CN116054154A - Double-layer optimization energy-saving control method and system considering voltage safety - Google Patents

Double-layer optimization energy-saving control method and system considering voltage safety Download PDF

Info

Publication number
CN116054154A
CN116054154A CN202310172980.2A CN202310172980A CN116054154A CN 116054154 A CN116054154 A CN 116054154A CN 202310172980 A CN202310172980 A CN 202310172980A CN 116054154 A CN116054154 A CN 116054154A
Authority
CN
China
Prior art keywords
voltage
control
node
power
energy
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310172980.2A
Other languages
Chinese (zh)
Inventor
郭培志
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Chuanggu Smart Environmental Technology Co ltd
Original Assignee
Nanjing Chuanggu Smart Environmental Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Chuanggu Smart Environmental Technology Co ltd filed Critical Nanjing Chuanggu Smart Environmental Technology Co ltd
Priority to CN202310172980.2A priority Critical patent/CN116054154A/en
Publication of CN116054154A publication Critical patent/CN116054154A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • 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
    • 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]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/30—Reactive power compensation

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention discloses a double-layer optimization energy-saving control method and a system considering voltage safety, wherein the method comprises the following steps: the upper optimizing layer establishes an energy-saving control model of the power distribution system considering voltage safety, and the power regulation and control quantity of the optimizing system is used for ensuring the voltage safety of the power distribution system on one hand and ensuring the minimum power regulation and control quantity of the system on the other hand; the lower layer optimization is based on a deep reinforcement learning algorithm, optimizes control parameters of the upper layer, realizes automatic determination of parameters set by human in the traditional process, optimizes equipment control quantity under a long time scale, and ensures that corresponding equipment parameters are not required to be switched excessively frequently, thereby ensuring the service life of equipment; solving an upper-layer optimized power distribution system energy-saving control model to obtain the optimal power regulation quantity of the system, and minimizing the power regulation of the system on the premise of ensuring the voltage safety so as to achieve the purpose of energy saving. The invention solves the problems of unreasonable control parameter setting, no consideration of safety risk and the like in the existing power distribution optimization control algorithm.

Description

Double-layer optimization energy-saving control method and system considering voltage safety
Technical Field
The invention belongs to the technical field of safety energy conservation of intelligent power distribution systems, and particularly relates to a double-layer optimization energy-saving control method considering voltage safety.
Background
With the wide accounting of new energy sources, the problems of safe operation and energy saving and electricity consumption of a power distribution system are getting more and more attention. As disclosed in CN113191549a, a method for planning and optimizing a power distribution network by combining a power distribution network source network, which considers network loss and simplifying ac power flow, obtains a power distribution network planning result according to acquired parameter data and a planning model constructed by taking the total cost of the power distribution system as a goal, and the existing energy-saving technology only considers the operation cost of the system to be minimum in an economic dispatch layer to optimize the power distribution system, lacks the consideration of voltage safety in the optimization process, and generally adopts a method of artificial experience setting to determine optimization and control parameters in the optimization process, thereby lacking rationality. The control method is needed, the optimal control parameters of the system can be effectively and intelligently set, meanwhile, the voltage safety problem in the power adjustment process is considered, the energy conservation of the system is ensured, the voltage can be stabilized in a safety range, and finally the safety and the economy of the system are effectively improved.
Disclosure of Invention
The invention aims to provide a double-layer optimized energy-saving control method considering voltage safety, which aims to solve the problems that safety risks are not considered in the existing energy-saving process of a power distribution system and system control parameters are unreasonably set.
In order to solve the technical problems, the invention provides a double-layer optimized energy-saving control method considering voltage safety, which comprises the following steps:
step one: the upper layer establishes a power distribution system energy-saving control model considering voltage safety;
step two: the lower layer optimizes the control parameters of the upper layer based on a deep reinforcement learning algorithm, including a control domain N c And prediction domain N p Thereby determining setting parameters, and simultaneously optimizing equipment control quantity under a long time scale to adjust switching frequency of corresponding equipment parameters;
step three: and solving an upper-layer optimized power distribution system energy-saving control model based on a particle swarm optimization algorithm to obtain the optimal power regulation and control quantity of the system.
A double-layer optimized energy-saving control system considering voltage safety comprises the following modules:
and a model building module: the upper layer establishes a power distribution system energy-saving control model considering voltage safety;
the equipment control amount optimizing module: the lower layer optimizes the control parameters of the upper layer based on a deep reinforcement learning algorithm, including a control domain N c And prediction domain N p Thereby determining setting parameters, and simultaneously optimizing equipment control quantity under a long time scale to adjust switching frequency of corresponding equipment parameters;
and a solving module: and solving an upper-layer optimized power distribution system energy-saving control model based on a particle swarm optimization algorithm to obtain the optimal power regulation and control quantity of the system.
A computer readable storage medium for storing the functional modules of the above-described voltage safety-considered two-layer optimized energy-saving control system.
A processor for executing operations according to the instructions in the above-described voltage safety-considered double-layer optimized energy-saving control method.
The invention achieves the beneficial technical effects that: the double-layer optimization energy-saving control method considering voltage safety can realize effective energy saving of the power distribution system based on the double-layer optimization algorithm while guaranteeing voltage safety, wherein the lower-layer optimization algorithm realizes intelligent optimization of upper-layer control parameters and long-time scale equipment control, and the upper-layer optimization algorithm realizes overall power optimization of the system after receiving the lower-layer optimal control parameters, so that the economy and safety of the system are effectively improved.
Drawings
Fig. 1 is a flow chart of a double-layer optimized energy-saving control method considering voltage safety according to an embodiment of the invention.
Detailed Description
The invention is further described below in connection with specific embodiments. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
A double-layer optimized energy-saving control method considering voltage safety, as shown in figure 1, comprises the following steps:
step one: the upper layer establishes an energy-saving control model of the power distribution system considering voltage safety, and the voltage safety of the power distribution system is ensured by optimizing the power regulation and control quantity of the system, and the power regulation and control quantity of the system is ensured to be minimum, and the method specifically comprises the following steps:
step 1-1: definition of upper layer optimization control quantity mu (k) of a power distribution network system:
Figure BDA0004099862110000031
wherein ,
Figure BDA0004099862110000032
active power for distributed power supply, +.>
Figure BDA0004099862110000033
Reactive power for a distributed power supply;
step 1-2: constructing a system predictive control model:
Figure BDA0004099862110000034
s.t.
μ min ≤μ(k+i|k)≤μ max ,i＝0,1…N c -1 (3)
Δμ min ≤Δμ(k+i|k)≤Δμ max ,i＝0,1…N c -1 (4)
y min ≤y(k+i|k)≤y max ,i＝0,1…N p -1 (5)
Figure BDA0004099862110000041
SOC min ≤SOC(k)≤SOC max (7)
Figure BDA0004099862110000042
wherein ,Nc and Np A control domain and a prediction domain respectively; q and R are cost weight matrices for controlling the objective function; mu (mu) min and μmax Respectively upper and lower limit constraint of the control quantity; Δμ (k) =μ (k+1) - μ (k); Δμ min and Δμmax Is controlled byClimbing constraint of the preparation quantity; y (k+i|k) is the predicted value of the voltage at time k+i based on the measured value at time k; y is min and ymax Is the system bus voltage constraint;
Figure BDA0004099862110000043
is a sensitivity matrix of the bus voltage to the control variable μ (k); Δy (k) =y (k+1) -y (k); when i > N c When Δμ (k+i) =0; SOC (k+i|k) is a predicted value of the state of energy storage at time k+i based on the measured value at time k;
Figure BDA0004099862110000046
and
Figure BDA0004099862110000047
Representing the maximum and minimum states of charge of the ith distributed energy storage, respectively; delta (k) is a charge-discharge coefficient, delta (k) =1 is energy storage discharge, delta (k) =0 is energy storage charge; η (eta) c and ηd Charge and discharge efficiency of energy storage->
Figure BDA0004099862110000044
Representing 2 norms, +. s (k+i-1|k) represents the energy amount of the stored energy.
Further, the upper and lower limit constraints of the control amount are specifically expressed as:
Figure BDA0004099862110000045
wherein ,Pi max 、P i min 、Q i max 、Q i min Respectively representing the upper and lower limits of the active output and the lower limit of the reactive output of the ith distributed power supply, N s Is the number of distributed power sources.
Further, the climbing constraint of the control amount is specifically expressed as:
Figure BDA0004099862110000051
wherein ,
Figure BDA0004099862110000052
respectively represent the charge-discharge power limit, N of the ith distributed power supply s Is the number of distributed power sources.
Step two: the lower layer optimizes the control parameters of the upper layer based on a deep reinforcement learning algorithm, including a control domain N c And prediction domain N p Thereby determining the setting parameters, optimizing the equipment control quantity under a long time scale, adjusting the switching frequency of the corresponding equipment parameters, avoiding the frequent switching of the corresponding equipment parameters, and further ensuring the service life of the equipment. The method comprises the following specific steps:
step 2-1: defining an optimization objective function of the lower layer optimization as follows:
Figure BDA0004099862110000053
wherein ：Ui A node voltage which is an i-th node; u (U) N Rated voltage of the power distribution network; n is the number of PQ nodes; adopting +/-5% as the maximum safe range of node voltage;
the constraint conditions are as follows:
Figure BDA0004099862110000054
Figure BDA0004099862110000055
Figure BDA0004099862110000056
Figure BDA0004099862110000061
wherein formula (12) isFlow constraint, P t,i,l and Qt,i,l Active power and reactive power consumed by the load l on the node i at the moment t respectively; p (P) t,loss and Qt,loss Active and reactive losses of the distribution network line at the time t are respectively calculated; p (P) t,M and Qt,M Respectively transmitting active power and reactive power to a main network at the moment t; p (P) t,G and Qt,G Respectively generating active power and reactive power for the distributed power supply at the moment t; q (Q) t,CB Reactive power emitted by the capacitor bank at the moment t; formula (13) is node voltage constraint, U i,min and Ui,max Respectively the minimum value and the maximum value of the voltage of the node i; u (U) i and UN Respectively node i voltage and power distribution network rated voltage; formula (14) is node power constraint, P i and Qi Active power and reactive power fed in on node i respectively; p (P) Gi and QGi Respectively integrating active output and reactive output of a distributed power supply for the node i; p (P) li and Qli Load power on node i respectively; p (P) i,min 、P i,max 、Q i,min and Qi,max The minimum value and the maximum value of the active power and the minimum value and the maximum value of the reactive power of the node i are respectively; formula (15) is a distributed power supply output constraint, P Gi,min 、P Gi,max 、Q Gi,min and QGi,max And respectively integrating the minimum value and the maximum value of the active output and the minimum value and the maximum value of the reactive output of the distributed power supply for the node i.
Step 2-2: based on capacitor bank capacity of configuration nodes, the distributed power supply and capacitor bank participating power distribution network voltage control model is converted into a regulation capacitor bank and a control parameter control domain N of the regulation capacitor bank by using DQN algorithm c Prediction domain N p Comprises:
defining a state space as a set of current voltage, active power and reactive power of each PQ node;
determining the compensation amount of the configuration node parallel capacitor bank according to the capacity of the capacitor bank, and setting the action space as the compensation amount of the configuration node parallel capacitor bank and the control parameter control domain N c And prediction domain N p Is a step length of (2);
the bonus function is set to the sum of the quadratic form and the compensation amount of the capacitor bank for the more amount of voltage at each node.
Further, the state space is:
s:{v 1 ,…,v k ,…,v n ,p 1 ,…,p k ,…,p n ,q 1 ,…,q k ,…,q n } (16)
wherein ,vk 、p k and qk The voltage value, the active power and the reactive power observed by the kth node are respectively equal to or more than 1 and equal to or less than n, and n is the total number of PQ nodes;
adopting a multi-gear capacitor bank, taking the capacity of the obtained capacitor bank as the maximum compensation amount, and taking the capacity of each gear as a set value of an action space:
Figure BDA0004099862110000071
wherein ,CBmax For the maximum compensation amount of the capacitor bank,
Figure BDA0004099862110000072
and
Figure BDA0004099862110000073
Control parameter control domains N c And prediction domain N p Is a maximum value of (a).
The reward function is:
Reward＝-[Δv 1 ,…,Δv i ,…,Δv n ]Q[Δv 1 ,…,Δv i ,…,Δv n ] T -Ra k (18)
wherein ,Δvi The threshold amount of the voltage of the node i is exceeded; a, a k Compensating the capacitor bank of node k for an amount; q and R are weight matrix and weight coefficient; deltav i The method comprises the following steps:
Figure BDA0004099862110000074
wherein 5% is the selected voltage out-of-limit safety value range.
Further, the solving the voltage control model of the regulation capacitor bank by using the DQN algorithm to obtain a voltage optimal control strategy includes:
step A1: initializing a memory bank, initializing a Q network weight parameter to omega, initializing a target Q network weight parameter to omega' =omega, and observing the voltage value, active power and reactive power of each current node to serve as an initial state s;
step A2: generating and executing an action a epsilon A according to a greedy strategy, and obtaining a reward r and a new state s' through a formula (18);
step A3: storing the transfer samples (s, a, s') in a memory bank, randomly extracting a sample(s) of minimatch from the memory bank i ,a i ,r i ,s' i )；
Step A4: order the
Figure BDA0004099862110000081
The DQN network loss function L (θ) is calculated according to:
L(θ)＝E[(TargetQ-Q(s,a；θ)) 2 ] (20)
wherein E (·) is the desired value; targetQ is a target network target value; q (s, a; omega) is the predicted value of action a under the state s when the weight parameter is omega; gamma is a discount factor;
step A5: updating a target Q network weight parameter ω' =ω using a gradient descent method;
step A6: repeatedly executing the step A2 to the step A5 until the iteration is finished, and obtaining a voltage optimal control strategy and optimal control parameters;
step A7: and uploading the optimal control parameters to an upper optimization model.
Step three: based on a particle swarm optimization algorithm, solving an upper-layer optimized power distribution system energy-saving control model to obtain the optimal power regulation quantity of the system, and minimizing the power regulation of the system on the premise of ensuring the voltage safety so as to achieve the purpose of energy saving.
A double-layer optimized energy-saving control system considering voltage safety comprises the following modules:
and a model building module: the upper layer establishes a power distribution system energy-saving control model considering voltage safety;
the equipment control amount optimizing module: based on deep reinforcement learning algorithm, optimizing control parameter control domain N c And prediction domain N p Thereby determining tuning parameters while optimizing the device control over a long time scale;
and a solving module: and solving an upper-layer optimized power distribution system energy-saving control model based on a particle swarm optimization algorithm to obtain the optimal power regulation and control quantity of the system.
A computer readable storage medium for storing the functional modules of the above-described voltage safety-considered two-layer optimized energy-saving control system.
A processor for executing operations according to the instructions in the above-described voltage safety-considered double-layer optimized energy-saving control method.
It should be noted that each step/component described in the present application may be split into more steps/components, or two or more steps/components or part of the operations of the steps/components may be combined into new steps/components, as needed for implementation, to achieve the object of the present invention.
The above-described method according to the present invention may be implemented in hardware, firmware, or as software or computer code storable in a recording medium such as a CD ROM, RAM, floppy disk, hard disk, or magneto-optical disk, or as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium and to be stored in a local recording medium downloaded through a network, so that the method described herein may be stored on such software process on a recording medium using a general purpose computer, special purpose processor, or programmable or special purpose hardware such as an ASIC or FPGA. It is understood that a computer, processor, microprocessor controller, or programmable hardware includes a memory component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code that, when accessed and executed by the computer, processor, or hardware, implements the processing methods described herein. Further, when the general-purpose computer accesses code for implementing the processes shown herein, execution of the code converts the general-purpose computer into a special-purpose computer for executing the processes shown herein.
It will be readily appreciated by those skilled in the art that the foregoing is merely illustrative of the present invention and is not intended to limit the invention, but any modifications, equivalents, improvements or the like which fall within the spirit and principles of the present invention are intended to be included within the scope of the present invention.

Claims (10)

1. The double-layer optimized energy-saving control method considering voltage safety is characterized by comprising the following steps of:
step one: the upper layer establishes a power distribution system energy-saving control model considering voltage safety;
step two: the lower layer optimizes the control parameters of the upper layer based on a deep reinforcement learning algorithm, including a control domain N c And prediction domain N p Thereby determining setting parameters, and simultaneously optimizing equipment control quantity under a long time scale to adjust switching frequency of corresponding equipment parameters;
step three: and solving an upper-layer optimized power distribution system energy-saving control model based on a particle swarm optimization algorithm to obtain the optimal power regulation and control quantity of the system.
2. The voltage safety considered double-layer optimized energy-saving control method according to claim 1, characterized in that: in the first step, the method specifically comprises the following steps:
step 1-1: definition of upper layer optimization control quantity mu (k) of a power distribution network system:
Figure FDA0004099862100000011
wherein ,
Figure FDA0004099862100000012
active power for distributed power supply, +.>
Figure FDA0004099862100000013
Reactive power for a distributed power supply;
step 1-2: constructing a system predictive control model:
Figure FDA0004099862100000014
s.t.
μ min ≤μ(k+i|k)≤μ max ,i＝0,1…N c -1 (3)
Δμ min ≤Δμ(k+i|k)≤Δμ max ,i＝0,1…N c -1 (4)
y min ≤y(k+i|k)≤y max ,i＝0,1…N p -1 (5)
Figure FDA0004099862100000015
SOC min ≤SOC(k)≤SOC max (7)
Figure FDA0004099862100000021
wherein ,Nc and Np A control domain and a prediction domain respectively; q and R are cost weight matrices for controlling the objective function; mu (mu) min and μmax Respectively upper and lower limit constraint of the control quantity; Δμ (k) =μ (k+1) - μ (k); Δμ min and Δμmax Is the climbing constraint of the control quantity; y (k+i|k) is the predicted value of the voltage at time k+i based on the measured value at time k; y is min and ymax Is the system bus voltage constraint;
Figure FDA0004099862100000027
is a sensitivity matrix of the bus voltage to the control variable μ (k); Δy (k) =y (k+1) -y (k); when i > N c When Δμ (k+i) =0;SOC (k+i|k) is a predicted value of the state of energy storage at time k+i based on the measured value at time k;
Figure FDA0004099862100000022
and
Figure FDA0004099862100000023
Representing the maximum and minimum states of charge of the ith distributed energy storage, respectively; delta (k) is a charge-discharge coefficient, delta (k) =1 is energy storage discharge, delta (k) =0 is energy storage charge; η (eta) c and ηd Charge and discharge efficiency of energy storage->
Figure FDA0004099862100000024
Representing 2 norms, +. s (k+i-1|k) represents the energy amount of the stored energy.
3. The voltage safety considered double-layer optimized energy-saving control method according to claim 2, characterized in that: the upper and lower limit constraints of the control amount are specifically expressed as:
Figure FDA0004099862100000025
wherein ,Pi max 、P i min 、
Figure FDA0004099862100000026
Respectively representing the upper and lower limits of the active output and the lower limit of the reactive output of the ith distributed power supply, N s Is the number of distributed power sources.
4. The voltage safety considered double-layer optimized energy-saving control method according to claim 2, characterized in that: the climbing constraint of the control quantity is specifically expressed as follows:
Figure FDA0004099862100000031
wherein ,
Figure FDA0004099862100000032
respectively represent the charge-discharge power limit, N of the ith distributed power supply s Is the number of distributed power sources.
5. The voltage safety considered double-layer optimized energy-saving control method according to claim 2, characterized in that: in the second step, the specific steps are as follows:
step 2-1: defining an optimization objective function of the lower layer optimization as follows:
Figure FDA0004099862100000033
wherein ：Ui A node voltage which is an i-th node; u (U) N Rated voltage of the power distribution network; n is the number of PQ nodes; adopting +/-5% as the maximum safe range of node voltage;
the constraint conditions are as follows:
Figure FDA0004099862100000034
Figure FDA0004099862100000035
Figure FDA0004099862100000036
Figure FDA0004099862100000037
wherein formula (12) is a tidal current constraint, P t,i,l and Qt,i,l Respectively are provided withActive power and reactive power consumed by the load l on the node i at the moment t; p (P) t,loss and Qt,loss Active and reactive losses of the distribution network line at the time t are respectively calculated; p (P) t,M and Qt,M Respectively transmitting active power and reactive power to a main network at the moment t; p (P) t,G and Qt,G Respectively generating active power and reactive power for the distributed power supply at the moment t; q (Q) t,CB Reactive power emitted by the capacitor bank at the moment t; formula (13) is node voltage constraint, U i,min and Ui,max Respectively the minimum value and the maximum value of the voltage of the node i; u (U) i and UN Respectively node i voltage and power distribution network rated voltage; formula (14) is node power constraint, P i and Qi Active power and reactive power fed in on node i respectively; p (P) Gi and QGi Respectively integrating active output and reactive output of a distributed power supply for the node i; p (P) li and Qli Load power on node i respectively; p (P) i,min 、P i,max 、Q i,min and Qi,max The minimum value and the maximum value of the active power and the minimum value and the maximum value of the reactive power of the node i are respectively; formula (15) is a distributed power supply output constraint, P Gi,min 、P Gi,max 、Q Gi,min and QGi,max Respectively integrating the minimum value and the maximum value of the active output and the minimum value and the maximum value of the reactive output of the distributed power supply for the node i;
step 2-2: based on capacitor bank capacity of configuration nodes, the distributed power supply and capacitor bank participating power distribution network voltage control model is converted into a regulation capacitor bank and a control parameter control domain N of the regulation capacitor bank by using DQN algorithm c Prediction domain N p Comprises:
defining a state space as a set of current voltage, active power and reactive power of each PQ node;
determining the compensation amount of the configuration node parallel capacitor bank according to the capacity of the capacitor bank, and setting the action space as the compensation amount of the configuration node parallel capacitor bank and the control parameter control domain N c And prediction domain N p Is a step length of (2);
the bonus function is set to the sum of the quadratic form and the compensation amount of the capacitor bank for the more amount of voltage at each node.
6. The voltage safety considered double-layer optimized energy-saving control method according to claim 5, characterized in that: the state space is as follows:
s:{v 1 ,…,v k ,…,v n ,p 1 ,…,p k ,…,p n ,q 1 ,…,q k ,…,q n } (16)
wherein ,vk 、p k and qk The voltage value, the active power and the reactive power observed by the kth node are respectively equal to or more than 1 and equal to or less than n, and n is the total number of PQ nodes;
adopting a multi-gear capacitor bank, taking the capacity of the obtained capacitor bank as the maximum compensation amount, and taking the capacity of each gear as a set value of an action space:
Figure FDA0004099862100000051
wherein ,CBmax For the maximum compensation amount of the capacitor bank,
Figure FDA0004099862100000052
and
Figure FDA0004099862100000053
Control parameter control domains N c And prediction domain N p Is the maximum value of (2);
the reward function is:
Reward＝-[Δv 1 ,…,Δv i ,…,Δv n ]Q[Δv 1 ,…,Δv i ,…,Δv n ] T -Ra k (18)
wherein ,Δvi The threshold amount of the voltage of the node i is exceeded; a, a k Compensating the capacitor bank of node k for an amount; q and R are weight matrix and weight coefficient; deltav i The method comprises the following steps:
Figure FDA0004099862100000054
wherein 5% is the selected voltage out-of-limit safety value range.
7. The voltage safety considered double-layer optimized energy-saving control method according to claim 5, characterized in that: solving a voltage control model of the regulation capacitor bank by using an DQN algorithm to obtain a voltage optimal control strategy, wherein the method specifically comprises the following steps of:
step A1: initializing a memory bank, initializing a Q network weight parameter to omega, initializing a target Q network weight parameter to omega' =omega, and observing the voltage value, active power and reactive power of each current node to serve as an initial state s;
step A2: generating and executing an action a epsilon A according to a greedy strategy, and obtaining a reward r and a new state s' through a formula (18);
step A3: storing the transfer samples (s, a, s') in a memory bank, randomly extracting a sample(s) of minimatch from the memory bank i ,a i ,r i ,s i ')；
Step A4: order the
Figure FDA0004099862100000061
The DQN network loss function L (θ) is calculated according to:
L(θ)＝E[(TargetQ-Q(s,a；θ)) 2 ] (20)
wherein E (·) is the desired value; targetQ is a target network target value; q (s, a; omega) is the predicted value of action a under the state s when the weight parameter is omega; gamma is a discount factor;
step A5: updating a target Q network weight parameter ω' =ω using a gradient descent method;
step A6: repeatedly executing the step A2 to the step A5 until the iteration is finished, and obtaining a voltage optimal control strategy and optimal control parameters;
step A7: and uploading the optimal control parameters to an upper optimization model.
8. The double-layer optimized energy-saving control system considering voltage safety is characterized by comprising the following modules:
and a model building module: the upper layer establishes a power distribution system energy-saving control model considering voltage safety;
the equipment control amount optimizing module: the lower layer optimizes the control parameters of the upper layer based on a deep reinforcement learning algorithm, including a control domain N c And prediction domain N p Thereby determining setting parameters, and simultaneously optimizing equipment control quantity under a long time scale to adjust switching frequency of corresponding equipment parameters;
and a solving module: and solving an upper-layer optimized power distribution system energy-saving control model based on a particle swarm optimization algorithm to obtain the optimal power regulation and control quantity of the system.
9. A computer readable storage medium for storing the functional modules of the above-described voltage safety-considered two-layer optimized energy-saving control system.
10. A processor for executing operations according to the instructions in the above-described voltage safety-considered double-layer optimized energy-saving control method.
CN202310172980.2A 2023-02-27 2023-02-27 Double-layer optimization energy-saving control method and system considering voltage safety Pending CN116054154A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310172980.2A CN116054154A (en) 2023-02-27 2023-02-27 Double-layer optimization energy-saving control method and system considering voltage safety

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310172980.2A CN116054154A (en) 2023-02-27 2023-02-27 Double-layer optimization energy-saving control method and system considering voltage safety

Publications (1)

Publication Number Publication Date
CN116054154A true CN116054154A (en) 2023-05-02

Family

ID=86120209

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310172980.2A Pending CN116054154A (en) 2023-02-27 2023-02-27 Double-layer optimization energy-saving control method and system considering voltage safety

Country Status (1)

Country Link
CN (1) CN116054154A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113162053A (en) * 2021-03-18 2021-07-23 南京邮电大学 Power distribution network voltage control method and system based on load transfer and energy storage regulation
CN113541151A (en) * 2021-04-19 2021-10-22 江苏大学 Distribution microgrid layered voltage control method considering photovoltaic output uncertainty
CN114362196A (en) * 2022-03-21 2022-04-15 南京邮电大学 Multi-time-scale active power distribution network voltage control method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113162053A (en) * 2021-03-18 2021-07-23 南京邮电大学 Power distribution network voltage control method and system based on load transfer and energy storage regulation
CN113541151A (en) * 2021-04-19 2021-10-22 江苏大学 Distribution microgrid layered voltage control method considering photovoltaic output uncertainty
CN114362196A (en) * 2022-03-21 2022-04-15 南京邮电大学 Multi-time-scale active power distribution network voltage control method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HAOCHEN HUA 等: "Data-Driven Dynamical Control for Bottom-up Energy Internet System", IEEE TRANSACTIONS ON SUSTAINABLE ENERGY, vol. 13, no. 1, 31 January 2022 (2022-01-31), pages 315 - 327, XP011894473, DOI: 10.1109/TSTE.2021.3110294 *
RUONAN HU 等: "Coordinated active and reactive power control for distribution networks with high penetrations of photovoltaic systems", SOLAR ENERGY, vol. 231, 20 December 2021 (2021-12-20), pages 809 - 827, XP086917119, DOI: 10.1016/j.solener.2021.12.025 *

Similar Documents

Publication Publication Date Title
CN105846461B (en) Control method and system for large-scale energy storage power station self-adaptive dynamic planning
CN109103912B (en) Industrial park active power distribution system scheduling optimization method considering power grid peak regulation requirements
CN105680474B (en) Control method for restraining rapid power change of photovoltaic power station through energy storage
CN107994595A (en) A kind of system of peak load shifting control method and system and the application control method
CN107274085B (en) Optimal management method for energy storage equipment of double-electric ship
CN109995076B (en) Energy storage-based photovoltaic collection system power stable output cooperative control method
CN110518634A (en) Wind-powered electricity generation field control method is accessed based on the batteries to store energy system for improving exponential smoothing
CN117455183B (en) Comprehensive energy system optimal scheduling method based on deep reinforcement learning
CN113644671A (en) Dynamic power distribution control method for urban rail hybrid energy storage system based on deep reinforcement learning
CN117559526A (en) Router-simulated energy regulation and control method based on optical storage and charging integrated charging station
CN116722561A (en) Reactive power compensation system
CN118117568A (en) Composite energy management strategy suitable for off-grid wind-light-hydrogen storage coupling system
CN117713068A (en) User energy storage day-ahead optimal scheduling method and system considering application scene change
CN115147245A (en) Virtual power plant optimal scheduling method with industrial load participating in peak shaving auxiliary service
CN113629736B (en) Method for optimizing intra-day rolling based on power distribution network hydrogen energy storage system
CN113113916B (en) Multi-target energy regulation and control method in ship comprehensive power system
CN112271741B (en) Active power distribution network distributed voltage regulation method based on multi-energy storage
CN117613903A (en) User side energy storage dispatching optimization control method and device based on digital twin architecture
CN117913789A (en) Improved multi-objective optimization control method for distributed optical storage participated power distribution network
CN117674160A (en) Active power distribution network real-time voltage control method based on multi-agent deep reinforcement learning
CN113627807B (en) Network source load interaction risk control method
CN117293871A (en) Peak clipping and valley filling method and system for power distribution network by power curve battery energy storage charge and discharge
CN116054154A (en) Double-layer optimization energy-saving control method and system considering voltage safety
CN109004642B (en) Distribution network distributed energy storage evaluation method for stabilizing power fluctuation of distributed power supply
CN103679292A (en) Electricity collaborative optimization method for double batteries of intelligent micro power grid

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