CN112909979B - Multi-mode frequency modulation method, device, equipment and medium for cluster electric automobile - Google Patents
Multi-mode frequency modulation method, device, equipment and medium for cluster electric automobile Download PDFInfo
- Publication number
- CN112909979B CN112909979B CN202110419399.7A CN202110419399A CN112909979B CN 112909979 B CN112909979 B CN 112909979B CN 202110419399 A CN202110419399 A CN 202110419399A CN 112909979 B CN112909979 B CN 112909979B
- Authority
- CN
- China
- Prior art keywords
- state
- control
- power
- charging
- electric automobile
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention discloses a multi-mode frequency modulation method, a multi-mode frequency modulation device, multi-mode frequency modulation equipment and a multi-mode frequency modulation medium for a cluster electric automobile, wherein the method comprises the steps of constructing a multi-state switching control model of the cluster electric automobile based on a load dynamic transfer model of the electric automobile in a charging, discharging and idle state and a switching process among the charging, discharging and idle states; calculating the real-time adjustable power of the cluster electric automobile according to the multi-state switching control model, and performing sectional processing on the range of the real-time adjustable power; and judging the section to which the AGC command belongs according to the result of the section processing, designing a corresponding control mode, and constructing a multi-mode frequency modulation mode of the cluster electric automobile according to the control mode. The invention widens the real-time adjustable power range by considering the direct conversion of the charge-discharge state; and the charging and discharging power of the cluster electric automobile can be adjusted in real time, the direct switching times of the charging and discharging states are reduced, the charging and discharging power is prevented from being offset, and the AGC instruction is accurately and quickly tracked.
Description
Technical Field
The invention relates to the technical field of vehicle network interconnection, in particular to a multi-mode frequency modulation method, a multi-mode frequency modulation device, multi-mode frequency modulation equipment and a multi-mode frequency modulation medium for a cluster electric vehicle.
Background
With the rapid development of battery technology and vehicle-electricity interconnection technology, the mobile energy storage characteristic of the electric vehicle and the rapid response capability of the battery are utilized to make up for the deficiency of the frequency modulation resource of the new energy power system, and the method becomes a hot spot of current research. Different from a centralized energy storage power station, a single electric vehicle has low power and low capacity, so that the user using behavior is uncertain, and the difficulty of directly controlling the single electric vehicle is very high, so that the cluster electric vehicle is more suitable for participating in power grid regulation and control. However, the cluster electric vehicles have a large number of groups and wide distribution, and the disordered charging of the cluster electric vehicles can increase peak-valley difference, reduce the power quality due to too many charging and discharging conversion times, and even have negative effects on the safe and stable operation of the system. Therefore, how to provide a frequency modulation method to reasonably adjust and control multiple modes of the cluster electric vehicle is a technical problem to be solved urgently in the field.
Disclosure of Invention
The invention aims to provide a multi-mode frequency modulation method, a multi-mode frequency modulation device, multi-mode frequency modulation equipment and a multi-mode frequency modulation medium for a cluster electric automobile, and aims to solve the problems that in the prior art, the frequency modulation difficulty of the electric automobile is high, the electric energy quality is influenced, and the safety is low.
In order to overcome the defects in the prior art, the invention provides a multi-mode frequency modulation method for a cluster electric automobile, which comprises the following steps:
constructing a multi-state switching control model of the cluster electric automobile based on a load dynamic transfer model of the electric automobile in a charging, discharging and idle state and a switching process between the charging, discharging and idle states;
calculating the real-time adjustable power of the cluster electric automobile according to the multi-state switching control model, and performing sectional processing on the range of the real-time adjustable power;
and judging the section to which the AGC command belongs according to the result of the section processing, designing a corresponding control mode, and constructing a multi-mode frequency modulation mode of the cluster electric automobile according to the control mode.
Further, the load dynamic transfer model in the charging state of the electric vehicle is as follows:
x c (k+1)＝A c x c (k)+v c (k)
in the formula, x c (k) The vector is an N-dimensional column vector and represents the load quantity in each SOC interval under the charging state at the moment k; v. of c (k) The N-dimensional column vector represents the load amount newly added or separated from the charging state in each interval at the moment k; a. The c A state transition matrix in a charging state;
wherein x is c (k)，v c (k) And A c Respectively satisfy:
x c (k)＝[x c (k,1),…,x c (k,i),…,x c (k,N)] T
v c (k)＝[v c (k,1),v c (k,2),…,v c (k,N)] T
in the formula (I), the compound is shown in the specification,
indicates SOC interval i during charging c And i c A transition probability between + 1;
the load amount representing the charge completion is a proportion of the total load amount in the section N.
Further, the load dynamic transfer model in the electric vehicle discharge state is as follows:
x d (k+1)＝A d x d (k)+v d (k)
in the formula, x d (k) The N-dimensional column vector represents the load quantity in each SOC interval in the discharge state at the moment k; v. of d (k) The N-dimensional column vector represents the load quantity newly added or separated from the discharge state in each interval at the moment k; a. The d State transition matrix for discharge state:
in the formula (I), the compound is shown in the specification,
indicates SOC interval i in the discharging process d +1 and i d Transition probabilities between;
the load representing the completion of discharge accounts for the total load in section 1.
Further, the load dynamic transfer model in the idle state of the electric vehicle is as follows:
x l (k+1)＝A l x l (k)+v l (k)
in the formula, x l (k) The N-dimensional column vector represents the load quantity in each SOC interval under the idle state at the moment k; v. of l (k) The N-dimensional column vector represents the load quantity newly added or separated from the idle state in each interval at the moment k; a. The l Is an N-dimensional unit matrix and represents a state transition matrix in an idle state.
Further, the multi-state switching control model of the cluster electric vehicle is as follows:
in the formula u 1 (k)、u 2 (k)、u 3 (k) Are all N-dimensional column vectors; a is a 3 Nx 3N-dimensional state transition matrix; b is a 3 Nx 3N-dimensional state transition matrix used for describing a state switching process; c is an output row vector with 3N dimensions and is used for describing a power output part; v (k) = [ V = c (k),v l (k),v d (k)] T (ii) a Y (k) is the total output power.
Wherein A, B and C are respectively:
in the formula, 0 represents an N-dimensional zero matrix; i represents an N-dimensional unit array;
representing that the cluster electric vehicle adopts maximum power charging;
indicating that the cluster electric automobile adopts maximum power discharge.
Further, the step of performing segmentation processing on the range of the real-time adjustable power includes dividing the real-time adjustable power into a real-time down-adjustable power and a real-time up-adjustable power; wherein, the first and the second end of the pipe are connected with each other,
the real-time down-tunable power is divided into a first section, a second section and a third section:
when u is 1 (k+1)＝x c (k)，u 2 (k + 1) =0 and u 3 When (k + 1) =0, the power is adjusted down by P 1 - To obtain the interval [0 1 - ]Is a first section;
when u is 1 (k+1)＝x c (k)，u 2 (k+1) =0 and u 3 (k+1)＝x l (k) When the power is adjusted downwards by P 2 - Obtaining a section (P) 1 - ,P 2 - ]Is a second section;
when u is 1 (k+1)＝0，u 2 (k+1)＝x c (k) And u is 3 (k+1)＝x l (k) Make the power down regulated by P 3 - To obtain a section
Is a third stage;
the real-time adjustable power is divided into a fourth section, a fifth section and a sixth section:
when u is 1 (k+1)＝0，u 2 (k + 1) =0 and u 3 (k+1)＝-x d (k) While making the power up-regulated by P 1 + To obtain the interval [0 1 + ]Is a fourth segment;
when u is 1 (k+1)＝-x l (k)，u 2 (k + 1) =0 and u 3 (k+1)＝-x d (k) While making the power up-regulated by P 2 + Obtaining a section (P) 1 + ,P 2 + ]Is a fifth section;
when u is 1 (k+1)＝-x l (k)，u 2 (k+1)＝-x d (k) And u is 3 When (k + 1) =0, power is adjusted up by P 3 + To obtain a section
Is the sixth stage.
Further, the multi-mode frequency modulation mode of the cluster electric vehicle comprises:
a first control mode, a second control mode, a third control mode, a fourth control mode, a fifth control mode and a sixth control mode;
when the cluster electric automobile is in the first section, the first control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)
s.t.0≤u 1 (k+i|k)≤x c (k),i＝0,1,2…m-1
in the formula, x 1 (k) The total state variable in the first control mode is the sum of three state variables of charging, idling and discharging in the mode, namely
Y 1 (k) The output power of the system in the first control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t Is the weight occupied by the error term; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control;
when the cluster electric automobile is in the second section, the second control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.0≤u 1 (k+i|k)≤x c (k),i＝0,1,2…m-1
0≤u 3 (k+i|k)≤x l (k),i＝0,1,2…m-1
in the formula, x 2 (k) Indicates the general form in the second control modeThe state variable is the sum of three state variables of charging, idling and discharging in the mode, namely
Y 2 (k) The output power of the system in the second control mode; y is p (k) An output sequence of the prediction model at the k moment; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term;
when the cluster electric automobile is in the third section, the third control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.u 1 (k+i|k),u 2 (k+i|k),u 3 (k+i|k)≥0
u 1 (k+i|k)+u 2 (k+i|k)≤x c (k)
u 3 (k+i|k)≤x l (k),i＝1,2,3…m
in the formula, x 3 (k) The total state variable in the third control mode is the sum of three state variables of charging, idling and discharging in the mode, i.e.
Y 3 (k) The output power of the system in the third control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t Is the weight occupied by the error term; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term;
when the cluster electric automobile is in the fourth section, the fourth control mode is adopted, and the corresponding state space expression, the corresponding control target and the corresponding constraint condition are respectively as follows:
minJ＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)
s.t.-x d (k)≤u 3 (k+i|k)≤0,i＝0,1,2…m-1
in the formula, x 4 (k) The total state variable in the fourth control mode is the sum of three state variables of charging, idling and discharging in the fourth control mode, namely
Y 4 (k) The output power of the system in the fourth control mode; y is p (k) An output sequence of the prediction model at the k moment; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t Is a mistakeThe difference accounts for the weight; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control;
when the cluster electric automobile is in the fifth section, the fifth control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.-x l (k)≤u 1 (k+i|k)≤0,i＝0,1,2…m-1
-x d (k)≤u 3 (k+i|k)≤0,i＝0,1,2…m-1
in the formula, x 5 (k) The total state variable in the fifth control mode is the sum of three state variables of charging, idling and discharging in the fifth control mode, i.e. the total state variable is
Y 5 (k) The output power of the system in the fifth control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t Is the weight occupied by the error term; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term;
when the cluster electric automobile is in the sixth section, the sixth control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.u 1 (k+i|k),u 2 (k+i|k),u 3 (k+i|k)≤0
u 2 (k+i|k)+u 3 (k+i|k)≥-x d (k)
u 1 (k+i|k)≥-x l (k),i＝1,2,3…m
in the formula, x 6 (k) The total state variable in the sixth control mode is the sum of the three state variables of charging, idling and discharging in the sixth control mode, namely
Y 6 (k) The output power of the system in the sixth control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state proportion and reducing the collective electric vehicle charge and discharge power, Q o Is Q o The weight coefficient of the (Cx (k + 1)) term.
The invention also provides a multi-mode frequency modulation device of the cluster electric automobile, which comprises the following components:
the control model building unit is used for building a multi-state switching control model of the cluster electric automobile based on a load dynamic transfer model of the electric automobile in a charging, discharging and idle state and a switching process between the charging, discharging and idle states;
the segmentation processing unit is used for calculating the real-time adjustable power of the cluster electric automobile according to the multi-state switching control model and carrying out segmentation processing on the range of the real-time adjustable power;
and the frequency modulation mode establishing unit is used for judging the section to which the AGC command belongs according to the result of the section processing, designing a corresponding control mode and establishing a multi-mode frequency modulation mode of the cluster electric automobile according to the control mode.
The present invention also provides a terminal device, including:
one or more processors;
a memory coupled to the processor for storing one or more programs;
when executed by the one or more processors, the one or more programs cause the one or more processors to implement the method for multi-mode frequency modulation for trunked electric vehicles as described in any of the above.
The invention also provides a computer readable storage medium, on which a computer program is stored, the computer program being executed by a processor to implement the multi-mode frequency modulation method for the cluster electric vehicle as described in any one of the above.
Compared with the prior art, the invention has the beneficial effects that:
the invention constructs a cluster electric vehicle multi-state switching control model by combining the switching process of three states based on the load dynamic transfer model of three states of charging, discharging and idling of the electric vehicle. The model considers the direct conversion between the charging state and the discharging state, and can widen the real-time adjustable power range of the cluster electric automobile. Based on the multi-state switching control model, the invention also carries out sectional processing on the real-time adjustable power, designs a corresponding control mode for each power section, and provides a multi-mode frequency modulation mode. The mode can effectively reduce the direct charge-discharge conversion times of the cluster electric automobile, avoid the offset of charge-discharge power and quickly and accurately track the AGC command of the power grid.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings required to be used in the embodiments will be briefly described below, and obviously, the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic flowchart of a multi-mode frequency modulation method for a cluster electric vehicle according to an embodiment of the present invention;
fig. 2 is a schematic diagram of probability transition between adjacent SOC intervals in a charging process according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a probability transition between adjacent SOC regions during a discharging process according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of state switching according to an embodiment of the present invention;
FIG. 5 is a graph comparing real-time adjustable power ranges provided by an embodiment of the present invention;
FIG. 6 is a graph illustrating a comparison of direct charge-discharge conversion times in different manners according to an embodiment of the present invention;
FIG. 7 is a graph comparing the number of electric vehicles charged and discharged in different manners according to an embodiment of the present invention;
fig. 8 is a diagram of the tracking result of AGC commands provided by an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a multi-mode frequency modulation device of a cluster electric vehicle according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be understood that the step numbers used herein are for convenience of description only and are not intended as limitations on the order in which the steps are performed.
It is to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The terms "comprises" and "comprising" indicate the presence of the described features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The term "and/or" refers to and includes any and all possible combinations of one or more of the associated listed items.
In a first aspect:
referring to fig. 1, an embodiment of the present invention provides a multi-mode frequency modulation method for a cluster electric vehicle, including:
s10, constructing a multi-state switching control model of the cluster electric automobile based on a load dynamic transfer model of the electric automobile in a charging, discharging and idle state and a switching process between the charging, discharging and idle states;
in this embodiment, first, a load dynamic transfer model in a charging, discharging, and idle state of the electric vehicle needs to be constructed, where SOC of the electric vehicle is discretized into N intervals, and x represents a load amount of the corresponding interval, and then the load dynamic transfer model in the charging state of the electric vehicle is:
x c (k+1)＝A c x c (k)+v c (k) (1)
in the formula, x c (k) Is an N-dimensional column vector and satisfies x c (k)＝[x c (k,1),…,x c (k,i),…,x c (k,N)] T Representing the load quantity in each SOC interval under the charging state at the moment k; v. of c (k) Is an N-dimensional column vector and satisfies v c (k)＝[v c (k,1),v c (k,2),…,v c (k,N)] T Indicating the load amount newly added or separated from the charging state in each interval at the moment k; a. The c State transition matrix for charged state:
in the formula (I), the compound is shown in the specification,
indicates SOC interval i during charging c And i c A transition probability between + 1;
the load amount representing the charge completion is a proportion of the total load amount in the section N.
It should be noted that, in two adjacent time intervals, the load in each interval has only two transfer paths: remain in the original state interval or move to the next adjacent interval. At this time, section i c To the interval i c +1 amount of load transferred
Represents; while remaining in the interval i c The load of (2) is
And (4) showing.
In one embodiment, the transition probability in equation (2)
The specific solving process of (2) is as follows:
assuming that the battery capacity of the electric automobile participating in the aggregation conforms to a certain probability density distribution f (C) P ) From the inherent charging characteristics of the battery, one can derive:
wherein, P ch Represents a charging power; eta ch Represents the charging efficiency; Δ t represents a time interval; s (k + 1) and S (k) respectively represent the charge states of the electric vehicle at the k +1 moment and the k moment;
the critical capacity is represented as the maximum allowable capacity that can achieve a transition from S (k) to S (k + 1) in one step. Thus, the relation between the random process SOC and the probability distribution of the battery capacity can be established:
in the formula, P r Representing the conditional probability.
As shown in FIG. 2, to facilitate accurate solution of transition probabilities
Section i c Equally divided into n cells. Firstly, calculating the inter-cell directional interval i c +1 transition probability
Further, the section i is obtained c To the interval i c +1 desired transition probability.
Wherein the content of the first and second substances,
represents a section i c Middle j c Inter-cell directional interval i c Probability of +1 transition; s. the down (i + 1) represents the i-th c A lower bound value of +1 intervals;S down (i, j) represents a section i c Middle j c SOC lower bound value of each small interval;
and
respectively represent the interval i c The minimum battery capacity C can be obtained by using the 0 probability critical value and the 1 probability critical value min And maximum battery capacity C max The results are obtained by the following formula (3). Probability of expected transition
Further, the discharge process is similar to the charging process in modeling, and the SOC is equally divided into N parts from low to high, so that the load dynamic transfer model in the discharge state of the electric vehicle is as follows:
x d (k+1)＝A d x d (k)+v d (k) (7)
in the formula, x d (k) The N-dimensional column vectors represent load quantities in each SOC interval in a discharge state at the moment k; v. of d (k) The N-dimensional column vector represents the load quantity newly added or separated from the discharge state in each interval at the moment k; a. The d State transition matrix for discharge state:
in the formula (I), the compound is shown in the specification,
indicates the SOC interval i in the discharging process d +1 and i d A transition probability therebetween;
the load amount indicating completion of discharge is a proportion of the total load amount in the section 1.
In contrast to the charging process, the discharging is a load shift from the high SOC to the low SOC, and therefore, the section i is referred to herein d +1 to the interval i d The transition probability of (2). At this time, the section i d +1 to the interval i d For the transferred load
Represents; while remaining in the interval i d The load of (2) is
And (4) showing.
In one embodiment, the transition probability in equation (8)
The specific solving process of (2) is as follows:
first, it is noted that the critical capacity during discharge
The following can be obtained from the inherent discharge characteristics of the battery:
in the formula, P dis Represents the discharge power; eta dis Indicating the discharge efficiency; Δ t represents a time interval; s (k + 1) and S (k) respectively represent the charge states of the electric vehicle at the k +1 moment and the k moment;
the critical capacity is represented as the maximum allowable capacity that can achieve a transition from S (k) to S (k + 1) in one step.
As shown in FIG. 3, the transition probabilities are solved for convenience and accuracy
Section i d +1 is equally divided into n sub-intervals. Firstly, calculating the inter-cell directional interval i d Transition probability p of d (i+1,j)→i Further, the section i is obtained d +1 to the interval i d Desired transition probability of (c):
wherein, the first and the second end of the pipe are connected with each other,
represents a section i d +1 th j d Inter-cell directional interval i d The probability of a transition; s up (i) Denotes the ith d Upper bound values of the intervals; s up (i +1, j) represents a section i d +1 th j d SOC upper bound value among the cells;
and
respectively represent the interval i d The 0 probability threshold and the 1 probability threshold in +1 may be set to the minimum battery capacity C min And maximum battery capacity C max Determination of the expected transition probability by means of the respective equations (9)
Further, the load dynamic transfer model in the idle state of the electric vehicle is as follows:
x l (k+1)＝A l x l (k)+v l (k) (12)
in the formula, x l (k) And the N-dimensional column vector represents the load quantity in each SOC interval in the idle state at the moment k. v. of l (k) Is an N-dimensional columnThe vector represents the load quantity newly added or separated from the idle state in each interval at the moment k; a. The l Is a state transition matrix in an idle state, because in the idle state, the electric vehicle has neither charging power nor discharging power, and the SOC of the electric vehicle does not change at all, the load amount in each SOC interval does not change, i.e. a l Is an N-dimensional unit array.
Further, in this embodiment, according to the established dynamic transition models of the three states of the electric vehicle, a dynamic switching model between the three states of the electric vehicle can be developed. In the dynamic transfer model of the three states, the SOC discretization methods are the same, namely the SOC interval N is equally divided and sorted according to the increasing sequence. The state switching process means that the load in the same SOC interval is reasonably switched among 3 states of charging, discharging and idling. As shown in FIG. 4, FIG. 4 shows a schematic diagram of the status switching process u 1 Representing the amount of load transfer between the charging and idle states; u. of 2 Indicating the amount of load transfer between the charged and discharged states; u. of 3 Indicating the load transfer amount between the idle state and the discharge state; the superscript "+" indicates power up; "-" indicates a power down.
Based on the above process analysis, in combination with the load dynamic transfer model of the charging, discharging and idle states that has been established in the foregoing, the expression of the multi-state switching control model of the clustered electric vehicle is as follows:
in the formula u 1 (k)、u 2 (k)、u 3 (k) Are all N-dimensional column vectors; a is a 3 Nx 3N-dimensional state transition matrix; b is a 3 Nx3N-dimensional state transition matrix used for describing a state switching process; c is an output row vector with 3N dimensions and is used for describing a power output part; v (k) = [ V = c (k),v l (k),v d (k)] T (ii) a Y (k) is the total output power.
It should be emphasized that, after considering the dynamic load transfer process of charging, discharging and idling, a in the multi-state switching control model is:
wherein A is c 、A l 、A d As already mentioned above, 0 then represents an N-dimensional zero matrix.
Because the state switching process should satisfy the quantity conservation of the electric automobiles, only the running state of the electric automobiles is changed, and the quantity of the electric automobiles is not changed, namely the total net load change quantity is zero during each state switching. For this reason, the control matrix B can be derived as:
wherein I represents an N-dimensional unit matrix.
Making the cluster electric vehicles all adopt maximum power charging and discharging, and obtaining C when the idle state power is 0:
in the formula (I), the compound is shown in the specification,
the method comprises the following steps that a cluster electric automobile is charged by adopting maximum power;
indicating that the cluster electric automobile adopts maximum power discharge.
S20, calculating the real-time adjustable power of the cluster electric automobile according to the multi-state switching control model, and performing sectional processing on the range of the real-time adjustable power;
in this embodiment, in the multi-state switching control model, the electric vehicles at time k are divided into 3 different groups, that is, a charging state group, a discharging state group and an idle state group, and the load number of each state group at time k +1 can be obtained by combining the control quantity at time k, so that the real-time adjustable power of the electric vehicles at the next time can be predicted.
In particular, the real-time adjustable power P - Can be divided into three sections, namely I-, II-and III-sections;
i-section: when u is 1 (k+1)＝x c (k)，u 2 (k+1)＝0，u 3 When (k + 1) =0, namely, the electric vehicles in the charging state group are completely stopped from charging and are converted into an idle state, the power can be reduced by P 1 - Then [0 1 - ]To down-regulate the power I-section.
II-section: when u is 1 (k+1)＝x c (k)，u 2 (k+1)＝0，u 3 (k+1)＝x l (k) When is at P 1 - On the basis of the power regulation, all the electric vehicles in the idle state group are converted into a discharge state, so that the power is regulated down by P 2 - Then (P) 1 - ,P 2 - ]In order to adjust the power II-section downward.
III-section: when u is 1 (k+1)＝0，u 2 (k+1)＝x c (k)，u 3 (k+1)＝x l (k) That is, all the electric vehicles in the charging state group and the idle state group are converted into the discharging state, so as to obtain the maximum down-regulated power P 3 - Then (P) 2 - ,P 3 - ]For down-regulation of power III-section, greater than P 3 - The frequency modulation instruction cluster electric automobile reduces the power P at the maximum 3 - A response is made.
Further, the power P can be adjusted down in real time + Can be divided into three sections, i.e. I + 、II + And III + A segment;
I + section (2): when u is 1 (k+1)＝0，u 2 (k+1)＝0，u 3 (k+1)＝-x d (k) When the electric vehicle in the discharging state group stops discharging and is converted into an idle state, the power can be adjusted up to P 1 + Then [0, P ] 1 + ]For up-regulating power I + And (4) section.
II + A section: when u is 1 (k+1)＝-x l (k)，u 2 (k+1)＝0，u 3 (k+1)＝-x d (k) When is at P 1 + On the basis, all the electric vehicles in the idle state group are converted into a charging state, so that the power is up-regulated by P 2 + Then (P) 1 + ,P 2 + ]For up-regulating power II + And (4) section.
III + Section (2): when u is 1 (k+1)＝-x l (k)，u 2 (k+1)＝-x d (k)，u 3 (k + 1) =0, that is, all the electric vehicles in the discharging state group and the idle state group are converted into the charging state, and the maximum up-regulated power P can be obtained 3 + Then (P) 2 + ,P 3 + ]For up-regulating power III + Segment greater than P 3 + The frequency modulation instruction cluster electric automobile adjusts the power P to the maximum 3 + A response is made.
And S30, judging the section to which the AGC command belongs according to the result of the section processing, designing a corresponding control mode, and constructing a multi-mode frequency modulation mode of the cluster electric automobile according to the control mode.
In this embodiment, a multi-mode control method based on model predictive control is proposed by determining an applicable control mode of the frequency modulation command in combination with the foregoing real-time adjustable power segmentation method. The 6 power segments correspond to 6 control modes in total, and the state space expression, the control target and the constraint condition corresponding to each control mode are as follows:
control mode 1: the method is suitable for real-time power down-regulation I-section, and the corresponding state space expression, control target and constraint condition are respectively as follows:
minJ＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)
s.t.0≤u 1 (k+i|k)≤x c (k),i＝0,1,2…m-1 (18)
in the formula, x 1 (k) The total state variable in the first control mode is the sum of three state variables of charging, idling and discharging in the mode, namely
Y 1 (k) The output power of the system in the first control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control;
control mode 2: the method is suitable for real-time power down-regulation II-section, and the corresponding state space expression, control target and constraint condition are respectively as follows:
minJ＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.0≤u 1 (k+i|k)≤x c (k),i＝0,1,2…m-1
0≤u 3 (k+i|k)≤x l (k),i＝0,1,2…m-1 (20)
in the formula, x 2 (k) The total state variable in the second control mode is the sum of three state variables of charging, idling and discharging in the mode, namely
Y 2 (k) The output power of the system in the second control mode; y is p (k) An output sequence of the prediction model at the k moment; p is t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term;
control mode 3: the method is suitable for real-time power down regulation III-section, and the corresponding state space expression, control target and constraint condition are respectively as follows:
minJ＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.u 1 (k+i|k),u 2 (k+i|k),u 3 (k+i|k)≥0
u 1 (k+i|k)+u 2 (k+i|k)≤x c (k)
u 3 (k+i|k)≤x l (k),i＝1,2,3…m (22)
in the formula, x 3 (k) The total state variable in the third control mode is the sum of the three state variables of charging, idling and discharging in the mode, i.e. the total state variable
Y 3 (k) The output power of the system in the third control mode; y is p (k) An output sequence of the prediction model at the k moment; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Error terms for the output predictor and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is for reducing electric vehiclesControl constraints introduced by the number of state switches, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state proportion and reducing the collective electric vehicle charge and discharge power, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term; it is emphasized that Q is introduced here o The term (Cx (k + 1)) has a function of reducing the number of times of direct switching of the charge/discharge state in addition to preventing the charge/discharge power from being cancelled out.
Control mode 4: adapted to up-regulate power I in real time + And the corresponding state space expression, control target and constraint condition are respectively:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)
s.t.-x d (k)≤u 3 (k+i|k)≤0,i＝0,1,2…m-1 (24)
in the formula, x 4 (k) The total state variable in the fourth control mode is the sum of three state variables of charging, idling and discharging in the fourth control mode, i.e. the total state variable is
Y 4 (k) The output power of the system in the fourth control mode; y is p (k) An output sequence of the prediction model at the k moment; p is t (k) A sequence formed by frequency modulation power commands at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t Is the weight occupied by the error term; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control;
control mode 5: adaptation toUp-regulation of power II in real time + And the corresponding state space expression, control target and constraint condition are respectively:
minJ＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.-x l (k)≤u 1 (k+i|k)≤0,i＝0,1,2…m-1
-x d (k)≤u 3 (k+i|k)≤0,i＝0,1,2…m-1 (26)
in the formula, x 5 (k) The total state variable in the fifth control mode is the sum of the three state variables of charging, idling and discharging in the fifth control mode, i.e. the total state variable is
Y 5 (k) The output power of the system in the fifth control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power commands at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t Is the weight occupied by the error term; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term;
control mode 6: adapted to up-regulate power III in real time + And the corresponding state space expression, control target and constraint condition are respectively:
minJ＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.u 1 (k+i|k),u 2 (k+i|k),u 3 (k+i|k)≤0
u 2 (k+i|k)+u 3 (k+i|k)≥-x d (k)
u 1 (k+i|k)≥-x l (k),i＝1,2,3…m (28)
in the formula, x 6 (k) The total state variable in the sixth control mode is the sum of the three state variables of charging, idling and discharging in the sixth control mode, namely
Y 6 (k) The output power of the system in the sixth control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Error terms for the output predictor and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o The weight coefficient of the (Cx (k + 1)) term.
The embodiment of the invention considers the direct conversion between the charging state and the discharging state, and can widen the real-time adjustable power range of the cluster electric automobile. Meanwhile, based on a multi-state switching control model, the real-time adjustable power is processed in a segmented mode, a corresponding control mode is designed for each power segment, a multi-mode frequency modulation mode is provided, the direct charge-discharge conversion times of the cluster electric vehicle can be effectively reduced, the offset of charge-discharge power is avoided, and the AGC command of the power grid is quickly and accurately tracked.
In order to further understand the invention and verify the effectiveness of the multimode frequency modulation method based on the multi-state switching control model, in a certain embodiment, an actual AGC instruction in a certain region 18. The simulation step length is 0.1s, and the cluster parameters of 1000 electric vehicles participating in aggregation are shown in table 1.
TABLE 1 electric vehicle Cluster parameter settings
To illustrate that the multi-state switching control model provided by the present invention has a wider real-time adjustable power range, simulation comparison is performed on the multi-state switching control model with the switching model without considering direct charge-discharge state conversion, and the result is shown in fig. 5. Therefore, in the face of the same regulation and control instruction, after direct charge-discharge conversion is considered in the regulation and control process, the same number of electric vehicles is switched, the power change value of the multi-state switching control model is larger, and a wider real-time adjustable power range is obtained really.
In one embodiment, to analyze the advantages of the multimode frequency modulation method, the following two ways are compared and simulated: mode 1 multi-mode frequency modulation control mode; mode 2 does not distinguish between AGC commands and real-time adjustable power, and a single control mode is adopted, i.e. it is considered that 3 states can be switched at each time. As shown in fig. 6, fig. 6 shows the comparison result of the number of times of direct charge and discharge switching of the battery in 2 modes. It can be seen that the frequency of the direct switching in mode 2 is significantly higher than that in mode 1, and under the multi-mode regulation of mode 1, the direct switching occurs only in mode 3 and mode 6, limited by the constraint condition. It was calculated that the same AGC command was switched 926 times in the mode 1 and the number of times of switching in the mode 2 was as high as 2925 times. Therefore, the multi-mode frequency modulation method provided by the invention can effectively reduce the direct switching times of battery charging and discharging and reduce the loss of electric energy.
Referring to fig. 7, in one embodiment, the results of comparing the sum of the number of electric vehicles in the charged and discharged states in 2 ways are shown. It can be seen that the sum of the number of electric vehicles in the charging and discharging states in the mode 2 is obviously greater than that in the mode 1, and since the two track the same AGC instruction in the time interval, the sum of the power of the electric vehicles in the mode 2 which is greater than that in the mode 1 is inevitably zero at each moment, that is, the charging and discharging power in the mode 2 is cancelled out, and the mode 1 effectively avoids the phenomenon, so that the more efficient dispatching of the clustered electric vehicles is realized.
Referring to fig. 8, in one embodiment, the tracking result of the AGC command in the multi-mode control mode is shown. In the period of 18. This is because the AGC command exceeds the real-time adjustable power limit of the clustered electric vehicle during the circled period, resulting in a failure of the clustered electric vehicle to fully respond.
In a second aspect:
referring to fig. 9, in an embodiment, there is further provided a multi-mode frequency modulation apparatus for a cluster electric vehicle, including:
the control model building unit 01 is used for building a multi-state switching control model of the cluster electric vehicle based on a load dynamic transfer model of the electric vehicle in a charging, discharging and idle state and a switching process between the charging, discharging and idle states;
the segmentation processing unit 02 is used for calculating the real-time adjustable power of the cluster electric automobile according to the multi-state switching control model and performing segmentation processing on the range of the real-time adjustable power;
and the frequency modulation mode establishing unit 03 is configured to judge the segment to which the AGC instruction belongs according to the result of the segment processing, design a corresponding control mode, and establish a multi-mode frequency modulation mode of the cluster electric vehicle according to the control mode.
It can be understood that the functional modules 01 to 03 of the device are respectively used for executing the steps S10 to S30, and direct conversion between the charging state and the discharging state is considered when the steps are executed, so that the real-time adjustable power range of the cluster electric vehicle is widened. Meanwhile, the direct charge-discharge conversion times of the cluster electric automobile can be effectively reduced, the offset of charge-discharge power is avoided, and the AGC instruction of the power grid is quickly and accurately tracked.
In a third aspect:
an embodiment of the present invention further provides a terminal device, including: one or more processors; a memory coupled to the processor for storing one or more programs; when executed by the one or more processors, the one or more programs cause the one or more processors to implement the method for multi-mode frequency modulation for a cluster electric vehicle as described above.
The processor is used for controlling the overall operation of the terminal device so as to complete all or part of the steps of the multi-mode frequency modulation method of the cluster electric automobile. The memory is used to store various types of data to support operation at the terminal device, and these data may include, for example, instructions for any application or method operating on the terminal device, as well as application-related data. The Memory may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically Erasable Programmable Read-Only Memory (EEPROM), erasable Programmable Read-Only Memory (EPROM), programmable Read-Only Memory (PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk.
The terminal Device may be implemented by one or more Application Specific 1 integrated circuits (AS 1C), digital Signal Processors (DSP), digital Signal Processing Devices (DSPD), programmable Logic Devices (PLD), field Programmable Gate Arrays (FPGA), controllers, microcontrollers, microprocessors, or other electronic components, and is configured to perform the method for multi-mode frequency modulation of the clustered electric vehicles according to any one of the embodiments described above, and achieve the technical effects consistent with the above methods.
An embodiment of the present invention further provides a computer readable storage medium including program instructions, which when executed by a processor, implement the steps of the method for multi-mode frequency modulation of the cluster electric vehicle according to any one of the above embodiments. For example, the computer readable storage medium may be the above memory including program instructions, which are executable by the processor of the terminal device to perform the method for multi-mode frequency modulation of the cluster electric vehicle according to any one of the above embodiments, and achieve the technical effects consistent with the above method.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (6)
1. A multi-mode frequency modulation method of a cluster electric automobile is characterized by comprising the following steps:
constructing a multi-state switching control model of the cluster electric automobile based on a load dynamic transfer model of the electric automobile in a charging, discharging and idle state and a switching process between the charging, discharging and idle states; wherein, the first and the second end of the pipe are connected with each other,
the load dynamic transfer model in the charging state of the electric automobile is as follows:
x c (k+1)＝A c x c (k)+v c (k)；
in the formula, x c (k) The vector is an N-dimensional column vector and represents the load quantity in each SOC interval under the charging state at the moment k; v. of c (k) The N-dimensional column vector represents the load amount newly added or separated from the charging state in each interval at the moment k; a. The c A state transition matrix in a charging state;
wherein x is c (k)，v c (k) And A c Respectively satisfy:
x c (k)＝[x c (k,1),…,x c (k,i),…,x c (k,N)] T ；
v c (k)＝[v c (k,1),v c (k,2),…,v c (k,N)] T ；
in the formula, N represents the SOC interval division number; x is the number of c (k, i) represents the load amount at the k-th interval; v. of c (k, N) represents a load amount newly added or removed from the charged state at the nth section k;
indicates SOC interval i during charging c And i c A transition probability between + 1;
the proportion of the load quantity of the charging completion to the total load quantity of the interval N is represented;
the load dynamic transfer model in the electric automobile discharging state is as follows:
x d (k+1)＝A d x d (k)+v d (k)；
in the formula, x d (k) The N-dimensional column vectors represent load quantities in each SOC interval in a discharge state at the moment k; v. of d (k) The N-dimensional column vector represents the load quantity newly added or separated from the discharge state in each interval at the moment k; a. The d State transition matrix for discharge state:
in the formula (I), the compound is shown in the specification,
indicates the SOC interval i in the discharging process d +1 and i d Transition probabilities between;
to representThe load quantity after the discharge accounts for the proportion of the total load quantity of the interval 1;
the load dynamic transfer model of the electric automobile in the idle state is as follows:
x l (k+1)＝A l x l (k)+v l (k)；
in the formula, x l (k) The N-dimensional column vector represents the load quantity in each SOC interval under the idle state at the moment k; v. of l (k) The N-dimensional column vector represents the load quantity newly added or separated from the idle state in each interval at the moment k; a. The l The matrix is an N-dimensional unit matrix and represents a state transition matrix in an idle state;
the cluster electric automobile multi-state switching control model is as follows:
in the formula u 1 (k)、u 2 (k)、u 3 (k) Are all N-dimensional column vectors; u. u 1 (k) Representing the load quantity of mutual switching between the charging state and the idle state; u. of 2 (k) A load amount indicating switching between a charge state and a discharge state; u. of 3 (k) The load quantity representing the mutual switching between the discharge state and the idle state; a is a 3 Nx 3N-dimensional state transition matrix; b is a 3 Nx 3N-dimensional state transition matrix used for describing a state switching process; c is an output row vector with 3N dimensions and is used for describing a power output part; v (k) = [ V = c (k),v l (k),v d (k)] T (ii) a Y (k) is the total output power;
wherein A, B and C are respectively:
in the formula, 0 represents an N-dimensional zero matrix; i represents an N-dimensional unit array;
representing that the cluster electric vehicle adopts maximum power charging;
indicating that the cluster electric automobile adopts maximum power discharge;
calculating real-time adjustable power of the cluster electric automobile according to the multi-state switching control model, and performing segmented processing on the range of the real-time adjustable power;
and judging the section to which the AGC command belongs according to the result of the section processing, designing a corresponding control mode, and constructing a multi-mode frequency modulation mode of the cluster electric automobile according to the control mode.
2. The method of claim 1, wherein the step of segmenting the range of the real-time adjustable power comprises dividing the real-time adjustable power into a real-time down-adjustable power and a real-time up-adjustable power; wherein, the first and the second end of the pipe are connected with each other,
the real-time down-adjustable power is divided into a first section, a second section and a third section:
when u is 1 (k+1)＝x c (k)，u 2 (k + 1) =0 and u 3 (k + 1) =0, the power is adjusted downward by P 1 - To obtain a section
Is a first section;
when u is 1 (k+1)＝x c (k)，u 2 (k + 1) =0 and u 3 (k+1)＝x l (k) While making the power down
Obtaining the interval
Is a second section;
when u is 1 (k+1)＝0，u 2 (k+1)＝x c (k) And u is 3 (k+1)＝x l (k) Make the power down regulated by P 3 - Obtaining a section
Is a third stage;
the real-time adjustable power is divided into a fourth section, a fifth section and a sixth section:
when u is 1 (k+1)＝0，u 2 (k + 1) =0 and u 3 (k+1)＝-x d (k) Time-of-day, power up-regulation
Get the interval
Is a fourth segment;
when u is 1 (k+1)＝-x l (k)，u 2 (k + 1) =0 and u 3 (k+1)＝-x d (k) Time-of-day, power up-regulation
Get the interval
Is a fifth section;
when u is 1 (k+1)＝-x l (k)，u 2 (k+1)＝-x d (k) And u is 3 When (k + 1) =0, the power is adjusted up
Obtaining the interval
Is the sixth stage.
3. The method for multi-mode frequency modulation of the clustered electric vehicles according to claim 2, wherein the multi-mode frequency modulation mode of the clustered electric vehicles comprises:
a first control mode, a second control mode, a third control mode, a fourth control mode, a fifth control mode and a sixth control mode;
when the cluster electric automobile is in the first section, the first control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)
s.t.0≤u 1 (k+i|k)≤x c (k),i＝0,1,2…m-1；
in the formula, x 1 (k) The total state variable in the first control mode is the sum of three state variables of charging, idling and discharging in the first control mode, i.e. the total state variable is
Y 1 (k) The output power of the system in the first control mode; y is p (k) An output sequence of the prediction model at the time k; p is t (k) A sequence formed by frequency modulation power commands at the time k; | | Y p (k)-P t (k)|| 2 Error terms for the output predictor and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; u (k) represents a control prediction vector at time k; u. of 1 (k + i | k) represents the steering vector for the (k + i) th step, and its physical meaning represents k + iAt this moment, the forecast of the load switching between the charging state and the idle state; m represents a control step in model predictive control;
when the cluster electric automobile is in the second section, the second control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.0≤u 1 (k+i|k)≤x c (k),i＝0,1,2…m-1
0≤u 3 (k+i|k)≤x l (k),i＝0,1,2…m-1；
in the formula, x 2 (k) The total state variable in the second control mode is the sum of three state variables of charging, idling and discharging in the mode, i.e. the total state variable is
Y 2 (k) The output power of the system in the second control mode; y is p (k) An output sequence of the prediction model at the k moment; p t (k) A sequence formed by frequency modulation power commands at the time k; | | Y p (k)-P t (k)|| 2 Error terms for the output predictor and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term; cx (k + 1) represents the predicted output at the k +1 th time;
when the cluster electric automobile is in the third section, the third control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.u 1 (k+i|k),u 2 (k+i|k),u 3 (k+i|k)≥0
u 1 (k+i|k)+u 2 (k+i|k)≤x c (k)
u 3 (k+i|k)≤x l (k),i＝1,2,3…m；
in the formula, x 3 (k) The total state variable in the third control mode is the sum of the three state variables of charging, idling and discharging in the mode, i.e. the total state variable
Y 3 (k) The output power of the system in the third control mode; y is p (k) An output sequence of the prediction model at the time k;
u 2 (k + i | k) represents a control prediction vector of the (k + i) th step, and the physical meaning of the control prediction vector represents the prediction quantity of load switching between the charging state and the discharging state at the moment of k + i; u. u 3 (k + i | k) represents a control prediction vector of the (k + i) th step, and the physical meaning of the control prediction vector represents the prediction quantity of load switching between the discharge state and the idle state at the moment of k + i; p t (k) A sequence formed by frequency modulation power commands at the time k;
||Y p (k)-P t (k)|| 2 error terms for the output predictor and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term;
when the cluster electric automobile is in the fourth section, the fourth control mode is adopted, and the corresponding state space expression, the control target and the constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)
s.t.-x d (k)≤u 3 (k+i|k)≤0,i＝0,1,2…m-1；
in the formula, x 4 (k) The total state variable in the fourth control mode is the sum of three state variables of charging, idling and discharging in the fourth control mode, i.e. the total state variable is
Y 4 (k) The output power of the system in the fourth control mode; y is p (k) An output sequence of the prediction model at the time k; p t (k) A sequence formed by frequency modulation power commands at the time k; | | Y p (k)-P t (k)|| 2 Error terms for the output predictor and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control;
when the cluster electric automobile is in the fifth section, the fifth control mode is adopted, and the corresponding state space expression, the corresponding control target and the corresponding constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.-x l (k)≤u 1 (k+i|k)≤0,i＝0,1,2…m-1
-x d (k)≤u 3 (k+i|k)≤0,i＝0,1,2…m-1；
in the formula, x 5 (k) The total state variable in the fifth control mode is the sum of three state variables of charging, idling and discharging in the fifth control mode, i.e. the total state variable is
Y 5 (k) The output power of the system in the fifth control mode; y is p (k) An output sequence of the prediction model at the k moment; p t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Is an error term of the output quantity predicted value and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state ratio and reducing the charge and discharge power of the electric automobile in the cluster, Q o Is Q o A weight coefficient of the (Cx (k + 1)) term;
when the cluster electric automobile is in the sixth section, the sixth control mode is adopted, and the corresponding state space expression, the corresponding control target and the corresponding constraint condition are respectively as follows:
min J＝Q t ||Y p (k)-P t (k)|| 2 +U(k) T Q u U(k)+Q o (Cx(k+1)) 2
s.t.u 1 (k+i|k),u 2 (k+i|k),u 3 (k+i|k)≤0
u 2 (k+i|k)+u 3 (k+i|k)≥-x d (k)
u 1 (k+i|k)≥-x l (k),i＝1,2,3…m；
in the formula, x 6 (k) The total state variable in the sixth control mode is the sum of the three state variables of charging, idling and discharging in the sixth control mode, namely
Y 6 (k) The output power of the system in the sixth control mode; y is p (k) An output sequence of the prediction model at the k moment; p is t (k) A sequence formed by frequency modulation power instructions at the time k; | | Y p (k)-P t (k)|| 2 Error terms for the output predictor and the reference signal, and Q t The weight of the error term is taken; u (k) T Q u U (k) is a control constraint term introduced to reduce the number of state switching times of the electric vehicle, Q u Is U (k) T Q u A weight matrix of U (k) terms; m represents a control step in model predictive control; q o (Cx (k + 1)) is a mutually offsetting term for increasing the idle state proportion and reducing the collective electric vehicle charge and discharge power, Q o Is Q o The weight coefficient of the (Cx (k + 1)) term.
4. A multi-mode frequency modulation device of a cluster electric automobile is characterized by comprising:
the control model building unit is used for building a multi-state switching control model of the cluster electric automobile based on a load dynamic transfer model of the electric automobile in a charging, discharging and idle state and a switching process between the charging, discharging and idle states; wherein, the first and the second end of the pipe are connected with each other,
the load dynamic transfer model in the charging state of the electric automobile is as follows:
x c (k+1)＝A c x c (k)+v c (k)；
in the formula, x c (k) Is an N-dimensional columnA vector representing the load amount in each SOC interval in the charging state at the time k; v. of c (k) The N-dimensional column vector represents the load quantity newly added or separated from the charging state in each interval at the moment k; a. The c A state transition matrix in a charging state;
wherein x is c (k)，v c (k) And A c Respectively satisfy:
x c (k)＝[x c (k,1),…,x c (k,i),…,x c (k,N)] T ；
v c (k)＝[v c (k,1),v c (k,2),…,v c (k,N)] T ；
in the formula, N represents the SOC interval division number; x is the number of c (k, i) represents the load amount at the k time of the ith interval; v. of c (k, N) represents the amount of load newly added or removed from the charged state at the time of the nth section k;
indicates SOC interval i in the charging process c And i c A transition probability between + 1;
the proportion of the load quantity of the charging completion to the total load quantity of the interval N is represented;
the load dynamic transfer model in the electric automobile discharging state is as follows:
x d (k+1)＝A d x d (k)+v d (k)；
in the formula, x d (k) The N-dimensional column vector represents the load quantity in each SOC interval in the discharge state at the moment k; v. of d (k) The N-dimensional column vector represents the load quantity newly added or separated from the discharge state in each interval at the moment k; a. The d State transition matrix for discharge state:
in the formula (I), the compound is shown in the specification,
indicates the SOC interval i in the discharging process d +1 and i d Transition probabilities between;
the load amount representing the completion of discharge accounts for the proportion of the total load amount in the interval 1;
the load dynamic transfer model in the idle state of the electric automobile is as follows:
x l (k+1)＝A l x l (k)+v l (k)；
in the formula, x l (k) The N-dimensional column vector represents the load quantity in each SOC interval under the idle state at the moment k; v. of l (k) The N-dimensional column vector represents the load quantity newly added or separated from the idle state in each interval at the moment k; a. The l The matrix is an N-dimensional unit matrix and represents a state transition matrix in an idle state;
the cluster electric automobile multi-state switching control model is as follows:
in the formula u 1 (k)、u 2 (k)、u 3 (k) Are all N-dimensional column vectors; u. of 1 (k) Representing the load quantity of mutual switching between the charging state and the idle state; u. u 2 (k) A load amount indicating switching between a charge state and a discharge state; u. u 3 (k) Representing the load quantity of mutual switching between the discharge state and the idle state; a is a 3 Nx 3N-dimensional state transition matrix; b is a 3 Nx 3N-dimensional state transition matrix used for describing a state switching process; c is an output row vector with 3N dimensions and is used for describing a power output part; v (k) = [ V = c (k),v l (k),v d (k)] T (ii) a Y (k) is the total output power;
wherein A, B and C are respectively:
in the formula, 0 represents an N-dimensional zero matrix; i represents an N-dimensional unit array;
representing that the cluster electric vehicle adopts maximum power charging;
indicating that the cluster electric automobile adopts maximum power discharge;
the segmentation processing unit is used for calculating the real-time adjustable power of the cluster electric automobile according to the multi-state switching control model and carrying out segmentation processing on the range of the real-time adjustable power;
and the frequency modulation mode establishing unit is used for judging the section to which the AGC command belongs according to the result of the section processing, designing a corresponding control mode and establishing a multi-mode frequency modulation mode of the cluster electric automobile according to the control mode.
5. A terminal device, comprising:
one or more processors;
a memory coupled to the processor for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of multi-mode frequency modulation for a cluster electric vehicle of any of claims 1-3.
6. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program is executed by a processor to implement the method of multi-mode frequency modulation of a trunked electric vehicle according to any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110419399.7A CN112909979B (en) | 2021-04-19 | 2021-04-19 | Multi-mode frequency modulation method, device, equipment and medium for cluster electric automobile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110419399.7A CN112909979B (en) | 2021-04-19 | 2021-04-19 | Multi-mode frequency modulation method, device, equipment and medium for cluster electric automobile |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112909979A CN112909979A (en) | 2021-06-04 |
| CN112909979B true CN112909979B (en) | 2022-11-11 |
Family
ID=76108760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110419399.7A Active CN112909979B (en) | 2021-04-19 | 2021-04-19 | Multi-mode frequency modulation method, device, equipment and medium for cluster electric automobile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112909979B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113829934B (en) * | 2021-10-22 | 2023-03-21 | 华北电力大学 | Electric vehicle cluster aggregation response capacity determining method and scheduling method |
| CN115001054B (en) * | 2022-07-29 | 2022-11-01 | 东南大学溧阳研究院 | Model-based power system frequency control strategy for predicting electric vehicle |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108306311A (en) * | 2018-02-09 | 2018-07-20 | 南京工程学院 | The control system and method for DC load system by stages responsive electricity grid frequency modulation demand |
| CN110048406A (en) * | 2019-04-12 | 2019-07-23 | 东北大学 | A kind of control method for dividing group to participate in power grid frequency modulation based on extensive electric car |
| CN110466384A (en) * | 2019-07-05 | 2019-11-19 | 武汉新能源汽车工业技术研究院有限公司 | A kind of charging module group power distribution method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010125868A (en) * | 2008-11-25 | 2010-06-10 | Denso Corp | Charge and discharge planning device |
| CN107612048B (en) * | 2017-10-23 | 2021-03-16 | 重庆大学 | Electric automobile frequency modulation control strategy based on model prediction |
-
2021
- 2021-04-19 CN CN202110419399.7A patent/CN112909979B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108306311A (en) * | 2018-02-09 | 2018-07-20 | 南京工程学院 | The control system and method for DC load system by stages responsive electricity grid frequency modulation demand |
| CN110048406A (en) * | 2019-04-12 | 2019-07-23 | 东北大学 | A kind of control method for dividing group to participate in power grid frequency modulation based on extensive electric car |
| CN110466384A (en) * | 2019-07-05 | 2019-11-19 | 武汉新能源汽车工业技术研究院有限公司 | A kind of charging module group power distribution method and device |
Non-Patent Citations (1)
| Title |
|---|
| 参与电网频率调整的电动汽车集群控制研究;朱贤文;《中国优秀硕士学位论文全文数据库》;20170615(第6期);第1-55页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112909979A (en) | 2021-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112909979B (en) | Multi-mode frequency modulation method, device, equipment and medium for cluster electric automobile | |
| CN108964102B (en) | Optimal configuration method for position and capacity of distributed energy storage in power distribution network | |
| CN110703101B (en) | Lithium ion battery inter-partition cycle capacity decline prediction method | |
| Chun et al. | Adaptive exploration harmony search for effective parameter estimation in an electrochemical lithium-ion battery model | |
| CN115291116B (en) | Energy storage battery health state prediction method and device and intelligent terminal | |
| CN113541168A (en) | Electric vehicle cluster controllability determining method, scheduling method and system | |
| CN113829934B (en) | Electric vehicle cluster aggregation response capacity determining method and scheduling method | |
| US20240014651A1 (en) | Probability estimation method for photovoltaic power based on optimized copula function and photovoltaic power system | |
| CN114725969B (en) | Electric automobile load aggregation method based on continuous tracking of wind power curve | |
| CN114204547B (en) | Power distribution network multi-measure combination loss reduction optimization method considering source network load storage cooperative optimization | |
| CN113011101B (en) | Control method and system for energy storage to participate in frequency modulation auxiliary service optimization | |
| CN111781529B (en) | Battery pack monomer capacity estimation method and device based on cloud data of electric automobile | |
| US20230148201A1 (en) | Method and system for supplying power to device, and related device | |
| CN117371755A (en) | Multi-microgrid comprehensive energy system distributed optimization method, device, equipment and medium | |
| CN117172486A (en) | Reinforced learning-based virtual power plant optical storage resource aggregation regulation and control method | |
| CN105207220B (en) | A kind of tapping voltage regulation and control method based on progressive learning | |
| CN114757548A (en) | Wind power energy storage equipment adjusting performance evaluation method adopting scene construction | |
| CN114239350A (en) | Water cooling plate optimization design method and device based on multi-objective optimization and storage medium | |
| CN115360768A (en) | Power scheduling method and device based on muzero and deep reinforcement learning and storage medium | |
| CN104408531B (en) | A kind of uniform dynamic programming method of multidimensional multistage complicated decision-making problems | |
| CN115018379B (en) | Electric vehicle in-day response capability assessment method and system and computer storage medium | |
| Romero et al. | Fast charge of Li-ion batteries using a two-layer distributed MPC with electro-chemical and thermal constraints | |
| CN111446719B (en) | Distribution network voltage stability control method and system | |
| Zhang et al. | Estimation of State of Charge and State of Health of Marine Lithium Battery based on BP Neural Network | |
| CN117698505A (en) | Electric automobile group aggregation charge and discharge control method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |