CN114202121A - Low-carbon operation method of power grid system and related device thereof - Google Patents
Low-carbon operation method of power grid system and related device thereof Download PDFInfo
- Publication number
- CN114202121A CN114202121A CN202111519546.4A CN202111519546A CN114202121A CN 114202121 A CN114202121 A CN 114202121A CN 202111519546 A CN202111519546 A CN 202111519546A CN 114202121 A CN114202121 A CN 114202121A
- Authority
- CN
- China
- Prior art keywords
- information matrix
- node
- grid system
- optimal
- optimization
- 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
Links
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 93
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000011159 matrix material Substances 0.000 claims abstract description 112
- 238000005457 optimization Methods 0.000 claims abstract description 68
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims abstract description 61
- 230000007787 long-term memory Effects 0.000 claims abstract description 22
- 230000006403 short-term memory Effects 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 15
- 230000006870 function Effects 0.000 claims description 30
- 230000009471 action Effects 0.000 claims description 23
- 230000015654 memory Effects 0.000 claims description 20
- 238000003860 storage Methods 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 description 6
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 230000002787 reinforcement Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000005261 decarburization Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- 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
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application discloses a low-carbon operation method of a power grid system and a related device thereof, wherein after initial parameters are configured, active requirements and reactive requirements of current optimization tasks are input; classifying the current optimization task to obtain a source task and a new task, and randomly generating an initial information matrix of the source task; performing global optimization on the initial information matrix by taking the lowest active power loss, the lowest carbon emission and the lowest voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix; training the long-term and short-term memory network through the optimal information matrix and the information labels of the source tasks to obtain the trained long-term and short-term memory network; generating an initial information matrix of the new task through the trained long and short term memory network, and carrying out global optimization on the initial information matrix to obtain an optimal information matrix of the new task; and acquiring an optimal carbon emission strategy according to the optimal information matrix of the new task, and executing the optimal carbon emission strategy, so that the technical problem of overlong search time in the prior art is solved.
Description
Technical Field
The application relates to the technical field of power grids, in particular to a low-carbon operation method of a power grid system and a related device thereof.
Background
In the world energy prospect of 2019, the international energy agency proposes that governments must take urgent action to achieve the decarburization goal, and wind energy and solar energy are expected to become major power sources by 2040 years. In order to solve the contradiction between energy and environment, the power industry must change the energy structure and reduce the pollution and the atmospheric emission level at present with the increasing public awareness of environmental protection.
Because the artificial intelligence optimization algorithm is developed more and more mature at present, deep learning and reinforcement learning are generally applied to various engineering fields and good effects are achieved. In the prior art, through reinforcement learning, the economic cost is minimum, the active network loss is minimum, the carbon emission is minimum, and more favorable decision support is provided for the scheduling of the power system. When solving a reactive power optimization problem, the technical problem of overlong search time easily exists.
Disclosure of Invention
The application provides a low-carbon operation method of a power grid system and a related device thereof, which are used for solving the technical problem of overlong search time in the prior art.
In view of this, the first aspect of the present application provides a low-carbon operation method for a power grid system, including:
s1, inputting the active demand and the reactive demand of the current optimization task after configuring the initial parameters;
s2, classifying the current optimization task to obtain a source task and a new task, and randomly generating an initial information matrix of the source task;
s3, carrying out global optimization on the initial information matrix by taking the minimum active power loss, the minimum carbon emission and the minimum voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix;
s4, training a long-short term memory network through the optimal information matrix and the information labels of the source tasks to obtain a trained long-short term memory network;
s5, generating an initial information matrix of the new task through the trained long-short term memory network, and skipping to execute the step S3 to obtain an optimal information matrix of the new task;
and S6, acquiring an optimal carbon emission strategy according to the optimal information matrix of the new task, and executing the optimal carbon emission strategy.
Optionally, the performing global optimization on the initial information matrix by using the minimum active power loss, the minimum carbon emission, and the minimum voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix includes:
setting a fitness function according to an optimization target of the power grid system with the lowest active power loss, the lowest carbon emission and the lowest voltage stability value, and setting constraint conditions;
performing action selection through a plurality of global searchers, transferring from one state to another state, and calculating a fitness value when each global searcher is transferred to another state through the fitness function;
calculating the reward value of each global searcher in the state according to the fitness value;
updating the initial information matrix through the reward value;
and entering next iteration, returning to the step of selecting actions through the global searchers, transferring from one state to another state, calculating the fitness value of each global searcher when transferring to another state through the fitness function until the iteration times reach the maximum iteration times, and outputting an optimal information matrix.
Optionally, the fitness function is:
where f (x) is a fitness function, x is a controlled variable,
respectively carbon emission CgsActive power loss PlossVoltage stable value UdThe weight coefficient of (a);
the constraint conditions include:
PGi-PDi-Ui∑Uj(gijcosθij+bijsinθij)＝0；
QGi-QDi-Ui∑Uj(gijsinθij-bijcosθij)＝0；
in the formula, PGiIs the active power output of node i, QGiFor reactive power output of node i, PDiFor the active power demand of node i, QDiFor the reactive power requirement of node i, UiIs the voltage amplitude of node i, UjIs the voltage amplitude, g, of node jijIs a line LijConductance of (b)ijIs a line LijSusceptance of thetaijIs the phase angle difference between node i and node j, NGSet of generator nodes for grid system, NiIs a collection of nodes, QCiReactive compensator capacity, N, for node iCFor sets of reactive-load compensation devices, TkIs the transformation ratio of an on-load tap changer k, NTFor on-load tap changer branch set, SlFor the complex power of line l, max, min represent the upper and lower limits of the corresponding variable.
Optionally, the calculation formula of the reward value is as follows:
in the formula, RijFor the reward value, M is a penalty factor, WjThe number of nodes which do not satisfy the constraint condition.
The second aspect of the present application provides a low-carbon operation system of a power grid system, including:
the input unit is used for inputting the active demand and the reactive demand of the current optimization task after the initial parameters are configured;
the initialization unit is used for classifying the current optimization tasks to obtain source tasks and new tasks and randomly generating an initial information matrix of the source tasks;
the optimization unit is used for carrying out global optimization on the initial information matrix by taking the lowest active power loss, the lowest carbon emission and the lowest voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix;
the training unit is used for training the long-term and short-term memory network through the optimal information matrix and the information labels of the source tasks to obtain the trained long-term and short-term memory network;
the generating unit is used for generating an initial information matrix of the new task through the trained long-term and short-term memory network and triggering the optimizing unit to obtain an optimal information matrix of the new task;
and the acquisition unit is used for acquiring an optimal carbon emission strategy according to the optimal information matrix of the new task and executing the optimal carbon emission strategy.
Optionally, the optimization unit is specifically configured to:
setting a fitness function according to an optimization target of the power grid system with the lowest active power loss, the lowest carbon emission and the lowest voltage stability value, and setting constraint conditions;
performing action selection through a plurality of global searchers, transferring from one state to another state, and calculating a fitness value when each global searcher is transferred to another state through the fitness function;
calculating the reward value of each global searcher in the state according to the fitness value;
updating the initial information matrix through the reward value;
and entering next iteration, returning to the step of selecting actions through the global searchers, transferring from one state to another state, calculating the fitness value of each global searcher when transferring to another state through the fitness function until the iteration times reach the maximum iteration times, and outputting an optimal information matrix.
Optionally, the fitness function is:
where f (x) is a fitness function, x is a controlled variable,
respectively carbon emission CgsActive power loss PlossVoltage stable value UdThe weight coefficient of (a);
the constraint conditions include:
PGi-PDi-Ui∑Uj(gijcosθij+bijsinθij)＝0；
QGi-QDi-Ui∑Uj(gijsinθij-bijcosθij)＝0；
in the formula, PGiIs the active power output of node i, QGiFor reactive power output of node i, PDiFor the active power demand of node i, QDiFor the reactive power requirement of node i, UiIs the voltage amplitude of node i, UjIs the voltage amplitude, g, of node jijIs a line LijConductance of (b)ijIs a line LijSusceptance of thetaijIs the phase angle difference between node i and node j, NGSet of generator nodes for grid system, NiIs a collection of nodes, QCiReactive compensator capacity, N, for node iCFor sets of reactive-load compensation devices, TkIs the transformation ratio of an on-load tap changer k, NTFor on-load tap changer branch set, SlFor the complex power of line l, max, min represent the upper and lower limits of the corresponding variable.
Optionally, the calculation formula of the reward value is as follows:
in the formula, RijFor the reward value, M is a penalty factor, WjThe number of nodes which do not satisfy the constraint condition.
The third aspect of the application provides a low-carbon operation device of a power grid system, which comprises a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the power grid system low-carbon operation method according to any one of the first aspect according to instructions in the program code.
A fourth aspect of the present application provides a computer-readable storage medium for storing program code, which, when executed by a processor, implements the low-carbon operation method of the power grid system according to any one of the first aspects.
According to the technical scheme, the method has the following advantages:
the application provides a low-carbon operation method of a power grid system, which comprises the following steps: s1, inputting the active demand and the reactive demand of the current optimization task after configuring the initial parameters; s2, classifying the current optimization task to obtain a source task and a new task, and randomly generating an initial information matrix of the source task; s3, carrying out global optimization on the initial information matrix by taking the minimum active power loss, the minimum carbon emission and the minimum voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix; s4, training the long-short term memory network through the optimal information matrix and the information labels of the source tasks to obtain a trained long-short term memory network; s5, generating an initial information matrix of the new task through the trained long-short term memory network, and skipping to execute the step S3 to obtain an optimal information matrix of the new task; and S6, acquiring an optimal carbon emission strategy according to the optimal information matrix of the new task, and executing the optimal carbon emission strategy.
In the method, when the optimization training of the source task is carried out, an initial information matrix of the source task is randomly generated, the initial information matrix is globally optimized by taking the lowest active power loss, the lowest carbon emission and the lowest voltage stability value of a power grid system as an optimization target, an optimal information matrix of the source task is obtained, a long-term and short-term memory network is trained through the optimal information matrix of the source task, the initial information matrix of a new task is generated through the trained long-term and short-term memory network, the learning experience of a historical task is stored in the initial information matrix of the new task, the learning efficiency of the new task is accelerated, the searching speed is increased, the optimal information matrix of the new task is quickly obtained, an optimal carbon emission strategy is obtained, and the technical problem that the searching time is too long in the prior art is solved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
Fig. 1 is a schematic flow chart of a low-carbon operation method of a power grid system according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a low-carbon operation system of a power grid system according to an embodiment of the present application.
Detailed Description
The application provides a low-carbon operation method of a power grid system and a related device thereof, which are used for solving the technical problem of overlong search time in the prior art.
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, 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 application.
For convenience of understanding, please refer to fig. 1, an embodiment of the present application provides a low-carbon operation method for a power grid system, including:
and S1, inputting the active demand and the reactive demand of the current optimization task after the initial parameters are configured.
In the embodiment of the application, the configuration of the initial parameter comprises the configuration of a value range of the terminal voltage of the generator, which can be configured to be 1.00 to 1.06 of a reference value; the adjustment range of the reactive compensation capacity can be formed by five parts, namely-0.4, -0.2, 0, 0.2 and 0.4 of the reference value; the regulation range of the transformation ratio of the on-load tap changing transformer can be formed by 3 parts, which are respectively 0.98, 1 and 1.02 of the reference value. The carbon emission intensity of each unit is shown in table 1, and in the example, 50% of the carbon emission responsibility should be borne by the power generation side, so it is 0.5; the carbon emission responsibility corresponding to the grid loss is totally responsible for the grid side, namely 1. Three weight coefficients of fitness function
Are all set as1/3。
TABLE 1 carbon emission intensity of each unit
And S2, classifying the current optimization task to obtain a source task and a new task, and randomly generating an initial information matrix of the source task.
Taking 15 minutes as a unit, there are a total of 96 optimization tasks in a day. Thus, the available active power demand P for different task characteristic informationi＝[P1，P2，…，P96]And reactive power demand Qi＝[Q1，Q2，…，Q96]And (6) distinguishing. And classifying the current optimization task, and randomly generating an initial information matrix of the source task and an action probability matrix of the historical source task.
And S3, carrying out global optimization on the initial information matrix by taking the minimum active power loss, the minimum carbon emission and the minimum voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix.
And S31, setting a fitness function according to the optimization target of the power grid system with the lowest active power loss, the lowest carbon emission and the lowest voltage stability value, and setting a constraint condition.
And analyzing the power grid system, and calculating the whole network flow on the basis that the power grid system meets all constraint conditions and ensures the safe and normal operation of electric power.
For node i, the active power P flowing into it by generator node mimAnd the active power flowing into the node i from the node m of the generator and the total flowing active power from the node i are shown as follows:
Pim＝λimPGm；
in the formula, λimThe contribution coefficient for the active power of generator node m flowing into node i is numerically equal to the ratio of the active power flowing into node i through generator node m to the total active power generated by generator node m, PGmTotal active power, P, for generator node miThe total incoming active power for node i.
Branch active loss delta PijCan be expressed as:
in the formula of UiIs the voltage amplitude of node i, UjIs the voltage amplitude, g, of node jijIs a line LijConductance of (theta)ijIs the phase angle difference between node i and node j.
Assuming that the grid system has M generator sets in common, then for branch LijResulting active loss Δ PijCan be resolved into the following formula:
suppose the system has a set of generator nodes of NGAnd Z is a branch set, so that the total active power loss of the whole power grid in unit time is the sum of the active power losses of all the branches, namely:
according to the superposition theorem, then for branch LijIn other words, its carbon emission loss Δ CijComprises the following steps:
in the formula, deltamThe carbon emission intensity of the generator node m is expressed in (kg CO 2/kW.h) and is used for measuring the carbon emission corresponding to the unit electricity output.
Then, the total carbon emission C of the whole grid systemlossCan be expressed as:
carbon emission C to the power generation sideGThe value of which is the carbon emission C on the grid sidelossAnd carbon emission C on the user sideDCan be expressed as:
CG＝∑PGmδm＝Closs+CD；
the embodiment of the application redistributes the carbon emission responsibility of the power grid side and the user side according to a new distribution mechanism, namely:
the embodiment of the application selects three targets of lowest active power loss, lowest carbon emission and lowest voltage stability value as optimization targets, linear weighting is carried out on the optimization targets, a fitness function is obtained, and the final fitness function is as follows:
where f (x) is a fitness function, x is a controlled variable,
respectively carbon emission CgsActive power loss PlossVoltage stable value UdThe weight coefficient of (2).
UiIs the voltage amplitude of node i, the maximum value of which is
Minimum value of
The constraints of the optimization target in the embodiment of the application comprise active power balance constraint, reactive power balance constraint, upper and lower limit constraint of the capacity of the reactive compensation device, on-load tap changer constraint and the like, and are as follows:
PGi-PDi-Ui∑Uj(gijcosθij+bijsinθij)＝0；
QGi-QDi-Ui∑Uj(gijsinθij-bijcosθij)＝0；
in the formula, PGiIs the active power output of node i, QGiFor reactive power output of node i, PDiFor the active power demand of node i, QDiFor the reactive power requirement of node i, UiIs the voltage amplitude of node i, UjIs the voltage amplitude, g, of node jijIs a line LijConductance of (b)ijIs a line LijSusceptance of thetaijIs the phase angle difference between node i and node j, NGSet of generator nodes for grid system, NiIs a collection of nodes, QCiReactive compensator capacity, N, for node iCFor sets of reactive-load compensation devices, TkIs the transformation ratio of an on-load tap changer k, NTFor on-load tap changer branch set, SlFor the complex power of line l, max, min represent the upper and lower limits of the corresponding variable.
S32, the global searcher selects an operation to transition from one state to another state, and calculates a fitness value when each global searcher transitions to another state by a fitness function.
In the context of the Q-learning algorithm, the global searcher has the ability to learn a knowledge matrix that can feed back historical actions and select the best action at the time of the next selection. The global searcher selects an action under the action of a control strategy according to the state of the global searcher in each iteration process to select an unknown environment, and the environment generates an evaluation signal (reward or punishment) to react to the global searcher after receiving the selection action of the global searcher. The global searcher again generates the next action based on the evaluation signal (reward or penalty) and the new state of the environment, the principle of action selection being to maximize the total reward expectation. The controllable variables of each global searcher are generator terminal voltage, transformer transformation ratio and reactive compensation capacity, the value range of each variable is the action space, and the state space is the current value of each variable at present.
According to the method and the device, a Q learning method is adopted for global optimization, a global searcher group is used as a multi-main body to perform parallel search on an initial information matrix, each global searcher selects actions along a state-action space to determine actions of a previous variable, a next variable can select actions according to the state after the previous variable is selected, information transmission is completed once, and mutual relation among the variables is kept. And calculating the fitness value of each global searcher when the global searcher is transferred to another state through the fitness function.
And S33, calculating the reward value of each global searcher in the state through the fitness value.
The calculation formula of the reward value is as follows:
in the formula, RijFor the reward value, M is a penalty factor, WjThe number of nodes which do not satisfy the constraint condition.
And S34, updating the initial information matrix through the reward value.
The initial information matrix is updated through the reward value, and the specific updating process belongs to the prior art and is not described herein any more.
And S35, entering next iteration, returning to the step S32 until the iteration number reaches the maximum iteration number, and outputting the optimal information matrix.
All global searchers share the same initial information matrix, and can update a plurality of information elements in the initial information matrix at the same time during each iterative search, thereby greatly shortening the optimization time and improving the search efficiency.
And S4, training the long-short term memory network through the optimal information matrix and the information labels of the source tasks to obtain the trained long-short term memory network.
After the optimal information matrix of the source task is obtained, the long-term and short-term memory network can be trained through the optimal information matrix of the source task and the information labels capable of distinguishing the optimization tasks.
S5, generating an initial information matrix of the new task through the trained long-short term memory network, and skipping to execute the step S3 to obtain the optimal information matrix of the new task.
According to the information label of the new task, the initial information matrix of the new task can be generated through the trained long-short term memory network, and then the step S3 is skipped to optimize the initial information matrix of the new task to obtain the optimal information matrix of the new task. The method comprises the steps of training a long-term and short-term memory network through an optimal information matrix of a source task, carrying out nonlinear migration on information in the optimal information matrix, realizing network storage of the optimal information matrices of a plurality of historical tasks by introducing the long-term and short-term memory network, predicting the optimal information matrix of a new task as an initial information matrix of the new task through the trained long-term and short-term memory network, and fully utilizing historical optimization information to improve optimization speed, shorten optimization time and further quickly obtain the optimal information matrix of the new task.
And S6, acquiring an optimal carbon emission strategy according to the optimal information matrix of the new task, and executing the optimal carbon emission strategy.
And obtaining an optimal carbon emission strategy according to the optimal information matrix of the new task, and executing the optimal carbon emission strategy, thereby realizing low-carbon operation.
In the embodiment of the application, when the optimization training of the source task is carried out, an initial information matrix of the source task is randomly generated, the initial information matrix is globally optimized by taking the lowest active power loss, the lowest carbon emission and the lowest voltage stability value of a power grid system as an optimization target, an optimal information matrix of the source task is obtained, a long-term and short-term memory network is trained through the optimal information matrix of the source task, the initial information matrix of a new task is generated through the trained long-term and short-term memory network, the learning experience of a historical task is stored in the initial information matrix of the new task, the learning efficiency of the new task is accelerated, the searching speed is increased, the optimal information matrix of the new task is quickly obtained, an optimal carbon emission strategy is obtained, and the technical problem that the searching time is too long in the prior art is solved.
The foregoing is an embodiment of a low-carbon operation method of a power grid system provided by the present application, and the following is an embodiment of a low-carbon operation system of a power grid system provided by the present application.
Referring to fig. 2, an embodiment of the present application provides a low-carbon operation system of a power grid system, including:
the input unit is used for inputting the active demand and the reactive demand of the current optimization task after the initial parameters are configured;
the initialization unit is used for classifying the current optimization tasks to obtain source tasks and new tasks and randomly generating an initial information matrix of the source tasks;
the optimization unit is used for carrying out global optimization on the initial information matrix by taking the lowest active power loss, the lowest carbon emission and the lowest voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix;
the training unit is used for training the long-term and short-term memory network through the optimal information matrix and the information labels of the source tasks to obtain the trained long-term and short-term memory network;
the generating unit is used for generating an initial information matrix of the new task through the trained long-term and short-term memory network and triggering the optimizing unit to obtain an optimal information matrix of the new task;
and the acquisition unit is used for acquiring the optimal carbon emission strategy according to the optimal information matrix of the new task and executing the optimal carbon emission strategy.
As a further improvement, the optimization unit is specifically configured to:
setting a fitness function according to an optimization target of the power grid system with the lowest active power loss, the lowest carbon emission and the lowest voltage stability value, and setting constraint conditions;
performing action selection through a plurality of global searchers, and transferring from one state to another state;
calculating the fitness value of each global searcher when the global searcher is transferred to another state through a fitness function;
calculating the reward value of each global searcher in the state through the fitness value;
updating the initial information matrix through the reward value;
and entering next iteration, returning to the step of selecting actions through a plurality of global searchers and transferring from one state to another state until the iteration times reach the maximum iteration times, and outputting the optimal information matrix.
As a further improvement, the fitness function is:
where f (x) is a fitness function, x is a controlled variable,
respectively carbon emission CgsActive power loss PlossVoltage stable value UdThe weight coefficient of (a);
the constraint conditions include:
PGi-PDi-Ui∑Uj(gijcosθij+bijsinθij)＝0；
QGi-QDi-Ui∑Uj(gijsinθij-bijcosθij)＝0；
in the formula, PGiIs the active power output of node i, QGiFor reactive power output of node i, PDiFor the active power demand of node i, QDiFor the reactive power requirement of node i, UiIs the voltage amplitude of node i, UjIs the voltage amplitude, g, of node jijIs a line LijConductance of (b)ijIs a line LijSusceptance of thetaijIs the phase angle difference between node i and node j, NGSet of generator nodes for grid system, NiIs a collection of nodes, QCiReactive compensator capacity, N, for node iCFor sets of reactive-load compensation devices, TkIs the transformation ratio of an on-load tap changer k, NTFor on-load tap changer branch set, SlFor the complex power of line l, max, min represent the upper and lower limits of the corresponding variable.
As a further improvement, the calculation formula of the prize value is:
in the formula, RijFor the reward value, M is a penalty factor, WjThe number of nodes that do not satisfy the constraint.
In the embodiment of the application, when the optimization training of the source task is carried out, an initial information matrix of the source task is randomly generated, the initial information matrix is globally optimized by taking the lowest active power loss, the lowest carbon emission and the lowest voltage stability value of a power grid system as an optimization target, an optimal information matrix of the source task is obtained, a long-term and short-term memory network is trained through the optimal information matrix of the source task, the initial information matrix of a new task is generated through the trained long-term and short-term memory network, the learning experience of a historical task is stored in the initial information matrix of the new task, the learning efficiency of the new task is accelerated, the searching speed is increased, the optimal information matrix of the new task is quickly obtained, an optimal carbon emission strategy is obtained, and the technical problem that the searching time is too long in the prior art is solved.
The embodiment of the application also provides low-carbon operation equipment of the power grid system, which comprises a processor and a memory;
the memory is used for storing the program codes and transmitting the program codes to the processor;
the processor is used for executing the power grid system low-carbon operation method in the foregoing method embodiments according to instructions in the program code.
The embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium is used for storing program codes, and when the program codes are executed by a processor, the low-carbon operation method of the power grid system in the foregoing method embodiments is implemented.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The terms "first," "second," "third," "fourth," and the like in the description of the application and the above-described figures, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that in the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" for describing an association relationship of associated objects, indicating that there may be three relationships, e.g., "a and/or B" may indicate: only A, only B and both A and B are present, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single item(s) or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for executing all or part of the steps of the method described in the embodiments of the present application through a computer device (which may be a personal computer, a server, or a network device). And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. A low-carbon operation method of a power grid system is characterized by comprising the following steps:
s1, inputting the active demand and the reactive demand of the current optimization task after configuring the initial parameters;
s2, classifying the current optimization task to obtain a source task and a new task, and randomly generating an initial information matrix of the source task;
s3, carrying out global optimization on the initial information matrix by taking the minimum active power loss, the minimum carbon emission and the minimum voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix;
s4, training a long-short term memory network through the optimal information matrix and the information labels of the source tasks to obtain a trained long-short term memory network;
s5, generating an initial information matrix of the new task through the trained long-short term memory network, and skipping to execute the step S3 to obtain an optimal information matrix of the new task;
and S6, acquiring an optimal carbon emission strategy according to the optimal information matrix of the new task, and executing the optimal carbon emission strategy.
2. The low-carbon operation method of the power grid system according to claim 1, wherein the global optimization of the initial information matrix is performed by taking the minimum active power loss, the minimum carbon emission and the minimum voltage stability of the power grid system as optimization targets to obtain an optimal information matrix, and the method comprises the following steps:
setting a fitness function according to an optimization target of the power grid system with the lowest active power loss, the lowest carbon emission and the lowest voltage stability value, and setting constraint conditions;
performing action selection through a plurality of global searchers, transferring from one state to another state, and calculating a fitness value when each global searcher is transferred to another state through the fitness function;
calculating the reward value of each global searcher in the state according to the fitness value;
updating the initial information matrix through the reward value;
and entering next iteration, returning to the step of selecting actions through the global searchers, transferring from one state to another state, calculating the fitness value of each global searcher when transferring to another state through the fitness function until the iteration times reach the maximum iteration times, and outputting an optimal information matrix.
3. The power grid system low-carbon operation method according to claim 2, wherein the fitness function is:
where f (x) is a fitness function, x is a controlled variable,
respectively carbon emission CgsActive power loss PlossVoltage stable value UdThe weight coefficient of (a);
the constraint conditions include:
PGi-PDi-Ui∑Uj(gijcosθij+bijsinθij)＝0；
QGi-QDi-Ui∑Uj(gijsinθij-bijcosθij)＝0；
in the formula, PGiIs the active power output of node i, QGiFor reactive power output of node i, PDiFor the active power demand of node i, QDiFor the reactive power requirement of node i, UiIs the voltage amplitude of node i, UjIs the voltage amplitude, g, of node jijIs a line LijConductance of (b)ijIs a line LijSusceptance of thetaijIs the phase angle difference between node i and node j, NGSet of generator nodes for grid system, NiIs a collection of nodes, QCiReactive compensator capacity, N, for node iCFor sets of reactive-load compensation devices, TkIs the transformation ratio of an on-load tap changer k, NTFor on-load tap changer branch set, SlFor the complex power of line l, max, min represent the upper and lower limits of the corresponding variable.
4. The low-carbon operation method of the power grid system as claimed in claim 3, wherein the calculation formula of the reward value is as follows:
in the formula, RijFor the reward value, M is a penalty factor, WjThe number of nodes which do not satisfy the constraint condition.
5. The utility model provides a power grid system low carbon operation system which characterized in that includes:
the input unit is used for inputting the active demand and the reactive demand of the current optimization task after the initial parameters are configured;
the initialization unit is used for classifying the current optimization tasks to obtain source tasks and new tasks and randomly generating an initial information matrix of the source tasks;
the optimization unit is used for carrying out global optimization on the initial information matrix by taking the lowest active power loss, the lowest carbon emission and the lowest voltage stability value of the power grid system as optimization targets to obtain an optimal information matrix;
the training unit is used for training the long-term and short-term memory network through the optimal information matrix and the information labels of the source tasks to obtain the trained long-term and short-term memory network;
the generating unit is used for generating an initial information matrix of the new task through the trained long-term and short-term memory network and triggering the optimizing unit to obtain an optimal information matrix of the new task;
and the acquisition unit is used for acquiring an optimal carbon emission strategy according to the optimal information matrix of the new task and executing the optimal carbon emission strategy.
6. The power grid system low-carbon operation system according to claim 5, wherein the optimization unit is specifically configured to:
setting a fitness function according to an optimization target of the power grid system with the lowest active power loss, the lowest carbon emission and the lowest voltage stability value, and setting constraint conditions;
performing action selection through a plurality of global searchers, transferring from one state to another state, and calculating a fitness value when each global searcher is transferred to another state through the fitness function;
calculating the reward value of each global searcher in the state according to the fitness value;
updating the initial information matrix through the reward value;
and entering next iteration, returning to the step of selecting actions through the global searchers, transferring from one state to another state, calculating the fitness value of each global searcher when transferring to another state through the fitness function until the iteration times reach the maximum iteration times, and outputting an optimal information matrix.
7. The power grid system low-carbon operation system according to claim 6, wherein the fitness function is:
where f (x) is a fitness function, x is a controlled variable,
respectively carbon emission CgsActive power loss PlossVoltage stable value UdThe weight coefficient of (a);
the constraint conditions include:
PGi-PDi-Ui∑Uj(gijcosθij+bijsinθij)＝0；
QGi-QDi-Ui∑Uj(gijsinθij-bijcosθij)＝0；
in the formula, PGiIs the active power output of node i, QGiFor reactive power output of node i, PDiFor the active power demand of node i, QDiFor the reactive power requirement of node i, UiIs the voltage amplitude of node i, UjIs the voltage amplitude, g, of node jijIs a line LijConductance of (b)ijIs a line LijSusceptance of thetaijIs the phase angle difference between node i and node j, NGSet of generator nodes for grid system, NiIs a collection of nodes, QCiReactive compensator capacity, N, for node iCFor sets of reactive-load compensation devices, TkIs the transformation ratio of an on-load tap changer k, NTFor on-load tap changer branch set, SlFor the complex power of line l, max, min represent the upper and lower limits of the corresponding variable.
8. The power grid system low-carbon operation system as claimed in claim 7, wherein the calculation formula of the reward value is as follows:
in the formula, RijFor the reward value, M is a penalty factor, WjThe number of nodes which do not satisfy the constraint condition.
9. The low-carbon operation equipment of the power grid system is characterized by comprising a processor and a memory;
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is used for executing the low-carbon operation method of the power grid system according to any one of claims 1 to 4 according to instructions in the program code.
10. A computer-readable storage medium, wherein the computer-readable storage medium is configured to store program code, and the program code, when executed by a processor, implements the low-carbon operation method of the power grid system according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111519546.4A CN114202121A (en) | 2021-12-13 | 2021-12-13 | Low-carbon operation method of power grid system and related device thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111519546.4A CN114202121A (en) | 2021-12-13 | 2021-12-13 | Low-carbon operation method of power grid system and related device thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114202121A true CN114202121A (en) | 2022-03-18 |
Family
ID=80653071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111519546.4A Pending CN114202121A (en) | 2021-12-13 | 2021-12-13 | Low-carbon operation method of power grid system and related device thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114202121A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115145709A (en) * | 2022-07-19 | 2022-10-04 | 华南师范大学 | Low-carbon big-data artificial intelligence method and health-care state system |
-
2021
- 2021-12-13 CN CN202111519546.4A patent/CN114202121A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115145709A (en) * | 2022-07-19 | 2022-10-04 | 华南师范大学 | Low-carbon big-data artificial intelligence method and health-care state system |
| CN115145709B (en) * | 2022-07-19 | 2024-05-17 | 华南师范大学 | Low-carbon big data artificial intelligence method and medical health state system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Erlich et al. | Optimal dispatch of reactive sources in wind farms | |
| CN107437813B (en) | Power distribution network reactive power optimization method based on cuckoo-particle swarm | |
| Shih et al. | On-line coordination of directional overcurrent relays: Performance evaluation among optimization algorithms | |
| Abbasi | Investigation of simultaneous effect of demand response and load uncertainty on distribution feeder reconfiguration | |
| CN106779177A (en) | Multiresolution wavelet neutral net electricity demand forecasting method based on particle group optimizing | |
| CN111146821A (en) | DSTATCOM optimal configuration method considering photovoltaic uncertainty | |
| Gutiérrez‐Alcaraz et al. | Two‐stage heuristic methodology for optimal reconfiguration and Volt/VAr control in the operation of electrical distribution systems | |
| Ghasemi | Application of MOPSO for economic load dispatch solution with transmission losses | |
| Li et al. | Knee Point‐Guided Multiobjective Optimization Algorithm for Microgrid Dynamic Energy Management | |
| Li et al. | A derived heuristics based multi-objective optimization procedure for micro-grid scheduling | |
| CN115470987A (en) | Short-term photovoltaic power generation prediction method based on improved long-term and short-term memory neural network | |
| CN114202121A (en) | Low-carbon operation method of power grid system and related device thereof | |
| CN110516889B (en) | Load comprehensive prediction method based on Q-learning and related equipment | |
| Ebell et al. | Reinforcement learning control algorithm for a pv-battery-system providing frequency containment reserve power | |
| Mishra et al. | Multi-objective optimization of economic emission load dispatch incorporating load forecasting and solar photovoltaic sources for carbon neutrality | |
| Ida Evangeline et al. | Real-time optimal power flow solution for wind farm integrated power system using evolutionary programming algorithm | |
| Zhang et al. | Physical-model-free intelligent energy management for a grid-connected hybrid wind-microturbine-PV-EV energy system via deep reinforcement learning approach | |
| Tamilselvan et al. | Multi objective Flower Pollination Algorithm for solving capacitor placement in radial distribution system using data structure load flow analysis | |
| Mahela et al. | Optimal capacitor placement techniques in transmission and distribution networks to reduce line losses and voltage stability enhancement: A review | |
| Pandya et al. | CHAOS enhanced flower pollination algorithm for optimal scheduling of distributed energy resources in smart grid | |
| Niknam et al. | Reactive power pricing in deregulated environments using novel search methods | |
| Tyagi et al. | Comparative study of IHS and BF-DE algorithm for optimal DG placement | |
| Khelifi et al. | A hybrid cuckoo search and krill herd technique for solving problem of optimal power flow in power systems. | |
| CN115994612A (en) | Power distribution network operation optimization method and device based on business expansion planning and storage medium | |
| CN112994016A (en) | Recovery solvability adjustment method and system for power flow of power system |
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 |