CN113708376A - Power system probability optimal power flow calculation method, system, equipment and storage medium - Google Patents
Power system probability optimal power flow calculation method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN113708376A CN113708376A CN202110970945.6A CN202110970945A CN113708376A CN 113708376 A CN113708376 A CN 113708376A CN 202110970945 A CN202110970945 A CN 202110970945A CN 113708376 A CN113708376 A CN 113708376A
- Authority
- CN
- China
- Prior art keywords
- generator
- power system
- data samples
- power
- network model
- 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
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected 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
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention belongs to the field of power system automation, and discloses a method, a system, equipment and a storage medium for calculating probability optimal power flow of a power system, wherein the method comprises the following steps: acquiring a plurality of historical data samples of the power system at the current time period, generating a confrontation network model through a preset multi-generator according to the plurality of acquired historical data samples, and generating a preset number of extended data samples; and obtaining a probability optimal power flow calculation result of the power system at the current time period through a preset probability optimal power flow model according to a preset number of extended data samples. The obtained probability optimal power flow calculation result is more precise and accurate, so that the variation range of the power generation cost can be effectively reduced, and more accurate reference information is provided for uncertain analysis and scheduling operation of a circuit system.
Description
Technical Field
The invention belongs to the field of power system automation, and relates to a method, a system, equipment and a storage medium for calculating probability optimal power flow of a power system.
Background
At present, uncertainty of wind and photovoltaic sources is usually not considered in optimal power flow calculation of a power system, and deterministic optimal power flow is obtained. However, as the penetration rate of wind and photovoltaic sources in the power system is higher and higher, the safety and economic operation of the power system are challenged due to the high intermittency and uncertainty of the wind and photovoltaic sources.
Therefore, the deterministic optimal power flow calculation cannot accurately display the operation state of the power system, wind and photovoltaic source uncertainty factors must be considered in power grid scheduling, otherwise, the safe operation of the power grid is possibly endangered, and the economical efficiency of the system operation is reduced; therefore, probabilistic optimal power flow calculation becomes an important tool for evaluating the influence of the strong uncertainty source loads on the reliability and the economy of the power system.
However, at present, when the probabilistic optimal power flow is calculated, the distribution difference of wind, light and load scenes in different time periods is not considered, so that the result of the probabilistic optimal power flow calculation is not fine enough, the fluctuation range of the obtained power generation cost is large, and the reference significance to a dispatcher is limited.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a method, a system, equipment and a storage medium for calculating the probability-optimal power flow of a power system.
In order to achieve the purpose, the invention adopts the following technical scheme to realize the purpose:
in a first aspect of the present invention, a method for calculating a probabilistic optimal power flow of an electrical power system includes the following steps:
acquiring a plurality of historical data samples of the power system at the current time period;
generating a confrontation network model through a preset multi-generator according to the acquired plurality of historical data samples, and generating a preset number of extended data samples;
and obtaining a probability optimal power flow calculation result of the power system at the current time period through a preset probability optimal power flow model according to a preset number of extended data samples.
The method for calculating the probability optimal power flow of the power system is further improved in that:
the specific method for generating the preset number of extended data samples by generating the confrontation network model through the preset multi-generator according to the obtained plurality of historical data samples comprises the following steps: training a preset multi-generator to generate a confrontation network model according to the acquired plurality of historical data samples, and obtaining the multi-generator generated confrontation network model of the power system at the current time period; and generating a countermeasure network model through a multi-generator of the power system at the current time period, and generating a preset number of extended data samples.
The multi-generator generation confrontation network model comprises a classifier, a discriminator and a preset number of generators; the generator is used for mapping the input noise into an initial extended data sample and outputting the initial extended data sample to the discriminator and the classifier; the discriminator is used for acquiring historical data samples and initial extended data samples of all generators, distinguishing the initial extended data samples from the historical data samples, and taking the initial extended data samples with errors as extended data samples; the classifier is used for obtaining the initial extended data samples of each generator and classifying according to the generators.
The classifier, the discriminator and the generators with preset number are all composed of a plurality of neural network layers, the neural network layers of the discriminator and the classifier have the same structure, and the neural network layers of the generators have the same structure; when a preset multi-generator is trained to generate a confrontation network model according to the acquired historical data samples, parameters of other neural network layers except an output neural network layer in the discriminator and the classifier are shared; parameters of the rest of the neural network layers except the input neural network layer in each generator are shared.
And alternately updating parameters of the classifier, the discriminator and the generator when training a preset multi-generator to generate the confrontation network model according to the acquired plurality of historical data samples.
The multi-generator generates a loss function against the network model as follows:
wherein G is1:KFor K generators; c is a classifier; d is a discriminator;
is historical data sample x-PdataA mathematical expectation of (d); d (x) is a discriminator function;
for all generators, initially spreading the data samples x-PmodelA mathematical expectation of (d); beta is more than 0 and is a hyperparameter; p is a radical ofkGenerating a probability of an initial spread data sample for the kth generator;
for the initial spread data sample of the kth generator
A mathematical expectation of (d); ck(x) The probability generated by the kth generator for the initial extended data sample.
The preset probability optimal power flow model is composed of a target function, equality constraint conditions and inequality constraint conditions;
the objective function is as follows:
wherein f is the total fuel cost of the system thermal power generating unit, ai，bi，ciIs the power generation cost coefficient, P, of the generator at node i of the power systemGiThe active power of a generator at a node i of the power system is shown, and m is the number of nodes in the power system;
the equality constraints are as follows:
wherein, PGi、QGiActive power and reactive power P of the output of the generator at the node i of the power system respectivelyLi、QLiRespectively the active power and the reactive power of a load at a node i of the power system, Ui、UjRespectively representing the voltage amplitude of a node i, the voltage amplitude of a node j, thetaijIs the phase angle difference between node i and node j of the power system, Gij、BijRepresenting the conductance and susceptance, N, between nodes i and j, respectively, of the power systemBIs a power system node set;
the inequality constraints are as follows:
wherein, PGi max、PGi min、QGi max、QGi minRespectively an upper limit of active power, a lower limit of active power, an upper limit of reactive power, a lower limit of reactive power and a U of output of a generator at a node i of the power systemi max、Ui minRespectively an upper voltage amplitude limit, a lower voltage amplitude limit, U at a node i of the power systemiIs the voltage at node i, P, of the power systemlim axIs the branch active power upper limit, N, of the power systemGSet of generator nodes, P, for an electric power systemlFor branch l active power, N of the power systemlIs a set of branches l of the power system.
In a second aspect of the present invention, a system for calculating a probabilistic optimal power flow of an electrical power system includes:
the acquisition module is used for acquiring a plurality of historical data samples of the current time period of the power system, wherein the data samples comprise wind power output data, photovoltaic output data and load data at the same moment;
the extended data sample generation module is used for generating a confrontation network model through a preset multi-generator according to the acquired plurality of historical data samples and generating a preset number of extended data samples;
and the load flow calculation module is used for obtaining a probability optimal load flow calculation result of the power system at the current time period through a preset probability optimal load flow model according to a preset number of extended data samples.
The probability optimal power flow calculation system of the power system is further improved in that:
the extended data sample generation module comprises:
the training module is used for training a preset multi-generator generation countermeasure network model according to the acquired historical data samples to obtain a multi-generator generation countermeasure network model of the power system at the current time period;
and the generation module is used for generating a confrontation network model through a multi-generator of the power system at the current time period and generating a preset number of extended data samples.
The multi-generator generation confrontation network model comprises a classifier, a discriminator and a preset number of generators; the generator is used for mapping the input noise into an initial extended data sample and outputting the initial extended data sample to the discriminator and the classifier; the discriminator is used for acquiring historical data samples and initial extended data samples of all generators, distinguishing the initial extended data samples from the historical data samples, and taking the initial extended data samples with errors as extended data samples; the classifier is used for obtaining the initial extended data samples of each generator and classifying according to the generators.
The multi-generator generates a loss function against the network model as follows:
wherein G is1:KFor K generators; c is a classifier; d is a discriminator;
is historical data sample x-PdataA mathematical expectation of (d); d (x) is a discriminator function;
for all generators, initially spreading the data samples x-PmodelA mathematical expectation of (d); beta is more than 0 and is a hyperparameter; p is a radical ofkGenerating a probability of an initial spread data sample for the kth generator;
for the initial spread data sample of the kth generator
A mathematical expectation of (d); ck(x) The probability generated by the kth generator for the initial extended data sample.
In a third aspect of the present invention, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above power system probability optimal power flow calculation method when executing the computer program.
In a fourth aspect of the present invention, a computer-readable storage medium stores a computer program, which when executed by a processor implements the steps of the above power system probability optimal power flow calculation method.
Compared with the prior art, the invention has the following beneficial effects:
the method for calculating the probability optimal power flow of the power system is based on the characteristic that the probability distribution difference of data samples in different time periods is large, the time sequence correlation of the data samples is fully considered, the probability optimal power flow calculation is carried out in different time periods, a plurality of historical data samples in the current time period are obtained firstly, then a confrontation network model is generated based on a preset multi-generator, the correlation of the data samples in the current time period is accurately captured, various and accurate extended data samples are further generated, finally, the probability optimal power flow calculation result in the current time period of the power system is obtained by utilizing the preset probability optimal power flow model, the obtained probability optimal power flow calculation result is more precise and accurate, the variation range of power generation cost can be effectively reduced, and more accurate reference information is provided for uncertain analysis and dispatching operation of the circuit system.
Drawings
FIG. 1 is a graph of photovoltaic output and load data distribution over a period of 6:00 to 7:00 in a region of Belgium;
FIG. 2 is a wind power output and photovoltaic output and load data distribution diagram at a time period of 6:00-7:00 in Belgium;
FIG. 3 is a flow chart of a method for calculating a probabilistic optimal power flow of an electrical power system according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a multi-generator generation countermeasure network model of the present invention;
FIG. 5 is a flow chart of a method for calculating a probabilistic optimal power flow of an electrical power system according to another embodiment of the present invention;
FIG. 6 is a graph of photovoltaic output probability density over time according to an embodiment of the present invention;
FIG. 7 is a graph of photovoltaic output probability density at different time intervals according to an embodiment of the present invention;
FIG. 8 is a wind power output probability density graph of the embodiment of the invention at different time intervals;
FIG. 9 is a wind power output probability density graph at different time intervals according to an embodiment of the present invention;
FIG. 10 is a graph of load probability density over time for an embodiment of the present invention;
FIG. 11 is a graph of load probability density at different time intervals according to an embodiment of the present invention;
FIG. 12 is a graph comparing the distribution of photovoltaic output and load data with consideration correlation generated by a generative countermeasure network in accordance with an embodiment of the present invention with real data thereof;
FIG. 13 is a graph comparing the distribution of correlation-considered photovoltaic output and load data generated against a network model using multiple generators with real data for an embodiment of the present invention;
FIG. 14 is a distribution comparison graph of wind power output, photovoltaic output and load data generated by the generation countermeasure network and considering the correlation with real data thereof according to the embodiment of the present invention;
FIG. 15 is a distribution comparison graph of wind power output, photovoltaic output and load data generated by a countermeasure network model with consideration of correlation and real data thereof, generated by using multiple generators in the embodiment of the invention;
FIG. 16 is a graph comparing a probability density function curve of wind power 1 output power generated by a multi-generator-based confrontation network model with a probability density function curve of real data thereof according to an embodiment of the present invention;
FIG. 17 is a graph comparing a probability density function curve of wind power 2 output power generated by a multi-generator-based confrontation network model with a probability density function curve of real data thereof according to an embodiment of the present invention;
FIG. 18 is a graph comparing a probability density function curve of the output power of the photovoltaic 1 generated by the multi-generator-based generation countermeasure network model with a probability density function curve of the real data thereof according to the embodiment of the present invention;
FIG. 19 is a graph comparing a probability density function curve of the photovoltaic 2 output power generated by a multi-generator-based countermeasure network model with a probability density function curve of the real data thereof according to an embodiment of the present invention;
FIG. 20 is a graph comparing a probability density function curve of a load 1 generated based on a multi-generator generated confrontation network model with a probability density function curve of real data thereof according to an embodiment of the present invention;
FIG. 21 is a graph comparing a probability density function curve of a load 2 generated by a multi-generator-based confrontation network model with a probability density function curve of real data thereof according to an embodiment of the present invention;
FIG. 22 is a graph of the source-to-load correlation thermodynamic distribution of an embodiment of the present invention based on measured data;
FIG. 23 is a diagram of a thermal distribution of source load correlations for a countermeasure network model generation scenario generated based on multiple generators in accordance with an embodiment of the present invention;
FIG. 24 is a schematic structural diagram of an IEEE standard 118 node system in accordance with an embodiment of the present invention;
FIG. 25 is a timing probability density graph of system power generation cost according to an embodiment of the present invention;
FIG. 26 is a graph comparing probability density of system power generation cost for an embodiment of the present invention and a prior method;
fig. 27 is a block diagram of a system for calculating a probabilistic optimal power flow of an electric power system according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The invention is described in further detail below with reference to the accompanying drawings:
in an electric power system, wind power output, photovoltaic output and loads of general adjacent areas have certain correlation respectively and mutually, and time sequence difference exists in probability distribution of the wind power output, the photovoltaic output and the loads in different time periods. Referring to fig. 1, the data distribution of photovoltaic output and load in a period of 6:00-7:00 in a certain area in belgium is shown, and referring to fig. 2, the data distribution of wind power output, photovoltaic output and load is shown, and it can be seen that the data distribution is dispersed and has strong diversity. If the difference is ignored to perform probability optimal power flow calculation, taking the minimum generation cost of the objective function as an example, the fluctuation range of the generation cost is large, and the reference significance to a dispatcher is limited. Therefore, the probability optimal power flow calculation considering the time sequence and the correlation of the wind power output, the photovoltaic output and the load has important significance for evaluating the operation state of the power system.
Based on the above, in order to better capture the time-space correlation and the time sequence difference between the wind power output, the photovoltaic output and the load, referring to fig. 3, the invention provides a power system probability optimal power flow calculation method, firstly, a multi-source load scene (namely an extended data sample) of an anti-network model is generated based on a multi-generator, then, the time sequence probability optimal power flow calculation is carried out based on the generated extended data sample and with the aim of minimum fuel cost of a thermal power unit, the probability optimal power flow calculation result of the time sequence correlation is considered, and more accurate power grid operation information can be provided for a dispatcher.
Specifically, the method for calculating the probability optimal power flow of the power system comprises the following steps:
s1: a number of historical data samples are obtained for a current time period of the power system.
In this embodiment, the plurality of historical data samples in the current time period refer to data samples in the current time period in the history each day, where the time period may be divided into one hour, for example, 6:00 to 7:00 is a time period, and then the plurality of historical data samples in the current time period of 6:00 to 7:00 refer to: historically, data samples were taken daily over a period of 6:00-7:00, but not limited thereto. The data samples comprise wind power output data, photovoltaic output data and load data at the same moment, and relevance among the wind power output data, the photovoltaic output data and the load data is fully considered. The same time refers to a sampling time, where the historical data samples may be historical data samples over a period of time, such as historical data samples of the first three days, so that each sampling time has three historical data samples, and the first three days are only exemplified here. These historical data samples can be obtained directly from the grid company's measurement data.
S2: and generating a confrontation network model through a preset multi-generator according to the plurality of acquired historical data samples, and generating a preset number of extended data samples.
When data is generated, it is generally generated by generating a countermeasure network model. Generation of Antagonistic Networks (GANs) was proposed by Ian Goodfellow et al in 2014, Generated adaptive Networks. The problem of generation is treated as a countermeasure and game for two networks, the arbiter and the generator: the generator produces synthetic data from a given noise (typically a uniform or normal distribution) and the discriminator resolves the output of the generator from the true data. The former attempts to produce data that is closer to the true one, and correspondingly, the latter attempts to more perfectly distinguish the true data from the generated data. Therefore, the two networks progress in the countermeasure, and continue to fight after the progress, the data obtained by the generation type network is more and more perfect and approaches the real data, so that the data (pictures, sequences, videos and the like) which is expected to be obtained can be generated.
However, the wind power output data, the photovoltaic output data and the load data in the data sample are distributed dispersedly and have strong diversity, if the existing generation countermeasure network model is simply adopted for generation, the problem of mode collapse often exists in the calculation process, and the diversity and the accuracy of the generated data are not enough.
Based on this, in the embodiment, a multi-generator generation confrontation network model is designed, and is improved on the basis of the existing generation confrontation network model, and a plurality of generators are added. Specifically, the method comprises a classifier, a discriminator and a preset number of generators; the generator is used for mapping the input noise into an initial extended data sample and outputting the initial extended data sample to the discriminator and the classifier; the discriminator is used for acquiring historical data samples and initial extended data samples of all generators, distinguishing the initial extended data samples from the historical data samples, and taking the initial extended data samples with errors as extended data samples; the classifier is used for obtaining the initial extended data sample of each generator and classifying according to the generator. The multi-generator generation countermeasure network model is combined with the time-interval historical data samples to generate time-interval extended data samples, the result shows that compared with the common several generation countermeasure network models, the generated extended data samples have stronger diversity and accuracy, and the network structure of the multi-generator generation countermeasure network model is shown in the attached figure 4.
Specifically, the multi-generator generation countermeasure network model comprises K generators G1:K(G1(z)～Gk(z)), a discriminator D and a classifier C, each composed of a plurality of neural network layers, the network structure of each generator being the same, and the neural network layer structures of the discriminator and the classifier being the same.
Each generator GkThe noise z (gaussian noise is typically used) can be mapped to x-Gk(z) to form a monodispersion
The K generators form a mixture P of K distributionsmodel. Then according to the Monte Carlo random sampling of the multivariate normal distribution, P is sampledmodelSample size G ofu(z) as an output quantity. Wherein u is a polynomial distribution X-PN(N:p1,p2,…,pK) The amount of Monte Carlo sampling, here [ p ]1,p2,…,pK]The probability of each generator generating data is a preset fixed parameter for calculation, and in the embodiment, the probability of each generator generating data is 1/K, and then the probability is input into a discriminator and a classificationIn the device. The function of the discriminator is to distinguish the sample quantity Gu(z) (i.e., initial extended data samples) and historical data samples x-Pdata. The classifier is used for classifying the initial extended data sample according to the generator through the index mark of the corresponding generator.
The multi-generator generates a loss function against the network model as follows:
wherein G is1:KFor K generators; c is a classifier; d is a discriminator;
is historical data sample x-PdataA mathematical expectation of (d); d (x) is a discriminator function;
for all generators, initially spreading the data samples x-PmodelA mathematical expectation of (d); beta is more than 0 and is a hyperparameter; p is a radical ofkGenerating a probability of an initial spread data sample for the kth generator;
for the initial spread data sample of the kth generator
A mathematical expectation of (d); ck(x) The probability generated by the kth generator for the initial extended data sample.
The first two terms of the loss function represent the interaction between the generator and the arbiter, the last one is the standard SoftMax loss function for a multi-class setup, with the goal of maximizing the entropy of the classifier, which represents the interaction between the generator and the classifier, encouraging each generator to generate data that is distinguishable from the other generators.
Specifically, a countermeasure network model is generated based on multiple generators, the countermeasure network model is generated through a preset multiple generator according to the acquired multiple historical data samples, and a specific method for generating a preset number of extended data samples includes: training a preset multi-generator to generate a confrontation network model according to the acquired plurality of historical data samples, and obtaining the multi-generator generated confrontation network model of the power system at the current time period; the countermeasure network model is generated through a multi-generator of the power system at the current time period, and a preset number of extended data samples are generated, wherein the preset number can be set according to specific conditions.
The method comprises the steps of training a preset multi-generator to generate a confrontation network model, namely inputting a plurality of historical data samples into the multi-generator to generate the confrontation network model, enabling the multi-generator to generate the confrontation network model to learn the distribution rule of the historical data samples, and further simulating the distribution of the historical data samples. Then, a countermeasure network model is generated based on the trained multi-generator, generating a large number of extended data samples. In addition, when the distribution rule of the historical data samples is learned, the historical data samples in all time periods are learned, the distribution difference of the data samples in different time periods is fully considered, and the learned distribution rule is more in line with the actual situation.
Preferably, when the preset multi-generator is trained to generate the confrontation network model according to the acquired plurality of historical data samples, parameters of other neural network layers except the output neural network layer in the discriminator and the classifier are shared; parameters of the rest of the neural network layers except the input neural network layer in each generator are shared. By adopting a parameter sharing mechanism, the number of network parameters can be reduced, and the model can be trained efficiently.
Preferably, when the preset multi-generator is trained to generate the confrontation network model according to the acquired plurality of historical data samples, the parameters of the classifier, the discriminator and the generator are alternately updated.
S3: and obtaining a probability optimal power flow calculation result of the power system at the current time period through a preset probability optimal power flow model according to a preset number of extended data samples.
Specifically, in this embodiment, the minimum value of the fuel cost of the thermal power generating unit of the system is used as an objective function, MATPOWER (matlab-based power system power flow and optimal power flow calculation software) is used to perform probability optimal power flow calculation on the power system, and the purpose is to optimize the safety and economy of the operation of the power system. The optimal power flow model is composed of an objective function, equality constraint conditions and inequality constraint conditions.
The objective function is as follows:
wherein f is the total fuel cost of the system thermal power generating unit, ai，bi，ciIs the power generation cost coefficient, P, of the generator at node i of the power systemGiThe active power of a generator at a node i of the power system is shown, and m is the number of nodes in the power system;
the equality constraints are as follows:
wherein, PGi、QGiActive power and reactive power P of the output of the generator at the node i of the power system respectivelyLi、QLiRespectively the active power and the reactive power of a load at a node i of the power system, Ui、UjRespectively representing the voltage amplitude of a node i, the voltage amplitude of a node j, thetaijIs the phase angle difference between node i and node j of the power system, Gij、BijRepresenting the conductance and susceptance, N, between nodes i and j, respectively, of the power systemBIs a power system node set;
the inequality constraints are as follows:
wherein, PGi max、PGi min、QGi max、QGi minRespectively an upper limit of active power, a lower limit of active power, an upper limit of reactive power, a lower limit of reactive power and a U of output of a generator at a node i of the power systemi max、Ui minRespectively an upper voltage amplitude limit, a lower voltage amplitude limit, U at a node i of the power systemiIs the voltage at node i, P, of the power systemlim axIs the branch active power upper limit, N, of the power systemGSet of generator nodes, P, for an electric power systemlFor branch l active power, N of the power systemlIs a set of branches l of the power system.
And inputting a probability optimal power flow model by generating a preset number of extended data samples generated by the countermeasure network model by the multi-generator, and further obtaining a probability optimal power flow calculation result of the power system at the current time.
Preferably, referring to fig. 5, traversing each time period i in one day, dividing intervals by taking one hour as a time period, obtaining historical data samples of the ith hour, training the multiple generators to generate a confrontation network model, obtaining the trained multiple generators to generate the confrontation network model, then generating the confrontation network model through the trained multiple generators, generating extended data samples of the ith hour, bringing the extended data samples of the ith hour into the probabilistic optimal power flow model for calculation, obtaining a probabilistic optimal power flow calculation result of the ith hour, judging whether i <24 is true, and when i <24, making i +1 and repeating the above steps to further obtain the probabilistic optimal power flow calculation result of one day.
In summary, the method for calculating the probabilistic optimal power flow of the power system of the invention fully considers the time sequence correlation of the wind power output data, the photovoltaic output data and the load data based on the characteristic of large difference of the probability distribution of the wind power output data, the photovoltaic output data and the load data in different time periods, performs the probabilistic optimal power flow calculation in different time periods, firstly obtains a plurality of historical data samples in the current time period, then generates a confrontation network model based on a preset multi-generator, accurately captures the correlation of the wind power output data, the photovoltaic output data and the load data in the current time period, further generates various and accurate extended data samples, finally obtains the probabilistic optimal power flow calculation result in the current time period of the power system by using the preset probabilistic optimal power flow model, and obtains the obtained probabilistic optimal power flow calculation result is more precise and accurate, further can effectively reduce the variation range of the power generation cost, more accurate reference information is provided for uncertain analysis and scheduling operation of the circuit system.
The accuracy and significance of the probabilistic optimal power flow calculation method for the power system are verified through data of Belgium elia operators and an IEEE118 node system.
And (4) making source load probability distribution function curves of different periods based on historical wind power output data, photovoltaic output data and load data. Fig. 6 is a power output probability density curve of a certain photovoltaic power plant at different time intervals, and fig. 7 is a photovoltaic power output probability density curve at different time intervals. It can be seen that the photovoltaic output probability distributions in different time periods are greatly different, and the same is true in the case of wind power and load, see fig. 8 to 11. From the experimental results, if the probability optimal power flow of the power system is calculated by only using the wind power output data, the photovoltaic output data and the load data which are not divided into time intervals, the reference meaning of the scheduling operation of the circuit system is limited and is not fine enough.
Based on the historical wind power output data, photovoltaic output data and load data of 2019 and 2017 in a certain area in Belgium, a corresponding multi-generator generation countermeasure network is designed according to the historical data distribution characteristics of wind, photovoltaic source output and load in each time period. Referring to fig. 12 and 13, which are graphs comparing photovoltaic output data and load data distributions generated using a generative countermeasure network model and a multi-generator generative countermeasure network model for periods of 6:00-7:00, respectively, taking into account correlation. Referring to fig. 14 and 15, which are respectively time periods from 6:00 to 7:00, a data distribution comparison graph of wind power output data, photovoltaic output data and load data generated by the generation countermeasure network model and the multi-generator generation countermeasure network model considering correlation is adopted. Referring to fig. 17 to 21, in a period from 12:00 to 13:00, the probability density function curves of wind power output data, photovoltaic output data and load data generated by the countermeasure network model by using the multi-generator are compared with the probability density function curves of real data. Compared with the common generated countermeasure network model, the multi-generator generated countermeasure network model has a better fitting effect on wind power output data, photovoltaic output data and load data in each time period in one day, can effectively improve the local adaptability of the probability density function curve of the wind power output data, the photovoltaic output data and the load data, and more effectively reflects the uncertainty and the volatility of the wind power output, the photovoltaic output and the load.
The correlation between the generated multi-source scenes is analyzed below. See, e.g., 22 and 23 for the respective source-load-related thermodynamic distribution plots based on the measured data and for the respective source-load-related thermodynamic distribution plots of the antagonistic network model generation scenario based on the multi-generator generation, respectively.
And (4) performing simulation operation on the time sequence related probability optimal power flow by adopting a standard IEEE118 system. Referring to FIG. 24, the network node wiring of the standard IEEE118 system, the number of nodes N in the network B118, number of branches Nl186, the number of conventional generator nodes NG54. No. 10, No. 12, No. 15 and No. 18 nodes are respectively wind power and photovoltaic nodes with correlation, and No. 54 and No. 59 nodes are respectively load nodes with correlation with wind power and photovoltaic.
In order to analyze the influence of the wind power output, the photovoltaic output and the time sequence and the correlation characteristics of the load on the system, the probability optimal power flow calculation is carried out according to the wind power output, the photovoltaic output and the time sequence data of the load in 24 hours. Referring to fig. 25, which is a time-series probability density diagram of the power generation cost of the power system, it can be seen that the power generation cost curve of the power system is relatively steep at 1:00-6:00 and 20:00-24:00 at night, and the power generation cost curve of the power system is relatively gentle at 7:00-19:00, and the peak probability density thereof fluctuates with time. The reason is that only fire electricity and intermittent wind electricity exist at night, and the day system power generation consists of three parts, namely wind power generation, solar power generation and a traditional power generation mode.
The method for calculating the probability optimal power flow of the power system can reduce the power generation cost and adjust the traditional power generation output according to the time sequence wind-light output, so that the traditional power generation output data in the daytime is more dispersed relative to the night, and the power generation cost curve is more smooth. In addition, the variation trend of the 24-hour power generation cost of the system with time sequence is similar to the variation trend of the load with time sequence shown in fig. 11, because the power generation amount of the power system is correspondingly changed to meet the requirement of the load time sequence variation of the power system shown in fig. 11, and the power generation cost is further influenced.
Referring to fig. 26, a comparison graph of the power generation cost probability density of the power system is shown, wherein wind power output, photovoltaic output and load data of 4:00-5:00, 12:00-13:00 and 16:00-17:00 are taken as input in consideration of the time sequence characteristic curve, and wind power output and load data of all days (all) are taken as input in consideration of the time sequence characteristic curve. Referring to fig. 25 again, the variation range of the power generation cost curve without considering the time sequence characteristic is about 117500-135000 $/h, and the data distribution is more dispersive. According to the method for calculating the probability optimal power flow of the power system, the variation ranges of the power generation costs of the obtained probability optimal power flow results are 122500$/h-129000$/h, 124000$/h-131000$/h and 123000$/h-131000$/h respectively in the ranges of 4:00-5:00, 12:00-13:00 and 16:00-17:00, the variation ranges are greatly reduced, data distribution is concentrated, and the power generation cost difference between every hour is obvious. The probability optimal power flow calculation method for the power system can improve the data density set of the power generation cost per hour, effectively reduce the variation range of the power generation cost of the power system, provide more valuable reference information for power system scheduling and reduce the budget of the power generation cost.
The following are embodiments of the apparatus of the present invention that may be used to perform embodiments of the method of the present invention. For details of non-careless mistakes in the embodiment of the apparatus, please refer to the embodiment of the method of the present invention.
Referring to fig. 27, in a further embodiment of the present invention, a system for calculating a probabilistic optimal power flow of an electric power system is provided, which can be used to implement the method for calculating a probabilistic optimal power flow of an electric power system.
The acquisition module acquires a plurality of historical data samples of the current time period of the power system, wherein the data samples comprise wind power output data, photovoltaic output data and load data at the same moment; the extended data sample generation module is used for generating a confrontation network model through a preset multi-generator according to the acquired plurality of historical data samples and generating a preset number of extended data samples; the load flow calculation module is used for obtaining a probability optimal load flow calculation result of the power system at the current time period through a preset probability optimal load flow model according to a preset number of extended data samples.
Preferably, the extended data sample generation module includes a training module and a generation module. The training module is used for training a preset multi-generator to generate a confrontation network model according to the acquired historical data samples, and obtaining the multi-generator generated confrontation network model of the power system at the current time period; the generation module is used for generating a confrontation network model through a multi-generator of the power system at the current time period and generating a preset number of extended data samples.
Preferably, the multi-generator generation countermeasure network model comprises a classifier, a discriminator and a preset number of generators; the generator is used for mapping the input noise into an initial extended data sample and outputting the initial extended data sample to the discriminator and the classifier; the discriminator is used for acquiring historical data samples and initial extended data samples of all generators, distinguishing the initial extended data samples from the historical data samples, and taking the initial extended data samples with errors as extended data samples; the classifier is used for obtaining the initial extended data samples of each generator and classifying according to the generators.
Preferably, the multiple generator generates a loss function against the network model as follows:
wherein G is1:KFor K generators; c is a classifier; d is a discriminator;
is historical data sample x-PdataA mathematical expectation of (d); d (x) is a discriminator function;
for all generators, initially spreading the data samples x-PmodelA mathematical expectation of (d); beta is more than 0 and is a hyperparameter; p is a radical ofkGenerating a probability of an initial spread data sample for the kth generator;
for the initial spread data sample of the kth generator
A mathematical expectation of (d); ck(x) The probability generated by the kth generator for the initial extended data sample.
In yet another embodiment of the present invention, a computer device is provided that includes a processor and a memory for storing a computer program comprising program instructions, the processor for executing the program instructions stored by the computer storage medium. The Processor may be a Central Processing Unit (CPU), or may be other general-purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable gate array (FPGA) or other Programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., which is a computing core and a control core of the terminal, and is specifically adapted to load and execute one or more instructions in a computer storage medium to implement a corresponding method flow or a corresponding function; the processor provided by the embodiment of the invention can be used for operating the power system probability optimal power flow calculation method.
In yet another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a Memory device in a computer device and is used for storing programs and data. It is understood that the computer readable storage medium herein can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also, one or more instructions, which may be one or more computer programs (including program code), are stored in the memory space and are adapted to be loaded and executed by the processor. It should be noted that the computer-readable storage medium may be a high-speed RAM memory, or may be a non-volatile memory (non-volatile memory), such as at least one disk memory. One or more instructions stored in the computer-readable storage medium may be loaded and executed by a processor to implement the corresponding steps of the method for calculating the probabilistic optimal power flow of the power system according to the above embodiments.
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.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (13)
1. A method for calculating the probability optimal power flow of an electric power system is characterized by comprising the following steps:
acquiring a plurality of historical data samples of the power system at the current time period;
generating a confrontation network model through a preset multi-generator according to the acquired plurality of historical data samples, and generating a preset number of extended data samples;
and obtaining a probability optimal power flow calculation result of the power system at the current time period through a preset probability optimal power flow model according to a preset number of extended data samples.
2. The method for calculating the probabilistic optimal power flow of the power system according to claim 1, wherein the method for generating the confrontation network model through a preset multi-generator according to the acquired plurality of historical data samples comprises the following specific steps:
training a preset multi-generator to generate a confrontation network model according to the acquired plurality of historical data samples, and obtaining the multi-generator generated confrontation network model of the power system at the current time period; and generating a countermeasure network model through a multi-generator of the power system at the current time period, and generating a preset number of extended data samples.
3. The power system probabilistic optimal power flow calculation method according to claim 2, wherein the multi-generator generation countermeasure network model includes a classifier, a discriminator, and a preset number of generators;
the generator is used for mapping the input noise into an initial extended data sample and outputting the initial extended data sample to the discriminator and the classifier;
the discriminator is used for acquiring historical data samples and initial extended data samples of all generators, distinguishing the initial extended data samples from the historical data samples, and taking the initial extended data samples with errors as extended data samples;
the classifier is used for obtaining the initial extended data samples of each generator and classifying according to the generators.
4. The method for calculating the probabilistic optimal power flow of the electric power system according to claim 3, wherein the classifier, the discriminator and the preset number of generators are all composed of a plurality of neural network layers, the neural network layers of the discriminator and the classifier have the same structure, and the neural network layers of the generators have the same structure;
when a preset multi-generator is trained to generate a confrontation network model according to the acquired historical data samples, parameters of other neural network layers except an output neural network layer in the discriminator and the classifier are shared; parameters of the rest of the neural network layers except the input neural network layer in each generator are shared.
5. The method for calculating the probabilistic optimal power flow of the power system according to claim 3, wherein parameters of the classifier, the discriminator and the generator are alternately updated when a preset multi-generator is trained to generate the confrontation network model according to the acquired historical data samples.
6. The power system probabilistic optimal power flow calculation method of claim 3, wherein the multiple generator generates a loss function against the network model as follows:
wherein G is1:KFor K generators; c is a classifier; d is a discriminator;
is historical data sample x-PdataA mathematical expectation of (d); d (x) is a discriminator function;
for all generators, initially spreading the data samples x-PmodelA mathematical expectation of (d); beta is more than 0 and is a hyperparameter; p is a radical ofkGenerating a probability of an initial spread data sample for the kth generator;
for the initial spread data sample of the kth generator
A mathematical expectation of (d); ck(x) The probability generated by the kth generator for the initial extended data sample.
7. The method according to claim 1, wherein the predetermined probabilistic optimal power flow model is composed of an objective function, equality constraints and inequality constraints;
the objective function is as follows:
wherein f is the total fuel cost of the system thermal power generating unit, ai，bi，ciIs the power generation cost coefficient, P, of the generator at node i of the power systemGiThe active power of a generator at a node i of the power system is shown, and m is the number of nodes in the power system;
the equality constraints are as follows:
wherein, PGi、QGiActive power and reactive power P of the output of the generator at the node i of the power system respectivelyLi、QLiRespectively the active power and the reactive power of a load at a node i of the power system, Ui、UjRespectively representing the voltage amplitude of a node i, the voltage amplitude of a node j, thetaijIs the phase angle difference between node i and node j of the power system, Gij、BijRepresenting the conductance and susceptance, N, between nodes i and j, respectively, of the power systemBIs a power system node set;
the inequality constraints are as follows:
wherein, PGimax、PGimin、QGimax、QGiminRespectively the upper limit of active power, the lower limit of active power, the upper limit of reactive power and the lower limit of reactive power of the output of the generator at the node i of the power systemLimit, Uimax、UiminRespectively an upper voltage amplitude limit, a lower voltage amplitude limit, U at a node i of the power systemiIs the voltage at node i, P, of the power systemlimaxIs the branch active power upper limit, N, of the power systemGSet of generator nodes, P, for an electric power systemlFor branch l active power, N of the power systemlIs a set of branches l of the power system.
8. A power system probabilistic optimal power flow calculation system, comprising:
the acquisition module is used for acquiring a plurality of historical data samples of the current time period of the power system, wherein the data samples comprise wind power output data, photovoltaic output data and load data at the same moment;
the extended data sample generation module is used for generating a confrontation network model through a preset multi-generator according to the acquired plurality of historical data samples and generating a preset number of extended data samples;
and the load flow calculation module is used for obtaining a probability optimal load flow calculation result of the power system at the current time period through a preset probability optimal load flow model according to a preset number of extended data samples.
9. The power system probabilistic optimal power flow calculation system of claim 8, wherein the extended data sample generation module comprises:
the training module is used for training a preset multi-generator generation countermeasure network model according to the acquired historical data samples to obtain a multi-generator generation countermeasure network model of the power system at the current time period;
and the generation module is used for generating a confrontation network model through a multi-generator of the power system at the current time period and generating a preset number of extended data samples.
10. The power system probabilistic optimal power flow calculation system of claim 8, wherein the multi-generator generated countermeasure network model includes a classifier, a discriminator, and a preset number of generators; the generator is used for mapping the input noise into an initial extended data sample and outputting the initial extended data sample to the discriminator and the classifier; the discriminator is used for acquiring historical data samples and initial extended data samples of all generators, distinguishing the initial extended data samples from the historical data samples, and taking the initial extended data samples with errors as extended data samples; the classifier is used for obtaining the initial extended data samples of each generator and classifying according to the generators.
11. The power system probabilistic optimal power flow calculation system of claim 10, wherein the multiple generator generates a loss function against the network model as follows:
wherein G is1:KFor K generators; c is a classifier; d is a discriminator;
is historical data sample x-PdataA mathematical expectation of (d); d (x) is a discriminator function;
for all generators, initially spreading the data samples x-PmodelA mathematical expectation of (d); beta is more than 0 and is a hyperparameter; p is a radical ofkGenerating a probability of an initial spread data sample for the kth generator;
for the initial spread data sample of the kth generator
A mathematical expectation of (d); ck(x) The probability generated by the kth generator for the initial extended data sample.
12. A computer arrangement comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, carries out the steps of the method for probabilistic optimal power flow calculation of an electric power system according to any of the claims 1 to 7.
13. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for probabilistic optimal power flow calculation of an electric power system according to any of the claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110970945.6A CN113708376B (en) | 2021-08-23 | 2021-08-23 | Power system probability optimal power flow calculation method, system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110970945.6A CN113708376B (en) | 2021-08-23 | 2021-08-23 | Power system probability optimal power flow calculation method, system, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113708376A true CN113708376A (en) | 2021-11-26 |
| CN113708376B CN113708376B (en) | 2023-08-08 |
Family
ID=78654278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110970945.6A Active CN113708376B (en) | 2021-08-23 | 2021-08-23 | Power system probability optimal power flow calculation method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113708376B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114330563A (en) * | 2021-12-30 | 2022-04-12 | 山东浪潮科学研究院有限公司 | Power dispatching plan generation method, equipment and medium based on GAN model |
| CN114443593A (en) * | 2021-12-13 | 2022-05-06 | 北京邮电大学 | Multi-party data sharing method based on generation countermeasure network and related equipment |
| CN117254472A (en) * | 2023-11-16 | 2023-12-19 | 国网湖北省电力有限公司武汉供电公司 | Power system probability power flow calculation method and system based on data fusion |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107658878A (en) * | 2017-10-25 | 2018-02-02 | 华中科技大学 | A kind of Probabilistic optimal load flow method containing large-scale energy storage device |
| CN112217215A (en) * | 2020-09-10 | 2021-01-12 | 西安交通大学 | PSD-BPA-based large-scale power system random load flow calculation method |
| CN113131482A (en) * | 2019-12-30 | 2021-07-16 | 东北电力大学 | Probability optimal power flow calculation method and system considering photovoltaic output characteristics |
-
2021
- 2021-08-23 CN CN202110970945.6A patent/CN113708376B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107658878A (en) * | 2017-10-25 | 2018-02-02 | 华中科技大学 | A kind of Probabilistic optimal load flow method containing large-scale energy storage device |
| CN113131482A (en) * | 2019-12-30 | 2021-07-16 | 东北电力大学 | Probability optimal power flow calculation method and system considering photovoltaic output characteristics |
| CN112217215A (en) * | 2020-09-10 | 2021-01-12 | 西安交通大学 | PSD-BPA-based large-scale power system random load flow calculation method |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114443593A (en) * | 2021-12-13 | 2022-05-06 | 北京邮电大学 | Multi-party data sharing method based on generation countermeasure network and related equipment |
| CN114330563A (en) * | 2021-12-30 | 2022-04-12 | 山东浪潮科学研究院有限公司 | Power dispatching plan generation method, equipment and medium based on GAN model |
| CN117254472A (en) * | 2023-11-16 | 2023-12-19 | 国网湖北省电力有限公司武汉供电公司 | Power system probability power flow calculation method and system based on data fusion |
| CN117254472B (en) * | 2023-11-16 | 2024-01-30 | 国网湖北省电力有限公司武汉供电公司 | Power system probability power flow calculation method and system based on data fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113708376B (en) | 2023-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Teo et al. | Forecasting of photovoltaic power using extreme learning machine | |
| Kavousi-Fard et al. | A novel stochastic framework based on cloud theory and $\theta $-modified bat algorithm to solve the distribution feeder reconfiguration | |
| CN113708376A (en) | Power system probability optimal power flow calculation method, system, equipment and storage medium | |
| Ding et al. | Neural network-based model design for short-term load forecast in distribution systems | |
| Fioriti et al. | A novel stochastic method to dispatch microgrids using Monte Carlo scenarios | |
| Athari et al. | Modeling the uncertainties in renewable generation and smart grid loads for the study of the grid vulnerability | |
| Tang et al. | Efficient scenario generation of multiple renewable power plants considering spatial and temporal correlations | |
| Toubeau et al. | Capturing spatio-temporal dependencies in the probabilistic forecasting of distribution locational marginal prices | |
| Jiang et al. | Day‐ahead renewable scenario forecasts based on generative adversarial networks | |
| CN113505458A (en) | Cascading failure key trigger branch prediction method, system, equipment and storage medium | |
| Alhmoud et al. | Short-term load forecasting for Jordan power system based on NARX-ELMAN neural network and ARMA model | |
| Mandal et al. | A new recursive neural network algorithm to forecast electricity price for PJM day‐ahead market | |
| Chen et al. | Distributional analysis for model predictive deferrable load control | |
| Tavares et al. | Comparison of PV power generation forecasting in a residential building using ANN and DNN | |
| Manur et al. | Smart solar home system with solar forecasting | |
| Zhang et al. | Multi-objective interval prediction of wind power based on conditional copula function | |
| Buonanno et al. | Comprehensive method for modeling uncertainties of solar irradiance for PV power generation in smart grids | |
| Lokman et al. | Multi-verse optimization based evolutionary programming technique for power scheduling in loss minimization scheme | |
| Khwaja et al. | Short-term individual electric vehicle charging behavior prediction using long short-term memory networks | |
| Zahraoui et al. | ANN-LSTM Based Tool For Photovoltaic Power Forecasting. | |
| Arshi et al. | Prediction using LSTM networks | |
| Osório et al. | Hybrid evolutionary-adaptive approach to predict electricity prices and wind power in the short-term | |
| Diop et al. | Wind Power Forecasting Using Machine Learning Algorithms | |
| Giraldo et al. | Optimal operation of community energy storage using stochastic gradient boosting trees | |
| Gilbert et al. | A hierarchical approach to probabilistic wind power forecasting |
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 |