CN110336277B - Power system transient stability evaluation method based on deep belief network - Google Patents

Power system transient stability evaluation method based on deep belief network Download PDF

Info

Publication number
CN110336277B
CN110336277B CN201910623078.1A CN201910623078A CN110336277B CN 110336277 B CN110336277 B CN 110336277B CN 201910623078 A CN201910623078 A CN 201910623078A CN 110336277 B CN110336277 B CN 110336277B
Authority
CN
China
Prior art keywords
fault
phase angle
layer
power system
bus
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.)
Active
Application number
CN201910623078.1A
Other languages
Chinese (zh)
Other versions
CN110336277A (en
Inventor
王怀远
林楠
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fuzhou University
Original Assignee
Fuzhou University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Fuzhou University filed Critical Fuzhou University
Priority to CN201910623078.1A priority Critical patent/CN110336277B/en
Publication of CN110336277A publication Critical patent/CN110336277A/en
Application granted granted Critical
Publication of CN110336277B publication Critical patent/CN110336277B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Landscapes

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

Abstract

The invention relates to a power system transient stability evaluation method based on a deep belief network.

Description

Power system transient stability evaluation method based on deep belief network
Technical Field
The invention relates to the technical field of power systems, in particular to a power system transient stability evaluation method based on a deep belief network.
Background
With the gradual installation of PMUs in the power grid, the control center can acquire dynamic response information of each system from the WAMS in real time. With the rise of big data technology, the traditional pattern recognition method is expanded by a data-driven method, and a foundation is laid for workers to analyze the safety and stability of the power system from the big data perspective. The data-driven transient stability discrimination model can quickly acquire the state information of whether the system is stable or not by only constructing a mapping model between input quantity and output quantity and inputting the real-time response quantity of the system into the model.
The current characteristic quantity after the fault is mainly obtained through fixed delay, the identification degree of the obtained characteristic quantity to a critical sample is very low, and meanwhile, control information of controllers such as a system speed regulator, a pressure regulator and the like cannot be reflected too much, so that the evaluation precision of a model cannot meet the requirement.
Disclosure of Invention
In view of the above, the present invention provides a method for evaluating transient stability of an electric power system based on a deep belief network, which extracts a feature quantity after a fault according to a maximum phase angle difference value of a bus phase angle of the electric power system in a swing process, and inputs the feature into the deep belief network to realize rapid evaluation of the transient stability of the electric power system.
The invention is realized by adopting the following scheme: a method for evaluating transient stability of an electric power system based on a deep belief network comprises the steps of extracting feature quantity after a fault according to the maximum phase angle difference value of a bus phase angle of the electric power system in a swinging process, inputting the features into the deep belief network, and realizing rapid evaluation of the transient stability of the electric power system.
Further, the method comprises the following steps:
step S1: acquiring a large number of training samples through different fault information; obtaining a large number of training samples by permutation and combination of different operation conditions, power transmission lines, fault positions, fault duration and the like;
step S2: when a fault occurs, extracting characteristic quantities before the fault occurs, at the fault occurrence time and at the fault clearing time; the characteristic quantity comprises voltage, voltage phase angle and frequency information of each bus;
step S3: taking the maximum phase angle difference value of each bus in the power system as a threshold value, and extracting the characteristic quantity after the fault;
step S4: training a deep belief network model by combining unsupervised pre-training and supervised fine tuning;
step S5: and inputting the actually measured characteristic quantities before the fault, the moment when the fault occurs, the moment when the fault is cleared and after the fault into the deep confidence network model trained in the step S4, and evaluating the stability.
Further, step S3 is specifically: maximum phase angle difference delta theta of voltage phase angle on each bus after fault clearingtThe calculation method is as follows:
Δθt＝θmax,t-θmin,t； (1)
in the formula, thetamax,tAnd thetamin,tRespectively is a maximum phase angle and a minimum phase angle corresponding to the bus voltage phase angle in the system at the time t; when the following formula is satisfied, extracting the characteristic quantity after the fault, wherein the characteristic quantity after the fault comprises the voltage, the voltage angle and the frequency information of all bus nodes:
ΔθT≥θset； (2)
in the formula, thetasetIs the set phase angle threshold.
Further, step S4 is specifically: training is started from the bottommost layer of limited Boltzmann machine RBM, after the learning of the characteristics of the layer is fully completed, the hidden layer of the layer of RBM is used as the visible layer of the next layer of RBM, the training of the next layer of RBM is continued, and the steps are repeated until all RBMs are fully trained; and adding a back propagation neural network on the top layer, and propagating error information to each layer of RBM from top to bottom according to the label information, thereby performing global adjustment on the DBN parameters.
Preferably, in step S5, when a fault occurs in the actual system, information such as voltage, voltage phase angle, and frequency of each bus before the fault occurs, at the time of the fault occurrence, and at the time of clearing the fault is extracted; information such as the voltage, the voltage phase angle, and the frequency of each bus after the fault is extracted according to equation (2). And inputting the information into the deep confidence network model for stability evaluation.
Compared with the prior art, the invention has the following beneficial effects:
1. the characteristic quantity extracted by the method can better reflect the relevant information of the system stability, thereby laying a foundation for high-precision evaluation;
2. according to the method, an evaluation model based on a deep belief network is constructed, and the high classification performance of deep learning is utilized, so that the rapid evaluation of the transient stability of the power system is realized.
Drawings
Fig. 1 is a deep belief network model structure according to an embodiment of the present invention.
Fig. 2 is a wiring diagram of an IEEE39 node system according to an embodiment of the present invention.
Detailed Description
The invention is further explained below with reference to the drawings and the embodiments.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. 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 application 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 example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiment provides a power system transient stability evaluation method based on a deep belief network, which is characterized in that the extraction of the post-fault characteristic quantity is carried out according to the maximum phase angle difference value of the bus phase angle of the power system in the swing process, and the characteristic is input into the deep belief network, so that the rapid evaluation of the transient stability of the power system is realized.
In this embodiment, the method comprises the following steps:
step S1: acquiring a large number of training samples through different fault information; obtaining a large number of training samples by permutation and combination of different operation conditions, power transmission lines, fault positions, fault duration and the like;
step S2: when a fault occurs, extracting characteristic quantities before the fault occurs, at the fault occurrence time and at the fault clearing time; the characteristic quantity comprises voltage, voltage phase angle and frequency information of each bus;
step S3: taking the maximum phase angle difference value of each bus in the power system as a threshold value, and extracting the characteristic quantity after the fault;
step S4: training a deep belief network model by combining unsupervised pre-training and supervised fine tuning;
step S5: and inputting the actually measured characteristic quantities before the fault, the moment when the fault occurs, the moment when the fault is cleared and after the fault into the deep confidence network model trained in the step S4, and evaluating the stability.
In this embodiment, step S3 specifically includes: maximum phase angle difference delta theta of voltage phase angle on each bus after fault clearingtThe calculation method is as follows:
Δθt＝θmax,t-θmin,t； (1)
in the formula, thetamax,tAnd thetamin,tRespectively is a maximum phase angle and a minimum phase angle corresponding to the bus voltage phase angle in the system at the time t; when the following formula is satisfied, extracting the characteristic quantity after the fault, wherein the characteristic quantity after the fault comprises the voltage, the voltage angle and the frequency information of all bus nodes:
ΔθT≥θset； (2)
in the formula, thetasetIs the set phase angle threshold.
In this embodiment, as shown in fig. 1, step S4 specifically includes: training is started from the bottommost layer of limited Boltzmann machine RBM, after the learning of the characteristics of the layer is fully completed, the hidden layer of the layer of RBM is used as the visible layer of the next layer of RBM, the training of the next layer of RBM is continued, and the steps are repeated until all RBMs are fully trained; and adding a back propagation neural network on the top layer, and propagating error information to each layer of RBM from top to bottom according to the label information, thereby performing global adjustment on the DBN parameters.
Preferably, in this embodiment, in step S5, when a fault occurs in the actual system, information such as voltage, voltage phase angle, and frequency of each bus before the fault occurs, at the time of the fault occurrence, and at the time of clearing the fault is extracted; information such as the voltage, the voltage phase angle, and the frequency of each bus after the fault is extracted according to equation (2). And inputting the information into the deep confidence network model for stability evaluation.
The simulation software adopted in the embodiment is PSD-BPA, and the test system adopts an IEEE-39 node system (as shown in figure 2) and is provided with a speed regulator and a voltage regulator. The load level of the system considers the load conditions of 90%, 100% and 110%, the faults are three-phase short circuits, the fault positions are respectively located at 10%, 30%, 50%, 70% and 90% of the line, the fault duration time is respectively 100ms, 120ms, 130ms, 140ms, 150ms, 160ms, 180ms, 200ms, 250ms and 300ms, and 10 fault clearing times are obtained. A total of 4950 samples were generated by simulation, with 2933 samples remaining stable and 2017 samples not stable. And randomly extracting sample data from the stable samples and the unstable samples according to a proportion to form a training set and a testing set.
According to the method of the embodiment, characteristic quantity information before the fault, at the moment of the fault occurrence, at the moment of the fault clearing and after the fault is extracted.
The accuracy of the evaluation model based on the deep belief network when different phase angle thresholds were compared is shown in table 1.
TABLE 1 results of discrimination of different phase angle thresholds
Figure BDA0002126134530000061
As can be seen from the above table, the setting of the phase angle threshold can make the difference between the extracted feature quantities of the critical situation more obvious, and meanwhile, the evaluation model based on the deep confidence network can realize high-precision evaluation.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is directed to preferred embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow. However, any simple modification, equivalent change and modification of the above embodiments according to the technical essence of the present invention are within the protection scope of the technical solution of the present invention.

Claims (2)

1. A power system transient stability assessment method based on a deep belief network is characterized in that the method comprises the steps of extracting the feature quantity after a fault according to the maximum phase angle difference value of a bus phase angle of a power system in the swing process, inputting the feature into the deep belief network, and achieving the rapid assessment of the transient stability of the power system;
the method comprises the following steps:
step S1: acquiring a large number of training samples through different fault information;
step S2: when a fault occurs, extracting characteristic quantities before the fault occurs, at the fault occurrence time and at the fault clearing time; the characteristic quantity comprises voltage, voltage phase angle and frequency information of each bus;
step S3: taking the maximum phase angle difference value of each bus in the power system as a threshold value, and extracting the characteristic quantity after the fault;
step S4: training a deep belief network model by combining unsupervised pre-training and supervised fine tuning;
step S5: inputting the actually measured characteristic quantities before the fault, the moment when the fault occurs, the moment when the fault is cleared and after the fault into the deep confidence network model trained in the step S4, and evaluating the stability;
wherein, step S3 specifically includes: maximum phase angle difference delta theta of voltage phase angle on each bus after fault clearingtThe calculation method is as follows:
Δθt＝θmax,t-θmin,t；
in the formula, thetamax,tAnd thetamin,tRespectively is a maximum phase angle and a minimum phase angle corresponding to the bus voltage phase angle in the system at the time t; when the following formula is satisfied, extracting the characteristic quantity after the fault, wherein the characteristic quantity after the fault comprises the voltage, the voltage angle and the frequency information of all bus nodes:
ΔθT≥θset；
in the formula, thetasetIs the set phase angle threshold.
2. The method for evaluating transient stability of power system based on deep belief network as claimed in claim 1, wherein the step S4 specifically comprises: training is started from the bottommost layer of limited Boltzmann machine RBM, after the learning of the characteristics of the layer is fully completed, the hidden layer of the layer of RBM is used as the visible layer of the next layer of RBM, the training of the next layer of RBM is continued, and the steps are repeated until all RBMs are fully trained; and adding a back propagation neural network on the top layer, and propagating error information to each layer of RBM from top to bottom according to the label information, thereby performing global adjustment on the DBN parameters.
CN201910623078.1A 2019-07-11 2019-07-11 Power system transient stability evaluation method based on deep belief network Active CN110336277B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910623078.1A CN110336277B (en) 2019-07-11 2019-07-11 Power system transient stability evaluation method based on deep belief network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910623078.1A CN110336277B (en) 2019-07-11 2019-07-11 Power system transient stability evaluation method based on deep belief network

Publications (2)

Publication Number Publication Date
CN110336277A CN110336277A (en) 2019-10-15
CN110336277B true CN110336277B (en) 2021-03-30

Family

ID=68146361

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910623078.1A Active CN110336277B (en) 2019-07-11 2019-07-11 Power system transient stability evaluation method based on deep belief network

Country Status (1)

Country Link
CN (1) CN110336277B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008011630A (en) * 2006-06-28 2008-01-17 Central Res Inst Of Electric Power Ind Apparatus and method for determining amount of voltage increase control and system and method for controlling generator
WO2018020933A1 (en) * 2016-07-27 2018-02-01 株式会社日立製作所 Stability monitoring device and method
CN109886604A (en) * 2019-03-13 2019-06-14 华北电力大学 A kind of transient stability evaluation in power system method based on one-dimensional convolutional neural networks

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102324781B (en) * 2011-09-08 2014-03-12 天津理工大学 Power supply quick switching control system and switching method thereof
CN107391852B (en) * 2017-07-26 2020-07-10 清华大学 Transient stability real-time evaluation method and device based on deep belief network
CN107993012B (en) * 2017-12-04 2022-09-30 国网湖南省电力有限公司娄底供电分公司 Time-adaptive online transient stability evaluation method for power system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008011630A (en) * 2006-06-28 2008-01-17 Central Res Inst Of Electric Power Ind Apparatus and method for determining amount of voltage increase control and system and method for controlling generator
WO2018020933A1 (en) * 2016-07-27 2018-02-01 株式会社日立製作所 Stability monitoring device and method
CN109886604A (en) * 2019-03-13 2019-06-14 华北电力大学 A kind of transient stability evaluation in power system method based on one-dimensional convolutional neural networks

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Study on real-time clustering method for power system transient stability assessment;Fu Su 等;《2016 IEEE Power and Energy Society General Meeting (PESGM)》;20160721;第1-5页 *
基于响应的电压稳定性评估和控制方法;高飞翎等;《电网与清洁能源》;20181130;第34卷(第11期);第8-14页 *

Also Published As

Publication number Publication date
CN110336277A (en) 2019-10-15

Similar Documents

Publication Publication Date Title
CN109800660B (en) Voltage sag source identification method and system based on big data clustering
CN110108992B (en) Cable partial discharge fault identification method and system based on improved random forest algorithm
CN111325223B (en) Training method and device for deep learning model and computer readable storage medium
CN103177188A (en) Electric system load dynamic characteristic classifying method based on characteristic mapping
CN107220505A (en) Child development capability assessment method and device
CN113092900B (en) State detection method and device of photovoltaic inverter and computer readable storage medium
CN112017070A (en) Method and system for evaluating transient stability of power system based on data enhancement
CN104636748B (en) A kind of method and device of number plate identification
Le et al. Ensemble machine learning based adaptive arc fault detection for DC distribution systems
CN110336277B (en) Power system transient stability evaluation method based on deep belief network
CN115186564A (en) Analog circuit fault diagnosis method based on feature fusion and improved particle swarm algorithm
CN111651932A (en) Online dynamic security assessment method for power system based on integrated classification model
CN110610140A (en) Training method, device and equipment of face recognition model and readable storage medium
CN104049159A (en) Fault detection method and device of inverter
CN105868918A (en) Similarity index computing method of harmonic current type monitoring sample
Papadopoulos et al. Online parameter identification and generic modeling derivation of a dynamic load model in distribution grids
Grzechca et al. Evolutionary algorithms for global parametric fault diagnosis in analogue integrated circuits
CN105608460A (en) Method and system for fusing multiple classifiers
CN109001556B (en) Discrimination method and system for forced ultralow frequency oscillation
CN116204771A (en) Power system transient stability key feature selection method, device and product
CN112051479A (en) Power distribution network operation state identification method and system
Chen et al. Coherent clustering method based on weighted clustering of multi-indicator panel data
CN113794198B (en) Method, device, terminal and storage medium for suppressing broadband oscillation
CN106709598B (en) Voltage stability prediction and judgment method based on single-class samples
CN105116323A (en) Motor fault detection method based on RBF

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant