CN117374975A - Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming - Google Patents
Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming Download PDFInfo
- Publication number
- CN117374975A CN117374975A CN202311658252.9A CN202311658252A CN117374975A CN 117374975 A CN117374975 A CN 117374975A CN 202311658252 A CN202311658252 A CN 202311658252A CN 117374975 A CN117374975 A CN 117374975A
- Authority
- CN
- China
- Prior art keywords
- period
- power
- electric automobile
- voltage
- electric
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 98
- 238000009826 distribution Methods 0.000 title claims abstract description 83
- 238000005457 optimization Methods 0.000 claims abstract description 56
- 230000008569 process Effects 0.000 claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000012886 linear function Methods 0.000 claims abstract description 16
- 238000005070 sampling Methods 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 53
- 238000007599 discharging Methods 0.000 claims description 22
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000004069 differentiation Effects 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 3
- 238000002347 injection Methods 0.000 claims description 3
- 239000007924 injection Substances 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 230000001419 dependent effect Effects 0.000 claims description 2
- 239000000243 solution Substances 0.000 description 9
- 230000000694 effects Effects 0.000 description 6
- 230000001105 regulatory effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004880 explosion Methods 0.000 description 3
- 238000012614 Monte-Carlo sampling Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000004870 electrical engineering Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L53/00—Methods of charging batteries, specially adapted for electric vehicles; Charging stations or on-board charging equipment therefor; Exchange of energy storage elements in electric vehicles
- B60L53/60—Monitoring or controlling charging stations
- B60L53/63—Monitoring or controlling charging stations in response to network capacity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- H02J3/007—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources
- H02J3/0075—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources for providing alternative feeding paths between load and source according to economic or energy efficiency considerations, e.g. economic dispatch
-
- 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
- H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
- H02J3/16—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load by adjustment of reactive power
-
- 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
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Mechanical Engineering (AREA)
- Transportation (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Quality & Reliability (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A real-time cooperative voltage regulation method of a power distribution network based on approximate dynamic programming uses the minimum voltage fluctuation as an objective function, and establishes a power distribution network voltage cooperative optimization model according to electric automobile cluster scheduling characteristics; reconstructing the power distribution network voltage collaborative optimization model into a Markov decision process, and converting the power distribution network voltage collaborative optimization model into a real-time voltage regulation model which is solved time by time; performing value function approximation on a Belman equation in a Markov decision process by adopting a piecewise linear function, wherein the slope of the piecewise linear function can be obtained after a group of offline training scenes generated by predictive data sampling are trained; and solving the Belman equation after the approximation of the value function time-period to obtain an approximate global optimal decision. The invention improves the voltage stability of the power distribution network and ensures the optimization accuracy in a random environment; on the premise of meeting the operation limit of the power distribution network and the electric automobile, the method can cope with uncertainty of output and load of renewable energy sources and give out the optimal dispatching result of the power distribution network voltage regulation under random environment.
Description
Technical Field
The invention belongs to the field of electrical engineering, and particularly relates to a real-time collaborative voltage regulating method for a power distribution network based on approximate dynamic programming.
Background
As the permeability of renewable energy increases, the distribution network begins to shift towards an active and active form. However, on one hand, the grid connection of high-proportion renewable energy sources is easy to cause the problems of voltage out-of-limit, line overload, network loss increase and the like, so that the capacity of the distribution network for receiving the distributed power sources is reduced, and on the other hand, the randomness of the distribution network causes uncertainty on system power flow and voltage distribution. Therefore, voltage management has important significance for safe and stable operation of the power distribution network in a random environment.
The traditional voltage regulating means comprise a load voltage regulator, a switching parallel capacitor bank and the like. However, the conventional voltage regulation method has slow regulation speed, frequent actions affect the service life of the voltage regulator, and the discrete type regulation method still has difficulty in coping with the fluctuation of the distributed power supply. Future power grids will move towards a variety of "source-grid-load" linked interactive modes of operation, as there is some flexibility in the power supply, grid and load. Under the development trend, the electric automobile can be regarded as controllable load resources as an important component and participates in centralized scheduling of the power distribution network. The interaction concept of the electric automobile and the power grid provides a new adjusting means for voltage control of the power distribution network.
In the existing research of considering the traditional voltage regulation mode and the electric automobile cooperative voltage regulation, the traditional voltage regulation mode and the electric automobile cooperative voltage regulation are all deterministic scenes, and the influence of the output of renewable energy sources and the randomness of loads on the voltage optimization result is not considered. The result that the predicted value and the actual value of the new energy output are not consistent with each other can cause that the original scheduling strategy can not optimize the scheduling result. On the premise of meeting the operation limit of a power distribution network and an electric automobile, how to cope with uncertainty of output and load of renewable energy sources and provide an optimal dispatching result of power distribution network voltage regulation under random environments is a problem to be solved urgently at present.
Disclosure of Invention
Aiming at the problems of power distribution network voltage stability, new energy and load uncertainty, the invention provides a power distribution network real-time cooperation voltage regulating method based on approximate dynamic programming aiming at the defects and improvement demands of the prior art by using a real-time cooperation voltage regulating method of an on-load voltage regulator and an electric automobile, so as to cope with the uncertainty of renewable energy output and load and provide the optimal power distribution network voltage regulating dispatching result in a random environment.
The technical scheme of the invention comprises the following steps:
firstly, taking the minimum voltage fluctuation as an objective function, and establishing a power distribution network voltage collaborative optimization model according to electric vehicle cluster scheduling characteristics;
reconstructing a power distribution network voltage collaborative optimization model into a Markov decision process, and converting the power distribution network voltage collaborative optimization model into a real-time voltage regulation model which is solved time by time;
step three, performing value function approximation on a Belman equation in a Markov decision process by adopting a piecewise linear function, wherein the slope of the piecewise linear function can be obtained after a group of offline training scenes generated by predictive data sampling are trained;
and step four, solving the Belman equation after the approximation of the valued function time-period by time-period to obtain an approximate global optimal decision.
The objective function of the power distribution network voltage collaborative optimization model is as follows:
characterizing the fluctuation of the system voltage by the sum of the deviation amounts of the node voltages relative to the reference voltage in each period, wherein: c is the sum of the voltage deviation amounts of all the optimization periods;the square of the voltage amplitude of the node e in the t period;square the node reference voltage die value;t is the total optimization period, which is the number of system nodes.
The electric automobile cluster scheduling method is realized by the following steps:
wherein,is the total number of the electric automobiles,the number of the electric automobiles is the I type;andthe single variable and the group variable of the charging power of the electric automobile in the t period are respectively;andthe single variable and the group variable of the discharge power of the electric automobile in the t period are respectively;andthe single variable and the group variable of the electric quantity of the electric automobile in the period t are respectively.
The power distribution network voltage collaborative optimization model comprises an electric vehicle charging pile selection model, an electric vehicle charging and discharging power regulation model, an on-load voltage regulator model and a branch power flow model;
(1) Electric automobile fills electric pile selection model
The selection of the charging position of the electric automobile is limited by the position of the existing charging pile, meanwhile, one type of electric automobile is guaranteed to be charged only by one charging pile, and after the charging positions of various electric automobiles are decided by a first time period, the subsequent time period is not changed any more, so that the following constraint exists:
in the method, in the process of the invention,to represent the 0-1 variable of the charging position of the class I clustered electric vehicle,the position of the existing charging pile of the power distribution network is shown, and the charging position of the electric automobile is only thatCharging and discharging can be performed at the node where the charging pile exists;
(2) Electric automobile charge-discharge power regulation and control model
The charging and discharging power of the electric automobile is limited by the time of being connected into the charging station and the maximum charging and discharging power, and meanwhile, the electric automobile cannot be charged and discharged at the same time, as follows:
in the method, in the process of the invention,andcharging and discharging zone bits of the t-period I type cluster electric automobile respectively;accessing a 0-1 variable of a charging pile for the t-period I type cluster electric automobile;andmaximum charge and discharge power of I-class cluster electric vehicles respectively;
the electric quantity constraint of the electric automobile is as follows:
wherein:the expected electric quantity when the ith electric automobile leaves the charging pile is obtained;the initial electric quantity when the ith electric automobile reaches the charging pile is set;andthe time for the I-class cluster electric automobile to arrive and leave the charging pile is respectively,the number of the electric automobiles is I-class cluster;
when the electric vehicle leaves the charging station, the SOC should meet the user's expected SOC; the upper and lower limits of the SOC need to be dependent on the operating parameters (i.e、、、And) The boundary of the reformulated SOC is as follows:
wherein:calculating an adjustment margin for the lower limit of the SOC of the ith electric automobile, and ensuring that the SOC of the electric automobile meets the lower limit requirement;the SOC adjustment allowance of the ith electric automobile in the last period is more flexible when the electric automobile participates in voltage adjustment;
the electric quantity balance constraint of the electric automobile is as follows:
wherein:is I-class cluster electric automobileThe amount of electricity in the time period;the charge and discharge efficiency coefficient of the electric automobile;
(3) On-load voltage regulator model
The on-load voltage regulator is only in one gear at any moment:
the single adjustment amount of the on-load voltage regulator cannot exceed the maximum value:
wherein:the transformation ratio of the voltage regulator is t period;the gear is a loaded voltage regulator gear;a 0-1 variable of a gear n of the load voltage regulator in a t period;andthe variable is 0-1 of the ascending and descending of the gear of the on-load voltage regulator in the t period;the gear single maximum adjustment quantity is adopted;
(4) Network tide model
New energy output constraint:
wherein:andthe upper and lower output limits of the new energy unit g in the t period are respectively set;the output of the new energy unit g is t time period;
selecting a branch power flow model to perform power flow calculation, wherein the following power flow constraint exists:
wherein e, f, k are node numbers;andthe voltage amplitudes of the node e and the node f in the t period are respectively;andactive power and reactive power injected into the node f in the t period respectively;andreactance and resistance values of the branch ef;andactive power and reactive power flowing through the head end of the branch ef in the t period respectively;andthe active power and the reactive power respectively flow through the head end of the branch fk at the t period;a current flowing through the branch ef for the period t;
the injection power of the node is calculated by the load power, the new energy output and the root node power, and is as follows:
wherein:andactive power and reactive power sent by the root node in the t period are respectively;andwind power and photovoltaic output of the node f at the t period respectively;the normal load power of the node f in the t period;the power of the charging pile is the power of the node f in the t period;
the tide constraint contains a quadratic term, belongs to non-convexity constraint, and the conventional algorithm has poor solving effect on the optimization problem containing the non-convexity constraint. Therefore, the non-convex power flow constraint can be converted into the second order cone constraint through the phase angle relaxation and the second order cone relaxation, so that the mixed nonlinear programming problem is converted into the second order cone programming problem, the problem has good solving effect by adopting a general commercial solver, and the solution of the commercial solver to the second order cone programming problem is an accurate optimal solution under the condition that the objective function is a convex function. The non-convexly constrained phase angle relaxation and second order cone relaxation processes are described below.
Variables characterizing the square of current and voltage are introduced by:
wherein:andrespectively squaring the voltage amplitude of the node e in the t period and squaring the current of the branch ef;
to this end, the branch tidal current constraint may be converted to the following:
the second order cone constraint to relax the above to rotation is available:
the above can be written as a standard second order taper, namely:
the power distribution network operating conditions also need to meet the following constraints:
wherein:andthe upper and lower limits of the square of the voltage die value of the node e are respectively;andthe upper and lower limits of the current modulus square of the branch ef are respectively determined.
According to the method, the electric automobile and the on-load voltage regulator are used for participating in voltage regulation, the power distribution network voltage collaborative optimization model meets the charging and discharging constraint of the electric automobile, the running constraint of the on-load voltage regulator and the power distribution network tide constraint, and the method comprises the following steps:
the electric automobile charge-discharge constraint includes: variable conversion constraint, charge and discharge position constraint, charge and discharge power constraint, electric quantity balance constraint and electric quantity boundary constraint;
the on-load voltage regulator operating constraints include: gear position constraint and gear adjustment constraint;
the tide constraint comprises: branch flow constraint, second order cone relaxation constraint and new energy output constraint.
The Markov decision process of the invention is as follows:
wherein,as a state variable, a state variable is used,the maximum output value of the new energy is t time periods,is the electric quantity of various electric vehicles in the period t,for whether various electric vehicles are connected with the charging pile in the period t,andactive reactive load power of each node in t period,a gear of the load voltage regulator is a t period;in order to make a decision as to the variables,the actual output of the new energy is t time period,andactive and reactive power of the root node at time t,is used for charging the electric automobile in various positions,andrespectively the charging and discharging states of the electric automobile in the t period,andrespectively the charging and discharging power of the electric automobile in the t period,andthe up and down zone of the on-load voltage regulator gear in the t period are respectively,andthe square of the voltage value of each node in the t period and the square of the current of each branch are respectively,andactive power and reactive power transmitted by each branch in t time period respectively;as the external information, a content of the external information,andprediction errors of renewable energy output and conventional load for a period t;a model objective function is co-optimized for the distribution network voltage,andthe pre-decision and post-decision value functions, respectively. The real-time voltage regulation model is a single-period solving model, and the power distribution network voltage collaborative optimization model which is solved in all time periods is subjected to time period decoupling and then can be reconstructed into a Markov decision process, wherein the Markov decision process comprises state variablesDecision variablesAnd external informationThe relation among the three is determined by various constraints, so that a real-time voltage regulation model for solving in a single period is formed.
After modeling the power distribution network voltage collaborative optimization model as a Markov decision process, the optimal decision sequence of the problem is obtained by solving a Belman equation through a dynamic programming method, and the method is as follows:
in the method, in the process of the invention,a model objective function is co-optimized for the distribution network voltage,andthe pre-decision and post-decision value functions,the problem is a multi-period optimization problem and is solved by adopting a dynamic programming method.
In the markov decision process of the present invention, the piecewise linear function is:
the solving mode of the decision variables in the Markov decision process is as follows:
wherein,is thatA time period decision-making post-state value function,is an approximation function of t time period,Is thatThe approximate function of the electric quantity of the electric vehicles in the class I cluster in the period of time is that M is the segmentation number of the electric quantity of each class of electric vehicles;slope of the mth section of the t-period I-class cluster electric automobile;representing the electric quantity of the mth section of the I-class cluster electric automobile;the expected electric quantity when the I-class cluster electric automobile leaves the charging pile is obtained;the initial electric quantity of the I-class cluster electric automobile reaching the charging pile period;
the piecewise linear function provided by the invention, the solving of the slope comprises the following steps:
the slope of the approximation function will significantly affect the accuracy of the solution, so the slope needs to be updated by training to improve the solution accuracy. Defining the slope as the partial derivative of the value function to the SOC:
solving the slope of the approximation function in a differential manner:
the slope is updated in an iterative manner using:
wherein,the slope sampling value of the mth section of the class I cluster electric automobile in the nth training t period is obtained,to solve the differentiation of SOC at slope;nth iteration for t-period I-class cluster electric automobileThe slope value of the m-th segment of the time period,and (3) the slope value of the nth section of the t-time period I type cluster electric automobile and the mth section of the mth-1 time period iteration t, wherein alpha is the slope updating step length.
Step one optimization model is built under the condition that the accurate state information of the whole time period is known, and the condition is difficult to realize in the real-time optimization process. Therefore, a Markov decision process is needed to reconstruct the voltage regulation model of the power distribution network, so that the full-time-period optimization problem is converted into a single-time-period optimization problem, and then the single-time-period optimization problem is solved time by using a constraint coupling relation among time periods.
Generating a plurality of groups of offline training scenes by adopting a Monte Carlo sampling method according to the new energy output and the load day-ahead predicted value and the error distribution information thereof;
and (3) calculating the slope of the approximation function according to the method of the step three, and solving all optimization problems by using a Gurobi solver on a Matlab platform.
The piecewise linear function slope obtained after the training of the last offline scene is completed is used for real-time optimization of the actual scene.
According to the characteristics of electric automobile cluster dispatching, a regulation and control model of the charging position, the charging and discharging power and the on-load voltage regulator gear of the electric automobile is established, and the minimum system voltage fluctuation is used as a power distribution network voltage optimization target; solving the model by adopting an approximate dynamic programming algorithm, reconstructing the model into a Markov decision process, approximating a value function in a Belman equation by adopting a piecewise linear function, and solving the Belman equation time by time to obtain an approximate optimal decision for real-time voltage regulation of the power distribution network, thereby avoiding the problem of dimension explosion. The invention improves the voltage stability of the power distribution network and ensures the optimization accuracy in a random environment; on the premise of meeting the operation limit of the power distribution network and the electric automobile, the method can cope with uncertainty of output and load of renewable energy sources and give out the optimal dispatching result of the power distribution network voltage regulation under random environment.
Drawings
FIG. 1 is a flow of a real-time collaborative voltage regulation method of a power distribution network based on approximate dynamic programming;
FIG. 2 is an IEEE-33 node system;
FIG. 3 is a schematic diagram of a training scenario of wind power day-ahead predicted values according to an embodiment of the invention;
FIG. 4 is a schematic diagram of a training scenario of photovoltaic future prediction values according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a training scenario of a conventional load day-ahead predictor according to an embodiment of the present invention;
fig. 6 is a voltage curve of a terminal 18 node in each scene obtained by the real-time coordinated voltage regulation method of the power distribution network according to the embodiment of the present invention for the scenes shown in fig. 3, fig. 4, and fig. 5;
fig. 7 is a voltage curve of a terminal 33 node in each scene obtained by the real-time coordinated voltage regulation method of the power distribution network according to the embodiment of the present invention for the scenes shown in fig. 3, fig. 4, and fig. 5;
fig. 8 is a gear change curve of the on-load voltage regulator in each scene obtained by the real-time coordinated voltage regulation method of the power distribution network according to the embodiment of the present invention for the scenes shown in fig. 3, fig. 4 and fig. 5;
fig. 9 is a graph of charging power of each cluster of electric vehicles and SOC curves of each cluster of electric vehicles in scene 1, which is obtained by the method for real-time coordinated voltage regulation of the power distribution network according to the embodiment of the present invention for the scenes shown in fig. 3, fig. 4, and fig. 5;
fig. 10 is a graph of charging power of each cluster of electric vehicles and SOC curves of each cluster of electric vehicles in a scenario 2 obtained by the method for real-time coordinated voltage regulation of the power distribution network according to the embodiment of the present invention;
fig. 11 is a schematic diagram of charging power and SOC curves of each cluster of electric vehicles in a scenario 3 obtained by the real-time coordinated voltage regulation method of the power distribution network according to the embodiment of the present invention for the scenario 3 shown in fig. 3, fig. 4, and fig. 5.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. In addition, the technical features of the embodiments of the present invention described below may be combined with each other as long as they do not interfere with each other.
Fig. 1 is a flowchart of a power distribution network real-time cooperation voltage regulating method based on approximate dynamic programming, which is provided by the embodiment of the invention, and includes the following steps:
step one: and establishing a power distribution network voltage collaborative optimization model.
In this embodiment, an IEEE-33 node model is used to verify the effect of improving the voltage stability of the power distribution network by the proposed real-time coordinated voltage regulation method, and the specific model is shown in fig. 2.
In this embodiment, parameters of various cluster electric vehicles are shown in table 1:
TABLE 1
| EV cluster | Time of arrival | Departure time | Initial SOC/kWh | Expected SOC/kWh | Maximum charge-discharge power/kW | Quantity is/% |
| EV1 | Day 8:00 | Day 18:00 | 10 | 40 | 10 | 21.31 |
| EV2 | Day 9:00 | Day 18:00 | 5 | 35 | 10 | 28.69 |
| EV3 | Day 20:00 | The next day 7:00 | 5 | 55 | 10 | 26.25 |
| EV4 | Day 19:00 | The next day 7:00 | 10 | 50 | 10 | 23.75 |
Establishing an objective function of a power distribution network voltage collaborative optimization model:
using voltages of nodes of each period relative to a reference voltageThe sum of the deviations characterizes the ripple of the system voltage, wherein: c is the sum of the voltage deviation amounts of all the optimization periods;the square of the voltage amplitude of the node e in the t period;is the node reference voltage;t is the total optimization period, which is the number of system nodes. (this is the objective function value of the power distribution network voltage collaborative optimization model which is solved in the whole time period, namely the total objective function value of 24 hours, after the model is reconstructed into a Markov decision process, a real-time voltage regulation model which is solved in a single time period can be formed, so that the objective function value is the value of the single time period, namely。）
According to the electric automobile cluster scheduling characteristics, it is provided withThe electric vehicles are classified into I types, and the number of vehicles in each type isThen:
monomer variables of the electric steam can be converted into group variables by the following formula:
in the method, in the process of the invention,andrespectively isMonomer variables and group variables of charging power of the electric automobile in a t period;andthe single variable and the group variable of the discharge power of the electric automobile in the t period are respectively;andthe single variable and the group variable of the electric quantity of the electric automobile in the period t are respectively.
The power distribution network voltage cooperative regulation and control model comprises an electric automobile charging pile selection model, an electric automobile charging and discharging power regulation and control model, an on-load voltage regulator model and a branch tide model.
(1) Electric automobile fills electric pile selection model
The selection of the charging position of the electric automobile is limited by the position of the existing charging pile, meanwhile, one type of electric automobile is guaranteed to be charged only by one charging pile, and after the charging positions of various electric automobiles are decided by a first time period, the subsequent time period is not changed any more, so that the following constraint exists:
in the method, in the process of the invention,to represent the 0-1 variable of the charging position of the class I clustered electric vehicle,the position of the existing charging pile of the power distribution network is indicated, and the charging position of the electric automobile can be charged and discharged only at the node where the charging pile exists.
(2) Electric automobile charge-discharge power regulation and control model
The charging and discharging power of the electric automobile is limited by the time of being connected into the charging station and the maximum charging and discharging power, and meanwhile, the electric automobile cannot be charged and discharged at the same time, as follows:
in the method, in the process of the invention,andcharging and discharging zone bits of the t-period I type cluster electric automobile respectively;accessing a 0-1 variable of a charging pile for the t-period I type cluster electric automobile;andand the maximum charge and discharge power of the I-class cluster electric automobile is respectively.
The electric quantity constraint of the electric automobile is as follows:
wherein:the expected electric quantity when the ith electric automobile leaves the charging pile is obtained;the initial electric quantity when the ith electric automobile reaches the charging pile is set;andthe time for the I-class cluster electric automobile to arrive and leave the charging pile is respectively,the number of the I-type cluster electric automobiles.
When the electric vehicle leaves the charging station, the SOC should meet the user's expected SOC. However, the rolling decision characteristics of the real-time optimization process make it difficult to ensure that the above conditions are fulfilled, and therefore, the upper and lower limits of the SOC need to be determined according to the operating parameters (i.e.、、、And) And (5) reformulating. The boundaries of the reformulated SOC are shown below, with a schematic diagram shown in fig. 1.
Wherein:and calculating an adjustment margin for the lower limit of the SOC of the ith electric automobile, and ensuring that the SOC of the electric automobile meets the lower limit requirement.And the margin is adjusted for the SOC of the ith electric automobile at the last period, so that the electric automobile is more flexible when participating in voltage adjustment.
The electric quantity balance constraint of the electric automobile is as follows:
wherein:is I-class cluster electric automobileThe amount of electricity in the time period;is the charge and discharge efficiency coefficient of the electric automobile.
(3) On-load voltage regulator model
The on-load voltage regulator is only in one gear at any moment:
the single adjustment amount of the on-load voltage regulator cannot exceed the maximum value:
wherein:the transformation ratio of the voltage regulator is t period;the gear is a loaded voltage regulator gear;a 0-1 variable of a gear n of the load voltage regulator in a t period;andthe variable is 0-1 of the ascending and descending of the gear of the on-load voltage regulator in the t period;the gear is a single maximum adjustment.
(4) Network tide model
New energy output constraint:
wherein:andthe upper and lower output limits of the new energy unit g in the t period are respectively set;the output of the new energy unit g is t time period;
selecting a branch power flow model to perform power flow calculation, wherein the following power flow constraint exists:
wherein e, f, k are node numbers;andthe voltage amplitudes of the node e and the node f in the t period are respectively;andactive power and reactive power injected into the node f in the t period respectively;andreactance and resistance values of the branch ef;andactive power and reactive power flowing through the head end of the branch ef in the t period respectively;andthe active power and the reactive power respectively flow through the head end of the branch fk at the t period;a current flowing through the branch ef for the period t;
the injection power of the node is calculated by the load power, the new energy output and the root node power, and is as follows:
wherein:andactive power and reactive power sent by the root node in the t period are respectively;andwind power and photovoltaic output of the node f at the t period respectively;the normal load power of the node f in the t period;and the charging pile power of the node f in the t period.
The tide constraint contains a quadratic term, belongs to non-convexity constraint, and the conventional algorithm has poor solving effect on the optimization problem containing the non-convexity constraint. Therefore, the non-convex power flow constraint can be converted into the second order cone constraint through the phase angle relaxation and the second order cone relaxation, so that the mixed nonlinear programming problem is converted into the second order cone programming problem, the problem has good solving effect by adopting a general commercial solver, and the solution of the commercial solver to the second order cone programming problem is an accurate optimal solution under the condition that the objective function is a convex function. The non-convexly constrained phase angle relaxation and second order cone relaxation processes are described below.
Variables characterizing the square of current and voltage are introduced by:
wherein:andrespectively squaring the voltage amplitude of the node e in the t period and squaring the current of the branch ef;
to this end, the branch tidal current constraint may be converted to the following:
the second order cone constraint to relax the above to rotation is available:
the above can be written as a standard second order taper, namely:
the power distribution network operating conditions also need to meet the following constraints:
wherein:andthe upper and lower limits of the square of the voltage die value of the node e are respectively;andthe upper and lower limits of the current modulus square of the branch ef are respectively determined.
And step two, reconstructing the power distribution network voltage collaborative optimization model into a Markov decision process, and converting the power distribution network voltage collaborative optimization model into a real-time voltage regulation model which is solved time by time.
Step one optimization model is built under the condition that the accurate state information of the whole time period is known, and the condition is difficult to realize in the real-time optimization process. Therefore, a Markov decision process is needed to reconstruct the voltage regulation model of the power distribution network, so that the full-time-period optimization problem is converted into a single-time-period optimization problem, and then the single-time-period optimization problem is solved time by using a constraint coupling relation among time periods.
The Markov decision process of the power distribution network voltage optimization model comprises state variablesDecision variablesAnd external informationThe specific variable sets are shown below:
wherein,as a state variable, a state variable is used,the maximum output value of the new energy is t time periods,is the electric quantity of various electric vehicles in the period t,for whether various electric vehicles are connected with the charging pile in the period t,andactive reactive load power of each node in t period,a gear of the load voltage regulator is a t period;in order to make a decision as to the variables,the actual output of the new energy is t time period,andactive and reactive power of the root node at time t,is used for charging the electric automobile in various positions,andrespectively the charging and discharging states of the electric automobile in the t period,andrespectively the charging and discharging power of the electric automobile in the t period,andthe up and down zone of the on-load voltage regulator gear in the t period are respectively,andthe square of the voltage value of each node in the t period and the square of the current of each branch are respectively,andactive power and reactive power transmitted by each branch in t time period respectively;as the external information, a content of the external information,andprediction errors of renewable energy output and conventional load for a period t; state variablesDecision variablesAnd external informationThe relation between the two functions, namely the transfer function, is determined by the charge and discharge constraint of the electric automobile, the operation constraint of the on-load voltage regulator and the tide constraint.
Further, after modeling the power distribution network voltage optimization model as a markov decision process, an optimal decision sequence of the problem can be obtained by solving a bellman equation through a dynamic programming method, as follows:
in the method, in the process of the invention,a model objective function is co-optimized for the distribution network voltage,andthe method is characterized in that the method is a value function before and after decision, the problem is a multi-period optimization problem, and a dynamic programming method can be adopted for solving.
However, the state variable space and the decision variable space need to be traversed under the dynamic planning method, so that the calculated amount is large, and the problem of dimension explosion is easy to generate.
Step three, adopting a piecewise linear function related to electric quantity of the electric automobile to approximate a value function of a Bellman equation in the Markov decision process, wherein the slope of the piecewise linear function can be obtained after training a group of offline training scenes generated by predictive data sampling;
aiming at solving the 'dimension explosion' problem of the real-time collaborative voltage regulation model of the power distribution network, a power distribution network real-time collaborative voltage regulation method based on approximate dynamic programming is adopted. For the optimization problem containing energy storage, a good effect can be obtained by using a piecewise linear function approximation function, so that the piecewise linear function related to the electric quantity of the electric automobile is used for approximating a state value function, and further a Belman equation is solved, and an approximately optimal decision sequence is obtained.
The state value function may be rewritten as:
the decision variables of the problem can be solved by:
wherein,is thatA time period decision-making post-state value function,is an approximation function of t time period,Is thatThe approximate function of the electric quantity of the electric vehicles in the class I cluster in the period of time is that M is the segmentation number of the electric quantity of each class of electric vehicles;slope of the mth section of the t-period I-class cluster electric automobile;and the electric quantity of the mth section of the I-class cluster electric automobile is represented.The expected electric quantity when the I-class cluster electric automobile leaves the charging pile is obtained;the initial electric quantity of the I-class cluster electric automobile reaching the charging pile period;
the slope of the approximation function will significantly affect the accuracy of the solution, so the slope needs to be updated by training to improve the solution accuracy. First, the slope is defined as the partial derivative of the value function with respect to SOC, as follows:
solving the slope of the approximation function in a differential manner:
the slope is updated in an iterative manner using:
wherein,the slope sampling value of the mth section of the class I cluster electric automobile in the nth training t period is obtained,to solve the differentiation of SOC at slope;nth iteration for t-period I-class cluster electric automobileThe slope value of the m-th segment of the time period,and (3) the slope value of the nth section of the t-time period I type cluster electric automobile and the mth section of the mth-1 time period iteration t, wherein alpha is the slope updating step length.
And step 4, solving the Belman equation after the approximation of the valued function time-period by time-period to obtain an approximate global optimal decision.
According to the new energy output and load day-ahead predicted value and error distribution information thereof, a Monte Carlo sampling method is adopted to generate 200 groups of offline training scenes, the specific distribution of the wind-solar output and the prediction error of the conventional load is N (0,0.052) under the assumption that the wind-solar output and the prediction error of the conventional load obey normal distribution, and the generated offline training scenes are shown in fig. 3, fig. 4 and fig. 5.
And (3) calculating the slope of the approximation function according to the method of the step three, and solving all optimization problems by using a Gurobi solver on a Matlab platform.
The piecewise linear function slope obtained after the training of the last offline scene is completed is used for real-time optimization of the actual scene.
In order to further explain the real-time collaborative voltage regulation method for the power distribution network provided by the embodiment, 3 optimization scenes are set for analysis, wherein the number of electric vehicles is 300:
scene 1: the system has the conventional load and the electric vehicle load, the electric vehicles charge with equal power, the charging positions of the electric vehicles of each cluster are respectively nodes 4, 12,4 and 12, and the voltage regulation is carried out by adopting a load voltage regulator;
scene 2: the conventional load and the electric vehicle load exist in the system, so that the charging position and the charging power of the electric vehicle are optimized, and the voltage fluctuation is reduced;
scene 3: the system has conventional load and electric vehicle load, and the on-load voltage regulator and the electric vehicle are adopted to carry out combined voltage regulation.
To analyze the superiority of this strategy, 1 of the 200 sets of scenes above was chosen as the subsequent comparative example.
Simulation results are shown in the following table:
TABLE 2
| Scene(s) | Voltage deviation | Minimum voltage (minimum voltage) | Charging position |
| Scene 1 | 3.1900 | 0.9425 | 4,12,4,12 |
| Scene 2 | 2.1947 | 0.9706 | 25,21,21,25 |
| Scene 3 | 2.1333 | 0.9706 | 25,25,25,25 |
Fig. 6 and 7 show voltage curves at the end nodes 18 and 33 of the system in each scenario. Scene 1 does not optimize the charging power and charging position of the electric automobile, only adopts the on-load voltage regulator to regulate voltage, so the voltage deviation is the largest under the whole period, and in the load peak period 12:00-14: the condition that the voltage is out of limit appears in 00, and the voltage is 0.9425 at the minimum, seriously threatens the steady operation of distribution network. In scene 2, after optimizing electric automobile charging power and charging position, the voltage of root node 18, 33 promotes by a wide margin, and system voltage stability promotes by a wide margin 31.2% simultaneously, and electric automobile selects 21, 25 node fills electric pile in being close to the root node and charges, avoids voltage drop by a wide margin, and system minimum voltage promotes to 0.9706 simultaneously, has effectively avoided the condition of voltage out of limit. In the scene 3, the electric automobile and the on-load voltage regulator carry out the coordinated voltage regulation, the system voltage stability is further improved, meanwhile, the charging positions of various cluster electric automobiles are concentrated on the node 25 charging piles, and the situation that the voltage drops is avoided when the cluster electric automobiles are close to a root node is avoided. On the other hand, the novel energy is absorbed in situ by approaching to the photovoltaic power generation node.
Fig. 8 is a gear change curve of the on-load voltage regulator in scenario 1 and scenario 3, where the on-load voltage regulator is not optimized in scenario 2. In the scene 1, the charging power of the electric automobile is unchanged, and in order to reduce the occurrence of the voltage out-of-limit condition, the on-load voltage regulator is always in the highest gear. In scene 3, in the early morning load valley period 0:00-6:00, the charging power of the electric automobile is lower, and in order to reduce voltage fluctuation, the gear of the on-load voltage regulator is reduced.
Fig. 9, 10 and 11 show the charge power and SOC variation of electric vehicles in each cluster in three scenarios, and scenario 1 shows constant-power charging. Under the condition that the electric automobile participates in voltage regulation, the charging power of the EV is greatly reduced in the conventional load peak time period 12:00-14:00 and 20:00-22:00, so that the situation of peak-to-peak peaking is prevented.
It will be readily appreciated by those skilled in the art that the foregoing description is merely a preferred embodiment of the invention and is not intended to limit the invention, but any modifications, equivalents, improvements or alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (9)
1. The real-time cooperative voltage regulation method for the power distribution network based on the approximate dynamic programming is characterized by comprising the following steps of:
firstly, taking the minimum voltage fluctuation as an objective function, and establishing a power distribution network voltage collaborative optimization model according to electric vehicle cluster scheduling characteristics;
reconstructing a power distribution network voltage collaborative optimization model into a Markov decision process, and converting the power distribution network voltage collaborative optimization model into a real-time voltage regulation model which is solved time by time;
step three, performing value function approximation on a Belman equation in a Markov decision process by adopting a piecewise linear function, wherein the slope of the piecewise linear function can be obtained after a group of offline training scenes generated by predictive data sampling are trained;
and step four, solving the Belman equation after the approximation of the valued function time-period by time-period to obtain an approximate global optimal decision.
2. The method according to claim 1, characterized in that the objective function of the power distribution network voltage collaborative optimization model is:
；
characterizing the fluctuation of the system voltage by the sum of the deviation amounts of the node voltages relative to the reference voltage in each period, wherein: c is the sum of the voltage deviation amounts of all the optimization periods;the square of the voltage amplitude of the node e in the t period; />Square the node reference voltage die value; />T is the total optimization period, which is the number of system nodes.
3. The method according to claim 1, characterized in that the electric vehicle cluster scheduling is achieved by:
；
wherein,the total number of electric vehicles is->The number of the electric automobiles is the I type; />And->The single variable and the group variable of the charging power of the electric automobile in the t period are respectively; />And->The single variable and the group variable of the discharge power of the electric automobile in the t period are respectively; />And->The single variable and the group variable of the electric quantity of the electric automobile in the period t are respectively.
4. The method according to claim 1, wherein the distribution network voltage collaborative optimization model comprises an electric vehicle charging pile selection model, an electric vehicle charging and discharging power regulation model, an on-load voltage regulator model and a branch power flow model:
(1) Electric automobile fills electric pile selection model
The selection of the charging position of the electric automobile is limited by the position of the existing charging pile, meanwhile, one type of electric automobile is guaranteed to be charged only by one charging pile, and after the charging positions of various electric automobiles are decided by a first time period, the subsequent time period is not changed any more, and the following constraint exists:
；
in the method, in the process of the invention,for 0-1 variable representing charging position of I-class cluster electric automobile,/for the charging position>The method comprises the steps that the position of a charging pile existing in a power distribution network is indicated, and charging and discharging can be carried out only at nodes where the charging pile exists at the charging position of an electric automobile;
(2) Electric automobile charge-discharge power regulation and control model
The charging and discharging power of the electric automobile is limited by the time of being connected into the charging station and the maximum charging and discharging power, and meanwhile, the electric automobile cannot be charged and discharged at the same time, as follows:
；
in the method, in the process of the invention,and->Charging and discharging zone bits of the t-period I type cluster electric automobile respectively; />Accessing a 0-1 variable of a charging pile for the t-period I type cluster electric automobile; />And->Maximum charge and discharge power of I-class cluster electric vehicles respectively;
the electric quantity constraint of the electric automobile is as follows:
；
wherein:the expected electric quantity when the ith electric automobile leaves the charging pile is obtained; />The initial electric quantity when the ith electric automobile reaches the charging pile is set; />And->The time when I-class cluster electric vehicles arrive and leave the charging pile is respectively +.>The number of the I-type cluster electric automobiles;
when the electric vehicle leaves the charging station, the SOC should meet the user's expected SOC; the upper and lower limits of the SOC need to be dependent on the operating parameters (i.e、/>、/>、/>And->) Reformulating; the boundaries of the reformulated SOC are as follows:
；
wherein:calculating an adjustment margin for the lower SOC limit of the ith electric automobile; />The method comprises the steps that (1) the SOC adjustment margin of the ith electric automobile in the last period is provided;
the electric quantity balance constraint of the electric automobile is as follows:
；
wherein:for I-class cluster electric automobile +.>The amount of electricity in the time period; />The charge and discharge efficiency coefficient of the electric automobile;
(3) On-load voltage regulator model
The on-load voltage regulator is only in one gear at any moment:
；
the single adjustment amount of the on-load voltage regulator cannot exceed the maximum value:
；
wherein:the transformation ratio of the voltage regulator is t period; />The gear is a loaded voltage regulator gear; />A 0-1 variable of a gear n of the load voltage regulator in a t period; />And->The variable is 0-1 of the ascending and descending of the gear of the on-load voltage regulator in the t period; />The gear single maximum adjustment quantity is adopted;
(4) Network tide model
New energy output constraint:
；
wherein:and->The upper and lower output limits of the new energy unit g in the t period are respectively set; />The output of the new energy unit g is t time period;
selecting a branch power flow model to perform power flow calculation, wherein the following power flow constraint exists:
；
；
；
wherein e, f, k are node numbers;and->The voltage amplitudes of the node e and the node f in the t period are respectively; />And->Active power and reactive power injected into the node f in the t period respectively; />And->Reactance and resistance values of the branch ef;and->Active power and reactive power flowing through the head end of the branch ef in the t period respectively; />And->The active power and the reactive power respectively flow through the head end of the branch fk at the t period; />A current flowing through the branch ef for the period t;
the injection power of the node is calculated by the load power, the new energy output and the root node power, and is as follows:
；
wherein:and->Active power and reactive power sent by the root node in the t period are respectively; />And->Wind power and photovoltaic output of the node f at the t period respectively; />And->The conventional active and reactive load power of the node f in the t period is respectively;the power of the charging pile is the power of the node f in the t period;
the above tide constraint contains a quadratic term, belongs to non-convex constraint, and the phase angle relaxation and second order cone relaxation processes of the non-convex constraint are as follows:
variables characterizing the square of current and voltage are introduced by:
；
wherein:and->Respectively squaring the voltage amplitude of the node e in the t period and squaring the current of the branch ef;
the branch tidal current constraint may be converted to the following:
；
；
；
the second order cone constraint to relax the above to rotation is available:
；
the above can be written as a standard second order taper, namely:
；
the power distribution network operating conditions also need to meet the following constraints:
；
wherein:and->The upper and lower limits of the square of the voltage die value of the node e are respectively; />And->The upper and lower limits of the current modulus square of the branch ef are respectively determined.
5. The method of claim 1, wherein the power distribution network voltage collaborative optimization model satisfies an electric vehicle charge-discharge constraint, an on-load voltage regulator operation constraint, a power distribution network power flow constraint, wherein:
the electric automobile charge-discharge constraint includes: variable conversion constraint, charge and discharge position constraint, charge and discharge power constraint, electric quantity balance constraint and electric quantity boundary constraint;
the on-load voltage regulator operating constraints include: gear position constraint and gear adjustment constraint;
the tide constraint comprises: branch flow constraint, second order cone relaxation constraint and new energy output constraint.
6. The method of claim 1, wherein the markov decision process is:
；
；
；
；
wherein,is a state variable +.>For the maximum output value of the new energy source in the t period, < > of>For the electric quantity of various electric vehicles in t time period, < >>Whether various electric vehicles are connected with charging piles or not in t time period>And->Active reactive load power for each node in t period, < > for>A gear of the load voltage regulator is a t period; />For decision variables +.>The actual output of the new energy is t time period,and->Active and reactive power of t-period root node respectively, < ->Charging positions for various electric vehicles>And->Respectively is the charging and discharging states of the electric automobile in the t period>And->Charging and discharging power of the electric automobile in t time period respectively, < >>Andthe up and down sign positions of the on-load voltage regulator in the t period are respectively +.>And->The square of the voltage module value of each node in the t period and the square of the current module value of each branch are respectively +.>And->Active power and reactive power transmitted by each branch in t time period respectively; />Is external information->And->Prediction errors of renewable energy output and conventional load for a period t;optimizing a model objective function, < > for the power distribution network voltage synergy>And->The pre-decision and post-decision value functions, respectively.
7. The method according to claim 1, wherein after reconstructing the power distribution network voltage collaborative optimization model into a markov decision process, the optimal decision sequence is obtained by solving a bellman equation through a dynamic programming method, as follows:
；
in the method, in the process of the invention,optimizing a model objective function, < > for the power distribution network voltage synergy>And->Before and after decision value function, respectively, ">Is a state variable after decision-making.
8. The method of claim 1, wherein in the markov decision process, the piecewise linear function is:
；
the solving mode of the decision variables in the Markov decision process is as follows:
；
wherein,is->Status value function after time period decision, +.>Approximation function for period t, < >>Is->The approximate function of the electric quantity of the electric vehicles in the class I cluster in the period of time is that M is the segmentation number of the electric quantity of each class of electric vehicles; />Slope of the mth section of the t-period I-class cluster electric automobile; />Representing the electric quantity of the mth section of the I-class cluster electric automobile; />The expected electric quantity when the I-class cluster electric automobile leaves the charging pile is obtained; />And (5) the initial electric quantity of the I-class cluster electric automobile reaching the charging pile period.
9. The method according to claim 1 or 8, wherein the solving of the slope comprises the steps of:
defining the slope as the partial derivative of the value function to the SOC:
；
solving the slope of the approximation function in a differential manner:
；
；
the slope is updated in an iterative manner using:
；
wherein,for the slope sampling value of the mth section of the class I cluster electric vehicle in the nth training t period,/>To solve the differentiation of SOC at slope; />Nth iteration +.for t-period I-class cluster electric vehicle>Slope value of period m +.>And (3) the slope value of the nth section of the t-time period I type cluster electric automobile and the mth section of the mth-1 time period iteration t, wherein alpha is the slope updating step length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311658252.9A CN117374975B (en) | 2023-12-06 | 2023-12-06 | Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311658252.9A CN117374975B (en) | 2023-12-06 | 2023-12-06 | Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117374975A true CN117374975A (en) | 2024-01-09 |
| CN117374975B CN117374975B (en) | 2024-02-27 |
Family
ID=89391349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311658252.9A Active CN117374975B (en) | 2023-12-06 | 2023-12-06 | Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117374975B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118134228A (en) * | 2024-05-10 | 2024-06-04 | 浙江大学 | Online dispatching method for industrial park comprehensive energy system containing electric automobile load |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163443A (en) * | 2019-05-27 | 2019-08-23 | 西南石油大学 | Consider the micro- energy net Optimization Scheduling in the natural gas pressure regulating station of electric-gas integration requirement response |
| DE102020210465A1 (en) * | 2020-08-18 | 2022-02-24 | Volkswagen Aktiengesellschaft | Method and device for supporting maneuver planning for an at least partially automated vehicle or a robot |
| JP2023007058A (en) * | 2021-07-01 | 2023-01-18 | 富士電機株式会社 | Optimum setting device and optimum setting program for voltage adjustment apparatus |
| WO2023001531A1 (en) * | 2021-07-23 | 2023-01-26 | Robert Bosch Gmbh | Method and device for estimating a departure time for use in an intelligent charging process of electric vehicles |
| WO2023082697A1 (en) * | 2021-11-15 | 2023-05-19 | 中国电力科学研究院有限公司 | Coordination and optimization method and system for comprehensive electric-thermal energy system, and device, medium and program |
| CN116436019A (en) * | 2023-04-12 | 2023-07-14 | 国网江苏省电力有限公司电力科学研究院 | Multi-resource coordination optimization method, device and storage medium |
| CN117057533A (en) * | 2023-07-12 | 2023-11-14 | 南京邮电大学 | Method and device for optimizing subdivision period ordered charging strategy based on load evaluation |
-
2023
- 2023-12-06 CN CN202311658252.9A patent/CN117374975B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163443A (en) * | 2019-05-27 | 2019-08-23 | 西南石油大学 | Consider the micro- energy net Optimization Scheduling in the natural gas pressure regulating station of electric-gas integration requirement response |
| DE102020210465A1 (en) * | 2020-08-18 | 2022-02-24 | Volkswagen Aktiengesellschaft | Method and device for supporting maneuver planning for an at least partially automated vehicle or a robot |
| JP2023007058A (en) * | 2021-07-01 | 2023-01-18 | 富士電機株式会社 | Optimum setting device and optimum setting program for voltage adjustment apparatus |
| WO2023001531A1 (en) * | 2021-07-23 | 2023-01-26 | Robert Bosch Gmbh | Method and device for estimating a departure time for use in an intelligent charging process of electric vehicles |
| WO2023082697A1 (en) * | 2021-11-15 | 2023-05-19 | 中国电力科学研究院有限公司 | Coordination and optimization method and system for comprehensive electric-thermal energy system, and device, medium and program |
| CN116436019A (en) * | 2023-04-12 | 2023-07-14 | 国网江苏省电力有限公司电力科学研究院 | Multi-resource coordination optimization method, device and storage medium |
| CN117057533A (en) * | 2023-07-12 | 2023-11-14 | 南京邮电大学 | Method and device for optimizing subdivision period ordered charging strategy based on load evaluation |
Non-Patent Citations (4)
| Title |
|---|
| LEI YANG: "EV_Charging_Scheduling_Under_Demand_Charge_A_Block_Model_Predictive_Control_Approach", 《IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING》, pages 1 - 14 * |
| RONG YANG: "Energy_Management_of_the_Power-Split_Hybrid_Electric_City_Bus_Based_on_the_Stochastic_Model_Predictive_Control", 《IEEE ACCESS》, pages 2055 - 2071 * |
| 刘敦楠: "基于深度强化学习的大规模电动汽车充换电负荷优化调度", 《电力系统自动化》, vol. 46, no. 4, pages 36 - 46 * |
| 杨志淳: "考虑光伏消纳能力的配电网规划方案优化", 《电测与仪表》, vol. 55, no. 15, pages 133 - 137 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118134228A (en) * | 2024-05-10 | 2024-06-04 | 浙江大学 | Online dispatching method for industrial park comprehensive energy system containing electric automobile load |
| CN118134228B (en) * | 2024-05-10 | 2024-08-13 | 浙江大学 | Online dispatching method for industrial park comprehensive energy system containing electric automobile load |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117374975B (en) | 2024-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | A customized voltage control strategy for electric vehicles in distribution networks with reinforcement learning method | |
| CN110739725B (en) | Optimal scheduling method for power distribution network | |
| CN110782363A (en) | AC/DC power distribution network scheduling method considering wind power uncertainty | |
| CN110826880B (en) | Active power distribution network optimal scheduling method for large-scale electric automobile access | |
| CN111697578B (en) | Multi-target energy storage-containing regional power grid operation control method | |
| CN109034587B (en) | Active power distribution system optimal scheduling method for coordinating multiple controllable units | |
| CN117374975B (en) | Real-time cooperative voltage regulation method for power distribution network based on approximate dynamic programming | |
| CN110676849B (en) | Method for constructing islanding micro-grid group energy scheduling model | |
| CN111244988A (en) | Electric automobile considering distributed power supply and energy storage optimization scheduling method | |
| CN107730048A (en) | A kind of random robust Optimization Scheduling of wind-powered electricity generation electric automobile association system | |
| CN111614110B (en) | Receiving-end power grid energy storage optimization configuration method based on improved multi-target particle swarm optimization | |
| CN111786417A (en) | Distributed new energy consumption-oriented active power distribution network multi-target interval optimization scheduling method | |
| CN111641233A (en) | Electric power system day-based flexible peak regulation method considering new energy and load uncertainty | |
| CN113315168A (en) | Wind-storage hybrid power station operation simulation method based on sequential Monte Carlo simulation | |
| CN115099590B (en) | Active power distribution network economic optimization scheduling method and system considering light load uncertainty | |
| CN115663793A (en) | Electric automobile low-carbon charging and discharging scheduling method based on deep reinforcement learning | |
| CN114462854A (en) | Hierarchical scheduling method and system containing new energy and electric vehicle grid connection | |
| CN114548597A (en) | Optimization method for alternating current-direct current hybrid optical storage and distribution power grid | |
| CN114421502A (en) | Cooperative optimization method for micro-grid community | |
| Kaur et al. | Design of the ANFIS based optimized frequency control module for an electric vehicle charging station | |
| CN112803422A (en) | Power transmission network voltage double-layer control method based on active and reactive power coordinated optimization | |
| CN117335431A (en) | Distributed coordination optimization method for multi-microgrid interconnection system considering virtual energy storage | |
| CN114759616B (en) | Micro-grid robust optimization scheduling method considering characteristics of power electronic devices | |
| CN116384655A (en) | New energy consumption-oriented source network load side energy storage system optimization planning method and system | |
| Saren et al. | An Optimized Control Strategy for Distributed Energy Storage System to Reduce the Peak-valley Difference of Distribution Network |
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 |