CN112310980A - Safety and stability evaluation method and system for direct-current blocking frequency of alternating-current and direct-current series-parallel power grid - Google Patents
Safety and stability evaluation method and system for direct-current blocking frequency of alternating-current and direct-current series-parallel power grid Download PDFInfo
- Publication number
- CN112310980A CN112310980A CN202011253105.XA CN202011253105A CN112310980A CN 112310980 A CN112310980 A CN 112310980A CN 202011253105 A CN202011253105 A CN 202011253105A CN 112310980 A CN112310980 A CN 112310980A
- Authority
- CN
- China
- Prior art keywords
- frequency
- direct
- current
- stability
- safety
- 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.)
- Granted
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
- H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
- H02J3/241—The oscillation concerning frequency
-
- 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
- G06N3/045—Combinations of 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
- G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
- G06N3/061—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using biological neurons, e.g. biological neurons connected to an integrated circuit
-
- 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
-
- 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]
Abstract
The invention provides a safety and stability evaluation method and system for direct current blocking frequency of an alternating current-direct current series-parallel power grid, belonging to the technical field of frequency stability evaluation of power systems, and a training sample set is constructed based on historical tide data; taking an expected dynamic simulation output vector of a training sample set as a frequency safety and stability index, and constructing a direct current blocking frequency safety and stability rapid evaluation model; training a direct current blocking frequency safety and stability rapid evaluation model, and identifying a high-risk operation scene influencing the direct current blocking frequency safety and stability of the alternating current-direct current hybrid power grid on line; and updating the scene to be evaluated in real time according to the predicted load change and the scheduling plan, and refreshing the high-risk evaluation scene identified on line. The method increases the rationality of the system frequency safety and stability evaluation result, can quickly calculate the frequency safety and stability evaluation result in different operation modes due to the avoidance of time domain simulation, increases the adaptability to a large-scale power grid, improves the evaluation accuracy, and is more suitable for engineering practice.
Description
Technical Field
The invention relates to the technical field of power system frequency stability evaluation, in particular to a method and a system for evaluating the safety and stability of direct-current blocking frequency of an alternating-current and direct-current series-parallel power grid.
Background
Along with the intensive operation of the extra-high voltage direct current project and the change of the load composition and characteristics, the characteristics of extra-high voltage direct current cluster feed-in and direct current power receiving scale step-type lifting are presented. The structural contradiction of strong direct current and weak alternating current is increasingly prominent, the coupling relation between an alternating current power grid and a direct current power transmission channel is tighter, the safety of a receiving-end power grid faces new challenges, and the problems of the integrated operation characteristic and the global stability of the power grid become more obvious. The access of high-capacity extra-high voltage direct current transmission in an alternating current-direct current series-parallel receiving end power grid causes the constant reduction of equivalent moment of inertia in the system and the constant reduction of frequency regulation capacity. If large-capacity direct current blocking occurs, great power shortage can be caused, so that power flow transfer, out-of-limit of power transmission section and frequency drop are caused, and a series of problems such as cascading failure, system disconnection and the like can be caused seriously. Therefore, the safety and stability evaluation of the direct current locking frequency is the basis and the key for preventing the frequency instability of the system and mining the high-risk operation mode.
Aiming at the problem of safety evaluation of the direct-current blocking frequency, the analysis method mainly comprises a time domain simulation method, an equivalent model method and the like.
The time domain simulation method is characterized in that a detailed mathematical model and a system network equation of each power element in the power system are established, a load flow calculation result or a state quantity in a steady state of the power system is used as an initial value, a numerical solution method is used for gradually and iteratively solving a multi-dimensional nonlinear differential equation of the system, and a curve of the state quantity of the power system changing along with time is obtained. The calculation result is more accurate, but the defects are that the simulation calculation amount is large, the calculation time is long, and the online evaluation requirement is difficult to meet.
The equivalent model method is based on the angle of simplifying the physical model of the system, integrates all the electromechanical equations of the generators of the system into an equivalent generator rotor model, greatly reduces the complexity of the power system model through the equivalent method, and accordingly achieves the rapid evaluation of the frequency of the power system after disturbance. But the defect is that the accuracy is not high, and the method is difficult to be applied to large-scale alternating current and direct current power grids.
The artificial intelligence method is an advanced data processing method, does not need to establish a complex physical model, can reflect the complex mapping relation of input and output through the mathematical relation of a single hidden layer or multiple hidden layers, and has the advantages of high calculation precision and high operation speed. The shallow machine learning method can realize the rapid evaluation of the frequency safety and stability in a certain precision range by learning the training data.
In summary, the existing frequency safety and stability assessment method has at least the following disadvantages: (1) the mutual influence of the direct current system and the alternating current system is not fully considered: the high-voltage direct-current transmission system has higher and higher occupation ratio in a transmission network, a direct-current system is connected with an alternating-current system through a converter, power impact can be caused to the alternating-current system by direct-current locking, and the fault of the alternating-current system can cause the operation of the converter of the direct-current system, seriously affect the normal operation of the direct-current system and further cause cascading faults. (2) With the expansion of the scale of the power grid, the existing time domain simulation method is long in calculation time and difficult to apply and evaluate online in real time. (3) The traditional equivalent model method and the shallow machine learning method have relatively low evaluation accuracy and cannot be applied to a large-scale power grid.
Disclosure of Invention
The invention aims to provide a safety and stability evaluation method and system for the direct-current blocking frequency of an alternating-current and direct-current hybrid power grid, which take the mutual influence between a direct-current system and an alternating-current system into consideration, improve the reasonability of an evaluation result, can quickly calculate the evaluation result and improve the evaluation accuracy. To solve at least one technical problem in the background art.
In order to achieve the purpose, the invention adopts the following technical scheme:
on one hand, the invention provides a safety and stability evaluation method for direct current blocking frequency of an alternating current-direct current hybrid power grid, which comprises the following steps:
constructing a training sample set based on historical trend data;
taking an expected dynamic simulation output vector of a training sample set as a frequency safety and stability index, and constructing a direct current blocking frequency safety and stability rapid evaluation model;
training a direct current blocking frequency safety and stability rapid evaluation model by using a training sample set;
the trained direct-current blocking frequency safety and stability rapid evaluation model is used for identifying a high-risk operation scene influencing the safety and stability of the direct-current blocking frequency of the alternating-current and direct-current series-parallel power grid on line;
and updating the scene to be evaluated in real time according to the predicted load change and the scheduling plan, and refreshing the high-risk evaluation scene identified on line.
Preferably, the constructing of the training sample set based on the historical trend data includes:
obtaining tidal current steady-state information from historical tidal current data of an alternating-current and direct-current hybrid power grid and taking instantaneous disturbance information-direct-current blocking active power deficit amount obtained according to an expected fault set as an original feature set of a training sample set; the power flow steady-state information comprises load active power, a damping coefficient, generator active power output, a generator inertia time constant and key power transmission line power flow data;
and setting corresponding DC transmission line unipolar/bipolar latching faults according to the expected DC latching fault set, and performing time domain simulation on the unipolar/bipolar latching faults by using the original characteristic set to obtain the frequency maximum change rate and the frequency maximum offset of the inertial frequency dynamic curve of the AC-DC hybrid power grid system after the faults, wherein the frequency maximum change rate and the frequency maximum offset are used as expected dynamic simulation output vectors.
Preferably, the constructing of the direct current blocking frequency safety and stability rapid evaluation model includes:
the direct-current blocking frequency safety and stability rapid evaluation model is composed of a depth confidence network and a gradient descent algorithm;
the deep confidence network is used for feature extraction in a pre-training stage to obtain high-order feature expression of original features, and a deep confidence neural network comprising a plurality of hidden layers is formed;
the gradient descent algorithm is used for optimizing the weight and the bias of each neuron of the deep belief neural network, and the regression layer of the deep belief neural network updates the network weight and the bias by adopting error back propagation.
And (3) taking the frequency maximum change rate and the frequency maximum offset of the system inertia frequency dynamic curve after the direct current blocking fault as output, taking the original feature set as input, and establishing a direct current blocking frequency safety and stability rapid evaluation model based on a deep confidence network.
Preferably, the training the dc blocking frequency safety and stability rapid evaluation model by using the training sample set includes:
decomposing each two layers of hidden layers of the deep confidence network from bottom to top and using the hidden layers as an independent limited Boltzmann machine model, and training the limited Boltzmann machine model layer by using the original feature set of all training samples and using a contrast divergence algorithm, namely performing unsupervised pre-training layer by layer;
after the pre-training of the depth confidence network layer by layer is completed, the high-order characteristics of the sample can be obtained in the last hidden layer;
and then entering a reverse fine tuning stage, and in the reverse fine tuning stage, performing reverse weight value and bias weight updating on the direct current lock-up frequency safety and stability rapid evaluation model by using the expected frequency safety and stability index vectors of all the training samples and utilizing a gradient descent algorithm, namely supervised training.
Preferably, the method for identifying the high-risk operation scene influencing the safety and stability of the direct-current blocking frequency of the alternating-current and direct-current series-parallel power grid on line by using the trained direct-current blocking frequency safety and stability rapid evaluation model comprises the following steps:
the method comprises the steps of obtaining a real-time operation mode of an alternating current-direct current hybrid power grid dispatching system, predicting power grid load active power in a future period according to a load power prediction function of the dispatching system, obtaining real-time data of load active power and damping coefficients, generator active power output and inertia time constants and key power transmission line tide according to a dispatching plan, and obtaining instantaneous disturbance information according to an expected direct current blocking fault set to serve as a model input original feature set;
inputting the original feature set into a trained direct current blocking frequency safety and stability rapid evaluation model, outputting the maximum frequency change frequency and the maximum frequency offset of a system inertia frequency dynamic curve after a fault, and respectively recording the maximum frequency change frequency and the maximum frequency offset as
And fnair；
Determining different threshold values capable of causing the starting of a frequency relay protection device and a low-frequency load shedding device in the system aiming at different alternating current-direct current hybrid systems;
according to
And fnairAnd determining the risk grade of the system, and further identifying a high-risk operation scene influencing the frequency safety and stability of the alternating-current and direct-current hybrid power grid.
Preferably, updating the scene to be evaluated in real time according to the predicted load change and the scheduling plan and refreshing the high risk evaluation scene identified on line comprises:
taking the current latest high-risk scene evaluation result as a time origin, namely t is 0, acquiring latest load power prediction information, generator output plan information and new energy power generation prediction information at intervals according to the actual mastery condition of a scheduling department on load change, and updating related predicted load flow operation modes in a rolling manner;
generating a future state operation scene set to be evaluated based on the latest predicted trend operation mode, and carrying out scene reduction on the future state operation scene to be evaluated;
evaluating all running scenes to be evaluated by using the trained direct current blocking frequency safety and stability evaluation model to obtain an evaluation result; and performing rolling updating on the original evaluation result.
In a second aspect, the invention provides an ac/dc hybrid power grid dc blocking frequency safety and stability evaluation system based on the ac/dc hybrid power grid dc blocking frequency safety and stability evaluation method, including the following modules:
the training sample construction module is used for constructing a training sample set based on historical trend data; taking an expected output vector of the training sample set as a frequency safety and stability index;
the evaluation model building module is used for building a direct current blocking frequency safety and stability rapid evaluation model based on the frequency safety and stability index;
the model training module is used for training the direct current blocking frequency safe, stable and rapid evaluation model by utilizing a training sample set;
the high-risk operation scene identification module is used for identifying a high-risk operation scene influencing the safety and stability of the frequency of the alternating current-direct current power grid on line by utilizing the trained direct current blocking frequency safety and stability rapid evaluation model;
the future state operation scene generation module is used for acquiring latest load power prediction information, generator output plan information and new energy power generation prediction information at intervals according to the actual mastery condition of the scheduling department on the load change, and updating the related predicted tidal current operation mode in a rolling manner; generating a future state operation scene set to be evaluated based on the latest prediction information;
the future state operation scene evaluation module is used for carrying out scene reduction on the future state operation scene to be evaluated; evaluating all latest evaluation operation scenes by using the trained direct current blocking frequency safety and stability rapid evaluation model to obtain an evaluation result;
and the evaluation result rolling updating module is used for rolling updating the original evaluation result.
In a third aspect, the invention provides a non-transitory computer readable storage medium comprising instructions for performing the method as described above.
In a fourth aspect, the invention provides a terminal device comprising a non-transitory computer readable storage medium as described above; and one or more processors capable of executing the instructions of the non-transitory computer-readable storage medium.
In a fifth aspect, the present invention provides a terminal device comprising means for performing the method as described above.
The beneficial effects of the invention mainly comprise the following points:
firstly, the maximum frequency change rate and the maximum frequency offset of a dynamic curve of the system inertia frequency after direct current blocking are provided as indexes for measuring the degree of influence of the system frequency by the direct current blocking based on the interactive influence characteristic of the alternating current-direct current system, and the reasonability of the system frequency safety and stability evaluation result is improved.
Secondly, a frequency safety and stability rapid evaluation model is established by utilizing a deep learning technology, and the maximum frequency change rate and the maximum frequency offset of the system after direct current blocking in a given operation mode can be rapidly calculated for a large-scale alternating current and direct current power grid; due to the fact that time domain simulation is avoided, frequency safety and stability evaluation results under different operation modes can be calculated quickly.
And thirdly, high-order feature extraction is carried out on the input features by adopting a deep confidence network, and parameters are optimized by adopting a gradient descent algorithm.
Additional aspects and advantages 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
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a method for evaluating safety and stability of a dc blocking frequency of an ac/dc hybrid power grid according to an embodiment of the present invention.
Fig. 2 is a basic composition structure diagram of a direct-current blocking frequency safety and stability rapid evaluation model of an alternating-current and direct-current series-parallel power grid according to an embodiment of the present invention.
Fig. 3 is a structural diagram of a direct-current blocking frequency safety and stability rapid evaluation model of an alternating-current and direct-current series-parallel power grid according to an embodiment of the present invention.
Fig. 4 is a functional schematic block diagram of a safety and stability evaluation system for a dc blocking frequency of an ac/dc hybrid power grid according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below by way of the drawings are illustrative only and are not to be construed as limiting the invention.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including 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 will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
For the purpose of facilitating an understanding of the present invention, the present invention will be further explained by way of specific embodiments with reference to the accompanying drawings, which are not intended to limit the present invention.
It should be understood by those skilled in the art that the drawings are merely schematic representations of embodiments and that the elements shown in the drawings are not necessarily required to practice the invention.
Example 1
As shown in fig. 1, a method for evaluating safety and stability of a dc blocking frequency of an ac/dc hybrid power grid according to an embodiment of the present invention includes the following steps:
constructing a training sample set based on historical trend data; taking an expected dynamic simulation output vector of the training sample set as a frequency safety and stability index;
constructing a direct current blocking frequency safety and stability rapid evaluation model based on the frequency safety and stability index;
training a direct current blocking frequency safety and stability rapid evaluation model by using a training sample set;
and identifying a high-risk operation scene influencing the safety and stability of the direct-current blocking frequency of the alternating-current and direct-current series-parallel power grid on line by using the trained direct-current blocking frequency safety and stability rapid evaluation model.
And updating the scene to be evaluated in real time according to the predicted load change and the scheduling plan, and refreshing the high-risk evaluation scene identified on line.
And constructing an original feature set of a system direct-current blocking frequency safety and stability rapid evaluation model according to the operation mode characteristics related to the system frequency dynamic curve after direct-current blocking, wherein the original feature set comprises load active power and damping coefficients, generator active power output and generator inertia time constants, key transmission line tidal current data, tidal current steady state information such as direct-current blocking active power disturbance quantity and instantaneous disturbance information. And obtaining the frequency change rate and the frequency maximum offset of the system inertia frequency dynamic curve after direct current locking through time domain simulation, and forming a training sample set with the original feature set as an expected output vector.
The input characteristics of the direct current blocking frequency safety and stability rapid evaluation model need to include all factors influencing the dynamic change of the frequency after direct current blocking. The frequency stability of the system is closely related to the active power, when the direct current receiving proportion of the system is large, the active vacancy caused by direct current locking is large, a primary frequency modulation device of the generator after a fault, the damping response of a load and the power flow of a key alternating current transmission line can also change to adjust the frequency dynamic, and the frequency oscillation phenomenon is prevented to a certain extent.
Therefore, the input feature set of the dc blocking frequency safety and stability rapid evaluation model should include: load active power and damping coefficient, generator active output and generator inertia time constant, key transmission line tide data, direct current blocking active power disturbance quantity and other tide steady state information, transient disturbance information and the like; the output of the model is under the specific expected DC blocking fault
And fnair。
In this embodiment 1, the process of constructing the training sample set includes:
the method comprises the steps that 1-2 years of historical operation modes of a scheduling system can be utilized, load active power and damping coefficients, active power output of a generator, inertia time constants of the generator, tidal current steady-state information such as key power transmission line tidal current data and the like can be obtained from historical tidal current data of an alternating-current and direct-current hybrid power grid, and transient disturbance information-direct-current blocking active power deficit quantity obtained according to an expected fault set is used as an original characteristic set of a training sample set;
the method comprises the steps of determining a basic operation mode of a dispatching system, obtaining loads of load nodes under corresponding system load proportions at equal intervals in a system reference load range of 80% -120%, obtaining load active power and damping coefficients, generator active power output and generator inertia time constants, tidal current steady state information such as key transmission line tidal current data and instantaneous disturbance information-direct current blocking active power deficit quantity obtained according to an expected fault set in the equal proportion range, and taking the instantaneous disturbance information-direct current blocking active power deficit quantity as an original characteristic set of a training sample set;
and setting a single-pole blocking fault or a double-pole blocking fault of a corresponding direct-current transmission line according to the expected direct-current blocking fault set, and performing time domain simulation on the direct-current transmission line by using the original characteristic set to obtain the maximum frequency change rate and the maximum frequency offset of the inertia frequency dynamic curve of the alternating-current and direct-current hybrid power grid system after the fault, wherein the maximum frequency change rate and the maximum frequency offset are used as expected dynamic simulation output vectors.
The process of constructing the direct current blocking frequency safety and stability rapid evaluation model comprises the following steps: and (3) taking the frequency change rate and the frequency maximum offset of the system inertia frequency dynamic curve after the direct current blocking fault as output, taking the original feature set of the training sample set as input, and establishing a direct current blocking frequency safety and stability rapid evaluation model based on the deep confidence network.
As shown in fig. 2, the basic component unit of the direct current blocking frequency safety and stability rapid evaluation model is composed of a limited boltzmann machine; the Restricted Boltzmann Machine (RBM) is a modified form of the Boltzmann Machine (RBM).
The RBM comprises two layers-a visible layer and a hidden layer, the connections between neurons have the following characteristics: no connection exists in the layers, and the layers are all connected. No weight relation exists between neurons in the layers, and weight relation exists between neurons in the layers. The RBM has strong unsupervised learning capability, can deeply mine complex nonlinear features in data, remove redundant data features, and express key sample features forming low latitude by learning, so the RBM is often used as a basic component module of other deep learning models such as a stacked automatic encoder and a deep belief network.
FIG. 2 is a block diagram of a constrained Boltzmann machine, in which n, m represent the number of neurons contained in the visible layer and the hidden layer, respectively; v ═ v (v)1,v2,v3...vn)，h＝(h1,h2,h3...hm) State vectors representing the visible layer and the hidden layer, respectively; a ═ a1,a2,a3...an)，b＝(b1,b2,b3...bm) Bias vectors representing the visible layer and the hidden layer, respectively; w ═ Wij)∈Rn ×mW represents the connection weight matrix of the visible layer and the hidden layer, WijRepresenting visible unit i and hiddenThe connection weight of layer unit j.
The task of training the RBM is to find the value of the parameter θ, v ═ v1,v2,v3...vn) State vector representing visible layer, given a training sample set of S ═ v1,v2,v3…vn}. They are independent and equally distributed, and the parameter θ can be obtained by maximizing the log-likelihood function learning of the RBM on the training set:
where θ ═ W, a, b denotes the parameter of RBM.
Namely, the weight and weight bias iterative update formula is as follows:
eta in the formula is a learning rate, and a contrast divergence algorithm (CD-K) is further combined to approximate the expected probability through Gibbs sampling, so that a specific updating formula of the parameter theta is obtained.
As shown in fig. 3, the direct current blocking frequency safety and stability rapid evaluation model is composed of a Deep Belief Network (DBN) and an adam (adaptive moment) algorithm, where the Deep Belief Network is used for feature extraction in a pre-training phase to obtain a high-order feature expression of an original feature, so as to form a Deep Belief neural Network including multiple hidden layers; the ADAM algorithm is used for reversely trimming and optimizing the weight and the bias of each neuron of the deep belief neural network in the trimming stage of the deep belief network, and the regression layer of the deep belief neural network updates the network weight and the bias by adopting the idea of error back propagation.
The DBN is a typical deep learning algorithm, high-order abstract features are extracted from original input features by using a stacked restricted Boltzmann machine, the accuracy of a subsequent regressor can be improved, and the DBN has better stability and robustness when being applied to a large amount of data. Therefore, the DBN is used for feature extraction in a rapid evaluation model of safety and stability of the direct current blocking frequency.
(1) The DBN is formed by stacking limited Boltzmann machines (RBMs) in a stacked mode, the limited Boltzmann machines are special topological structures of the Boltzmann Machines (BMs), the limited Boltzmann machines are energy function-based modeling methods and are composed of visible layers and hidden layers, and network nodes are divided into visible units and hidden units.
And training the RBMs of the stack layer by layer, entering a reverse fine adjustment stage of the deep belief network after the training is finished, and finishing the reverse propagation and iterative updating of the error between the last regression layer and the expected vector.
(2) The back-propagated loss function is typically chosen as the loss function of the ridge regression, i.e., the MSE function plus a regularization term of the weight, regularization term L2Is introduced to prevent the training process from generating an overfitting, defined as follows:
in the formula, N is the number of the operation mode samples, i is the number of the samples, and N is the total number of the samples; y isiFor the i-th sample actual dynamic index value, yiAnd W is a weight matrix.
In this embodiment 1, the process of training the dc-lock frequency safety and stability rapid evaluation model includes:
decomposing each two layers of the depth confidence network from bottom to top and using the depth confidence network as an independent limited Boltzmann machine model, and performing layer-by-layer training of the limited Boltzmann machine model by using the original feature sets of all training samples and using a CD-K algorithm, namely performing layer-by-layer unsupervised pre-training;
after the pre-training of the deep belief network layer by layer is completed, the high-level input characteristics of the sample can be obtained at the last layer, and the characteristics can more abstractly express the dynamic response characteristics of the power grid frequency. Then entering a reverse fine tuning stage, and in the reverse fine tuning stage, performing reverse weight value and bias weight updating on the direct current lock-up frequency safety and stability rapid evaluation model by using the expected frequency safety and stability index vectors of all training samples and utilizing an ADAM algorithm, so that the model regression performance is better, namely supervised training is performed;
in this embodiment 1, the process of online identifying a high-risk operation scene that affects the safety and stability of the dc blocking frequency of the ac-dc hybrid power grid includes:
the method comprises the steps of obtaining a real-time operation mode of a dispatching system, predicting power grid load active power in a future period according to a load power prediction function of the dispatching system, obtaining real-time data of load active power and damping coefficients, generator active power output and inertia time constants and key power transmission line power flow according to a dispatching plan, obtaining instantaneous disturbance information according to an expected direct current blocking fault set, and inputting the instantaneous disturbance information as a model into an original feature set;
inputting the input feature set into the trained direct current blocking frequency safety and stability rapid evaluation model, and respectively recording the maximum frequency change frequency and the maximum frequency offset of the system inertia frequency dynamic curve after outputting the fault as
And fnair；
Aiming at different alternating current-direct current hybrid systems, different threshold values capable of causing the starting of devices such as a frequency relay protection device and a low-frequency load shedding device in the system are determined;
according to
And fnairAnd determining the risk grade of the system, and further identifying a high-risk operation scene influencing the frequency safety and stability of the alternating-current and direct-current hybrid power grid.
The process of updating the scene to be evaluated in real time according to the predicted load change and the scheduling plan and refreshing the high risk evaluation scene identified on line comprises the following steps:
taking the current latest high-risk scene evaluation result as a time origin, namely t is 0, acquiring latest load power prediction information, generator output plan information and new energy power generation prediction information at intervals of 15min according to the actual grasping condition of a scheduling department on load change, and updating a related predicted load flow operation mode in a rolling manner;
generating a future state operation scene set to be evaluated based on the latest predicted trend operation mode, and reducing the scene of the future state operation scene to be evaluated;
and evaluating all the operation scenes to be evaluated by using the trained direct current blocking frequency safety and stability evaluation model to obtain an evaluation result, and performing rolling update on the original evaluation result.
When the power system actually runs, the load power and the generator output plan information are updated in a rolling mode at certain time intervals, and available preventive control resources are changed continuously. The load power will become more accurate as the rolling update becomes more accurate and the uncertainty will gradually decrease. Calculation is performed based on the updated prediction information, and the accuracy of the evaluation result can be improved.
Example 2
As shown in fig. 4, the transient voltage stability evaluation system for the ac/dc large power grid provided in this embodiment 2 includes:
the training sample construction module is used for constructing a training sample set based on historical trend data; taking an expected output vector of the training sample set as a frequency safety and stability index;
the evaluation model building module is used for building a direct current blocking frequency safety and stability rapid evaluation model based on the frequency safety and stability index;
the model training module is used for training the direct current blocking frequency safe, stable and rapid evaluation model by utilizing a training sample set;
and the high-risk operation scene identification module is used for identifying the high-risk operation scene influencing the safety and stability of the frequency of the alternating current-direct current power grid on line by utilizing the trained direct current blocking frequency safety and stability rapid evaluation model.
The future state operation scene generation module is used for acquiring latest load power prediction information, generator output plan information and new energy power generation prediction information once every a period of time such as 15min according to the actual mastery condition of the scheduling department on the load change, and updating the related predicted tidal current operation mode in a rolling manner; generating a future state operation scene set to be evaluated based on the latest prediction information;
the future state operation scene evaluation module is used for carrying out scene reduction on the future state operation scene to be evaluated; evaluating all latest evaluation operation scenes by using the trained direct current blocking frequency safety and stability rapid evaluation model to obtain an evaluation result;
and the evaluation result rolling updating module is used for rolling updating the original evaluation result.
In this embodiment 2, the process of constructing the training sample set by the sample construction module is as follows:
the method comprises the steps that 1-2 years of historical operation modes of a scheduling system can be utilized, load active power and damping coefficients, active power output of a generator, inertia time constants of the generator, tidal current steady-state information such as key power transmission line tidal current data and the like can be obtained from historical tidal current data of an alternating-current and direct-current hybrid power grid, and transient disturbance information-direct-current blocking active power deficit quantity obtained according to an expected fault set is used as an original characteristic set of a training sample set;
the method comprises the steps of determining a basic operation mode of a dispatching system, obtaining loads of load nodes under corresponding system load proportions at equal intervals in a system reference load range of 80% -120%, obtaining load active power and damping coefficients, generator active power output and generator inertia time constants, tidal current steady-state information such as key transmission line tidal current data and instantaneous disturbance information, namely direct current blocking active power deficit quantity, obtained according to an expected fault set in the equal proportion range, and taking the instantaneous disturbance information as an original characteristic set of a training sample set;
and setting a single-pole blocking fault or a double-pole blocking fault of a corresponding direct-current transmission line according to the expected direct-current blocking fault set, and performing time domain simulation on the direct-current transmission line by using the original characteristic set to obtain the maximum frequency change rate and the maximum frequency offset of the inertia frequency dynamic curve of the alternating-current and direct-current hybrid power grid system after the fault, wherein the maximum frequency change rate and the maximum frequency offset are used as expected dynamic simulation output vectors.
The process of constructing the direct current blocking frequency safety and stability rapid evaluation model comprises the following steps: and (3) taking the frequency change rate and the frequency maximum offset of the system inertia frequency dynamic curve after the direct current blocking fault as output, taking the original feature set of the training sample set as input, and establishing a direct current blocking frequency safety and stability rapid evaluation model based on the deep confidence network.
The process of training the direct current lock-up frequency safety and stability rapid evaluation model comprises the following steps:
decomposing each two layers of the deep confidence network from bottom to top to serve as an independent restricted Boltzmann machine, and performing unsupervised pre-training layer by using all training samples;
after the pre-training of the deep belief network layer by layer is completed, taking the extracted high-order features as new sample features, and then performing supervised training by using all training samples in a reverse fine-tuning stage;
and reversely updating the weight and the weight bias of the direct current blocking frequency safety and stability rapid evaluation model according to the expected frequency safety and stability index vector of the training sample.
The process of online identifying the high-risk operation scene influencing the frequency safety and stability of the alternating current-direct current hybrid power grid is as follows:
the method comprises the steps of obtaining a real-time operation mode of a dispatching system, predicting the active power of a power grid load in a future period according to a load power prediction function of the dispatching system, obtaining real-time data of the active power of the load, a damping coefficient, the active power output and inertia time constant of a generator and the power flow of a key power transmission line according to a dispatching plan, and obtaining instantaneous disturbance information as an input feature set according to an expected direct current blocking fault set;
inputting the input feature set into the trained direct current blocking frequency safety and stability rapid evaluation model, and respectively recording the maximum frequency change frequency and the maximum frequency offset of the system inertia frequency dynamic curve after outputting the fault as
And fnair；
Aiming at different alternating current-direct current hybrid systems, different threshold values capable of causing the starting of devices such as a frequency relay protection device and a low-frequency load shedding device in the system are determined;
according to
And fnairAnd determining the risk grade of the system, and further identifying a high-risk operation scene influencing the frequency safety and stability of the alternating-current and direct-current hybrid power grid.
The alternating current-direct current large power grid transient voltage stability evaluation system further comprises:
the future state operation scene generation module is used for acquiring latest load power prediction information, generator output plan information and new energy power generation prediction information once every a period of time such as 15min according to the actual mastery condition of the scheduling department on the load change, and updating the related predicted tidal current operation mode in a rolling manner; generating a future state operation scene set to be evaluated based on the latest prediction information;
the future state operation scene evaluation module is used for carrying out scene reduction on the future state operation scene to be evaluated; evaluating all latest evaluation operation scenes by using the trained direct current blocking frequency safety and stability rapid evaluation model to obtain an evaluation result;
and the evaluation result rolling updating module is used for rolling updating the original evaluation result.
When the power system actually runs, the load power and the generator output plan information are updated in a rolling mode at certain time intervals, and available preventive control resources are changed continuously. The load power will become more accurate as the rolling update becomes more accurate and the uncertainty will gradually decrease. Calculation is performed based on the updated prediction information, and the accuracy of the evaluation result can be improved.
Example 3
In this embodiment 3, a computer-readable storage medium is provided, on which a computer program is stored, where the program is suitable for being loaded by a processor of a terminal device and executing instructions for implementing the dc blocking frequency safety and stability assessment method for the ac/dc hybrid grid.
Example 4
In this embodiment 4, a terminal device is provided, where the terminal device includes a processor and a computer-readable storage medium, and the processor is used to implement a program of a method for evaluating safety and stability of a dc blocking frequency of an ac-dc hybrid grid; the computer-readable storage medium is used for storing a plurality of program instructions, and the program instructions are suitable for being loaded by the processor and executing the safety and stability evaluation method for the direct-current blocking frequency of the alternating-current and direct-current hybrid power grid.
In summary, according to the method and system for evaluating the safety and stability of the direct current blocking frequency of the alternating current-direct current hybrid power grid in the embodiments of the present invention, the maximum frequency change rate and the maximum frequency offset of the system inertia center frequency change curve after the fault are used as the expected dynamic simulation output vector, so as to establish a rapid evaluation model of the safety and stability of the direct current blocking frequency, and the load active power and damping coefficient obtained by rolling, the generator active power output and the generator inertia time constant, the tidal current data of the key power transmission line, the tidal current steady state information and the transient disturbance information such as the disturbance amount of the direct current blocking frequency of the alternating current-direct current hybrid power grid are used as the inputs of the model, so as to perform online evaluation of the safety.
Based on the interactive influence characteristic of the alternating current-direct current system, the frequency maximum change rate and the frequency maximum offset of the system inertia frequency dynamic curve after direct current blocking are provided as indexes for measuring the influence degree of the system frequency by the direct current blocking, and the reasonability of the system frequency safety and stability evaluation result is improved.
A frequency safety and stability rapid evaluation model is established by utilizing a deep learning technology, and the maximum frequency change rate and the maximum frequency offset of a system after direct current blocking in a given operation mode can be rapidly calculated for a large-scale alternating current and direct current power grid; due to the fact that time domain simulation is avoided, frequency safety and stability evaluation results under different operation modes can be calculated quickly.
The method adopts the deep belief network to extract the high-order features of the input features, and adopts the ADAM algorithm to optimize the parameters, compared with the traditional equivalent model method and the shallow machine learning method, the method increases the adaptability to the large-scale power grid, improves the evaluation accuracy, and is more fit with the engineering practice.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention 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 invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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 above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Although the present disclosure has been described with reference to the specific embodiments shown in the drawings, it is not intended to limit the scope of the present disclosure, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive faculty based on the technical solutions disclosed in the present disclosure.
Claims (10)
1. A safety and stability assessment method for direct current blocking frequency of an alternating current-direct current hybrid power grid is characterized by comprising the following steps:
constructing a training sample set based on historical trend data;
taking an expected dynamic simulation output vector of a training sample set as a frequency safety and stability index, and constructing a direct current blocking frequency safety and stability rapid evaluation model;
training a direct current blocking frequency safety and stability rapid evaluation model by using a training sample set;
the trained direct-current blocking frequency safety and stability rapid evaluation model is used for identifying a high-risk operation scene influencing the safety and stability of the direct-current blocking frequency of the alternating-current and direct-current series-parallel power grid on line;
and updating the scene to be evaluated in real time according to the predicted load change and the scheduling plan, and refreshing the high-risk evaluation scene identified on line.
2. The method for evaluating the safety and stability of the direct-current blocking frequency of the alternating-current and direct-current series-parallel power grid according to claim 1, wherein the step of constructing a training sample set based on historical power flow data comprises the steps of:
obtaining tidal current steady-state information from historical tidal current data of an alternating-current and direct-current hybrid power grid and taking instantaneous disturbance information-direct-current blocking active power deficit amount obtained according to an expected fault set as an original feature set of a training sample set; the power flow steady-state information comprises load active power, a damping coefficient, generator active power output, a generator inertia time constant and key power transmission line power flow data;
and setting corresponding DC transmission line unipolar/bipolar latching faults according to the expected DC latching fault set, and performing time domain simulation on the unipolar/bipolar latching faults by using the original characteristic set to obtain the frequency maximum change rate and the frequency maximum offset of the inertial frequency dynamic curve of the AC-DC hybrid power grid system after the faults, wherein the frequency maximum change rate and the frequency maximum offset are used as expected dynamic simulation output vectors.
3. The safety and stability assessment method for the direct-current blocking frequency of the alternating-current and direct-current hybrid power grid according to claim 2, wherein the construction of the rapid assessment model for the safety and stability of the direct-current blocking frequency comprises the following steps:
the direct-current blocking frequency safety and stability rapid evaluation model is composed of a depth confidence network and a gradient descent algorithm;
the deep confidence network is used for feature extraction in a pre-training stage to obtain high-order feature expression of original features, and a deep confidence neural network comprising a plurality of hidden layers is formed;
the gradient descent algorithm is used for optimizing the weight and the bias of each neuron of the deep belief neural network, and the regression layer of the deep belief neural network updates the network weight and the bias by adopting error back propagation.
And (3) taking the frequency maximum change rate and the frequency maximum offset of the system inertia frequency dynamic curve after the direct current blocking fault as output, taking the original feature set as input, and establishing a direct current blocking frequency safety and stability rapid evaluation model based on a deep confidence network.
4. The AC-DC hybrid power grid DC blocking frequency safety and stability assessment method according to claim 3, wherein training the DC blocking frequency safety and stability rapid assessment model by using the training sample set comprises:
decomposing each two layers of hidden layers of the deep confidence network from bottom to top and using the hidden layers as an independent limited Boltzmann machine model, and training the limited Boltzmann machine model layer by using the original feature set of all training samples and using a contrast divergence algorithm, namely performing unsupervised pre-training layer by layer;
after the pre-training of the depth confidence network layer by layer is completed, the high-order characteristics of the sample can be obtained in the last hidden layer;
and then entering a reverse fine tuning stage, and in the reverse fine tuning stage, performing reverse weight value and bias weight updating on the direct current lock-up frequency safety and stability rapid evaluation model by using the expected frequency safety and stability index vectors of all the training samples and utilizing a gradient descent algorithm, namely supervised training.
5. The method for evaluating the safety and stability of the direct-current blocking frequency of the alternating-current and direct-current hybrid power grid according to claim 4, wherein the step of identifying the high-risk operation scene influencing the safety and stability of the direct-current blocking frequency of the alternating-current and direct-current hybrid power grid on line by using the trained direct-current blocking frequency safety and stability rapid evaluation model comprises the following steps:
the method comprises the steps of obtaining a real-time operation mode of an alternating current-direct current hybrid power grid dispatching system, predicting power grid load active power in a future period according to a load power prediction function of the dispatching system, obtaining real-time data of load active power and damping coefficients, generator active power output and inertia time constants and key power transmission line tide according to a dispatching plan, and obtaining instantaneous disturbance information according to an expected direct current blocking fault set to serve as a model input original feature set;
inputting the original feature set into a trained direct current blocking frequency safety and stability rapid evaluation model, outputting the maximum frequency change frequency and the maximum frequency offset of a system inertia frequency dynamic curve after a fault, and respectively recording the maximum frequency change frequency and the maximum frequency offset as
And fnair；
Determining different threshold values capable of causing the starting of a frequency relay protection device and a low-frequency load shedding device in the system aiming at different alternating current-direct current hybrid systems;
according toAnd fnairDetermining the risk level of the system and further identifying the influence of alternating current and direct currentAnd (4) a high-risk operation scene of safe and stable frequency of the hybrid power grid.
6. The safety and stability assessment method for the direct-current blocking frequency of the alternating-current and direct-current hybrid power grid according to claim 5, wherein the real-time updating of the scene to be assessed according to the predicted load change and the scheduling plan and the refreshing of the high-risk assessment scene identified on line comprises the following steps:
taking the current latest high-risk scene evaluation result as a time origin, namely t is 0, acquiring latest load power prediction information, generator output plan information and new energy power generation prediction information at intervals according to the actual mastery condition of a scheduling department on load change, and updating related predicted load flow operation modes in a rolling manner;
generating a future state operation scene set to be evaluated based on the latest predicted trend operation mode, and carrying out scene reduction on the future state operation scene to be evaluated;
evaluating all running scenes to be evaluated by using the trained direct current blocking frequency safety and stability evaluation model to obtain an evaluation result; and performing rolling updating on the original evaluation result.
7. The safety and stability evaluation system for the direct-current blocking frequency of the alternating-current and direct-current hybrid power grid based on the safety and stability evaluation method for the direct-current blocking frequency of the alternating-current and direct-current hybrid power grid according to any one of claims 1 to 6 is characterized by comprising the following modules:
the training sample construction module is used for constructing a training sample set based on historical trend data; taking an expected output vector of the training sample set as a frequency safety and stability index;
the evaluation model building module is used for building a direct current blocking frequency safety and stability rapid evaluation model based on the frequency safety and stability index;
the model training module is used for training the direct current blocking frequency safe, stable and rapid evaluation model by utilizing a training sample set;
the high-risk operation scene identification module is used for identifying a high-risk operation scene influencing the safety and stability of the frequency of the alternating current-direct current power grid on line by utilizing the trained direct current blocking frequency safety and stability rapid evaluation model;
the future state operation scene generation module is used for acquiring latest load power prediction information, generator output plan information and new energy power generation prediction information at intervals according to the actual mastery condition of the scheduling department on the load change, and updating the related predicted tidal current operation mode in a rolling manner; generating a future state operation scene set to be evaluated based on the latest prediction information;
the future state operation scene evaluation module is used for carrying out scene reduction on the future state operation scene to be evaluated; evaluating all latest evaluation operation scenes by using the trained direct current blocking frequency safety and stability rapid evaluation model to obtain an evaluation result;
and the evaluation result rolling updating module is used for rolling updating the original evaluation result.
8. A non-transitory computer-readable storage medium characterized in that: the non-transitory computer readable storage medium comprising instructions for performing the method of any of claims 1-6.
9. A terminal device characterized by: comprising the non-transitory computer-readable storage medium of claim 8; and one or more processors capable of executing the instructions of the non-transitory computer-readable storage medium.
10. A terminal device characterized by: the apparatus comprising means for performing the method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011253105.XA CN112310980B (en) | 2020-11-11 | 2020-11-11 | Safety and stability evaluation method and system for direct-current blocking frequency of alternating-current and direct-current series-parallel power grid |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011253105.XA CN112310980B (en) | 2020-11-11 | 2020-11-11 | Safety and stability evaluation method and system for direct-current blocking frequency of alternating-current and direct-current series-parallel power grid |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112310980A true CN112310980A (en) | 2021-02-02 |
| CN112310980B CN112310980B (en) | 2022-09-30 |
Family
ID=74325726
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011253105.XA Active CN112310980B (en) | 2020-11-11 | 2020-11-11 | Safety and stability evaluation method and system for direct-current blocking frequency of alternating-current and direct-current series-parallel power grid |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112310980B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113449968A (en) * | 2021-06-09 | 2021-09-28 | 重庆锦禹云能源科技有限公司 | New energy power grid frequency risk assessment method and device based on deep learning |
| CN113991760A (en) * | 2021-10-20 | 2022-01-28 | 山东大学 | Direct-current locking multi-resource emergency control optimization method for alternating-current and direct-current series-parallel receiving-end power grid |
| CN114330884A (en) * | 2021-12-29 | 2022-04-12 | 山东大学 | Multi-time scale coordinated alternating current and direct current power grid dynamic safety early warning method and system |
| CN116151130A (en) * | 2023-04-19 | 2023-05-23 | 国网浙江新兴科技有限公司 | Wind power plant maximum frequency damping coefficient calculation method, device, equipment and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017092527A1 (en) * | 2015-12-02 | 2017-06-08 | 国电南瑞科技股份有限公司 | Online evaluation method of safety margin of operating reserve in power system |
| CN111475915A (en) * | 2020-03-05 | 2020-07-31 | 国电南瑞科技股份有限公司 | Successive fault online evaluation method based on fault probability and time domain simulation quasi-steady state |
| CN111628501A (en) * | 2020-06-18 | 2020-09-04 | 国网山东省电力公司济南供电公司 | AC/DC large power grid transient voltage stability assessment method and system |
-
2020
- 2020-11-11 CN CN202011253105.XA patent/CN112310980B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017092527A1 (en) * | 2015-12-02 | 2017-06-08 | 国电南瑞科技股份有限公司 | Online evaluation method of safety margin of operating reserve in power system |
| CN111475915A (en) * | 2020-03-05 | 2020-07-31 | 国电南瑞科技股份有限公司 | Successive fault online evaluation method based on fault probability and time domain simulation quasi-steady state |
| CN111628501A (en) * | 2020-06-18 | 2020-09-04 | 国网山东省电力公司济南供电公司 | AC/DC large power grid transient voltage stability assessment method and system |
Non-Patent Citations (2)
| Title |
|---|
| 朱乔木 等: "基于深度置信网络的电力系统暂态稳定评估方法", 《中国电机工程学报》 * |
| 李常刚 等: "大电网动态安全风险智能评估系统", 《电力系统自动化》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113449968A (en) * | 2021-06-09 | 2021-09-28 | 重庆锦禹云能源科技有限公司 | New energy power grid frequency risk assessment method and device based on deep learning |
| CN113449968B (en) * | 2021-06-09 | 2023-08-29 | 重庆锦禹云能源科技有限公司 | New energy power grid frequency risk assessment method and device based on deep learning |
| CN113991760A (en) * | 2021-10-20 | 2022-01-28 | 山东大学 | Direct-current locking multi-resource emergency control optimization method for alternating-current and direct-current series-parallel receiving-end power grid |
| CN114330884A (en) * | 2021-12-29 | 2022-04-12 | 山东大学 | Multi-time scale coordinated alternating current and direct current power grid dynamic safety early warning method and system |
| CN114330884B (en) * | 2021-12-29 | 2024-04-16 | 山东大学 | Multi-time scale coordinated AC/DC power grid dynamic safety early warning method and system |
| CN116151130A (en) * | 2023-04-19 | 2023-05-23 | 国网浙江新兴科技有限公司 | Wind power plant maximum frequency damping coefficient calculation method, device, equipment and medium |
| CN116151130B (en) * | 2023-04-19 | 2023-08-15 | 国网浙江新兴科技有限公司 | Wind power plant maximum frequency damping coefficient calculation method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112310980B (en) | 2022-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112310980B (en) | Safety and stability evaluation method and system for direct-current blocking frequency of alternating-current and direct-current series-parallel power grid | |
| CN111628501B (en) | AC/DC large power grid transient voltage stability assessment method and system | |
| Cui et al. | Reinforcement learning for optimal primary frequency control: A Lyapunov approach | |
| EP3889855A1 (en) | Ac/dc bulk power grid dynamic security risk situation rolling prospective early warning method and system | |
| CN112508442B (en) | Transient stability assessment method and system based on automatic and interpretable machine learning | |
| CN109638815B (en) | Method for determining safety and stability prevention control strategy of medium-and-long-term voltage of power system | |
| CN110009141B (en) | Climbing event prediction method and system based on SDAE feature extraction and SVM classification model | |
| CN109066651B (en) | Method for calculating limit transmission power of wind power-load scene | |
| Wen et al. | Data‐driven transient frequency stability assessment: A deep learning method with combined estimation‐correction framework | |
| CN111652478B (en) | Umbrella algorithm-based power system voltage stability evaluation misclassification constraint method | |
| CN112862065A (en) | Power grid branch parameter identification method based on multitask graph neural network | |
| Nhung et al. | Load shedding in microgrids with dual neural networks and AHP algorithm | |
| Oladeji et al. | Density-based clustering and probabilistic classification for integrated transmission-distribution network security state prediction | |
| CN114336632A (en) | Method for correcting alternating current power flow based on model information assisted deep learning | |
| CN114006370A (en) | Power system transient stability analysis and evaluation method and system | |
| Razmi et al. | Neural network based on a genetic algorithm for power system loading margin estimation | |
| CN110675276B (en) | Method and system for inversion droop control of direct current power transmission system | |
| Hossain et al. | Machine learning accelerated real-time model predictive control for power systems | |
| CN115276067A (en) | Distributed energy storage voltage adjusting method adaptive to topological dynamic change of power distribution network | |
| CN112510685B (en) | Urban power grid risk avoidance prediction control method and device | |
| CN113659636A (en) | N-k distribution robust unit control method and device under distribution uncertainty condition | |
| CN112865118B (en) | Deep learning model generation method for power grid dynamic reactive power reserve demand calculation | |
| Wang et al. | Model-Data Integration Driven Based Power System Frequency Response Model | |
| CN113852082A (en) | Method and device for preventing and controlling transient stability of power system | |
| Wang et al. | Transient stability assessment in bulk power grids using sequential minimal optimization based support vector machine with pinball loss |
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 |