CN112632840A - Power grid transient stability evaluation method based on adaptive differential evolution algorithm and ELM - Google Patents
Power grid transient stability evaluation method based on adaptive differential evolution algorithm and ELM Download PDFInfo
- Publication number
- CN112632840A CN112632840A CN202011398561.3A CN202011398561A CN112632840A CN 112632840 A CN112632840 A CN 112632840A CN 202011398561 A CN202011398561 A CN 202011398561A CN 112632840 A CN112632840 A CN 112632840A
- Authority
- CN
- China
- Prior art keywords
- power grid
- transient stability
- elm
- differential evolution
- evolution algorithm
- 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
- 230000001052 transient effect Effects 0.000 title claims abstract description 55
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 43
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 29
- 238000011156 evaluation Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 35
- 239000002245 particle Substances 0.000 claims abstract description 31
- 238000005457 optimization Methods 0.000 claims abstract description 22
- 238000013210 evaluation model Methods 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000004088 simulation Methods 0.000 claims abstract description 13
- 230000008859 change Effects 0.000 claims abstract description 11
- 230000007246 mechanism Effects 0.000 claims abstract description 11
- 230000009467 reduction Effects 0.000 claims description 6
- 230000008030 elimination Effects 0.000 claims description 5
- 238000003379 elimination reaction Methods 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 4
- 238000013097 stability assessment Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000006978 adaptation Effects 0.000 claims 2
- 230000006835 compression Effects 0.000 claims 1
- 238000007906 compression Methods 0.000 claims 1
- 238000004458 analytical method Methods 0.000 abstract description 12
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 239000010977 jade Substances 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000007418 data mining Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000001360 synchronised effect Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000012098 association analyses Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/25—Design optimisation, verification or simulation using particle-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Economics (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a power grid transient stability evaluation method based on a self-adaptive differential evolution algorithm and ELM, which comprises the following steps: acquiring disturbed dynamic data and disturbed steady-state data of a power grid simulation disturbed track so as to construct a sample set; optimizing an extreme learning machine by adopting a self-adaptive differential evolution algorithm comprising an improved variation strategy and an optimal particle local optimization mechanism; training the optimized extreme learning machine by adopting a sample set to obtain a transient stability evaluation model; and rapidly judging the transient change after the power grid disturbance according to the transient stability evaluation model. The method comprises the steps of establishing analysis models of different fault disturbance scenes and stability relations, determining the relation between historical change trends of different positions and different monitoring quantities and system stability, extracting the hierarchical key features of transient stability, and optimizing an ELM transient stability evaluation model based on an adaptive differential evolution algorithm to realize rapid stability judgment of transient changes after power grid disturbance.
Description
Technical Field
The invention relates to the technical field of power grid safety, in particular to a power grid transient stability evaluation method based on an adaptive differential evolution algorithm and ELM.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The impact on the power grid caused by large-scale rapid growth of source-grid-load, large access of fluctuating new energy and renewable energy and the like is increasing; meanwhile, due to factors such as transition and evolution of a power grid structure, association constraint of a provincial power grid and a large-area power grid, mutual influence of alternating current, direct current and multiple direct currents and the like, especially when large-scale power grid faults and uncontrollable chain reaction occur, safety and stability analysis and dispatching operation control of a power system face more severe tests.
The traditional power grid monitoring System (SCADA) is difficult to acquire system fault information in real time, the transient stability analysis decision of the power grid is usually performed off-line analysis aiming at an expected fault set, the actual operation condition is difficult to be completely matched with the off-line analysis mode due to the complex and changeable operation mode of the current power grid, and the problems that the screening and the real-time performance of the fault set are difficult to guarantee and the like are also faced by adopting on-line analysis and calculation. When occasional faults or large-capacity load random fluctuation and the like occur in the power grid, the stability of the system can be further reduced, if effective measures are not actively taken to restore the stability level of the system, the system loses stability after a certain follow-up event, and even a large-area power failure accident is caused.
The traditional time domain simulation method needs to determine mathematical models, parameters and simulation scenes of all physical elements before each simulation calculation, so that the calculated amount is large, and the requirement on the accuracy of the models is high. The power system is a nonlinear complex dynamic network system in nature, coupling exists between each element and external factors, and the establishment of a physical model is generally based on certain assumptions and simplifications. The dynamic behavior of the power electronic power system is more complex, the accurate modeling of elements is difficult, and the assumed conditions of the traditional research may fail, so that the numerical simulation result sometimes cannot sufficiently reflect the actual operation condition of the power grid. The direct method, as another power grid stability analysis method, is difficult to accurately calculate load dynamics and apply to a complex system, so that the application in an actual power grid is limited.
As a high-dimensional complex energy transmission network, a large amount of data is generated in the process of energy transmission, and the traditional analysis method cannot effectively utilize the data, namely cannot timely and accurately perform system analysis and assistant decision by taking data driving as a way, so that resource waste is caused. In recent years, a wide area measurement system based on a synchrophasor measurement unit (PMU) is rapidly developed, which uses GPS Global Positioning System (GPS) second pulses as a synchronous clock, assembles positive-sequence phasors, time stamps, and the like into messages, and transmits the messages to a remote data concentrator through a dedicated channel. The data concentrator collects information from each PMU, the information covers tens of thousands of power grid synchronous phase angles and dynamic response data, the real-time operation condition of the power grid is reflected to a great extent, the real-time operation condition of the power grid is contained in the information, the information comprises abundant characteristic steady-state and transient-state process characteristic sets, massive data of the power system are fully mined and analyzed by using the advantages of strong learning capacity and high calculation speed of a big data technology, a stability assessment model with high accuracy, strong stability and high calculation speed is established, transient stability assessment of the power system is rapidly carried out according to power grid monitoring data and historical events, and a new way is provided for practically ensuring safe and stable operation of the power grid. In order to meet the safe operation requirement of the current complex power grid, massive data of a dispatching center is fully mined, and no unified description method exists for the characteristics of the complex power grid, which are closely related to the stability of a system.
Disclosure of Invention
In order to solve the problems, the invention provides a power grid transient stability evaluation method based on an adaptive differential evolution algorithm and an ELM (element-free mass modeling), which is characterized in that by establishing analysis models of different fault disturbance scenes and stability relations, the relation between historical change trends of different positions and different monitoring quantities and system stability is determined, the hierarchical key features of transient stability are extracted, and meanwhile, the ELM transient stability evaluation model is optimized based on the adaptive differential evolution algorithm, so that the rapid stability judgment of transient changes after power grid disturbance is realized.
In order to achieve the purpose, the invention adopts the following technical scheme:
in a first aspect, the invention provides a power grid transient stability evaluation method based on an adaptive differential evolution algorithm and an ELM, which includes:
acquiring disturbed dynamic data and disturbed steady-state data of a power grid simulation disturbed track so as to construct a sample set;
optimizing an extreme learning machine by adopting a self-adaptive differential evolution algorithm comprising an improved variation strategy and an optimal particle local optimization mechanism;
training the optimized extreme learning machine by adopting a sample set to obtain a transient stability evaluation model;
and rapidly judging the transient change after the power grid disturbance according to the transient stability evaluation model.
In a second aspect, the present invention provides a power grid transient stability evaluation system based on an adaptive differential evolution algorithm and an ELM, including:
the data acquisition module is used for acquiring disturbed dynamic data and disturbed steady-state data of the power grid simulation disturbed track so as to construct a sample set;
the model optimization module is used for optimizing the extreme learning machine by adopting a self-adaptive differential evolution algorithm comprising an improved variation strategy and an optimal particle local optimization mechanism;
the model training module is used for training the optimized extreme learning machine by adopting a sample set to obtain a transient stability evaluation model;
and the fast stability judging module is used for fast judging the transient change after the power grid disturbance according to the transient stability evaluation model.
In a third aspect, the present invention provides an electronic device comprising a memory and a processor, and computer instructions stored on the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method of the first aspect is performed.
In a fourth aspect, the present invention provides a computer readable storage medium for storing computer instructions which, when executed by a processor, perform the method of the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
according to the method, the key feature set of the power grid is constructed and the input features of the power grid are subjected to dimension reduction, the dimension of identification of the power grid stability association rule is simplified to the greatest extent, the key association features of the power grid operation and stability features are obtained, the redundant features are eliminated, the most valuable information is kept, the problem of dimension disaster in big data mining is avoided, and the classification effect is further improved.
The intelligent evaluation model based on machine learning fully excavates real-time synchronous power grid data, establishes an incidence relation between a power grid operation mode and transient stability through characteristic selection and dimension reduction of input characteristic quantities, determines key factors influencing the stability level of a system, quickly evaluates the current stability level according to the operation state of the system, and improves the intellectualization, standardization, rapidity and self-adaption capability of power grid stability evaluation.
Compared with the traditional evolutionary difference algorithm, the method can improve the global optimization capability and the convergence speed and overcome the defects of premature convergence and the like by utilizing the JADE algorithm, and meanwhile, the local optimization operation is introduced into the JADE algorithm, and the activity of the optimal individual can be increased and the local search capability of the algorithm can be enhanced by applying the Gaussian disturbance quantity near the optimal individual.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
Fig. 1 is an application block diagram of a power grid transient stability evaluation method based on an adaptive differential evolution algorithm and ELM according to embodiment 1 of the present invention;
fig. 2 is a flowchart of a power grid transient stability evaluation method based on an adaptive differential evolution algorithm and ELM according to embodiment 1 of the present invention;
fig. 3 is a flowchart of a recursive feature elimination algorithm provided in embodiment 1 of the present invention.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, and it should be understood that the terms "comprises" and "comprising", and any variation 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.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example 1
According to the technical idea of information acquisition and classification, information aggregation and integration, association analysis model construction and power grid stable state rapid identification, through intelligent technologies such as data mining, key feature extraction and intelligent classifier construction, information is extracted from historical records of different power grid events, the relation between the information and the power grid stability is acquired, the association analysis model is obtained through abstraction and synthesis, online updating is carried out according to real-time data, and rapid early warning of instability is finally achieved, so that a dispatcher can take measures quickly to improve the system stability level.
As shown in fig. 1-2, the present embodiment provides a power grid transient stability evaluation method based on an adaptive differential evolution algorithm and an ELM, which specifically includes the following steps:
s1: acquiring disturbed dynamic data and disturbed steady-state data of a power grid simulation disturbed track so as to construct a sample set;
s2: optimizing an extreme learning machine by adopting a self-adaptive differential evolution algorithm comprising an improved variation strategy and an optimal particle local optimization mechanism;
s3: training the optimized extreme learning machine by adopting a sample set to obtain a transient stability evaluation model;
s4: and rapidly judging the transient change after the power grid disturbance according to the transient stability evaluation model.
In the step S1, simulating the disturbed process of the power grid, obtaining disturbed dynamic data of off-line simulation and disturbed EMS/WAMS power grid steady-state operation data, and storing the collected multisource power grid data into a database by means of multisource data storage;
preferably, characteristics which can be directly measured or are convenient to combine and analyze, such as power grid bus characteristics, line characteristics and power grid system characteristics, are selected; the method specifically comprises the following steps:
the bus bar feature includes: the system comprises a bus voltage, a generator active power, a generator reactive power, a load active level, a load reactive level, a generator coefficient, a bus voltage amplitude value, a phase angle and the like;
the line features include: line active level, line reactive level, line active loss, line reactive loss, etc.;
the system is generally characterized by: system active load, system reactive load level, system active output, reactive output, and the like.
Because a unified description method is not available for the characteristics of a complex power grid, which are closely related to the stability of the power grid, in the existing method, the description of the incidence relation is firstly carried out, the key characteristics representing the stability level of the power grid are given from the perspective of model analysis, the automatic mapping between the specific dynamic time variable of the power grid and the stability of the power grid is established, the key state characteristic set of the power grid is obtained, and the transient stability index of the power grid is quickly obtained through the static and dynamic characteristics such as the output of a generator node, the section tidal current condition and the angular speed of a generator rotor in the key state;
then, extracting transient stability hierarchical key features of the obtained sample set by adopting a data reduction and recursive feature elimination method, which specifically comprises the following steps:
the method comprises the steps of considering screening and dimension reduction of input characteristics of transient stability indexes of the power grid, using power grid historical data or simulation data as training samples, determining a strong association relation between the power grid input characteristics and the stability characteristics by using a characteristic recursive elimination method, sequencing and combining the strongly associated power grid operation characteristics, achieving dimension reduction of the input characteristics, reducing complexity of identification of the association relation between the power grid operation characteristics and the stability characteristics, removing redundant information, reducing operation scale, and improving model training efficiency and algorithm accuracy.
It can be understood that the selection of the power grid input characteristics and the stability index is not limited to the scope designed in this embodiment, and time series data between PMU measurement points and time series characteristics of sections of a line may also be selected as an information source.
The characteristic recursive elimination method is shown in fig. 3, and specifically includes:
(1) inputting a training set Dtrain, a test set Dtest, an implicit layer node Ltest and an activation function S;
(2) judging whether i is larger than n;
(3) if not, eliminating the ith dimension characteristic, training and testing the model, enabling the test accuracy to be Gi, enabling the mu-G0-Gi to be the characteristic importance index, enabling i +1, and returning to the step (2);
(4) if yes, sorting the importance indexes of i-1, 2 and … N, marking the sorted importance indexes as indexes, and sorting the original features by the indexes to obtain a new feature set;
(5) judging whether j is larger than N;
(6) if not, taking the previous j-dimensional feature for training, recording the test accuracy Gj', making j +1, and returning to the step (5);
(7) if yes, j corresponding to the best performance in the accuracy Gj' is assigned to value;
(8) and outputting the reserved feature number value and the index value table index of the reserved feature number value in the original feature set.
In this embodiment, in step S2, the extreme learning machine ELM is an improved single hidden layer feedforward neural network, overcomes the disadvantages that the traditional neural network gradient descent algorithm needs multiple iterative solutions and the network structure is complex, and has the advantages of simple parameter selection, fast training speed, strong generalization capability, and the like, and is suitable for online application. In the embodiment, an extreme learning machine is used as a classifier to train a key sample set with labels, and the performance of the model is evaluated through a test sample set with unknown labels; meanwhile, the embodiment provides an improved adaptive differential evolution algorithm which is combined with an extreme learning machine to optimize the input weight and the threshold vector of the ELM and further improve the performance of the model.
In the step S2, a self-adaptive differential evolution algorithm JADE optimization extreme learning machine including an improved mutation strategy and an optimal particle local optimization mechanism is adopted; the method specifically comprises the following steps:
(1) determining an ELM network structure through cross validation and a grid search method, initializing network parameters, setting K ELM hidden layer nodes, setting a particle variable dimension D to be m multiplied by k + k, setting m to be the number of ELM inputs, and setting an activation function to be s (x).
(2) Setting initial parameters of adaptive evolution differential algorithm JADE, setting JADE population size NP, maximum iteration number maxIteration and initialization parameter muCR、μFAnd local variation probability PA。
(3) Let the training sample set be [ x ]i,yi](i 1, 2.., N), y is a data tag, and a primary population is randomly generated, wherein the population theta is a group of a plurality of generationsGNP×DWhere θ is ═ a11，a12，…，a1k，a21，a22，…，a2k，…，an1，an2，…，ank，b1，b2，…，bk]D represents the dimension of the population theta;
and converting the initial generation population particles into an input weight and hidden layer bias of the ELM, calculating an output matrix of the middle layer to obtain an output weight, and taking the estimated accuracy as a fitness function to further obtain an individual adaptive value.
(4) Setting the current iteration Generation as 1 according to muCRAnd muFRespectively calculating the scaling factors F corresponding to each population in the current algebraiAnd cross probability CRi(i∈[1,NP]) As shown in formula (1):
wherein randn is a Gaussian distribution and randc is a Cauchy distribution.
(5) Adopting a new mutation strategy, executing cross operation and selection operation, and selecting the particle with a better adaptive value as a new parent particle; the method specifically comprises the following steps: storing the generated offspring population in a matrix U, calculating the adaptive value of each population particle, comparing the adaptive value with the corresponding particles in the parent population X according to a greedy algorithm criterion, selecting the particles with better adaptive values as new parent particles to be retained in X, and recording the optimal particles Xbest。
Compared with the traditional differential evolution algorithm, the DE/current _ to _ best/1 variation strategy is adopted by JADE in the embodiment, as shown in formula (2):
wherein, Xbest,gIndividuals with the top p NP in the population of the current generation ranked in fitness (p is typically taken to be 5%) were randomly selected. Xr1,gIs a randomly selected individual in the population,
randomly selecting individuals in a set of the current population and an external archive A, wherein the external archive A stores the individuals with better offspring than parents in history, and the scale of the individuals is fixed as NP;
after the selection operation, F corresponding to the individual with effective cross variation of the current generationi、CRiRespectively stored in the set SFAnd SCRIn (1).
(6) After the selection operation is executed, a local searching mechanism of the optimal particles is added to perform local optimization of the optimal individual;
generating a random number rand if rand>PAThen to the optimal solution XbestPerforming Gaussian variation operation, namely applying Gaussian disturbance random quantity to each dimension of the optimal individual, as shown in formulas (3) to (4):
wherein:
the particles are the optimal particles after Gaussian variation; n (0,1) is a Gaussian distribution random quantity with a mean value of 0 and a variance of 1.
(7) Adaptive update of muCRAnd muF：
Wherein S isFAnd SCRTo successfully evolveSet of parameters corresponding to the body, c is a given value, meanAIs an arithmetic mean, meanLThe Lehmer mean.
(8) When iteration number Generation>When maxIteration, the optimization is over, XbestNamely, the optimal input weight value and the hidden layer bias of the ELM are obtained, and the optimized ELM model is obtained.
In this embodiment, in step S3, the optimized extreme learning machine is trained by using a sample set after extracting the key features, so as to obtain a transient stability evaluation model, and the model is used to quickly determine the transient stability change of the power grid and assist in controlling the strategy.
In the embodiment, by establishing an analysis model of the relationship between different fault disturbance scenes and the system stability, the relationship between the historical change trends of different positions and different monitoring quantities and the system stability is determined, the correlation between the power grid time sequence characteristic quantity and the system stability level is analyzed, data mining is performed through a large amount of accumulated online historical data and typical examples, the power grid transient state information is extracted, the important characteristics of the power grid dynamic process are further identified, the system stable state is rapidly identified, and guidance is provided for an optimization control strategy for improving the system stability.
Example 2
The embodiment provides a power grid transient stability evaluation system based on an adaptive differential evolution algorithm and an ELM, which includes:
the data acquisition module is used for acquiring disturbed dynamic data and disturbed steady-state data of the power grid simulation disturbed track so as to construct a sample set;
the model optimization module is used for optimizing the extreme learning machine by adopting a self-adaptive differential evolution algorithm comprising an improved variation strategy and an optimal particle local optimization mechanism;
the model training module is used for training the optimized extreme learning machine by adopting a sample set to obtain a transient stability evaluation model;
and the fast stability judging module is used for fast judging the transient change after the power grid disturbance according to the transient stability evaluation model.
It should be noted that the above modules correspond to steps S1 to S4 in embodiment 1, and the above modules are the same as the examples and application scenarios realized by the corresponding steps, but are not limited to the disclosure in embodiment 1. It should be noted that the modules described above as part of a system may be implemented in a computer system such as a set of computer-executable instructions.
In further embodiments, there is also provided:
an electronic device comprising a memory and a processor and computer instructions stored on the memory and executed on the processor, the computer instructions when executed by the processor performing the method of embodiment 1. For brevity, no further description is provided herein.
It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general purpose processors, digital signal processors DSP, application specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and may provide instructions and data to the processor, and a portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.
A computer readable storage medium storing computer instructions which, when executed by a processor, perform the method described in embodiment 1.
The method in embodiment 1 may be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, among other storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor. To avoid repetition, it is not described in detail here.
Those of ordinary skill in the art will appreciate that the various illustrative elements, i.e., algorithm steps, described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A power grid transient stability assessment method based on an adaptive differential evolution algorithm and ELM is characterized by comprising the following steps:
acquiring disturbed dynamic data and disturbed steady-state data of a power grid simulation disturbed track so as to construct a sample set;
optimizing an extreme learning machine by adopting a self-adaptive differential evolution algorithm comprising an improved variation strategy and an optimal particle local optimization mechanism;
training the optimized extreme learning machine by adopting a sample set to obtain a transient stability evaluation model;
and rapidly judging the transient change after the power grid disturbance according to the transient stability evaluation model.
2. The power grid transient stability evaluation method based on the adaptive differential evolution algorithm and the ELM as claimed in claim 1, wherein a recursive feature elimination method is adopted to extract key features from a sample set, determine a strong association relationship between power grid operation features and stability features, and perform sequencing combination and compression dimension reduction on the strongly associated power grid operation features.
3. The power grid transient stability evaluation method based on the adaptive differential evolution algorithm and the ELM as claimed in claim 1, wherein the adaptive differential evolution algorithm optimization extreme learning machine comprises: after an improved mutation strategy, a crossover operation and a selection operation are sequentially adopted, population particles with the optimal fitness value are used as new parent particles, after an optimal particle local optimization mechanism is adopted to carry out optimal individual local optimization, next generation scaling factors and crossover probabilities are updated, and optimal input weights and hidden layer bias are output.
4. The power grid transient stability evaluation method based on the adaptive differential evolution algorithm and the ELM as claimed in claim 3, wherein the initial population is initialized, the initial population particles are converted into the input weight and the hidden layer bias of the ELM, the middle layer output matrix is calculated to obtain the output weight, the evaluation accuracy is used as a fitness function, and further the individual fitness value is obtained.
5. The adaptive differential evolution algorithm and ELM-based power grid transient stability evaluation method according to claim 3, wherein the selecting operation comprises calculating an adaptation value of the particle of the child population, comparing the particle of the child population with a corresponding particle in the parent population according to a greedy algorithm, and selecting a particle with a better adaptation value as a new parent particle.
6. The power grid transient stability evaluation method based on the adaptive differential evolution algorithm and the ELM as claimed in claim 1, wherein the optimal particle local optimization mechanism comprises generating a random number, and if the random number is greater than the local variation probability, performing a Gaussian variation operation applying a Gaussian disturbance random quantity on the population particle with the optimal fitness value after the selection operation to perform optimal individual local optimization.
7. The adaptive differential evolution algorithm and ELM-based power grid transient stability assessment method according to claim 1, wherein the sample set comprises bus characteristics, line characteristics and power grid system characteristics;
the bus bar feature includes: the system comprises a bus voltage, a generator active power, a generator reactive power, a load active level, a load reactive level, a generator coefficient, a bus voltage amplitude and a phase angle;
the line features include: line active level, line reactive level, line active loss, line reactive loss;
the grid system features include: system active load, system reactive load level, system active output and reactive output.
8. A power grid transient stability evaluation system based on an adaptive differential evolution algorithm and ELM is characterized by comprising:
the data acquisition module is used for acquiring disturbed dynamic data and disturbed steady-state data of the power grid simulation disturbed track so as to construct a sample set;
the model optimization module is used for optimizing the extreme learning machine by adopting a self-adaptive differential evolution algorithm comprising an improved variation strategy and an optimal particle local optimization mechanism;
the model training module is used for training the optimized extreme learning machine by adopting a sample set to obtain a transient stability evaluation model;
and the fast stability judging module is used for fast judging the transient change after the power grid disturbance according to the transient stability evaluation model.
9. An electronic device comprising a memory and a processor and computer instructions stored on the memory and executed on the processor, the computer instructions when executed by the processor performing the method of any of claims 1-7.
10. A computer-readable storage medium storing computer instructions which, when executed by a processor, perform the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011398561.3A CN112632840A (en) | 2020-12-04 | 2020-12-04 | Power grid transient stability evaluation method based on adaptive differential evolution algorithm and ELM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011398561.3A CN112632840A (en) | 2020-12-04 | 2020-12-04 | Power grid transient stability evaluation method based on adaptive differential evolution algorithm and ELM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112632840A true CN112632840A (en) | 2021-04-09 |
Family
ID=75307576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011398561.3A Pending CN112632840A (en) | 2020-12-04 | 2020-12-04 | Power grid transient stability evaluation method based on adaptive differential evolution algorithm and ELM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112632840A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113642766A (en) * | 2021-07-08 | 2021-11-12 | 南方电网科学研究院有限责任公司 | Method, device, equipment and medium for predicting power failure number of power system station |
| CN114329702A (en) * | 2021-12-21 | 2022-04-12 | 中国地震局工程力学研究所 | High-stability method and device for calibrating and designing reaction spectrum based on improved differential evolution algorithm |
| CN114389265A (en) * | 2022-03-23 | 2022-04-22 | 山东大学 | Power grid emergency load optimization method and system based on proliferation and reduced evolution |
| CN116031879A (en) * | 2023-02-28 | 2023-04-28 | 四川大学 | Hybrid intelligent feature selection method suitable for transient voltage stability evaluation of power system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109379780A (en) * | 2018-10-23 | 2019-02-22 | 华南理工大学 | Wireless sensor network locating method based on adaptive differential evolution algorithm |
| CN109960860A (en) * | 2019-03-13 | 2019-07-02 | 华北电力大学 | Transient stability evaluation in power system method based on differential evolution extreme learning machine |
-
2020
- 2020-12-04 CN CN202011398561.3A patent/CN112632840A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109379780A (en) * | 2018-10-23 | 2019-02-22 | 华南理工大学 | Wireless sensor network locating method based on adaptive differential evolution algorithm |
| CN109960860A (en) * | 2019-03-13 | 2019-07-02 | 华北电力大学 | Transient stability evaluation in power system method based on differential evolution extreme learning machine |
Non-Patent Citations (2)
| Title |
|---|
| 张林林等: "基于极限学习机的电力***暂态稳定评估方法", 《上海交通大学学报》 * |
| 王东升等: "基于JADE-ELM的煤巷围岩稳定性预测", 《煤矿安全》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113642766A (en) * | 2021-07-08 | 2021-11-12 | 南方电网科学研究院有限责任公司 | Method, device, equipment and medium for predicting power failure number of power system station |
| CN113642766B (en) * | 2021-07-08 | 2024-01-30 | 南方电网科学研究院有限责任公司 | Method, device, equipment and medium for predicting power outage number of power system station |
| CN114329702A (en) * | 2021-12-21 | 2022-04-12 | 中国地震局工程力学研究所 | High-stability method and device for calibrating and designing reaction spectrum based on improved differential evolution algorithm |
| CN114329702B (en) * | 2021-12-21 | 2022-07-22 | 中国地震局工程力学研究所 | High-stability method and device for calibrating and designing reaction spectrum based on improved differential evolution algorithm |
| CN114389265A (en) * | 2022-03-23 | 2022-04-22 | 山东大学 | Power grid emergency load optimization method and system based on proliferation and reduced evolution |
| CN116031879A (en) * | 2023-02-28 | 2023-04-28 | 四川大学 | Hybrid intelligent feature selection method suitable for transient voltage stability evaluation of power system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112632840A (en) | Power grid transient stability evaluation method based on adaptive differential evolution algorithm and ELM | |
| CN111179121B (en) | Power grid emergency control method based on expert system and deep reverse reinforcement learning | |
| Liu et al. | A systematic approach for dynamic security assessment and the corresponding preventive control scheme based on decision trees | |
| Zhang et al. | Real-time assessment of fault-induced delayed voltage recovery: A probabilistic self-adaptive data-driven method | |
| CN109614997A (en) | A kind of stealing Risk Forecast Method and device based on deep learning | |
| CN103488869A (en) | Wind power generation short-term load forecast method of least squares support vector machine | |
| CN111523785A (en) | Power system dynamic security assessment method based on generation countermeasure network | |
| CN104155574A (en) | Power distribution network fault classification method based on adaptive neuro-fuzzy inference system | |
| CN104617574A (en) | Assessment method for transient voltage stabilization of load area of electrical power system | |
| CN108876163A (en) | The transient rotor angle stability fast evaluation method of comprehensive causality analysis and machine learning | |
| An et al. | A data-driven method for transient stability margin prediction based on security region | |
| CN106485594A (en) | A kind of main distribution integration incident response decision method | |
| Vasconcelos et al. | Online security assessment with load and renewable generation uncertainty: The itesla project approach | |
| CN111695666A (en) | Wind power ultra-short term conditional probability prediction method based on deep learning | |
| Brahimi et al. | Prediction of wind speed distribution using artificial neural network: The case of Saudi Arabia | |
| Kharlamova et al. | Evaluating machine-learning-based methods for modeling a digital twin of battery systems providing frequency regulation | |
| Shahriyari et al. | A Deep Learning-Based Approach for Comprehensive Rotor Angle Stability Assessment | |
| Alharbi et al. | Short-term wind speed and temperature forecasting model based on gated recurrent unit neural networks | |
| Gupta et al. | Wind ramp event prediction with parallelized gradient boosted regression trees | |
| CN113033898A (en) | Electrical load prediction method and system based on K-means clustering and BI-LSTM neural network | |
| Zhang et al. | Data-driven risk assessment early-warning model for power system transmission congestions | |
| CN115564310A (en) | Reliability evaluation method for new energy power system based on convolutional neural network | |
| Wang et al. | Short-term wind power forecasting based on BOMLS K-means similar hours Clustering method | |
| Jianli et al. | Wind power forecasting by using artificial neural networks and Grubbs criterion | |
| Xiao et al. | The short-time traffic flow prediction at ramp junction based on wavelet neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210409 |
|
| RJ01 | Rejection of invention patent application after publication |