CN110021929B - Electromagnetic transient time domain simulation modeling method for fast switch type fault current limiter - Google Patents
Electromagnetic transient time domain simulation modeling method for fast switch type fault current limiter Download PDFInfo
- Publication number
- CN110021929B CN110021929B CN201811641975.7A CN201811641975A CN110021929B CN 110021929 B CN110021929 B CN 110021929B CN 201811641975 A CN201811641975 A CN 201811641975A CN 110021929 B CN110021929 B CN 110021929B
- Authority
- CN
- China
- Prior art keywords
- current limiter
- current
- network
- branch
- equivalent
- 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
Links
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H9/00—Emergency protective circuit arrangements for limiting excess current or voltage without disconnection
- H02H9/02—Emergency protective circuit arrangements for limiting excess current or voltage without disconnection responsive to excess current
-
- 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
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
An electromagnetic transient time domain simulation modeling method for a fast switch type fault current limiter. Based on a big data means, carrying out feature analysis on test data of a large number of current limiters and operation data of a power system in various current limiting scenes, and combining inherent parameters and structural features of the rapid switching type fault current limiter to obtain a standardized data set suitable for machine learning; subsequently, extracting a characteristic vector in a data set, constructing a training sample, constructing a BP neural network algorithm flow, and obtaining a correction function of the current limiter Norton equivalent circuit through training; and finally, building a current limiter electromagnetic transient simulation module based on the obtained current limiter Noton equivalent circuit correction function, designing an electromagnetic transient simulation flow, and simulating the electromagnetic transient characteristics of the current limiter. The electromagnetic transient simulation model obtained by the method can simulate the electromagnetic time domain transient characteristic response of each branch in the quick switch type fault current limiter with different performance parameters under different current limiting scenes.
Description
Technical Field
The invention relates to an EMTP electromagnetic transient time domain simulation modeling method for a fault current limiter.
Background
With the continuous and rapid growth of national economy, the power grid of China is also rapidly developed. The scale of the power grid is gradually enlarged, the power consumption demand is sharply increased, and the short-circuit current level of the system is continuously increased along with the load increase of the power system, the operation of large-capacity units and the interconnection of each large-area power grid and even a transnational power grid.
The excessive short-circuit current brings a series of negative effects to the power system, greatly threatens the safe and stable operation of the power system, destroys the energy balance of the system, causes the oscillation of the power system, seriously leads to the loss of the synchronization of the generator, causes the disconnection of the system and causes large-area power failure accidents. In addition, the heat effect generated when the short-circuit current flows through the conductor equipment enables the short-circuit loop and the current of nearby branches to greatly rise, and the insulation between the electrical equipment and a system is damaged; short-circuit faults reduce system voltage, affect power supply quality and bring huge economic loss to industrial production; short-circuit current generates considerable electrodynamic force effect through the conductor, so that the conductor is deformed and damaged, and accidents are enlarged; the asymmetric grounding short circuit fault can generate induced electromotive force on nearby communication lines, so that normal communication is interfered, and the safety of communication equipment is threatened; the increase of the short-circuit current level can improve the requirement of the thermodynamic stability of the power system and greatly increase the investment cost of power grid construction.
In certain economically developed areas of China, the expected short-circuit current level of a power grid exceeds the breaking capacity of a circuit breaker which can be produced in China. If the short-circuit current exceeds the breaking capacity of the breaker, the fault cannot be isolated from the system, and the fault accident is expanded and spread. In order to ensure the safe and stable operation of the power system, the increase of the short-circuit current level of the system must be restrained. The traditional measures for limiting the short-circuit current are mainly divided into three aspects: adjusting the structure of a power grid, changing the operation mode of a system, and additionally installing power equipment to limit short-circuit current. Although conventional current limiting measures reduce the short circuit current level to some extent, there are a number of problems and disadvantages.
The economical current limiter is a reasonable, effective and economical current limiting measure when additionally arranged in the power system. The quick switching type fault current limiter is a fault current limiter which is quick in action, high in reliability and good in economical efficiency, almost has zero loss when a system normally operates, meets the energy-saving requirement of the system, is small in occupied area and simple to maintain, and can be flexibly applied to various current-limiting occasions of a power system. In addition, the quick switch type fault current limiter can be upgraded and modified by installing a quick switch on the basis of the existing current limiting reactor of a power grid, and the technical economy is good.
Currently, there is less research on simulation models describing the electromagnetic transient process of fast switching fault current limiters. Transient process of the fault current limiter is very short and is influenced by various factors, transient and complexity enable a traditional method to be incapable of describing complex target electromagnetic characteristics in an accurate and quantitative mode, and a fault current limiter model of the complex target electromagnetic characteristics is difficult to establish. In most of the existing researches, the electromagnetic transient process of the fault current limiter is idealized from the aspect of power system operation, the power system is integrated for simulation analysis, and the transient characteristics of complex electromagnetic coupling before and after the action of the fault current limiter are not considered. The fault current limiter lacks a detailed electromagnetic transient process simulation model, a series of engineering problems are brought to site selection optimization and capacity configuration of the economic fault current limiter and refitting configuration of an inherent current limiting reactor, and engineering popularization and use of the current limiter are hindered.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a simulation modeling method for electromagnetic transient time domain characteristics of a fast switch type fault current limiter based on a big data means.
The invention establishes a fast switch type fault current limiter electromagnetic transient simulation model which accords with the actual situation by carrying out big data characteristic analysis on test data of a large number of current limiters and operation data of an electric power system under various current limiting scenes and combining the intrinsic parameters and the structural characteristics of the fast switch type fault current limiter. The electromagnetic transient simulation model obtained by the method can simulate the electromagnetic time domain transient characteristic response of each branch in the quick switch type fault current limiter with different performance parameters under different current limiting scenes, has important effects on the manufacturing and the engineering practical application of the quick switch type fault current limiter, and has high reference value on the topological structure design and the parameter selection of a novel fault current limiter.
The technical scheme adopted by the invention is as follows:
firstly, acquiring big data of installation and operation environments of the current limiter, and carrying out standardization processing on environmental parameters, operation parameters and intrinsic parameters of the current limiter to form a standard data set; then extracting a characteristic vector in the data set, constructing a training sample, constructing a BP neural network algorithm flow, and obtaining a correction function of the current limiter Norton equivalent circuit through training; and finally, building a current limiter electromagnetic transient simulation model based on the obtained current limiter Noton equivalent circuit correction function, designing an electromagnetic transient simulation flow, and simulating the electromagnetic transient characteristics of the current limiter.
The method comprises the following specific steps:
1. building a standard data set
Firstly, acquiring data such as air pressure, temperature, humidity and the like of a current limiter installation environment; obtaining intrinsic parameters of the restrictor mechanism, such as: intrinsic parameters of a current-limiting reactance, a closing excitation coil, an opening excitation coil, a repulsion copper disc, an energy storage capacitor, a vacuum arc-extinguishing chamber, a contact and a disc spring retaining mechanism, such as geometry, electromagnetism, machinery and the like; obtaining external equivalent network parameters under different fault scenes: such as an equivalent voltage source, an equivalent current source, an equivalent resistor, etc.; and acquiring corresponding electromagnetic transient process test data before and after the action of the current limiter, such as pulse current, current-limiting reactance current, fast switching current, current-limiting current, current limiter voltage and other waveform data, current on-off phase, on-off current and other initial conditions. And (3) carrying out big data standardization processing on the environment parameters and the intrinsic parameters of the current limiter, unifying the environment parameters and the intrinsic parameters as an environment influence factor set, and recording the environment influence factor set as:
in the formula, M c Represents a table of influence factors and is used for determining the influence factors,
representing an influence factor, the superscript i representing the index number of the influence factor, and the subscript c representing the index number of the data set sample;
and marking the time sampling sequences of the current and voltage waveforms inside and outside the current limiter as follows:
in the formula, O represents a branch current and voltage array, i represents a branch current, u represents a branch voltage, subscript k represents an internal network, subscript s represents an external network, and superscript t represents time;
norton equivalent network substitution is made for the current limiter external port and is noted as:
in the formula, N represents a branch Noton equivalent parameter array, I represents a branch Noton equivalent current, G represents a branch Noton equivalent conductance, subscript s represents an external network, and superscript t represents time;
according to the data set, calculating the Norton equivalent network parameters of the internal branch of the current limiter, and recording as:
in the formula, N represents a branch norton equivalent parameter array, I represents a branch norton equivalent current, G represents a branch norton equivalent conductance, subscript k represents an internal network, superscript t represents time, and the norton equivalent parameter is determined by the following formula:
wherein, delta t is the simulation calculation step length.
Establishing a time sample data set based on the data extraction result;
in the formula, A represents a sample data table, M represents a factor table, N represents a branch Norton equivalent parameter, O represents a branch current and voltage array, c represents a sample data index number, t represents time, k represents an internal network, and s represents an external network.
2. Construction of a current limiter Norton equivalent circuit correction function
Firstly, a multi-layer perceptron is established, namely a learning function of a supervised learning algorithm is established.
f(·):R m →R n
In the formula R m Representing the value of the input variable as an m-dimensional vector space, R n The value of the representative output variable is n-dimensional vector space, and the training function f is the mapping from m-dimensional vector space to n-dimensional vector space.
From the above data set
Extracting feature vectors as an input layer of the BP neural network, and recording as:
in the formula, X represents an input variable, M represents a factor table, O represents a branch current and voltage array, c represents a sample data index number, k represents an internal network, s represents an external network, delta t is a simulation calculation step length, t 0 And the action triggering moment of the current limiter, t is the current simulation moment, and m represents the dimension of an input vector, and the size of the dimension is related to the number of network branches and the number of influence factors of the data set.
From the above data set
Extracting feature vectors as supervision and guidance samples of the BP neural network output layer, and recording as follows:
Y＝{y 1 ,y 2 ,…,y k ,…,y n } n
wherein Y is an output supervisory variable,
g represents the branch norton equivalent conductance, delta t is the simulation calculation step length, n is the output vector dimension, and the size of the output vector dimension is related to the number of branches in the model.
And (3) through a supervised function regression learning algorithm, recording a current training feedback result as:
obtained by BP neural network algorithm training
Substituting the internal network parameter modification function as a simulation model into a function obtained by training to modify a current limiter Noton equivalent circuit;
I k (t+Δt)＝i k (t)
G k (t+Δt)＝f(X(t))·Δt
in the formula I k ,G k For the norton equivalent current and conductance, i, of the internal branch of the current limiter k The branch current is used, X is an input variable, an equivalent circuit parameter is obtained through simultaneous external network port calculation, namely, the network state at the time of t + delta t is obtained, and then the next network state is obtained:
i k (t+Δt),u k (t+Δt),i s (t+Δt),u s (t+Δt)
in the formula i k ,u k ,i s ,u s For branches of internal or external networks of current limitersThe current voltage.
3. Establishing a current limiter electromagnetic transient simulation model
Firstly, a logic judgment module is established:
(1) A flow device state judgment module: controlling the flow of a simulation algorithm according to the current commissioning state of the current limiter, and if the current limiter is in the commissioning state, changing the topology structure of the whole network, generating a current limiter sub-network and solving internal variables; if the current limiter is in the cut-off state, the influence of the current limiter is not considered, the current limiter jumps out of the sub-network for calculation, and the simulation calculation task of the external network is normally executed;
(2) The switching triggering module of the current limiter: when the current limiter is in the cut-off state, throwing out a commissioning trigger signal to enable the current limiter to enter the commissioning state, generating a current limiter sub-network and solving internal variables; when the current limiter is in a commissioning state, a cutting-off trigger signal is thrown out, so that the current limiter enters a cutting-off action state, and the action state identifier is changed to control the calculation mode of a subsequent calculation module.
Subsequently, a sub-network core computation module is established:
(1) A current limiter Noton equivalent parameter calculation module: substituting the current voltage of the external port branch of the current limiter, the current voltage of the internal branch, an action instruction and other environmental parameters into a norton parameter correction function of the current limiter obtained by the BP neural network training based on the big data to correct the norton equivalent parameters of the internal accompanying circuit of the current limiter;
(2) A current limiter internal variable calculation module: calculating the outer port network of the current limiter to obtain the parameters of the sub-network of the current limiter when the parameters of the current limiter are parallelly connected to form a node admittance matrix Y m×m And solving the node voltage equation Y m×m U m ＝I m Wherein, U m For the voltage vector of the sub-network nodes within the current limiter, I m And calculating internal branch voltage, current and other variables of each element for the current limiter internal sub-network node current vector.
And finally, establishing an EMTP electromagnetic transient simulation flow, and circularly calculating the system states of a series of equally-spaced discrete time points, wherein the calculation process is as follows:
(1) Network initialization: importing the logic judgment module and the sub-network core calculation module;
(2) And operating a logic judgment module, and selecting and calling a current limiter Noton equivalent parameter calculation module to calculate the equivalent parameters of the current limiter model. If the current limiter acts, the network topology structure and the calculation flow need to be changed, a sub-network is established at the port of the current limiter based on the current limiter topology, and the Norton equivalent parameters of the accompanying circuits of each branch circuit are updated according to the initial action value and the network state at the moment. Calculating the Norton equivalent current I of other network elements l And norton's equivalent admittance vector G l ；
(3) Operating a logic judgment module, selectively calling a current limiter internal variable calculation module to calculate the current of the current limiter internal node, and forming a node current vector I according to the network topology at the moment n Forming a node admittance matrix Y n×n Solving the node voltage equation Y n×n U n ＝I n Wherein, U n Is a node voltage vector. Then, calculating variables such as internal branch voltage, current and the like of each element;
(4) And operating a logic judgment module to judge whether the set simulation step length is reached or not and calculating the network state at the next moment.
Drawings
Fig. 1 is a calculation flow chart of the EMTP algorithm of the fast switching type fault current limiter of the present invention.
Detailed Description
The invention is further described with reference to the following drawings and detailed description.
The construction implementation mode of the electromagnetic transient time domain simulation modeling method of the rapid switch type fault current limiter is as follows:
1. first, a standard data set is constructed
Acquiring data such as air pressure, temperature, humidity and the like of the installation environment of the current limiter; obtaining intrinsic parameters of the restrictor mechanism, such as: intrinsic parameters of a current-limiting reactance, a closing excitation coil, an opening excitation coil, a repulsion copper disc, an energy storage capacitor, a vacuum arc-extinguishing chamber, a contact and a disc spring retaining mechanism, such as geometry, electromagnetism, machinery and the like; obtaining external equivalent network parameters under different fault scenes: such as an equal voltage source, an equal current source, an equal resistance, etc.; and acquiring corresponding electromagnetic transient process test data before and after the action of the current limiter, such as pulse current, current-limiting reactance current, fast switching current, current-limiting current, current limiter voltage and other waveform data, current on-off phase, on-off current and other initial conditions. And (3) carrying out big data standardization processing on the environment parameters and the intrinsic parameters of the current limiter, unifying the environment parameters and the intrinsic parameters as an environment influence factor set, and recording the environment influence factor set as:
in the formula, M c Represents a table of influence factors and is used for determining the influence factors,
representing an influence factor, the superscript i representing the index number of the influence factor, and the subscript c representing the index number of the data set sample;
time sampling sequences of current and voltage waveforms inside and outside the current limiter are marked as follows:
in the formula, O represents a branch current and voltage array, i represents a branch current, u represents a branch voltage, subscript k represents an internal network, subscript s represents an external network, and superscript t represents time;
norton equivalent network substitution is made for the current limiter external port and is noted as:
in the formula, N represents a branch Noton equivalent parameter array, I represents a branch Noton equivalent current, G represents a branch Noton equivalent conductance, subscript s represents an external network, and superscript t represents time;
according to the data set, calculating the norton equivalent network parameters of the internal branch of the current limiter, and recording the parameters as:
in the formula, N represents a branch norton equivalent parameter array, I represents a branch norton equivalent current, G represents a branch norton equivalent conductance, subscript k represents an internal network, superscript t represents time, and the norton equivalent parameter is determined by the following formula;
wherein, delta t is the simulation calculation step length.
Based on the data extraction result, establishing a time sample data set;
in the formula, A represents a sample data table, M represents a factor table, N represents a branch Norton equivalent parameter, O represents a branch current and voltage array, c represents a sample data index number, t represents time, k represents an internal network, and s represents an external network.
2. Then constructing a current limiter Noton equivalent circuit correction function;
from the above data set
Extracting feature vectors as an input layer of the BP neural network, and recording as:
in the formula, X represents an input variable, M represents a factor table, O represents a branch current and voltage array, c represents a sample data index number, k represents an internal network, s represents an external network, delta t is a simulation calculation step length, t 0 And the action triggering moment of the current limiter, t is the current simulation moment, and m represents the dimension of an input vector, and the size of the dimension is related to the number of network branches and the number of influence factors of the data set.
From the above data set
Extracting feature vectors as supervision and guidance samples of the BP neural network output layer, and recording as follows:
Y＝{y 1 ,y 2 ,…,y k ,…,y n } n
wherein Y is an output supervisory variable,
g represents the branch norton equivalent conductance, delta t is the simulation calculation step length, n is the output vector dimension, and the size of the output vector dimension is related to the number of branches in the model.
And (3) through a supervised function regression learning algorithm, recording a current training feedback result as:
obtained by BP neural network algorithm training
Substituting the internal network parameter modification function as a simulation model into the function obtained by training to modify the current limiter Noton equivalent circuit
I k (t+Δt)＝i k (t)
G k (t+Δt)＝f(X(t))·Δt
In the formula I k ,G k To limitNorton equivalent current and conductance i of internal branch of current transformer k The branch current is used, X is an input variable, an equivalent circuit parameter is obtained through simultaneous external network port calculation, namely, the network state at the time of t + delta t is obtained, and then the next network state is obtained:
i k (t+Δt),u k (t+Δt),i s (t+Δt),u s (t+Δt)
in the formula i k ,u k ,i s ,u s The current voltage of each branch of the internal and external network of the current limiter.
3. Then establishing a current limiter electromagnetic transient simulation model;
establishing a logic judgment module:
(1) Establishing a current limiter state judgment module: controlling the flow of a simulation algorithm according to the current commissioning state of the current limiter, and if the current limiter is in the commissioning state, changing the topology structure of the whole network, generating a current limiter sub-network and solving internal variables; and if the current limiter is in the cut-off state, the influence of the current limiter is not considered, the sub-network calculation is carried out, and the simulation calculation task of the external network is normally executed.
(2) Establishing a current limiter switching trigger module: when the current limiter is in the cut-off state, throwing out a commissioning trigger signal to enable the current limiter to enter the commissioning state, generating a current limiter sub-network and solving internal variables; when the current limiter is in a commissioning state, a cutting trigger signal is thrown out, so that the current limiter enters a cutting action state, and an action state identifier is changed to control the calculation mode of a subsequent calculation module.
Establishing a sub-network core computing module:
(1) Establishing a current limiter Noton equivalent parameter calculation module: substituting the current voltage of the external port branch of the current limiter, the current voltage of the internal branch, an action instruction and other environmental parameters into a norton parameter correction function of the current limiter obtained by the BP neural network training based on the big data to correct the norton equivalent parameters of the internal accompanying circuit of the current limiter;
(2) Establishing a current limiter internal variable calculation module: external port network meter for current limiterCalculating the parameters of the current limiter sub-network when the accompanying circuit carries out the Norton equivalence and connects in parallel to form a node admittance matrix Y m×m And solving the node voltage equation Y m×m U m ＝I m Wherein, U m For the voltage vector of the sub-network nodes within the current limiter, I m And calculating internal branch voltage, current and other variables of each element for the current limiter internal sub-network node current vector.
And finally, establishing an EMTP electromagnetic transient simulation flow. As shown in fig. 1, the system status at a series of equally spaced discrete time points is calculated in a loop, and the calculation process is as follows:
(1) Network initialization: importing the logic judgment module and the sub-network core calculation module;
(2) And operating a logic judgment module, and selecting and calling a current limiter Noton equivalent parameter calculation module to calculate the equivalent parameters of the current limiter model. If the current limiter acts, the network topology structure and the calculation flow need to be changed, a sub-network is established at the port of the current limiter based on the current limiter topology, and the Norton equivalent parameters of the accompanying circuits of each branch circuit are updated according to the initial action value and the network state at the moment. Calculating the Norton equivalent current I of other network elements l And norton's equivalent admittance vector G l ；
(3) Operating a logic judgment module, selectively calling a current limiter internal variable calculation module to calculate the current of the internal node of the current limiter, and forming a node current vector I according to the network topology at the moment n Forming a node admittance matrix Y n×n Solving the node voltage equation Y n×n U n ＝I n Wherein, U n Is a node voltage vector. Then, calculating variables such as internal branch voltage, current and the like of each element;
(4) And operating a logic judgment module to judge whether the set simulation step length is reached or not and calculating the network state at the next moment.
Claims (4)
1. A modeling method for electromagnetic transient time domain simulation of a fast switch type fault current limiter is characterized in that the modeling method firstly obtains big data of installation and operation environments of the current limiter and carries out standardization processing on environment parameters, operation parameters and intrinsic parameters of the current limiter to form a standard data set; then extracting a characteristic vector in the data set, constructing a training sample, constructing a BP neural network algorithm flow, and obtaining a correction function of the current limiter Norton equivalent circuit through training; and finally, building a current limiter electromagnetic transient simulation model based on the obtained current limiter Noton equivalent circuit correction function, designing an electromagnetic transient simulation flow, and simulating the electromagnetic transient characteristics of the current limiter.
2. A method for modeling an electromagnetic transient time domain of a fast switching fault current limiter according to claim 1, wherein said standard data set is established by:
firstly, acquiring air pressure, temperature and humidity data of a current limiter installation environment; obtaining intrinsic parameters of the current limiter mechanism: the device comprises a current-limiting reactance, a closing excitation coil, an opening excitation coil, a repulsion copper disc, an energy storage capacitor, a vacuum arc-extinguishing chamber, a contact and geometric, electromagnetic and mechanical parameters of a disc spring retaining mechanism; obtaining external equivalent network parameters under different fault scenes: an equivalent voltage source, an equivalent current source and an equivalent resistor; acquiring electromagnetic transient process test data before and after the action of a corresponding current limiter: pulse current, current-limiting reactance current, fast switching current, current-limiting current, current limiter voltage, data and current on-off phase and on-off current; and (3) carrying out big data standardization processing on the environment parameters and the intrinsic parameters of the current limiter, unifying the environment parameters and the intrinsic parameters as an environment influence factor set, and recording the environment influence factor set as:
in the formula, M c Represents a table of influence factors and is used for determining the influence factors,
representing an influence factor, the superscript i representing the index number of the influence factor, and the subscript c representing the index number of the data set sample;
time sampling sequences of current and voltage waveforms inside and outside the current limiter are marked as follows:
in the formula, O represents a branch current and voltage array, i represents a branch current, u represents a branch voltage, subscript k represents an internal network, subscript s represents an external network, and superscript t represents time;
norton equivalent network substitution is made for the current limiter external port and is noted as:
in the formula, N represents a branch Noton equivalent parameter array, I represents a branch Noton equivalent current, G represents a branch Noton equivalent conductance, subscript s represents an external network, and superscript t represents time;
according to the data set, calculating the norton equivalent network parameters of the internal branch of the current limiter, and recording the parameters as:
in the formula, N represents a branch norton equivalent parameter array, I represents a branch norton equivalent current, G represents a branch norton equivalent conductance, subscript k represents an internal network, superscript t represents time, and the norton equivalent parameter is determined by the following formula:
in the formula, delta t is a simulation calculation step length;
establishing a time sample data set based on the data extraction result;
in the formula, A represents a sample data table, M represents an influence factor table, N represents a branch Norton equivalent parameter, O represents a branch current and voltage array, c represents a sample data index number, t represents time, k represents an internal network, and s represents an external network.
3. The electromagnetic transient time domain modeling method for the fast switch type fault current limiter according to claim 2 is characterized in that feature vectors in a data set are extracted, a training sample is constructed, a BP neural network algorithm flow is constructed, and the method for obtaining the correction function of the Norton equivalent circuit of the current limiter through training comprises the following steps:
firstly, establishing a multilayer perceptron, namely establishing a learning function of a supervised learning algorithm;
f(·):R m →R n
in the formula R m Representing the value of the input variable as an m-dimensional vector space, R n The value of the representative output variable is n-dimensional vector space, and the training function f is the mapping from m-dimensional vector space to n-dimensional vector space;
from the above data set
Extracting feature vectors as an input layer of the BP neural network, and recording as:
in the formula, X represents an input variable, M represents an influence factor table, O represents a branch current and voltage array, c represents a sample data index number, k represents an internal network, s represents an external network, delta t is a simulation calculation step length,t 0 the method comprises the steps that a current limiter acts as a trigger moment, t is a current simulation moment, m represents an input vector dimension, and the size of the input vector dimension is related to the number of network branches and the number of influence factors of a data set;
from the above data set
Extracting feature vectors as supervision and guidance samples of the BP neural network output layer, and recording as follows:
Y＝{y 1 ,y 2 ,…,y k ,…,y n } n
in the formula, Y is an output supervision variable,
g represents the branch norton equivalent conductance, delta t is the simulation calculation step length, n is the output vector dimension, and the size of the output vector dimension is related to the number of branches in the model;
and (3) through a supervised function regression learning algorithm, recording a current training feedback result as:
obtained by BP neural network algorithm training
Substituting the internal network parameter modification function as a simulation model into the function obtained by training to modify the current limiter Noton equivalent circuit:
I k (t+Δt)＝i k (t)
G k (t+Δt)＝f(X(t))·Δt
in the formula I k ，G k For the norton equivalent current and conductance, i, of the internal branch of the current limiter k The branch current is used, X is an input variable, an equivalent circuit parameter is obtained through simultaneous external network port calculation, namely, the network state at the time of t + delta t is obtained, and then the next network state is obtained:
i k (t+Δt)，u k (t+Δt)，i s (t+Δt)，u s (t+Δt)
in the formula i k ，u k ，i s ，u s The current and voltage of each branch of the internal and external networks of the current limiter.
4. The electromagnetic transient time domain modeling method of the fast switching fault current limiter according to claim 1, characterized in that the electromagnetic transient simulation model method of the current limiter is as follows:
firstly, a logic judgment module is established:
(1) A current limiter state judgment module: controlling the flow of a simulation algorithm according to the current commissioning state of the current limiter, and if the current limiter is in the commissioning state, changing the topology structure of the whole network, generating a current limiter sub-network and solving internal variables; if the current limiter is in the cut-off state, the influence of the current limiter is not considered, the current limiter jumps out of the sub-network for calculation, and the simulation calculation task of the external network is normally executed;
(2) The switching triggering module of the current limiter: when the current limiter is in the cut-off state, throwing out a commissioning trigger signal to enable the current limiter to enter the commissioning state, generating a current limiter sub-network and solving internal variables; when the current limiter is in a commissioning state, throwing a 'cutting-off' trigger signal to enable the current limiter to enter a 'cutting-off action state', and changing an action state identifier to control a calculation mode of a subsequent calculation module;
subsequently, a sub-network core computation module is set up:
(1) A current limiter Noton equivalent parameter calculation module: substituting the current voltage of the external port branch of the current limiter, the current voltage of the internal branch, an action instruction and other environmental parameters into a norton parameter correction function of the current limiter obtained by the BP neural network training based on the big data to correct the norton equivalent parameters of the internal accompanying circuit of the current limiter;
(2) A current limiter internal variable calculation module: calculating the outer port network of the current limiter to obtain the parameters of the sub-network of the current limiter when the parameters are connected in parallel to form the node admittance momentMatrix Y m×m And solving the node voltage equation Y m×m U m ＝I m Wherein, U m For the voltage vector of the sub-network nodes within the current limiter, I m Calculating variables such as internal branch voltage, current and the like of each element for a node current vector of an internal sub-network of the current limiter;
and finally, establishing an EMTP electromagnetic transient simulation flow, and circularly calculating the system states of a series of equally-spaced discrete time points, wherein the calculation process is as follows:
(1) Network initialization: importing the logic judgment module and the sub-network core calculation module;
(2) Running a logic judgment module, and selecting and calling a current limiter Norton equivalent parameter calculation module to calculate equivalent parameters of a current limiter model; if the current limiter acts, the network topology structure and the calculation flow need to be changed, a sub-network is established at the port of the current limiter based on the current limiter topology, and the Norton equivalent parameters of the accompanying circuits of each branch are updated according to the initial action value and the network state at the moment; calculating the Norton equivalent current I of other network elements l And norton's equivalent admittance vector G l ；
(3) Operating a logic judgment module, selectively calling a current limiter internal variable calculation module to calculate the current of the current limiter internal node, and forming a node current vector I according to the network topology at the moment n Forming a node admittance matrix Y n×n Solving the node voltage equation Y n×n U n ＝I n Wherein, U n Is a node voltage vector; then, calculating variables such as internal branch voltage, current and the like of each element;
(4) And operating a logic judgment module to judge whether the set simulation step length is reached or not and calculating the network state at the next moment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811641975.7A CN110021929B (en) | 2018-12-29 | 2018-12-29 | Electromagnetic transient time domain simulation modeling method for fast switch type fault current limiter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811641975.7A CN110021929B (en) | 2018-12-29 | 2018-12-29 | Electromagnetic transient time domain simulation modeling method for fast switch type fault current limiter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110021929A CN110021929A (en) | 2019-07-16 |
| CN110021929B true CN110021929B (en) | 2023-01-13 |
Family
ID=67188724
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811641975.7A Active CN110021929B (en) | 2018-12-29 | 2018-12-29 | Electromagnetic transient time domain simulation modeling method for fast switch type fault current limiter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110021929B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110879915B (en) * | 2019-10-21 | 2022-09-02 | 内蒙古电力(集团)有限责任公司内蒙古电力科学研究院分公司 | Method and system for external equivalence of power systems |
| CN112966463A (en) * | 2021-03-03 | 2021-06-15 | 通号（长沙）轨道交通控制技术有限公司 | Efficient simulation method for distributed grid-connected system of converter group |
| CN113077049B (en) * | 2021-06-04 | 2021-08-03 | 国网浙江省电力有限公司金华供电公司 | Intelligent regulation and control method for mobile terminal |
| CN114048699B (en) * | 2021-11-22 | 2024-04-26 | 深圳技术大学 | Method for establishing and calculating electrothermal and electromagnetic transient simulation model of comprehensive energy system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1993019515A1 (en) * | 1992-03-23 | 1993-09-30 | Creos International Ltd. | Continuous, in-phase switched-mode resonant converter power supply |
| CN103778479A (en) * | 2014-01-10 | 2014-05-07 | 国网上海市电力公司 | Adaptive information fault-tolerant protection method |
| CN104679936A (en) * | 2014-12-05 | 2015-06-03 | 国家电网公司 | PMFCL (permanent-magnet-biased fault current limiter) magnetic current thermal coupling modeling method |
| CN106296451A (en) * | 2016-08-08 | 2017-01-04 | 国家电网公司 | A kind of fault current limiter Optimizing collocation method based on genetic algorithm |
| CN108923401A (en) * | 2018-08-27 | 2018-11-30 | 广东电网有限责任公司 | A kind of calculation method and device of superconductive current limiter fault resstance and cut-off current |
| CN110210152A (en) * | 2019-06-06 | 2019-09-06 | 福州大学 | A kind of ultra harmonics source modeling method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7538673B2 (en) * | 2005-08-26 | 2009-05-26 | Texas Instruments Incorporated | Voltage regulation circuit for RFID systems |
| US20210334441A1 (en) * | 2020-04-28 | 2021-10-28 | The Texas A&M University System | Apparatus and systems for power system protective relay control using reinforcement learning |
-
2018
- 2018-12-29 CN CN201811641975.7A patent/CN110021929B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1993019515A1 (en) * | 1992-03-23 | 1993-09-30 | Creos International Ltd. | Continuous, in-phase switched-mode resonant converter power supply |
| CN103778479A (en) * | 2014-01-10 | 2014-05-07 | 国网上海市电力公司 | Adaptive information fault-tolerant protection method |
| CN104679936A (en) * | 2014-12-05 | 2015-06-03 | 国家电网公司 | PMFCL (permanent-magnet-biased fault current limiter) magnetic current thermal coupling modeling method |
| CN106296451A (en) * | 2016-08-08 | 2017-01-04 | 国家电网公司 | A kind of fault current limiter Optimizing collocation method based on genetic algorithm |
| CN108923401A (en) * | 2018-08-27 | 2018-11-30 | 广东电网有限责任公司 | A kind of calculation method and device of superconductive current limiter fault resstance and cut-off current |
| CN110210152A (en) * | 2019-06-06 | 2019-09-06 | 福州大学 | A kind of ultra harmonics source modeling method |
Non-Patent Citations (5)
| Title |
|---|
| Modeling Superconductive Fault Current Limiter Using Constructive Neural Networks;Behrooz Makki;《IEEE》;20071231;第1-5页 * |
| Simulation of Superconductive Fault Current Limiter (SFCL);Behrooz Makki;《IEEE》;20061231;第1-5页 * |
| 基于人工智能的故障限流器EMTP 建模;王俊生;《高压电气》;20210831;第54-60页 * |
| 基于人工神经网络控制的短路限流器;贾贵玺;《天津大学学报》;20061231;第1-5页 * |
| 电力系统电磁暂态建模仿真及应用;方毅;《中国优秀硕士论文全文数据库工程科技辑》;20111231;第1-80页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110021929A (en) | 2019-07-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110021929B (en) | Electromagnetic transient time domain simulation modeling method for fast switch type fault current limiter | |
| CN105096694B (en) | A kind of electrical equipment Virtual maintenance of nuclear emulation mode | |
| CN106786423B (en) | Service restoration method after failure under a kind of movement of power distribution network differential protection | |
| CN105425118A (en) | Multi-information fusion fault arc detection method and device | |
| CN104716646B (en) | A kind of node Coupling Degrees method based on Injection Current | |
| CN105322519A (en) | Big data fusion analysis and running state monitoring method for intelligent power distribution network | |
| CN105071771A (en) | Neural network-based distributed photovoltaic system fault diagnosis method | |
| CN106130781B (en) | Transformer fault cumulative effect appraisal procedure based on power distribution network topological model | |
| Gupta et al. | Neural network based early warning system for an emerging blackout in smart grid power networks | |
| CN112003272A (en) | Power system frequency safety control method based on convolutional neural network | |
| CN104538222B (en) | High-voltage switch gear phase-controlled device based on artificial neural network and method | |
| CN116562618A (en) | Method for obtaining dynamic security assessment model | |
| CN113887157A (en) | Active distribution network automatic terminal network closed-loop test method based on RTDS | |
| CN109066612A (en) | The switching overvoltage calculation method combined based on electromechanical transient and electro-magnetic transient | |
| CN110146744B (en) | Intelligent analysis method for impedance of power supply loop of low-voltage transformer area | |
| CN106294966B (en) | A kind of single telegram in reply cable core transient state temperature rise acquisition methods not depending on skin temperature | |
| Liu et al. | An adaptive inverse time overcurrent relay model implementation for real time simulation and hardware-in-the-loop testing | |
| CN108921280A (en) | A kind of high-voltage circuit-breaker status appraisal procedure based on improvement chaotic particle swarm optimization neural network | |
| CN111933459A (en) | Method for detecting electrical wear state of breaker contact by utilizing arc power | |
| CN108564112B (en) | Unified fault positioning method for comprehensive energy system | |
| Qiao et al. | Graph neural network based transient stability assessment considering topology changes | |
| CN107728002B (en) | Distribution network fault online positioning method considering distributed power access | |
| Ivanković et al. | Multifunctional system protection for transmission lines based on phasor data | |
| Lingjie et al. | Research on the model of HV SF6 circuit breaker fault diagnosis based on fuzzy theory | |
| Zhang et al. | Classification of building electrical system faults based on Probabilistic Neural Networks |
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 |