective function is as follows:
min f 3 ＝αf 1 +(1-α)f 2
wherein: alpha denotes the objective function f 2 The weight of (c); (1-. alpha.) represents the objective function f 3 The weight of (c).
FIG. 3 is a schematic diagram of MPC based real-time load leveling control.
Step e, after the optimization model is established and solved in step B, C, D, based on the MPC theory, since the forecast information is prone to be biased, the load forecast information is updated in real time, and the model is solved by rolling optimization to enhance the scheduling accuracy and effectiveness, as shown in fig. 3, the load data before the t period is read, and [ t, t + H ] is forecasted, and then the rolling optimization is performed on [ t, t + H ], and the result of the solution is only applied to the t period, but not executed in the following period, and the rolling is repeated until one day is finished.
The real-time power control method for participating in load stabilization of the electric automobile based on the MPC fully considers the system whole consisting of the electric automobile, the thermal power generating unit, the power grid and the aggregator, brings the travel constraint condition of the electric automobile, the charge and discharge constraint condition of the battery and the output constraint condition of the thermal power generating unit into the consideration range, and provides a day-ahead-day load power stabilization method based on model predictive control. The load power stabilizing model established by the invention is different from the traditional method of independently considering the optimal cost target at the power grid side, the load target tracking or the maximum profit target at the user side, and more fully and comprehensively considers the influence of main body participation in various aspects and is more fit with the actual regulation and control scene; the load stabilizing-cost control model established at the same time endows different weight values to the load stabilizing target and the cost optimal target, so that the deviation condition of the load stabilizing target and the total system cost optimal target can be controlled, and the regulation and control performance is stronger.
The method has generality in a modeling mode, does not have any special application condition, has wide application range, and is easy to popularize in the fields of electric automobiles or participating in wind power consumption, load stabilization, peak shaving and the like of power grids and energy storage.
The above embodiments are only preferred embodiments of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are also within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (6)

1. The invention relates to a real-time power control method for an electric automobile participating in load stabilization based on MPC, which comprises the following steps:
step A, collecting travel, load power, operation of a thermal power generating unit and time-of-use electricity price information of the electric vehicle;
b, predicting the load by adopting a neural network algorithm according to various collected historical information data to obtain a load predicted value of the next day;
c, establishing an intra-day regulation and control model considering load stabilization, system power balance constraint, thermal power unit constraint and various costs;
and D, solving by adopting a weight multi-objective method based on the model established in the step C, and obtaining a scheduling result.
And E, performing rolling optimization on the load information and updating the scheduling result in real time.
2. The real-time power control method for load stabilization participation of an electric vehicle based on MPC as claimed in claim 1, wherein: the basic information of the cluster electric vehicle in the step A comprises EV arrival time, predicted departure time, battery capacity, vehicle-mounted battery charge state when arriving and an expected SOC value required to be reached when leaving, the load power information is load power data of a plurality of historical days in a certain area, the operation information of the thermal power unit comprises information such as maximum output, minimum output, climbing rate constraint and operation cost coefficient of the thermal power unit, and the time-sharing electricity price information comprises charging electricity prices and discharging electricity prices of different time periods in a certain area day.
3. The real-time power control method for load stabilization of the electric vehicle based on the MPC as claimed in claim 2, wherein: in the step B, load is predicted by adopting a neural network algorithm according to various collected historical information data to obtain a load predicted value of the next day, and the output relation of the input of the LSTM unit is as follows:
Figure FDA0003695405260000011
S(t),S(t-1),...,S(t-96+1),
C(t),C(t-1),...,C(t-96+1))
in the formula: r p (t) represents the charge and discharge power vectors of all EVs in the t-th scheduling period, and the magnitude of the charge and discharge power vectors is N x 1; s (t) a presence status vector representing all EVs of the t-th scheduling period; and C (t) represents a charge and discharge electricity price vector of the t scheduling period, and the magnitude of the charge and discharge electricity price vector is 2 x 1.
The LSTM unit carries out accurate prediction through a door mechanism (a forgetting door, an input door and an output door), and each door has the following functions:
1) forgetting the door: selecting weak related information in data such as load power, EV field state, EV charge state, charge and discharge price and the like
f t ＝σ(W f ·[h t-1 ,x t ]+b f )
In the formula: h is t-1 The output (information such as storage load power, EV in-field state, EV charge state, electricity price and the like) of the hidden layer at the time t-1 is represented; x is the number of t The input load power value at the time t is represented; sigma represents a sigmoid activation function; f. of t Indicating a forgotten gate output; w is a group of f Representing a forgetting gate weight matrix; b f A matrix representing forgotten gate offset values;
2) an input gate: updating the state of the memory unit together with the sigmoid layer and the hidden layer
i t ＝σ(W i ·[h t-1 ,x t ]+b i )
Figure FDA0003695405260000021
Figure FDA0003695405260000022
In the formula: i.e. i t Representing the input gate output; w i Representing an input gate weight matrix; b is a mixture of i Representing a matrix of input gate offset values;
Figure FDA0003695405260000023
the updated value of the information of the state of the unit at the time t, such as load power, the on-site state of the EV, the state of charge of the EV, the price of electricity and the like; w C 、b C Respectively representing a neuron weight matrix and an offset value matrix; c t And important information representing data such as load power, EV presence state, EV state of charge, price and the like which are reserved in the unit state at the time t.
3) An output gate: and outputting the load information prediction result.
4. The real-time power control method for load stabilization participation of electric vehicles based on MPC as claimed in claim 3, wherein: in the step B, the load power stabilization is taken as a target, various physical constraints and EV travel constraints of each unit of the system are fully considered, a mixed nonlinear integer programming model before the day is constructed, and the constraint conditions comprise:
Figure FDA0003695405260000031
P plant,min <P plant (t)<P plant,max
-ε 1 P plant,max <P plant (t)-P plant (t-1)<ε 2 P plant,max
Figure FDA0003695405260000032
Figure FDA0003695405260000033
δ dischg +δ chg ＝0
Figure FDA0003695405260000034
Figure FDA0003695405260000035
wherein, P plant (t) representing the output of the thermal power generating unit at the moment t; p is plant,min 、P plant,max Respectively representing theoretical minimum output power and theoretical maximum output power of the thermal power generating unit; epsilon 1 、ε 2 Respectively representing the downward and upward climbing rates of the unit, limiting the power adjustment amount of the unit in unit time,
Figure FDA0003695405260000036
respectively representing the maximum charge and discharge of the nth EV in the period tAn amount of electricity; delta chg 、δ dischg Representing EV charge-discharge state 0-1 variable to ensure that charge and discharge are not performed simultaneously, SOC n,min Represents the SOC minimum value of the nth EV; SOC n,arr 、SOC n,dep Respectively representing an on-site SOC value and an off-site SOC value of the nth EV; eta chg,n 、η dischg,n The charge and discharge efficiencies of the nth EV are shown, respectively.
5. The real-time power control method for load stabilization participation of electric vehicles based on MPC as claimed in claim 4, wherein: the target function of the built in-day regulation model is as follows, and the load stabilizing target is as follows:
Figure FDA0003695405260000037
min f 1 ＝K var,day
t∈[t,t+3]
the cost optimal target is:
Figure FDA0003695405260000041
Figure FDA0003695405260000042
Figure FDA0003695405260000043
L ET ＝L c E s DoD
Figure FDA0003695405260000044
min f 2 ＝C ev +C B +C plant +C wind
wherein, P ev (t) represents the total net charge-discharge power of all electric vehicles at time t;
Figure FDA0003695405260000045
respectively representing the charging electric quantity and the discharging electric quantity of the nth EV in a time period t; k var,day Respectively representing the variance of the optimization and the reference value in the day;
Figure FDA0003695405260000046
predicting power for the load;
Figure FDA0003695405260000047
the representation of the optimized reference power is sent by a dispatching center; lambda tou，t 、λ feed，t Represents a charging power rate and a discharging power rate at a time period t; b is Cost Representing the depreciation cost of the battery generated by the discharge capacity of the EV unit; c bat Represents the base cost of the battery; l is ET Represents battery life in kWh; l is c Battery life in cycles; e s Represents the total stored energy of the battery; DoD denotes EV depth of discharge; a. b and c are quadratic fitting coefficients of the operation cost of the thermal power generating unit; s c Is the unit coal price.
6. The method for predicting the dispatchable capacity of the cluster electric vehicles according to claim 5, wherein the method comprises the following steps: in the step D, the weight multi-objective method sets different weight values for the two objective functions respectively, so as to perform different weight biased objective algorithm solving.
CN202210673083.5A 2022-06-15 2022-06-15 Real-time power control method for electric vehicle participating in load stabilization based on MPC Pending CN115036952A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210673083.5A CN115036952A (en) 2022-06-15 2022-06-15 Real-time power control method for electric vehicle participating in load stabilization based on MPC

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210673083.5A CN115036952A (en) 2022-06-15 2022-06-15 Real-time power control method for electric vehicle participating in load stabilization based on MPC

Publications (1)

Publication Number Publication Date
CN115036952A true CN115036952A (en) 2022-09-09

Family

ID=83125938

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210673083.5A Pending CN115036952A (en) 2022-06-15 2022-06-15 Real-time power control method for electric vehicle participating in load stabilization based on MPC

Country Status (1)

Country Link
CN (1) CN115036952A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116073418A (en) * 2023-02-14 2023-05-05 燕山大学 Electric automobile charging and discharging scheduling method based on dynamic electricity price

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116073418A (en) * 2023-02-14 2023-05-05 燕山大学 Electric automobile charging and discharging scheduling method based on dynamic electricity price

Similar Documents

Publication Publication Date Title
CN109492815B (en) Energy storage power station site selection and volume fixing optimization method for power grid under market mechanism
CN103241130B (en) Energy management method and system for electric bus charging and swap station
CN110378548B (en) Electric automobile virtual power plant multi-time scale response capability assessment model construction method
CN112238781B (en) Electric automobile ordered charging control method based on layered architecture
CN110138006A (en) Consider more micro electric network coordination Optimization Schedulings containing New-energy electric vehicle
CN112865190A (en) Optimal scheduling method and system for photovoltaic and charging demand-based optical storage charging station
CN113794199B (en) Maximum benefit optimization method of wind power energy storage system considering electric power market fluctuation
Li et al. Online battery protective energy management for energy-transportation nexus
CN116345577A (en) Wind-light-storage micro-grid energy regulation and optimization method, device and storage medium
Dong et al. Optimal scheduling framework of electricity-gas-heat integrated energy system based on asynchronous advantage actor-critic algorithm
CN114744662A (en) Power grid peak regulation method and system based on multiple types of electric automobiles
Han et al. An optimization scheduling method of electric vehicle virtual energy storage to track planned output based on multiobjective optimization
CN108110801A (en) Consider electric vehicle and the active power distribution network multilevel redundancy control method for coordinating of energy storage
CN113972645A (en) Power distribution network optimization method based on multi-agent depth determination strategy gradient algorithm
CN115036952A (en) Real-time power control method for electric vehicle participating in load stabilization based on MPC
CN114723230A (en) Micro-grid double-layer scheduling method and system for new energy power generation and energy storage
Li et al. Optimal dispatch for PV-assisted charging station of electric vehicles
CN202650066U (en) Coordination control system for electric-automobile service network based on Multi-Agent system
CN115115145B (en) Demand response scheduling method and system for distributed photovoltaic intelligent residence
Zhang et al. Transfer deep reinforcement learning-based large-scale V2G continuous charging coordination with renewable energy sources
CN113283166B (en) Retired power battery residual value optimization method
CN115147244A (en) Method for achieving wind curtailment and accommodation by considering charging load-electricity price response of electric automobile
Huang et al. The impact of electric vehicle development on grid load power and electricity consumption
Xu et al. Control Strategy of Electric Vehicle Participating in Power Grid Peak Regulation Based on V2G
CN114742453A (en) Micro-grid energy management method based on Rainbow deep Q network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication