CN113381405A - Micro-grid local feedback control method considering frequency recovery and voltage adjustment - Google Patents

Micro-grid local feedback control method considering frequency recovery and voltage adjustment Download PDF

Info

Publication number
CN113381405A
CN113381405A CN202110685988.XA CN202110685988A CN113381405A CN 113381405 A CN113381405 A CN 113381405A CN 202110685988 A CN202110685988 A CN 202110685988A CN 113381405 A CN113381405 A CN 113381405A
Authority
CN
China
Prior art keywords
power
frequency
voltage
frequency recovery
establishing
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
CN202110685988.XA
Other languages
Chinese (zh)
Inventor
容春艳
李军阔
郝军魁
林荣
申永鹏
王中亮
吴志
柳伟
彭杨
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.)
State Grid Corp of China SGCC
Southeast University
Economic and Technological Research Institute of State Grid Hebei Electric Power Co Ltd
Original Assignee
State Grid Corp of China SGCC
Southeast University
Economic and Technological Research Institute of State Grid Hebei Electric Power 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 State Grid Corp of China SGCC, Southeast University, Economic and Technological Research Institute of State Grid Hebei Electric Power Co Ltd filed Critical State Grid Corp of China SGCC
Priority to CN202110685988.XA priority Critical patent/CN113381405A/en
Publication of CN113381405A publication Critical patent/CN113381405A/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
    • H02J3/008—Circuit arrangements for ac mains or ac distribution networks involving trading of energy or energy transmission rights
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/12—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
    • 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
    • H02J3/241—The oscillation concerning frequency
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/388—Islanding, i.e. disconnection of local power supply from the network
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • 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
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
    • 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 microgrid local feedback control method considering frequency recovery and voltage adjustment, which comprises the following steps: step S1: analyzing an unbalanced power range based on the load frequency voltage variation characteristic; step S2: calculating the active power regulation capacity after the isolated network stably operates; step S3: establishing a local feedback construction strategy considering frequency recovery and voltage adjustment; step S4: and establishing an in-situ feedback construction strategy considering reactive power average. The microgrid local feedback control method can deduce the active power unbalance range which does not cause high-low voltage and high-low frequency protection actions according to the voltage characteristics of the load, and provides the microgrid local feedback control method for realizing frequency recovery, voltage regulation and reactive power sharing, thereby realizing local calculation of feedback income values.

Description

Micro-grid local feedback control method considering frequency recovery and voltage adjustment
Technical Field
The invention relates to the technical field of reinforcement learning local feedback construction, in particular to a micro-grid local feedback control method considering frequency recovery and voltage adjustment.
Background
Among MAS, Reinforcement Learning (RL) is the main method currently studied. Multi-agent coordination is mainly based on the learning ability of agents in a distributed environment. The reinforcement learning is an on-line learning method, which adopts a process similar to the trial-error-modification process frequently existing in human thinking, reinforces good behaviors and weak behaviors by feedback information and a proper algorithm, converges to optimal behaviors and is suitable for problems of little knowledge of learners about the environment or dynamic environment. Unlike the supervised learning technology which informs what kind of behavior is taken through positive examples and negative examples, the reinforcement learning obtains the improvement of the strategy through 'trial and error' and environment interaction, and the characteristics of self-learning and online learning make the reinforcement learning become an important branch of machine learning research.
The reinforcement learning continuously improves the performance of the intelligent agent in compliance with the environmental change in the interaction of the intelligent agent and the environment, and is an excellent algorithm for researching a multi-intelligent-agent system. In the field of electric power systems, a reinforcement learning algorithm is mainly applied in the fields of load/power prediction, power grid fault detection and diagnosis, and the role of the reinforcement learning algorithm in deeper fields such as operation control and the like is yet to be excavated. Reinforcement learning schemes use reward feedback to assess the quality of a solution. The learner proposes a multi-agent combined action reinforcement learning algorithm to solve the problem of the wind-solar hybrid power generation system in a distributed view. The learner proposes an MAS reinforcement learning model and a Q algorithm based on the element countermeasures, overcomes the defects of the Q learning algorithm under Nash balance, and realizes the MAS learning problem of the element countermeasure model of the nonzero and Markov countermeasures. In response to this situation, a method for local feedback control of a microgrid involving frequency recovery and voltage adjustment is proposed.
Disclosure of Invention
The invention aims to provide a microgrid on-site feedback control method considering frequency recovery and voltage regulation, which analyzes an unbalanced power range based on load frequency and voltage change characteristics, calculates the active power regulation capability after isolated grid stable operation, and establishes an on-site feedback construction strategy considering frequency recovery and voltage regulation and an on-site feedback construction strategy considering reactive power sharing.
The purpose of the invention can be realized by the following technical scheme:
a microgrid on-site feedback control method taking frequency recovery and voltage regulation into account, comprising the steps of:
step S1: analyzing an unbalanced power range based on the frequency and voltage change characteristics of the load according to the power change between the power supply and the load when the isolated network is formed;
step S2: calculating the active power regulation capacity after the isolated network stably operates according to the unbalanced power range obtained in the step S1;
step S3: establishing a local feedback construction strategy considering frequency recovery and voltage adjustment according to the active power regulation capacity obtained in the step S2;
step S4: and establishing an in-situ feedback construction strategy considering reactive power sharing according to the active power and reactive power needing to be adjusted obtained in the step S3.
Further, in step S1, the unbalanced power range is an active unbalanced range and a reactive unbalanced range.
Further, prior to arc network formation there are:
Figure BDA0003124687590000031
after the isolated network is formed, the voltage at the two ends of the load is:
Figure BDA0003124687590000032
substituting formula (2) into formula (1), wherein V' is in the range of [ Vmin,Vmax]Obtaining:
Figure BDA0003124687590000033
the range calculated by equation (3) is:
Figure BDA0003124687590000034
the ratio of the active power shortage to the output of the generator in the isolated network is in the range of the formula (4), and voltage instability can not occur after the isolated network is formed;
in an actual power grid, the output of the generator changes along with the magnitude of unbalanced power, and the power P of the generator is the actual active output after the output changes.
Further, the reactive imbalance is represented as:
Figure BDA0003124687590000035
after formation of isolated network, f0Has a range of [ fmin,fmax]To obtain:
Figure BDA0003124687590000036
the range of Δ Q is calculated from equation (6):
Figure BDA0003124687590000037
further, the active power regulating capacity calculation process after the isolated network stably operates is divided into active power regulating capacity calculation after the single isolated network stably operates and active power regulating capacity calculation after the multi-machine isolated network stably operates.
Further, in the single isolated grid, the active power-frequency static characteristic relation of the system is as follows:
Figure BDA0003124687590000041
steam turbine generator
Figure BDA0003124687590000042
A value of 16.6 to 25, defined as:
Figure BDA0003124687590000043
Figure BDA0003124687590000044
the frequency adjustment effect coefficient of the load is 1-3, and the frequency adjustment effect coefficient is defined as:
Figure BDA0003124687590000045
obtaining the active power regulating capacity delta P for defining the isolated networkL0Comprises the following steps:
Figure BDA0003124687590000046
further, the active regulation capacity of the multiple isolated networks can be defined as the sum of the regulation capacities of all generators in the isolated networks;
Figure BDA0003124687590000047
the regulation effect of the isolated network cannot make up for the power shortage, the frequency can be rapidly reduced, the load must be timely reduced, and the redundant load is cut off. .
Further, in step S3, the in-situ feedback construction strategy process for establishing frequency recovery and voltage adjustment is as follows:
step S31, establishing a frequency recovery and voltage regulation local feedback construction strategy state set and an action set;
step S32, build in-place feedback for frequency recovery and voltage regulation to build a policy reward function.
Further, in step S31, when performing frequency recovery, the state of the microgrid is defined based on the frequency deviation of the system:
1) the frequency is lower than 49.80Hz or higher than 50.20Hz, and the requirement of the operating frequency of the power system is not met;
2) the requirement of the operating frequency of the power system is met, but the requirement of the operating frequency of the power system adopting automatic control is not met, and the frequency is between 49.80Hz and 49.85Hz, and between 50.15Hz and 50.20 Hz;
3) the frequency is between 49.99Hz and 50.01Hz, which is a stable interval and does not need to be adjusted;
4) the frequency is required to be adjusted between 49.85 Hz-49.99 Hz and 50.01 Hz-50.15 Hz;
7 states are divided for the state of the microgrid, and the state space sf is defined as:
{(-∞,-0.2),[-0.2,-0.15),[-0.15,-0.01),[-0.01,0.01],(0.01,0.15],(0.15,0.2],(0.2,∞)}；
when the frequency is recovered, the environment state input variable is the frequency deviation delta f of the microgrid, the output action variable of the controller is secondary control power delta P, the delta P can be discretized into a series of tiny values { delta P1, delta P2, … and delta Pm }, and the corresponding action set A is { a1, a2, … and am };
setting the target of frequency quadratic control to be (50 +/-0.00008) Hz;
the controllable power supply in the microgrid can be used for the capacity of frequency modulation control, and an action set A1 of the microgrid frequency modulation control is designed, namely the power variation of secondary control: correction Δ P for P;
the designed action set a1 is: a1 { -2000, -1000, -200, -100, -20, -2,0,2,20,100,200,1000,2000} W;
the adjustment quantity delta P in the secondary control process is the action quantity of secondary adjustment of DG each time, and is the final result obtained in the secondary control, and the frequency generated by droop control can be recovered;
when voltage is adjusted, the environment state input variable is frequency deviation delta u of the microgrid, the output action variable of the controller is secondary control power delta Q, the delta Q can be discretized into a series of tiny values of { delta Q1, delta Q2, … and delta Qn }, and the corresponding action set A is { a1, a2, … and am };
the reactive action set a2 is designed to be a2 ═ a1, a2, …, a19 [ -0.08, -0.07, -0.06, -0.05, -0.04, -0.03, -0.02, -0.015, -0.01, -0.005, 0.0005, 0.0005, 0.005, 0.01, 0.015, 0.02, 0.04, 0.06, 0.08 ].
The secondary control power change Δ Q is calculated according to the following equation:
ΔQ＝KQ*K∈A2 (13)
wherein: q is the rated reactive power of each DG inverter.
Further, in the step S32, the secondary control is targeted to:
1) restoring the frequency to a nominal value;
2) regulate and control the voltage to return to the optimum state.
Two reward functions are set to respectively realize the two goals, and the local return function of reinforcement learning is defined as the problem of coordinating frequency recovery and voltage regulation.
When frequency recovery is carried out, the local reward value rj (si, ai) represents a value obtained by the jth distributed power supply taking ai action in the si state, and is defined as:
Figure BDA0003124687590000061
Δ f represents the difference between the actual frequency of the system and the specified frequency of 50Hz, α 1, α 2, α 3, β 1 and β 2 are all fixed parameters, α 3> α 2> α 1, β 1< β 2, the larger the frequency deviation, the smaller the reward.
Based on the local reward definition above, the global reward Rg is defined as follows:
Figure BDA0003124687590000062
and Rg selects the current action to control each inverter when the global reward value is maximum in each learning.
Further, in the step S4, the process of establishing the local feedback construction strategy considering the reactive power share includes:
step S41, establishing an in-situ feedback construction strategy state set and an action set considering reactive power equipartition;
and step S42, establishing an in-situ feedback construction strategy reward function considering reactive power average.
Further, in the step S41, the reactive power generated by each distributed photovoltaic in the microgrid is regarded as an environmental state, the reactive deviation Δ Qreac is divided into a series of discrete intervals, which are { Δ Q1, Δ Q2, …, Δ Qm }, and the corresponding environmental state set S is { S1, S2, … sm };
the state set for each DG is represented as:
S＝{s1,s2,...,sm}
＝{|Q1_reac-Qavg|,|Q2_reac-Qavg|,...,|Qm_reac-Qavg|} (16)
Qavg＝Qload/N (17)
wherein: qm _ reac is the reactive output value of PVi; qavg is the average reactive power of the micro-grid during stable operation; qload is the reactive load of the system; n is the total number of the distributed power supplies;
the more the set S setting elements are, the finer and more precise the micro-grid operation state is divided; the excessive number of elements in the set can prolong the learning period and influence the online analysis control;
the alternating-current micro-grid is separated from the main grid, the frequency under the steady-state working condition is a global quantity, the active power output of the distributed power supply which operates in parallel is accurately distributed according to the droop coefficient, and the active power is irrelevant to the equivalent impedance;
the output voltage is a local variable, and the reactive deviation among the PVs is determined by the PV output voltage amplitude difference, the line resistance difference and the micro-grid structure;
the reactive voltage droop control action set A is a set of action strategies for enabling a certain state s of the microgrid to be changed into a better state s' at the current moment;
the set of actions for each distributed power source is defined as:
A＝{a1,a2,...,aL}＝{kqQ1,kqQ2,...,kqQL} (18)
wherein: ajj kqQij, jj ∈ {1, …, L }; subscript L is the number of actions.
Further, in step S42, the value of the reward immediately after the action is executed directly affects the Q value, and the Q value can directly reflect the performance of the selected action, which produces two results with significant differences:
1) if the system still has the condition that the reactive power sharing is not realized, which indicates that Q learning is not finished, setting the reward value ri corresponding to the total reactive power deviation delta Qtotal as a negative value, and punishing;
2) the system realizes reactive power equalization and sets the reward value corresponding to delta Qtotal to zero.
The prize value is defined using Δ Qtotal as follows:
Figure BDA0003124687590000081
Figure BDA0003124687590000082
and delta Qtotal is the sum of absolute values of differences between Qi, reac and the microgrid mean value Qavg of all the distributed photovoltaics.
Furthermore, a microgrid local feedback controller considering frequency recovery and voltage adjustment stores a microgrid local feedback control method program for operating the microgrid local feedback controller considering frequency recovery and voltage adjustment.
The invention has the beneficial effects that:
the microgrid local feedback control method can deduce the active power unbalance range which does not cause high-low voltage and high-low frequency protection actions according to the voltage characteristics of the load, and provides the microgrid local feedback control method for realizing frequency recovery, voltage regulation and reactive power sharing, thereby realizing local calculation of feedback income values.
Drawings
The invention will be further described with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a microgrid local feedback control method.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A micro-grid local feedback control method considering frequency recovery and voltage adjustment comprises the following steps:
step S1: the unbalanced power range based on the load frequency voltage variation characteristics is analyzed. When the isolated network is formed, due to the fact that power changes exist between the power supply and the load, frequency and voltage in the microgrid change, and when the frequency and voltage change is too large, the protection device acts immediately, so that the microgrid cannot run stably. Therefore, the unbalanced power range should be determined when the microgrid is formed so as to satisfy the microgrid formation condition.
Step S2: and calculating the active power regulation capacity after the isolated network stably operates. Step S1 shows that the unbalanced power within a certain range can make the isolated grid operate stably after forming, and at this time, the microgrid has a certain active power regulation capability to automatically recover the frequency. When disturbance occurs, the frequency can be recovered to be stable under the action of the microgrid, but the previous steady-state operation frequency is not returned.
Step S3: an in-place feedback construction strategy is established that accounts for frequency recovery and voltage regulation. After disturbance occurs, in addition to the inertial regulation in step S2, the microgrid also needs to adjust its own active and reactive power outputs to perform frequency modulation and voltage regulation, so that the system further recovers to the optimal state (the frequency recovers to the rated value, and the voltage recovers to the proper state). This strategy is established to control the power supply to output corresponding active and reactive power, depending on the frequency and voltage variations in the ambient conditions.
Step S4: and establishing an in-situ feedback construction strategy considering reactive power average. The real and reactive power that needs to be regulated has been derived in step S3. Due to the fact that the number of distributed power sources in the new energy microgrid is large, a reactive power sharing strategy is established, and a proper state set is divided according to the number of the power sources, so that the controller can rapidly and finely correspond to the reactive power state needing to be adjusted.
In step S1, the step of analyzing the unbalanced power range based on the load frequency voltage variation characteristic is as follows:
step S11, analyzing the active imbalance range;
and step S12, analyzing the reactive unbalance range.
In step S11, before the isolated network is formed, there are:
Figure BDA0003124687590000101
after the isolated network is formed, the voltage at the two ends of the load is:
Figure BDA0003124687590000102
formula (2) is substituted for formula (1), and the range of V' is considered as [ Vmin,Vmax]It is possible to obtain:
Figure BDA0003124687590000103
in the formula, V is the voltage of the load when the system operates normally, and is generally 1 pu; vminAnd VmaxRespectively taking 0.8pu and 1.2pu as the minimum and maximum voltage values of the load allowed by stable operation after the isolated network is formed; and P is the total active power generated when all the generators in the isolated network are networked to run. The range that can be calculated from equation (3) is:
Figure BDA0003124687590000104
therefore, when the ratio of the active deficit to the output of the generator in the isolated network is within the range of the formula (4), voltage instability can not occur after the isolated network is formed. In an actual power grid, the output of the generator changes along with the magnitude of unbalanced power, and the power P of the generator is the actual active output after the output changes.
In step S12, the reactive imbalance may be expressed as:
Figure BDA0003124687590000105
after isolated network formation, consider f0Has a range of [ fmin,fmax]To obtain:
Figure BDA0003124687590000111
in the formula, f is the frequency of the system in normal operation, and is generally 50 Hz; f. ofminAnd fmaxThe allowable frequency ranges for stable operation after the isolated network is formed are 47Hz and 55 Hz. Get QfAs 1.5, Δ Q can be calculated from equation (6) in the range:
Figure BDA0003124687590000112
as can be seen from equations (4) - (7), the voltage and frequency changes after the isolated network is formed are related to reactive power imbalance, and when the voltage and frequency changes beyond the allowable range, high/low frequency and high/low voltage protection actions are triggered, and the isolated network cannot stably operate. When the voltage and frequency change values are within the allowable range, the stable operation can be stably carried out at a new stable operating point through the adjustment of the isolated network. Comparing equations (7) and (4) it can be seen that reactive power imbalances have a more sensitive effect on voltage and frequency variations than active imbalances.
In step S2, the process of calculating the active power adjustment capability after the isolated grid stable operation is as follows:
step S21, calculating the active power regulation capacity after the single machine isolated network stably operates;
and step S22, calculating the active power regulation capacity after the multi-machine isolated network stably operates.
In step S21, the active power-frequency static characteristic relation of the system is:
Figure BDA0003124687590000113
Figure BDA0003124687590000114
the power is regulated for the system unit (per unit value), which represents the amount of load change that causes the frequency unit change when the regulating effect of the genset and load is taken into account, according to
Figure BDA0003124687590000115
The magnitude of the value can determine the amount of change in the load that the system can withstand within the allowable frequency range,
Figure BDA0003124687590000116
the larger the value, the smaller the frequency change caused by the load increase and decrease, the more stable the system frequency,
Figure BDA0003124687590000117
frequency regulation effect of load
Figure BDA0003124687590000118
Unit regulation power of generator set
Figure BDA0003124687590000119
And (4) summing.
kr＝PGN/PLNIs a system spare coefficient, which expresses the ratio of rated capacity of a generating set to the total active load at rated frequency of the system, and under the condition of spare capacity (k)rMore than 1), the unit regulating power of the system is correspondingly increased, and the output of the active power source of the isolated network meets the requirement of the system on the active power under the rated frequency and also has certain active spare capacity in order to adapt to the increase of the load in the isolated network.
Figure BDA0003124687590000121
The power is regulated for the unit of the generator set, the value being regulated by the generatorA speed characteristic determination which reflects the amount of change in the output power of the generator per unit change in frequency,
Figure BDA0003124687590000122
the larger the value, the smaller the frequency offset corresponding to the same power change. The steam turbine generator is generally 16.6-25. It is defined as:
Figure BDA0003124687590000123
Figure BDA0003124687590000124
the frequency adjustment effect coefficient of the load is generally 1-3, which is defined as:
Figure BDA0003124687590000125
thus, the active regulation capacity Δ P of the isolated network is definedL0Comprises the following steps:
Figure BDA0003124687590000126
wherein P isGN、PLNThe installed capacity and initial load level of the system;
ΔPG、ΔPLthe change of the mechanical power of the isolated network generator and the change of the load active power along with the frequency are realized;
fNis a rated frequency;
and deltaf is the frequency deviation after isolated network stabilization, and the value is related to the regulation range of the primary frequency modulation and is generally less than 0.2 Hz.
In step S22, when a plurality of generators are included in the grid, the disturbance instantaneous power shortage is distributed to each generator rotor according to the rotational inertia, the load rotational inertia also bears a part of the load change, the rotation speeds of all the rotors change according to the same acceleration, in the process from the disturbance instantaneous to the stabilization, the generators bear load increment according to the inverse ratio of the respective difference adjustment coefficients, and the active power adjustment capability of the multi-machine grid can be defined as the sum of the adjustment capabilities of all the generators in the grid.
Figure BDA0003124687590000131
If the regulation effect of the isolated network is not enough to compensate the power shortage, the frequency can be rapidly reduced, and in order to avoid frequency collapse, a load reduction measure must be taken in time to cut off redundant load.
In step S3, the in-situ feedback construction strategy process for establishing frequency recovery and voltage adjustment is as follows:
step S31, establishing a frequency recovery and voltage regulation local feedback construction strategy state set and an action set;
step S32, build in-place feedback for frequency recovery and voltage regulation to build a policy reward function.
In step S31, when frequency recovery is performed, the state of the microgrid is defined based on the frequency deviation of the system:
wherein the frequency deviation is out of the national allowable range, i.e. greater than 0.2 Hz;
the method meets the requirement of the operating frequency of the power system, but does not meet the requirement of the operating frequency of the power system adopting automatic control, namely the frequency deviation is 0.2Hz to 0.15 Hz.
Finally, the frequency deviation of 0.01Hz is taken as a demarcation point, the frequency deviation smaller than 0.01Hz is taken as a stable interval, and the frequency adjustment is not needed; frequency deviation between 0.01Hz and 0.15Hz requires some frequency adjustment to prevent the system frequency deviation from the specified range.
Therefore, the micro-grid state division standard is used, and 7 states are defined in total. The state space sf is defined as:
{(-∞,-0.2),[-0.2,-0.15),[-0.15,-0.01),[-0.01,0.01],(0.01,0.15],(0.15,0.2],(0.2,∞)}。
when the power grid has serious faults, the frequency and the voltage are rapidly reduced, the stability control device integrates the frequency and the voltage value of the power grid during the faults, partial loads are cut off according to a control strategy, the system frequency and the voltage value are recovered in time, and the secondary control of the micro-grid realizes the reduction of the frequency deviation of the system by controlling each distributed power supply in the micro-grid to output active power.
When the frequency is recovered, the environment state input variable is the frequency deviation delta f of the microgrid, and the output action variable of the controller is the secondary control power delta P. Δ P may be discretized into a series of fractional values { Δ P1, Δ P2, …, Δ Pm }, with the corresponding action set A being { a1, a2, …, am }.
If the division of the discrete value is too fine, the dimension of the state-action pair is too high, the system is subjected to dimension disaster in real-time optimization, and the requirement of rapid convergence of the algorithm cannot be met; over-sparseness, i.e., reducing the dimension of the state-action pair, is not favorable for frequency recovery, and reduces the precision of the controller, so that it is critical to effectively and reasonably discretize the state space and the action space, and the target of frequency quadratic control is set to (50 ± 0.00008) Hz.
Then, according to the capacity of the controllable power supply in the microgrid, which can be used for frequency modulation control, an action set A1 of the microgrid frequency modulation control is designed, namely the variation of the secondary control power: correction Δ P for P.
The action set A1 designed by the invention is as follows: a1 { -2000, -1000, -200, -100, -20, -2,0,2,20,100,200,1000,2000} W.
The adjustment amount Δ P in the secondary control process is an operation amount of secondary adjustment for each DG, that is, a final result of constant correction obtained by using reinforcement learning in the secondary control, and such secondary control is significant in order to recover the frequency of occurrence of droop control.
When voltage adjustment is carried out, the environment state input variable is the frequency deviation delta u of the microgrid, the output action variable of the controller is secondary control power delta Q, the delta Q can be discretized into a series of tiny values of { delta Q1, delta Q2, … and delta Qn }, and the corresponding action set A is { a1, a2, … and am }.
Taking the above factors and the V-Q characteristic parameters of each inverter into consideration, the designed reactive action set a2 is a2 ═ a1, a2, …, a19 [ -0.08, -0.07, -0.06, -0.05, -0.04, -0.03, -0.02, -0.015, -0.01, -0.005, 0.0005, 0.0005, 0.005, 0.01, 0.015, 0.02, 0.04, 0.06, 0.08 ].
The secondary control power change Δ Q is then calculated according to the following equation:
ΔQ＝KQ*K∈A2 (13)
in formula (13): q is the rated reactive power of each DG inverter.
In step S32: for autonomous droop control MMG, which may result in frequency and voltage deviations during main control, the goals of secondary control are:
1) restoring the frequency to a nominal value;
2) regulate and control the voltage to return to the optimum state.
Two reward functions are set to achieve the two goals respectively, and therefore, the local reward function of reinforcement learning is defined as a problem of coordinating frequency recovery and voltage regulation.
When frequency recovery is performed, the local reward value rj (si, ai) represents a value obtained by the jth distributed power supply taking ai action in the si state, and is defined as:
Figure BDA0003124687590000151
in the formula (14), Δ f represents the difference between the actual frequency of the system and the specified frequency of 50Hz, and α 1, α 2, α 3, β 1 and β 2 are all fixed parameters, wherein α 3> α 2> α 1, β 1< β 2, i.e. the larger the frequency deviation, the smaller the reward.
Based on the local reward definition above, the global reward Rg is defined as follows:
Figure BDA0003124687590000152
in formula (15): rf is a global frequency reward; ru is the global voltage reward; ri, f, ri and u are local rewards, and Rg means that the current action is selected to control each inverter when the global reward value is maximum in each learning.
In step S4, the process of establishing the in-situ feedback construction strategy considering the reactive power average is as follows:
step S41, establishing an in-situ feedback construction strategy state set and an action set considering reactive power equipartition;
and step S42, establishing an in-situ feedback construction strategy reward function considering reactive power average.
In step S41, the reactive power generated by each distributed photovoltaic in the microgrid is regarded as an environmental condition, and the reactive deviation Δ Qreac may be divided into a series of discrete intervals, such as { Δ Q1, Δ Q2, …, Δ Qm }, and the corresponding environmental condition set S is { S1, S2, … sm }.
The state set for each DG is represented as:
S＝{s1,s2,...,sm}
＝{|Q1_reac-Qavg|,|Q2_reac-Qavg|,...,|Qm_reac-Qavg|} (16)
Qavg＝Qload/N (17)
in formulae (16) and (17): qm _ reac is the reactive output value of PVi; qavg is the average reactive power of the micro-grid during stable operation; qload is the reactive load of the system; and N is the total number of the distributed power supplies.
The more the set S setting elements are, the finer and more precise the micro-grid operation state is divided; however, the number of elements in the set is too large, which greatly prolongs the learning period and further affects the online analysis control.
When the alternating-current micro-grid is separated from the main grid, the frequency under the steady-state working condition is a global quantity, the active power output of the distributed power supplies running in parallel is accurately distributed according to the droop coefficient, namely the active power is unrelated to the equivalent impedance. Considering that the output voltage is a local variable, the reactive deviation among the PVs is determined by the PV output voltage amplitude difference, the line resistance difference, the micro-grid structure and other factors.
The reactive voltage droop control action set A means that: a set of action strategies that transition the microgrid to a better state s' over a certain state s at the present moment.
The action set for each distributed power supply is defined as:
A＝{a1,a2,...,aL}＝{kqQ1,kqQ2,...,kqQL} (18)
in formula (18): ajj kqQij, jj ∈ {1, …, L }; subscript L is the number of actions.
In step S42, the Q value is directly affected by the reward value immediately after the action is executed, and the Q value can directly reflect the performance of the selected action, and for the problem of reactive power sharing considering that the voltage is not out of limit, two different results are generated after the action selected based on the Q algorithm is executed:
(1) the system still has the reactive power sharing which is not realized, which indicates that Q learning is not finished, and for this reason, the reward value ri corresponding to the total reactive power deviation delta Qtotal is set as a negative value, namely punishment is carried out;
(2) and the system realizes reactive power sharing, and sets the reward value corresponding to the delta Qtotal to zero.
The prize value is defined using Δ Qtotal, as shown in equations (19) and (20):
Figure BDA0003124687590000171
Figure BDA0003124687590000172
in formulae (19) and (20): b 1-b 4 and mu 1-mu 4 are fixed parameters, and b1< b2< b3< b 4;
and delta Qtotal is the sum of absolute values of differences between Qi, reac and the microgrid mean value Qavg of all the distributed photovoltaics.
In the description herein, references to the description of "one embodiment," "an example," "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.

Claims (10)

1. A microgrid local feedback control method considering frequency recovery and voltage adjustment is characterized by comprising the following steps:
step S1: analyzing an unbalanced power range based on the frequency and voltage change characteristics of the load according to the power change between the power supply and the load when the isolated network is formed;
step S2: calculating the active power regulation capacity after the isolated network stably operates according to the unbalanced power range obtained in the step S1;
step S3: establishing a local feedback construction strategy considering frequency recovery and voltage adjustment according to the active power regulation capacity obtained in the step S2;
step S4: and obtaining the active power and the reactive power which need to be regulated according to the in-situ feedback construction strategy which is established in the step S3 and takes the frequency recovery and the voltage regulation into account, and establishing the in-situ feedback construction strategy which takes the reactive power sharing into consideration.
2. The method of claim 1, wherein in step S1, the unbalanced power ranges are an active unbalanced range and a reactive unbalanced range.
3. The method of claim 1, wherein in step S2, the calculation process of the active power regulation capability after isolated network stable operation is divided into the calculation of the active power regulation capability after single isolated network stable operation and the calculation of the active power regulation capability after multiple isolated network stable operation.
4. The method of claim 1, wherein in step S3, the local feedback construction strategy for frequency recovery and voltage adjustment comprises:
step S31, establishing a frequency recovery and voltage regulation local feedback construction strategy state set and an action set;
and step S32, establishing a local feedback construction strategy reward function of frequency recovery and voltage regulation according to the local feedback construction strategy state set and the action set established in the step S31.
5. The method of claim 4, wherein the state space sf in the step S31 is defined as: { (∞, -0.2), [ -0.2, -0.15), [ -0.15, -0.01), [ -0.01,0.01], (0.01,0.15], (0.15,0.2], (0.2, ∞) };
the designed action set a1 is: a1 { -2000, -1000, -200, -100, -20, -2,0,2,20,100,200,1000,2000} W;
the designed reactive action set a2 is a2 ═ a1, a2, …, a19 [ -0.08, -0.07, -0.06, -0.05, -0.04, -0.03, -0.02, -0.015, -0.01, -0.005, 0.0005, 0.0005, 0.005, 0.01, 0.015, 0.02, 0.04, 0.06, 0.08 ].
6. The method as claimed in claim 4, wherein when the frequency recovery is performed in step S32, the local reward value rj is defined as:
Figure FDA0003124687580000021
the global reward Rg is defined as:
Figure FDA0003124687580000022
7. the method of claim 1, wherein in step S4, the method of establishing the feedback strategy considering the reactive power average comprises:
step S41, establishing an in-situ feedback construction strategy state set and an action set considering reactive power equipartition;
and step S42, establishing an in-situ feedback construction strategy state set and an action set according to the step S41, and establishing an in-situ feedback construction strategy reward function considering reactive power sharing.
8. The method of claim 7, wherein in step S41, the state set of each DG is represented as:
Figure FDA0003124687580000031
Qavg＝Qload/N (17)
the set of actions for each distributed power source is defined as:
A＝{a1,a2,...,aL}＝{kqQ1,kqQ2,...,kqQL} (18)。
9. the method of claim 7, wherein in step S42, Δ Qtotal is used to define the reward value:
Figure FDA0003124687580000032
Figure FDA0003124687580000033
10. a microgrid local feedback controller in consideration of frequency recovery and voltage regulation, storing a program for operating the microgrid local feedback control method in consideration of frequency recovery and voltage regulation according to any one of claims 1 to 9.
CN202110685988.XA 2021-06-21 2021-06-21 Micro-grid local feedback control method considering frequency recovery and voltage adjustment Pending CN113381405A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110685988.XA CN113381405A (en) 2021-06-21 2021-06-21 Micro-grid local feedback control method considering frequency recovery and voltage adjustment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110685988.XA CN113381405A (en) 2021-06-21 2021-06-21 Micro-grid local feedback control method considering frequency recovery and voltage adjustment

Publications (1)

Publication Number Publication Date
CN113381405A true CN113381405A (en) 2021-09-10

Family

ID=77578046

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110685988.XA Pending CN113381405A (en) 2021-06-21 2021-06-21 Micro-grid local feedback control method considering frequency recovery and voltage adjustment

Country Status (1)

Country Link
CN (1) CN113381405A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113890057A (en) * 2021-09-18 2022-01-04 国网河北省电力有限公司经济技术研究院 Control method and device based on multi-microgrid collaborative optimization and storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111355234A (en) * 2020-03-18 2020-06-30 国网浙江嘉善县供电有限公司 Micro-grid frequency control method based on reinforcement learning
CN111431216A (en) * 2020-03-18 2020-07-17 国网浙江嘉善县供电有限公司 High-proportion photovoltaic microgrid reactive power sharing control method adopting Q learning

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111355234A (en) * 2020-03-18 2020-06-30 国网浙江嘉善县供电有限公司 Micro-grid frequency control method based on reinforcement learning
CN111431216A (en) * 2020-03-18 2020-07-17 国网浙江嘉善县供电有限公司 High-proportion photovoltaic microgrid reactive power sharing control method adopting Q learning

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
史建勋 等: "高比例光伏微网无功均分控制中的 Q 学习方法", 《电力系统及其自动化学报》, 18 August 2020 (2020-08-18), pages 1 - 2 *
吴作君: "钢铁厂自备电厂孤网稳定运行研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, 15 November 2013 (2013-11-15), pages 11 - 13 *
沈珺 等: "基于强化学习的多微电网分布式二次优化控制", 《电力系统自动化》, 10 March 2020 (2020-03-10), pages 198 - 206 *
赵凤贤等: "微网孤岛模式下负荷分配的改进控制策略", 《中国电力》, no. 04, 5 April 2013 (2013-04-05) *
魏金花: "基于PSASP的甘南电网孤网运行频率稳定控制研究", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》, 31 March 2017 (2017-03-31), pages 5 - 13 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113890057A (en) * 2021-09-18 2022-01-04 国网河北省电力有限公司经济技术研究院 Control method and device based on multi-microgrid collaborative optimization and storage medium

Similar Documents

Publication Publication Date Title
Chen et al. Model-free emergency frequency control based on reinforcement learning
Hu et al. Grid-forming inverter enabled virtual power plants with inertia support capability
CN109256801B (en) Virtual inertia and virtual damping coefficient self-adaptive control method for virtual synchronous generator
EP2703914A1 (en) Hybrid control system and method for automatic voltage regulator
Moafi et al. Energy management system based on fuzzy fractional order PID controller for transient stability improvement in microgrids with energy storage
Vetoshkin et al. Dynamic stability improvement of power system by means of STATCOM with virtual inertia
US20160141877A1 (en) Method and device for controlling power generators of a subgrid within an interconnected grid
Shan et al. A switching event-triggered resilient control scheme for primary and secondary levels in AC microgrids
Hasan et al. An improved adaptive hybrid controller for battery energy storage system to enhance frequency stability of a low inertia grid
CN113381405A (en) Micro-grid local feedback control method considering frequency recovery and voltage adjustment
Hajizadeh et al. Intelligent robust control of hybrid distributed generation system under voltage sag
Ghasemi et al. Intelligent and optimal energy management strategy to control the Micro-Grid voltage and frequency by considering the load dynamics and transient stability
Oshnoei et al. Intelligent power control of inverter air conditioners in power systems: A brain emotional learning-based approach
Beus et al. A model predictive control approach to operation optimization of an ultracapacitor bank for frequency control
Lyu et al. Fuzzy control based virtual synchronous generator for self-adaptative control in hybrid microgrid
Umrao et al. Load frequency control using polar fuzzy controller
CN105356472B (en) Online frequency synthesis control method based on source lotus characteristic
Jafari et al. Adaptive intelligent secondary control of microgrids using a biologically-inspired reinforcement learning
Karthikeyan et al. Load frequency control for three area system with time delays using fuzzy logic controller
Oğuz Fuzzy PI Control with Parallel Fuzzy PD Control for Automatic Generation Control of a Two-Area Power System
CN109713664B (en) Network source coordination control strategy calculation method and system with stable direct current island frequency
CN107515532B (en) System self-adaptive control method considering action threshold
Sengupta et al. Delay dependent wide area damping controller using deep learning technique
Renduchintala et al. Fuzzy theory based distributed power flow conditioner to enhance power system dynamic stability
Chen et al. A transient stability analysis method for wind power system with thyristor controlled phase shifting transformer based on branch potential energy index

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210910