[go: up one dir, main page]

CN118246326A - Distributed distribution network reliability assessment method and system based on deep neural network - Google Patents

Distributed distribution network reliability assessment method and system based on deep neural network Download PDF

Info

Publication number
CN118246326A
CN118246326A CN202410369522.2A CN202410369522A CN118246326A CN 118246326 A CN118246326 A CN 118246326A CN 202410369522 A CN202410369522 A CN 202410369522A CN 118246326 A CN118246326 A CN 118246326A
Authority
CN
China
Prior art keywords
neural network
distributed
distribution network
wind speed
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.)
Pending
Application number
CN202410369522.2A
Other languages
Chinese (zh)
Inventor
易辰颖
周杨珺
张炜
周柯
周远超
潘俊涛
黄伟翔
李珊
秦丽文
奉斌
颜丽娟
伍红文
阳浩
许士锦
李林穗
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Electric Power Research Institute of Guangxi Power Grid Co Ltd
Original Assignee
Electric Power Research Institute of Guangxi Power Grid Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Electric Power Research Institute of Guangxi Power Grid Co Ltd filed Critical Electric Power Research Institute of Guangxi Power Grid Co Ltd
Priority to CN202410369522.2A priority Critical patent/CN118246326A/en
Publication of CN118246326A publication Critical patent/CN118246326A/en
Pending legal-status Critical Current

Links

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
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • 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
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • 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
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/04—Power grid distribution networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
    • H02J2101/24—
    • H02J2101/28—
    • H02J2103/30—

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Power Engineering (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • Artificial Intelligence (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Water Supply & Treatment (AREA)
  • General Business, Economics & Management (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Strategic Management (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

本发明公开了基于深度神经网络的分布式配电网可靠性评估方法及系统，涉及电力系统自动化技术领域，包括获取风速数据，构建风电机组出力模型和光伏出力模型；为配电网中每个节点分配一个最大可接受分布式电源容量；通过改进的蒙特卡罗方法获得原始数据集，训练神经网络并进行预测。本发明提供的基于深度神经网络的分布式配电网可靠性评估方法通过出力模型考虑了环境因素的影响，增强了模型的实际适用性，采用改进的蒙特卡罗方法计算可靠性指标，提高了计算的效率和精度，通过BP神经网络对可靠性指标进行快速预测，显著减少了分析时间，提高了配电网可靠性评估的效率和准确性，本发明在分析时间、评估准确率和评估效率方面都取得更加良好的效果。

The present invention discloses a distributed distribution network reliability assessment method and system based on a deep neural network, which relates to the technical field of power system automation, including obtaining wind speed data, constructing a wind turbine output model and a photovoltaic output model; allocating a maximum acceptable distributed power supply capacity to each node in the distribution network; obtaining the original data set through an improved Monte Carlo method, training the neural network and performing prediction. The distributed distribution network reliability assessment method based on a deep neural network provided by the present invention takes into account the influence of environmental factors through the output model, enhances the practical applicability of the model, calculates the reliability index using an improved Monte Carlo method, improves the efficiency and accuracy of the calculation, and quickly predicts the reliability index through a BP neural network, significantly reduces the analysis time, and improves the efficiency and accuracy of the distribution network reliability assessment. The present invention achieves better results in terms of analysis time, assessment accuracy and assessment efficiency.

Description

Distributed power distribution network reliability assessment method and system based on deep neural network
Technical Field
The invention relates to the technical field of power system automation, in particular to a distributed power distribution network reliability assessment method based on a deep neural network.
Background
With the growth of energy demand and the advancement of energy conversion, distributed power sources have become an important component of modern power distribution networks. Meanwhile, with the improvement of industrialization degree, the requirements of people on the reliability of power supply are increasing. However, due to randomness and volatility of distributed power generation such as wind power generation and photovoltaic power generation, particularly under extreme weather conditions such as typhoons and heavy rain, the power demand of power users cannot be met, and the access of a large-scale distributed power supply makes safe operation of a power grid face a great challenge. However, the existing reliability analysis method generally requires a long calculation time, and cannot be adjusted in time according to the real-time change condition. Therefore, the reliability analysis of the power distribution network accessed by the large-scale distributed power supply has great significance.
At present, four methods for calculating the reliability index of the power distribution network are respectively a classical analysis method, an analog method, a hybrid method and an artificial intelligence algorithm. The classical analysis method is a method for quantitatively analyzing based on the structure and the operation parameters of the power distribution network, and generally comprises mathematical deduction and calculation of reliability indexes, such as failure rate, average repair time, availability and the like, and the method needs to fully know the grid structure and the parameters of the power distribution network and conduct reliability analysis through a mathematical model and a statistical theory; the simulation method comprises the steps of performing simulation by using a computer, evaluating the reliability of the power distribution network by generating a large number of random samples, wherein common simulation methods comprise Monte Carlo simulation and event-driven simulation, and obtaining the reliability level of the power distribution network under different conditions by performing multiple simulation on the state and the event of the power distribution network; the two methods have certain defects although each method has the characteristics, and for the method, the mixed method combines the advantages of the classical analysis method and the analog method, the reliability analysis is carried out by combining the two methods, firstly, the reliability parameters of key equipment in the power distribution network are obtained by using the classical analysis method, and then the parameters are input into the analog model for simulation calculation, so that a more comprehensive reliability evaluation result is obtained; the artificial intelligence algorithm comprises a neural network, a genetic algorithm, fuzzy logic and other methods, and can be used for processing a large amount of data and complex relations of the power distribution network, so that calculation and prediction of reliability indexes are realized, rules can be learned from the data and prediction can be made by the artificial intelligence algorithm, and understanding and prediction capabilities of the reliability of the power distribution network can be improved.
Disclosure of Invention
The present invention has been made in view of the above-described problems.
Therefore, the technical problems solved by the invention are as follows: the existing reliability analysis method has the problems of larger limitation of a calculation method, insufficient artificial intelligence algorithm and optimization of how to comprehensively consider the influence of distributed power supply access on the power distribution network.
In order to solve the technical problems, the invention provides the following technical scheme: the distributed power distribution network reliability assessment method based on the deep neural network comprises the steps of obtaining wind speed data and constructing a wind turbine output model and a photovoltaic output model; distributing a maximum acceptable distributed power capacity to each node in the power distribution network; the raw data set is obtained by a modified monte carlo method, the neural network is trained and predicted.
As a preferable scheme of the reliability evaluation method of the distributed power distribution network based on the deep neural network, the invention comprises the following steps: the wind turbine generator output model comprises the steps of modeling and predicting wind speed by adopting an ARMA model, calculating a wind speed average value and a standard deviation of wind speed distribution based on original wind speed data, and processing the original wind speed data to obtain standardized data, wherein the relationship between the standardized data and the original wind speed data is expressed as follows:
Wherein y t is standardized data, V t is raw wind speed data, μ is an average value of wind speeds, σ is a standard deviation of wind speed distribution, and the ARMA model is expressed as:
yt＝a1yt-1+a2yt-2+…+amyt-m+
αt+c1αt-1+c2αt-2+…cnαt-n
Wherein a i is an autoregressive coefficient, c i is a moving average coefficient, alpha t is normally distributed white noise, and the relation between the output and wind speed of the wind turbine generator set is expressed as:
Wherein a, b, c are related to cut-in wind speed and rated wind speed, and a, b, c are expressed as:
v a is the cut-in wind speed of the wind turbine, V b is the rated wind speed of the wind turbine, V c is the cut-out wind speed of the wind turbine, and P e is the rated power of the wind turbine.
As a preferable scheme of the reliability evaluation method of the distributed power distribution network based on the deep neural network, the invention comprises the following steps: the photovoltaic output model comprises an output model for obtaining a photovoltaic power generation system through simulated solar radiation intensity, the output of the photovoltaic power generation system is positively correlated with the solar radiation intensity, and the functional relation is expressed as follows:
P＝rAηK
Where r is solar radiation intensity, A is the area of the photovoltaic array, η is the conversion efficiency of the module to convert light energy into electrical energy, and K is the correction coefficient.
As a preferable scheme of the reliability evaluation method of the distributed power distribution network based on the deep neural network, the invention comprises the following steps: the maximum acceptable distributed power supply capacity comprises Zeno polyhedrons, zeno polyhedron is a special symmetric polyhedron, defined as the minkowski sum of finite segments, expressed as:
Wherein c epsilon R n is the center of Zeno polyhedron, g (i)∈Rn is the generation vector of Zeno polyhedron, and the uncertain output of the distributed power supply is represented by Zeno polyhedron.
As a preferable scheme of the reliability evaluation method of the distributed power distribution network based on the deep neural network, the invention comprises the following steps: the maximum acceptable distributed power supply capacity further comprises constructing an uncertain input set, and setting the predicted power of three distributed power supplies asThe output of each distributed power supply has uncertainty of +/-30%, and three directional line segments are obtained and expressed as:
g1＝[0.12 0 0]T
g3＝[0 0 0.15]T
A Zeno polyhedron was obtained, the power uncertainty set being a rectangular prism centered on [ 0.4.0.5.0.6 ] T, each of side lengths 0.24,0.18,0.3.
As a preferable scheme of the reliability evaluation method of the distributed power distribution network based on the deep neural network, the invention comprises the following steps: the improved monte carlo method includes obtaining, by mathematical sampling, a sampled version of the time to failure TTF and time to repair TTR of the component, expressed as:
R1 and R2 are random numbers in intervals (0 and 1), original parameters of a power distribution network are read and data are initialized, N power elements are arranged in a system, a group of random numbers delta 1,δ2,...,δn are generated, the number of the random numbers is the same as that of the elements in the power distribution network, the range of the random numbers is between (0 and 1), based on an element time sampling formula, the non-fault working time TTF of each element is calculated according to the generated random numbers, the non-fault working time is compared, the power element with the minimum non-fault working time in the fault element is determined, the trend of the system after the fault is calculated, the node voltage of each node after the fault is determined, whether the distributed power supply is offline is determined, the overvoltage influence range caused by the fault is determined if the distributed power supply is not offline, the running state of each load point is determined, the random running time of the system is calculated, the evaluation process is repeated if the running time of the system is smaller than a simulation period, the fault times and the fault time of each load point in the simulation period are calculated, and the reliability index of the system is calculated if the running time of the system is larger than or equal to the simulation period.
As a preferable scheme of the reliability evaluation method of the distributed power distribution network based on the deep neural network, the invention comprises the following steps: the method comprises the steps of training a neural network and predicting the BP neural network, selecting the capacity of each distributed power supply as an input parameter of the BP neural network, selecting a system average outage frequency index, a system average outage duration index and an expected outage amount as the output of the BP neural network, for input characteristics, enabling all input parameters to be in the same competition status, reducing calculation errors caused by the data, firstly, establishing an output model of the distributed power supply, determining a possible output range of the distributed power supply at each moment according to the uncertainty modeling based on Zeno polyhedrons, obtaining the maximum accessible capacity of the distributed power supply through power flow calculation according to the requirement of electric energy quality, obtaining an original dataset for training through a Monte Carlo simulation process, training the neural network model by using a normalized sample after determining the input and output parameters, obtaining a trained neural network model, and predicting by using the trained neural network model.
Another object of the present invention is to provide a distributed power distribution network reliability evaluation system based on a deep neural network, which can train the neural network and perform reliability prediction through a training prediction module, so as to solve the problem of low efficiency when dealing with large-scale and complex power distribution networks at present.
As a preferred scheme of the distributed power distribution network reliability evaluation system based on the deep neural network, the invention comprises the following steps: the system comprises a model building module, a power capacity module and a training prediction module; the model building module is used for obtaining wind speed data and building a wind turbine generator set output model and a photovoltaic output model; the power capacity module is used for distributing a maximum acceptable distributed power capacity to each node in the power distribution network; the training prediction module is used for obtaining an original data set through an improved Monte Carlo method, training a neural network and predicting.
A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that execution of the computer program by the processor is the step of implementing a method for evaluating reliability of a distributed power distribution network based on a deep neural network.
A computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of a method for evaluating reliability of a distributed power distribution network based on a deep neural network.
The invention has the beneficial effects that: according to the distributed power distribution network reliability assessment method based on the deep neural network, the influence of environmental factors is considered through the output models of the wind turbine generator system and the photovoltaic power generation system, the practical applicability of the models is enhanced, the reliability index is calculated by adopting the improved Monte Carlo method, the calculation efficiency and accuracy are improved, the reliability index is rapidly predicted through the BP neural network, the analysis time is remarkably shortened, the real-time assessment of the reliability of the large-scale distributed power supply access to the power distribution network is realized, bad weather conditions are effectively met, the reliability assessment efficiency and accuracy of the power distribution network are improved, the safe and stable operation of the power grid is promoted, the access and management of the distributed power supply are optimized, and the method has better effects in the analysis time, the assessment accuracy and the assessment efficiency.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without the need of creative efforts for a person of ordinary skill in the art. Wherein:
fig. 1 is an overall flowchart of a distributed power distribution network reliability evaluation method based on a deep neural network according to a first embodiment of the present invention.
Fig. 2 is a simulation flow chart of an improved monte carlo method of the distributed power distribution network reliability evaluation method based on the deep neural network according to the first embodiment of the present invention.
Fig. 3 is a block diagram of an example of a power distribution system according to a method for evaluating reliability of a distributed power distribution network based on a deep neural network according to a second embodiment of the present invention.
Fig. 4 is a graph of influence of single DG access on a power distribution network according to a method for evaluating reliability of a distributed power distribution network based on a deep neural network according to a second embodiment of the present invention.
Fig. 5 is a graph of the impact of line fault No. 22 on DG access points in the reliability evaluation method for a distributed power distribution network based on a deep neural network according to a second embodiment of the present invention.
Fig. 6 is a graph of DG capacity versus reliability of a distributed power distribution network based on a deep neural network reliability evaluation method according to a second embodiment of the present invention.
Fig. 7 is an overall flowchart of a distributed power distribution network reliability evaluation system based on a deep neural network according to a third embodiment of the present invention.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Example 1
Referring to fig. 1-2, for one embodiment of the present invention, a method for evaluating reliability of a distributed power distribution network based on a deep neural network is provided, including:
S1: and acquiring wind speed data, and constructing a wind turbine generator output model and a photovoltaic output model.
Further, the wind turbine generator output model comprises the steps of modeling and predicting wind speed by adopting an ARMA model.
It should be noted that, based on the original wind speed data, the standard deviation of the wind speed average value and the wind speed distribution is calculated, and the original wind speed data is processed to obtain standardized data, and the relationship between the standardized data and the original wind speed data is expressed as:
Wherein y t is standardized data, V t is raw wind speed data, μ is an average value of wind speeds, σ is a standard deviation of wind speed distribution, and the ARMA model is expressed as:
yt＝a1yt-1+a2yt-2+…+amyt-m+
αt+c1αt-1+c2αt-2+…cnαt-n
Wherein a i is an autoregressive coefficient, c i is a moving average coefficient, alpha t is normally distributed white noise, and the relation between the output and wind speed of the wind turbine generator set is expressed as:
Wherein a, b, c are related to cut-in wind speed and rated wind speed, and a, b, c are expressed as:
v a is the cut-in wind speed of the wind turbine, V b is the rated wind speed of the wind turbine, V c is the cut-out wind speed of the wind turbine, and P e is the rated power of the wind turbine.
Further, the photovoltaic output model comprises an output model of the photovoltaic power generation system obtained through the simulated solar radiation intensity.
It should be noted that, the output of the photovoltaic power generation system is positively correlated with the solar radiation intensity, and the functional relationship is expressed as follows:
P＝rAηK
Where r is solar radiation intensity, A is the area of the photovoltaic array, η is the conversion efficiency of the module to convert light energy into electrical energy, and K is the correction coefficient.
S2: each node in the distribution network is allocated a maximum acceptable distributed power supply capacity.
Still further, the maximum acceptable distributed power capacity comprises a Zeno polyhedron.
It should be noted that Zeno polyhedrons are a special symmetrical polyhedron, defined as minkowski sums of finite segments, expressed as:
Wherein c epsilon R n is the center of Zeno polyhedron, g (i)∈R n is the generation vector of Zeno polyhedron, and the uncertain output of the distributed power supply is represented by Zeno polyhedron.
Still further, the maximum acceptable distributed power supply capacity further includes constructing an uncertainty input set.
It should be noted that, let the predicted power of three distributed power sources beThe output of each distributed power supply has uncertainty of +/-30%, and three directional line segments are obtained and expressed as:
g1＝[0.12 0 0]T
g2＝[0 0.09 0]T
g3＝[0 0 0.15]T
A Zeno polyhedron was obtained, the power uncertainty set being a rectangular prism centered on [ 0.4.0.5.0.6 ] T, each of side lengths 0.24,0.18,0.3.
It should also be noted that, assuming that there is an uncertainty of ±30% per hour of the wind turbine output, an hour-by-hour wind turbine output model is obtained using the ARMA model, and assuming that the rated capacity of the wind turbine is 1MW, if at a certain moment the predicted power of the wind turbine is 60% of the rated capacity, then the possible output power is in the range of [0.42,0.78] MW, in which case it is necessary to ensure that the node voltage of the wind turbine access point meets the power quality requirement, i.e. the voltage deviation cannot exceed ±7%.
S3: the raw data set is obtained by a modified monte carlo method, the neural network is trained and predicted.
Still further, the improved monte carlo method includes sampling by mathematics.
It should be noted that a sampled version of the time to failure TTF and time to repair TTR of the element is obtained, expressed as:
R1 and R2 are random numbers in intervals (0 and 1), original parameters of a power distribution network are read and data are initialized, N power elements are arranged in a system, a group of random numbers delta 1,δ2,...,δn are generated, the number of the random numbers is the same as that of the elements in the power distribution network, the range of the random numbers is between (0 and 1), based on an element time sampling formula, the non-fault working time TTF of each element is calculated according to the generated random numbers, the non-fault working time is compared, the power element with the minimum non-fault working time in the fault element is determined, the trend of the system after the fault is calculated, the node voltage of each node after the fault is determined, whether the distributed power supply is offline is determined, the overvoltage influence range caused by the fault is determined if the distributed power supply is not offline, the running state of each load point is determined, the random running time of the system is calculated, the evaluation process is repeated if the running time of the system is smaller than a simulation period, the fault times and the fault time of each load point in the simulation period are calculated, and the reliability index of the system is calculated if the running time of the system is larger than or equal to the simulation period.
Still further, training the neural network and making predictions includes BP neural network.
It should be noted that, the capacity of each distributed power supply is selected as the input parameter of the BP neural network, the average power failure frequency index of the system, the average power failure duration index of the system and the expected power failure amount are selected as the output of the BP neural network, for the input characteristics, the data of the input parameters are in the same order of magnitude, all the input parameters are in the same competition status, the calculation error caused by the data is reduced, firstly, the output model of the distributed power supply is established, the possible output range of each moment of the distributed power supply is determined according to the uncertainty modeling based on Zeno polyhedrons, the accessible maximum capacity of the distributed power supply is obtained through the calculation of the power flow according to the requirement of the electric energy quality, the original dataset for training is obtained through the monte Carlo simulation process, after the input and output parameters are determined, the trained neural network model is obtained through the normalized sample set training, and the trained neural network model is used for prediction.
It should also be noted that the BP neural network is a multi-layer feedforward neural network, the weight adjustment adopts a back propagation algorithm, the firing function of the neurons is a sigmoid type function, and the output is a continuous value between [0,1 ].
It should also be noted that the learning process of the BP neural network may be divided into two main stages, a known training set is input, and the neurons of each layer are obtained by recursion layer by layer from the neurons of the first layer through the set network structure and the weights and thresholds set in the last iteration or initial setting; the influence of the weight and the threshold value of each neuron on the output of the last layer of neurons is calculated, namely the weight and the threshold value are modified by transferring gradients and errors layer by layer, and the learning process is repeated until the total error reaches the required error performance.
It should also be noted that for output parameters, if the magnitude of the phase difference between the output parameters is too large, then an output parameter of a larger magnitude will have a larger absolute error, and an output parameter of a smaller magnitude will have a smaller absolute error of the output data, so that the raw data must be normalized before the BP neural network is trained using the raw data.
Example 2
Referring to fig. 3-6, for one embodiment of the present invention, a method for evaluating reliability of a distributed power distribution network based on a deep neural network is provided, and in order to verify the beneficial effects of the present invention, scientific demonstration is performed through economic benefit calculation and simulation experiments.
An improved IEEE 33 node distribution system is adopted as a calculation example analysis of the reliability analysis of a distribution network, the system comprises 32 load nodes, 32 transformers, 5 circuit breakers and 3 distributed power supplies, and the structure of the calculation example is shown in figure 3.
As can be seen from fig. 4, when node 17 (LP 17) is connected to a distributed power supply, the voltage at each node in the distribution network changes, and the voltage at each node is increased by the switching-in of the distributed power supply, wherein the effect of the lifting of the branch end nodes is most obvious, and fig. 4 (b) is the maximum capacity of the distributed power supply that each node can accommodate under the power quality requirement.
When three distributed power supplies are connected in a grid mode, as each distributed power supply has a boosting function, the capacity of each distributed power supply cannot be simply calculated according to the maximum capacity of a single distributed power supply when the distributed power supplies are connected in the grid mode, the maximum accessible capacity of the 3 distributed power supplies is selected in various ways on the premise that the power quality of a line is met, in the invention, the maximum capacity of DG1 is 3.6MW, the maximum capacity of DG2 is 1.35MW, the maximum capacity of DG3 is 2.5MW, and the uncertainty of 30% is assumed in the invention.
As can be seen from fig. 6, when the line 22 fails, the possible ranges of node voltages of the node 17, the node 21 and the node 32 are the voltage ranges without failure, the solid line is the voltage range after failure, and the occurrence of failure may cause the node voltage to rise, which may cause the generation of overvoltage, in which case the distributed power supply needs to be disconnected from the power grid, the overvoltage node needs to be cut off, and the system reliability will be affected.
As can be seen from fig. 7, when the capacities of the distributed power sources are different, the reliability of the distribution network changes, in general, as the access capacity of the distributed power sources is closer to the maximum allowable capacity, the reliability of the distribution network is lower, when the access capacity is not more than 50% of the maximum accessible capacity, the reliability of the distribution network is not significantly affected by the change of the capacity of the distributed power sources, and when the access capacity is close to the maximum accessible capacity, the reliability of the distribution network is significantly affected by the change of the capacity of the distributed power sources.
When the capacities of the distributed power sources are different, the reliability index value of the power distribution network is quantitatively calculated, the capacity of the distributed power sources is represented by the percentage of the maximum accessible capacity, when the capacity of the distributed power sources is changed, the reliability index calculation result of the power distribution network is shown in table 1, in order to realize the prediction of the reliability index of the power distribution network by the BP neural network, firstly, the original data obtained from a simulation experiment is required to be normalized so as to obtain a data set for training the BP neural network, finally, the trained BP neural network model is utilized for prediction, and the accuracy and the prediction error of the prediction are compared and analyzed, wherein part of the original data is shown in table 1, and the real data, the prediction result and the error are shown in table 2.
The BP neural network model mainly comprises network layer number design, namely the number of hidden layers, the number of hidden layer nodes, an activation function and a training method, wherein a single hidden layer can finish nonlinear mapping by increasing the number of nodes on the hidden layer, but increasing the hidden layer number can lead to the reduction of an analyzable network scale, the number of hidden layer nodes can not be obtained through an analysis method, the approximate range of the hidden layer nodes can only be obtained through an empirical formula, and then the optimal node number is obtained through continuous experiments. According to the result in the table, the reliability index value of the power distribution network predicted by the BP neural network is basically consistent with the reliability index value of the power distribution network directly obtained by the improved Monte Carlo method, the error is smaller, meanwhile, in the calculation example, when the reliability index is calculated by adopting the improved Monte Carlo method, the calculation time and cost of the method are greatly increased, the running time is about 29.56 seconds, when the reliability index is calculated by adopting the trained BP neural network, only about 0.06 seconds is needed, the rapidity of the reliability analysis of the power distribution network based on the BP neural network is reflected, when the power distribution network structure is complex, more time is needed for calculating the reliability index of the power distribution network by using the Monte Carlo method, at the moment, more time can be saved by calculating the reliability index by utilizing the BP neural network, and the superiority of an artificial intelligent algorithm is better embodied.
TABLE 1 reliability index Table for different distributed Capacity distribution networks
DG capacity DG2 capacity DG3 capacity SAIFI SAIDI EENS
20％ 40％ 40％ 1.0239 5.8677 20.4164
30％ 50％ 80％ 1.0310 5.9283 20.9339
40％ 70％ 80％ 1.0327 5.9232 20.8973
50％ 60％ 50％ 1.0279 5.9078 20.5910
50％ 80％ 80％ 1.0415 5.9999 21.0698
60％ 50％ 50％ 1.0310 5.9299 20.6954
70％ 80％ 80％ 1.0554 6.1563 21.6529
80％ 60％ 80％ 1.0676 6.2851 22.1225
100％ 100％ 100％ 1.1344 6.9246 24.6625
TABLE 2 prediction results and error Table for BP neural network
Real data Prediction Error of Real data Prediction Error of
SAIFI 1.066 1.068 0.21％ 1.081 1.091 0.92％
SAIDI 6.276 6.298 0.36％ 6.414 6.527 1.77％
EENS 22.325 22.151 0.78％ 22.987 23.060 0.32％
Example 3
Referring to fig. 7, for one embodiment of the present invention, there is provided a distributed power distribution network reliability evaluation system based on a deep neural network, including: the system comprises a model construction module, a power capacity module and a training prediction module.
The model building module is used for obtaining wind speed data and building a wind turbine generator set output model and a photovoltaic output model; the power capacity module is used for distributing a maximum acceptable distributed power capacity to each node in the power distribution network; the training prediction module is used for obtaining an original data set through the improved Monte Carlo method, training the neural network and predicting.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method of the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium may even be paper or other suitable medium upon which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like. It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.

Claims (10)

1.基于深度神经网络的分布式配电网可靠性评估方法，其特征在于，包括：1. A distributed distribution network reliability assessment method based on deep neural network, characterized by comprising: 获取风速数据，构建风电机组出力模型和光伏出力模型；Obtain wind speed data and build wind turbine output model and photovoltaic output model; 为配电网中每个节点分配一个最大可接受分布式电源容量；Assign a maximum acceptable distributed generation capacity to each node in the distribution network; 通过改进的蒙特卡罗方法获得原始数据集，训练神经网络并进行预测。The original data set is obtained through the improved Monte Carlo method, the neural network is trained and prediction is performed. 2.如权利要求1所述的基于深度神经网络的分布式配电网可靠性评估方法，其特征在于：所述风电机组出力模型包括采用ARMA模型对风速进行建模和预测，基于原始风速数据计算风速平均值和风速分布的标准差，并对原始风速数据进行处理，得到标准化的数据，标准化数据与原始风速数据间的关系表示为：2. The distributed distribution network reliability assessment method based on deep neural network according to claim 1 is characterized in that: the wind turbine output model includes using ARMA model to model and predict wind speed, calculating the wind speed average and the standard deviation of wind speed distribution based on original wind speed data, and processing the original wind speed data to obtain standardized data, and the relationship between the standardized data and the original wind speed data is expressed as: 其中，yt为标准化的数据，Vt为原始风速数据，μ为风速的平均值，σ为风速分布的标准差，ARMA模型表示为：Among them, yt is the standardized data, Vt is the original wind speed data, μ is the mean value of wind speed, σ is the standard deviation of wind speed distribution, and the ARMA model is expressed as: yt＝a1yt-1+a2yt-2+…+amyt-m+y t =a 1 y t-1 +a 2 y t-2 +…+a m y tm + αt+c1αt-1+c2αt-2+…cnαt-n α t +c 1 α t-1 +c 2 α t-2 +…c n α tn 其中，ai为自回归系数，ci为移动平均系数，αt为正态分布的白噪声，风电机组的出力与风速函数关系表示为：Among them, a i is the autoregressive coefficient, c i is the moving average coefficient, α t is the normally distributed white noise, and the functional relationship between the output of the wind turbine and the wind speed is expressed as: 其中，a、b、c与切入风速和额定风速相关，a、b、c表示为：Among them, a, b, and c are related to the cut-in wind speed and the rated wind speed, and a, b, and c are expressed as: 其中，Va为风电机组的切入风速，Vb为风电机组的额定风速，Vc为风电机组的切出风速，Pe为风电机组的额定功率。Wherein, Va is the cut-in wind speed of the wind turbine, Vb is the rated wind speed of the wind turbine, Vc is the cut-out wind speed of the wind turbine, and Pe is the rated power of the wind turbine. 3.如权利要求2所述的基于深度神经网络的分布式配电网可靠性评估方法，其特征在于：所述光伏出力模型包括通过模拟的太阳辐射强度获得光伏发电系统的出力模型，光伏发电系统的出力与太阳辐射强度正相关，函数关系表示为：3. The distributed distribution network reliability assessment method based on deep neural network according to claim 2 is characterized in that: the photovoltaic output model includes an output model of the photovoltaic power generation system obtained by simulating the solar radiation intensity, and the output of the photovoltaic power generation system is positively correlated with the solar radiation intensity, and the functional relationship is expressed as: P＝rAηKP＝rAηK 其中，r为太阳辐射强度，A为光伏阵列的面积，η为模块将光能转换成电能的转换效率，K为修正系数。Among them, r is the solar radiation intensity, A is the area of the photovoltaic array, η is the conversion efficiency of the module to convert light energy into electrical energy, and K is the correction factor. 4.如权利要求3所述的基于深度神经网络的分布式配电网可靠性评估方法，其特征在于：所述最大可接受分布式电源容量包括Zeno多面体，Zeno多面体是一种特殊对称多面体，被定义为有限线段的闵可夫斯基和，表示为：4. The distributed distribution network reliability assessment method based on deep neural network according to claim 3 is characterized in that: the maximum acceptable distributed power generation capacity includes a Zeno polyhedron, which is a special symmetric polyhedron defined as the Minkowski sum of finite line segments, expressed as: 其中，c∈Rn为Zeno多面体的中心，g(i)∈Rn为Zeno多面体的生成向量，通过Zeno多面体表示分布式电源的不确定出力。Among them, c∈Rn is the center of the Zeno polyhedron, g (i) ∈Rn is the generating vector of the Zeno polyhedron, and the uncertain output of distributed generation is represented by the Zeno polyhedron. 5.如权利要求4所述的基于深度神经网络的分布式配电网可靠性评估方法，其特征在于：所述最大可接受分布式电源容量还包括构建不确定输入集，设三个分布式电源的预测功率为每个分布式电源的出力都有±30％的不确定性，得到三条有向线段表示为：5. The distributed distribution network reliability assessment method based on deep neural network according to claim 4 is characterized in that: the maximum acceptable distributed power supply capacity also includes constructing an uncertain input set, assuming that the predicted power of the three distributed power supplies is The output of each distributed power source has an uncertainty of ±30%, resulting in three directed line segments represented as: g1＝[0.12 0 0]T g 1 = [0.12 0 0] T g3＝[0 0 0.15]T g 3 =[0 0 0.15] T 得到一个Zeno多面体，电力不确定性集为一个以[0.4 0.5 0.6]T为中心，分别以0.24，0.18，0.3为边长的矩形棱柱。A Zeno polyhedron is obtained, and the power uncertainty set is a rectangular prism with [0.4 0.5 0.6] T as the center and side lengths of 0.24, 0.18, and 0.3 respectively. 6.如权利要求5所述的基于深度神经网络的分布式配电网可靠性评估方法，其特征在于：所述改进的蒙特卡罗方法包括通过数学抽样，获得元件的无故障工作时间TTF和修复时间TTR的抽样形式，表示为：6. The distributed distribution network reliability assessment method based on deep neural network according to claim 5, characterized in that: the improved Monte Carlo method includes obtaining the sampling form of the fault-free working time TTF and the repair time TTR of the component through mathematical sampling, expressed as: 其中，R1、R2分别为区间(0,1)内的随机数，首先读取配电网的原始参数并初始化数据，设系统中有N个功率元件，生成一组随机数δ1,δ2,...,δn，与配电网中的元件数量相同，随机数的范围在(0,1)之间，基于元件时间抽样公式，根据生成的随机数计算每个元件的无故障工作时间TTF，通过比较无故障工作时间，确定故障元件中最小无故障工作时间的电力元件，计算故障后系统的潮流，并确定故障后每个节点的节点电压，确定分布式电源是否离线，若分布式电源脱网，则确定故障引起的过电压影响范围，若分布式电源未脱网，确定每个负载点的运行状态，并计算系统的随机运行时间，若系统运行时间小于仿真周期，则重复评估过程，若系统运行时间大于或等于模拟周期，则计算模拟期内每个负载点的故障次数和故障时间，并计算系统可靠性指标。Among them, R1 and R2 are random numbers in the interval (0,1) respectively. First, the original parameters of the distribution network are read and the data is initialized. Assume that there are N power components in the system, and generate a set of random numbers δ 1 ,δ 2 ,...,δ n , which is the same as the number of components in the distribution network, and the range of random numbers is between (0,1). Based on the component time sampling formula, the fault-free working time TTF of each component is calculated according to the generated random numbers. By comparing the fault-free working time, the power component with the minimum fault-free working time among the faulty components is determined, the power flow of the system after the fault is calculated, and the node voltage of each node after the fault is determined. It is determined whether the distributed power source is offline. If the distributed power source is offline, the impact range of the overvoltage caused by the fault is determined. If the distributed power source is not offline, the operating status of each load point is determined, and the random operating time of the system is calculated. If the system operating time is less than the simulation period, the evaluation process is repeated. If the system operating time is greater than or equal to the simulation period, the number of failures and failure time of each load point during the simulation period are calculated, and the system reliability index is calculated. 7.如权利要求6所述的基于深度神经网络的分布式配电网可靠性评估方法，其特征在于：所述训练神经网络并进行预测包括BP神经网络，选择每个分布式电源的容量作为BP神经网络的输入参数，选择系统平均停电频率指标、系统平均停电持续时间指标和期望缺供电量作为BP神经网络的输出，对于输入特征，输入参数的数据处于相同数量级，使所有输入参数处于相同竞争地位，减少数据导致的计算误差，首先，建立分布式电源的出力模型，根据基于Zeno多面体的不确定性建模，确定分布式电源每一时刻的可能出力范围，根据电能质量的要求，通过潮流计算得到分布式电源可接入的最大容量，并通过蒙特卡罗模拟过程，获得用于训练的原始数据集，确定输入和输出参数后，使用归一化的样本集训练神经网络模型，获得训练好的神经网络模型，使用训练好的神经网络模型进行预测。7. The distributed distribution network reliability assessment method based on deep neural network as described in claim 6 is characterized in that: the training neural network and predicting include BP neural network, selecting the capacity of each distributed power source as the input parameter of the BP neural network, selecting the system average power outage frequency index, the system average power outage duration index and the expected power shortage as the output of the BP neural network, for the input characteristics, the data of the input parameters are in the same order of magnitude, so that all input parameters are in the same competitive position, reducing the calculation error caused by the data, first, establish a distributed power output model, and determine the possible output range of the distributed power source at each moment according to the uncertainty modeling based on Zeno polyhedron, according to the requirements of power quality, the maximum capacity that the distributed power source can access is obtained through flow calculation, and through the Monte Carlo simulation process, the original data set for training is obtained, after determining the input and output parameters, the neural network model is trained using the normalized sample set to obtain the trained neural network model, and the trained neural network model is used for prediction. 8.一种采用如权利要求1～7任一所述的基于深度神经网络的分布式配电网可靠性评估方法的系统，其特征在于：包括模型构建模块、电源容量模块、训练预测模块；8. A system using the distributed distribution network reliability assessment method based on deep neural network as claimed in any one of claims 1 to 7, characterized in that it comprises a model building module, a power capacity module, and a training prediction module; 所述模型构建模块用于获取风速数据，构建风电机组出力模型和光伏出力模型；The model building module is used to obtain wind speed data and build a wind turbine output model and a photovoltaic output model; 所述电源容量模块用于为配电网中每个节点分配一个最大可接受分布式电源容量；The power capacity module is used to allocate a maximum acceptable distributed power capacity to each node in the distribution network; 所述训练预测模块用于通过改进的蒙特卡罗方法获得原始数据集，训练神经网络并进行预测。The training prediction module is used to obtain the original data set through the improved Monte Carlo method, train the neural network and perform prediction. 9.一种计算机设备，包括存储器和处理器，所述存储器存储有计算机程序，其特征在于，所述处理器执行所述计算机程序时实现权利要求1至7中任一项所述的基于深度神经网络的分布式配电网可靠性评估方法的步骤。9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein when the processor executes the computer program, the steps of the distributed distribution network reliability assessment method based on a deep neural network as described in any one of claims 1 to 7 are implemented. 10.一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现权利要求1至7中任一项所述的基于深度神经网络的分布式配电网可靠性评估方法的步骤。10. A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the distributed distribution network reliability assessment method based on a deep neural network described in any one of claims 1 to 7 are implemented.
CN202410369522.2A 2024-03-28 2024-03-28 Distributed distribution network reliability assessment method and system based on deep neural network Pending CN118246326A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410369522.2A CN118246326A (en) 2024-03-28 2024-03-28 Distributed distribution network reliability assessment method and system based on deep neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410369522.2A CN118246326A (en) 2024-03-28 2024-03-28 Distributed distribution network reliability assessment method and system based on deep neural network

Publications (1)

Publication Number Publication Date
CN118246326A true CN118246326A (en) 2024-06-25

Family

ID=91564370

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410369522.2A Pending CN118246326A (en) 2024-03-28 2024-03-28 Distributed distribution network reliability assessment method and system based on deep neural network

Country Status (1)

Country Link
CN (1) CN118246326A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN120337003A (en) * 2025-03-28 2025-07-18 国网四川省电力公司营销服务中心 A voltage metering reliability assessment method in complex environments based on knowledge distillation

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN120337003A (en) * 2025-03-28 2025-07-18 国网四川省电力公司营销服务中心 A voltage metering reliability assessment method in complex environments based on knowledge distillation
CN120337003B (en) * 2025-03-28 2025-11-11 国网四川省电力公司营销服务中心 A Reliability Assessment Method for Voltage Measurement in Complex Environments Based on Knowledge Distillation

Similar Documents

Publication Publication Date Title
Li et al. An intelligent transient stability assessment framework with continual learning ability
Ghadimi et al. PSO based fuzzy stochastic long-term model for deployment of distributed energy resources in distribution systems with several objectives
CN112685961A (en) Method and system for predicting remaining service life of analog circuit
CN113659565B (en) An Online Prediction Method for Frequency Situation of New Energy Power System
CN119378733A (en) A method and system for optimizing power dispatching based on deep reinforcement learning
CN114548762A (en) Real-time power system cascading failure risk assessment method and system based on space-time diagram neural network
CN114759579A (en) Power grid active power optimization control system, method and medium based on data driving
CN119647898A (en) A power dispatch optimization method and system based on reinforcement learning
CN119129860A (en) Distributed photovoltaic power generation intelligent prediction system and method
Fang et al. A two-stage deep learning approach for solving microgrid economic dispatch
CN118246326A (en) Distributed distribution network reliability assessment method and system based on deep neural network
CN116093927B (en) A distribution network reliability assessment method and device taking into account source and load uncertainty
CN118199078A (en) A robust reinforcement learning reactive power optimization method for unobservable distribution networks
CN118504813A (en) Comprehensive evaluation method for reliability of power distribution network considering large-scale distributed power supply access
CN120879577B (en) Regional power grid prediction scheduling method and system based on topological balance
CN117239722A (en) A short-term prediction method of system wind load considering the influence of multiple loads
CN120728593B (en) Power load forecasting method based on dynamic correlation of industrial user groups
CN116702988A (en) Carbon neutral calculation cost optimization method and equipment for economic dispatch of carbon capture power plants in smart grid
CN119726938A (en) New energy consumption capacity evaluation method and system based on network security and stability constraints
CN119416076A (en) Distribution network topology identification method, device, equipment and medium based on ARAM graph convolution
CN119109127A (en) A reliability analysis method and system for distribution network considering large-scale distributed power access
CN113922369B (en) Method, terminal and storage medium for reliability assessment of transmission and distribution network
CN116307838A (en) A precise equivalence method, system and medium of distribution network in grid reliability assessment
CN113592565A (en) Method for predicting importance of nodes of power system
CN120746209B (en) Power grid resource allocation method and system based on generation type artificial intelligence

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