CN113471982B - Cloud edge cooperation and power grid privacy protection distributed power supply in-situ voltage control method - Google Patents
Cloud edge cooperation and power grid privacy protection distributed power supply in-situ voltage control method Download PDFInfo
- Publication number
- CN113471982B CN113471982B CN202110839904.3A CN202110839904A CN113471982B CN 113471982 B CN113471982 B CN 113471982B CN 202110839904 A CN202110839904 A CN 202110839904A CN 113471982 B CN113471982 B CN 113471982B
- Authority
- CN
- China
- Prior art keywords
- distribution network
- power distribution
- training
- sub
- power supply
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- 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
- H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
- H02J3/16—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load by adjustment of reactive power
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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
- H02J2300/22—The renewable source being solar energy
- H02J2300/24—The renewable source being solar energy of photovoltaic origin
-
- 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
- H02J2300/28—The renewable source being wind energy
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
- Y02E10/56—Power conversion systems, e.g. maximum power point trackers
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A cloud edge cooperation and power grid privacy protection distributed power supply local voltage control method comprises the following steps: dividing the region under the jurisdiction of each edge computing device of the active power distribution network and inputting parameters of the power distribution network; building a training set of a power distribution network proxy model based on a graph convolution neural network at the cloud, building the power distribution network proxy model based on the graph convolution neural network, and obtaining the power distribution network proxy model after training; the trained power distribution network agent model is sent to an edge computing device on the edge side of the distribution network to serve as a power distribution network sub-agent model; setting local voltage control curve parameters of the distributed power supplies in each region according to each distribution network sub-agent model; updating the training objects to form an edge side sub-training set; updating to obtain an updated power distribution network sub-agent model; and re-setting the local voltage control curve parameters of the distributed power supplies in each region, and outputting the local voltage control curve setting result. The method fully excavates massive multi-source data information and improves the intelligent level of operation regulation and control of the power distribution network.
Description
Technical Field
The invention relates to a distributed power supply local voltage control method. In particular to a distributed power supply local voltage control method based on cloud edge cooperation and power grid privacy protection.
Background
The large-scale and high-proportion access of Distributed Generators (DGs) realizes the low carbon of the energy supply of the distribution network, but also causes the problem of voltage out-of-limit of the distribution network, and makes the optimization and control mode of the distribution network more complicated. Meanwhile, with the fact that large power distribution and power distribution data information is more and more numerous, data of multiple benefit subjects on the distribution network side exist in an isolated island mode, and the data concentration faces many problems of communication pressure, privacy disclosure and the like, so that the revolution of a power distribution network operation management mechanism is accelerated.
By scheduling various reactive devices, voltage violations can be mitigated. Fast voltage control is difficult because conventional regulation devices (e.g., on-load tap changers and capacitor banks) have slow response speeds and voltage cannot be continuously regulated. The remaining capacity of the distributed power inverter can be used for continuous regulation of the voltage, so that there is great potential to achieve fast voltage control using a DG interface inverter.
For a power distribution network with a distributed power supply with high penetration rate access, a centralized control mode is generally adopted to uniformly allocate controllable resources to realize global optimization of the system, but the method has large measurement data volume and heavy communication burden. In order to reduce the communication data volume, the distributed control mode realizes the integral optimization of the system operation by reasonably partitioning the power distribution network and interacting boundary information between adjacent areas, but in the method, a large amount of information communication is still required between area controllers, and a final optimization scheme can be determined by a plurality of iterative processes. Compared with a centralized control mode and a distributed control mode, the local control mode can complete the control of the local adjustable resources by only utilizing local information, and has the advantages of high response speed, low investment cost and small communication data volume. The Q-V curve is used for carrying out local control, so that quick response to uncertain output of the distributed power supply can be realized, and the robustness is good. Most of the current researches set the Q-V curve in a mode of solving an optimization model based on the whole network parameters and accurate prediction data, and accurate network parameters are difficult to obtain in the actual operation of a power distribution network. Meanwhile, the running state of the power distribution network changes along with the change of the output of the distributed power supply, so that the anti-interference capability of the model is poor, and the advantage of local control is reduced to a certain extent.
The rapid development of information and communication technologies has increased the level of digitization of power distribution networks. A large amount of heterogeneous operating data can be measured and acquired by using a data acquisition and monitoring device of the power distribution network. How to fully mine massive multi-source data information becomes the key for improving the intelligent level of operation regulation and control of the power distribution network. Data-driven technology is attracting attention as a method for achieving optimal control based on only historical data without requiring an accurate model. The artificial intelligence is used as a branch of the data driving technology to analyze and learn the collected historical data, and valuable information is extracted from the historical data and used for guiding the operation of the power distribution network. The Graph Convolutional Neural network (GCN) is a new multi-layer image data processing framework capable of processing non-european space structures, and in recent years, has many application cases in many fields such as computer vision, chemical engineering and the like.
In the power distribution network, each distributed power source may belong to different owners, so that the distributed power sources can not obtain data of the whole power distribution network, nor can the distributed power sources obtain private data of other distributed power sources. Model-based in-place control inevitably causes the problem of user privacy disclosure due to the need to collect information of each user. Therefore, a method for quickly setting the local control curve of the distributed power supply of the active power distribution network under the premise of considering user data privacy is needed.
Disclosure of Invention
The invention aims to solve the technical problem of providing a distributed power supply on-site voltage control method for cloud-edge cooperation and power grid privacy protection, which can fully mine massive multi-source data information and improve the intelligent level of power distribution network operation regulation.
The technical scheme adopted by the invention is as follows: a distributed power supply local voltage control method for cloud edge cooperation and power grid privacy protection comprises the following steps:
1) dividing the governed area of each edge computing device according to the selected active power distribution network, and inputting power distribution network parameters including network topology connection relation of the power distribution network, distributed power supply access position, capacity and parameters; input distribution network NdHistorical power flow data of a typical day, and a predicted sunrise force curve of the distributed power supply; setting a typical topology total number NsThe total sampling duration delta T and the sampling time interval delta T of the training object; setting parameters of a power distribution network proxy model based on a graph convolution neural network, wherein the parameters comprise a learning rate gamma, a regularization coefficient beta and times E of completely traversing a training set;
2) according to the parameters of the power distribution network and N in the step 1)dHistorical load flow data of a typical day, constructing a training set of a power distribution network agent model based on a graph convolution neural network at a cloud end, and determining the number of training objects in the training set, the composition of each training object and preprocessing the training objects; constructing a power distribution network proxy model based on a graph convolution neural network at the cloud end, and finishing the training of the power distribution network proxy model based on the graph convolution neural network to obtain the trained power distribution network proxy model;
3) The trained power distribution network agent model is sent to edge computing devices on the edge side of the distribution network to serve as power distribution network sub-agent models of all the edge computing devices;
4) according to each distribution network subagent model, on the basis of a predicted distributed power supply daily output curve and daily load output curve and voltage and net load power measurement information in the area governed by each edge computing device, on the edge side of a distribution network, setting local voltage control curve parameters of the distributed power supply in each area by taking the minimum node voltage deviation in the area governed by each edge computing device as a target function;
5) updating the training objects based on the sub-agent model of the power distribution network to form an edge side sub-training set according to the local voltage control curves of the distributed power supplies in the regions obtained in the step 4); respectively updating the weight coefficients of the sub-agent models of the distribution network at each edge side by using a federal learning framework to obtain updated sub-agent models of the distribution network;
6) and (5) setting the local voltage control curve parameters of the distributed power supply in each region again according to the updated power distribution network sub-agent model in the step 5), and outputting a local voltage control curve setting result.
The invention discloses a distributed power supply on-site voltage control method based on cloud edge cooperation and power grid privacy protection, which aims to solve the setting problem of a distributed power supply on-site control curve for protecting user data privacy, fully considers the user data privacy, constructs a power distribution network proxy model for simulating the dynamic characteristics of a power distribution network, sets the on-site control curve parameters of the distributed power supply by taking the minimum voltage deviation of an active power distribution network as a target function, fuses and updates a sub-proxy model of the power distribution network at the edge side by using a federal learning framework, updates the on-site control curve parameters of the distributed power supply, and obtains a distributed power supply on-site voltage reactive power control strategy. The method fully excavates massive multi-source data information, improves the intelligent level of operation regulation and control of the power distribution network, effectively solves the problem of data privacy disclosure in the Q-V curve setting process based on the cloud-edge cooperative process of the Federal learning framework, improves the on-site setting effect through updating of the proxy model, and protects data and user behaviors of the power distribution network.
Drawings
FIG. 1 is a flow chart of a distributed power supply in-place voltage control method of cloud-edge coordination and power grid privacy protection of the present invention;
FIG. 2 is a modified IEEE 33 node distribution network example structure diagram;
FIG. 3 is a predicted distributed power, load out curve;
FIG. 4 is a comparison graph of the distribution of extreme voltage values of the system according to scheme I and scheme III;
FIG. 5 is a comparison graph of the distribution of extreme voltage values for the systems of case II and case III;
FIG. 6 is a comparison graph of the distribution of extreme values of system voltages for case III and case IV.
Detailed Description
The invention provides a distributed power supply local voltage control method for cloud edge coordination and power grid privacy protection, which is described in detail below with reference to embodiments and drawings.
The invention discloses a cloud-edge cooperation and power grid privacy protection distributed power supply local voltage control method, which comprises the following steps of:
1) dividing the region under the jurisdiction of each edge computing device according to the selected active power distribution network, and inputting parameters of the power distribution network, including the network topology connection relation of the power distribution network, the access positions, the capacities and the parameters of the distributed power supplies; input distribution network NdHistorical power flow data of a typical day, and a predicted sunrise output curve and a predicted sunrise load output curve of the distributed power supply; setting a typical topological total number N sThe total sampling duration delta T and the sampling time interval delta T of the training object; setting parameters of a power distribution network proxy model based on a graph convolution neural network, wherein the parameters comprise a learning rate gamma, a regularization coefficient beta and times E of completely traversing a training set;
(1) the distribution network NdHistorical trend data for a typical day, is: in the distribution network NdIn the course of a typical day,a typical day distributed power supply does not participate in reactive power regulation,a typical daily distributed power supply is given as [1.0,1.0 ]]Adjusting reactive power for the dead zone base curve to obtain distribution network NdHistorical trend data for a typical day.
For this embodiment, the modified IEEE 33 node power distribution network example structure is shown in fig. 2, and the topological connection relationship of the power distribution network, the access positions, capacities, and parameters of the distributed power sources, the predicted distributed power sources, the load output curves, and the historical power flow data of the power distribution network are input, and the detailed parameters are shown in tables 1 and 2. Node 9, node 10, node 11, node 18, node 20, node 21, node 23, node 24, node 25, node 31, node 32 andthe nodes 33 are respectively connected to a group of photovoltaic systems, and the capacity of each photovoltaic system is 0.18 MVA; the node 15, the node 16, the node 17, the node 22, the node 29 and the node 30 are respectively connected to a group of fans, and the capacities are all 0.36 MVA; the predicted distributed power supply and load output curve is shown in fig. 3; the upper and lower safe operation limits of the voltage amplitude (per unit value) of each node are respectively 1.10 and 0.90; setting typical number of days N dTotal number of topologies N200 s2, the total sampling time duration Δ T of the training object is 24h, and the sampling time interval Δ T is 5 min.
TABLE 1 IEEE 33 node sample load Access location and Power
TABLE 2 IEEE 33 node example line parameters
2) According to the parameters of the power distribution network in the step 1) and NdHistorical load flow data of a typical day, constructing a training set of a power distribution network agent model based on a graph convolution neural network at a cloud end, and determining the number of training objects in the training set, the composition of each training object and preprocessing the training objects; and constructing a power distribution network proxy model based on the graph convolution neural network at the cloud end, and finishing the training of the power distribution network proxy model based on the graph convolution neural network to obtain the trained power distribution network proxy model.
(1) The number of the training objects in the training set is as follows: based on distribution network NdHistorical trend data of a typical day, and considering NsA typical topological structure is used for constructing a training set of a power distribution network agent model based on a graph convolution neural network, and the training set comprisesAnd (4) training the subject.
(2) The composition of each training object comprises:
each training object consists of three matrixes, namely a characteristic information matrix X, a topological information matrix A and a label matrix Z, and the label matrix Z is fitted by taking the characteristic information matrix X and the topological information matrix A as input based on a power distribution network agent model of the graph convolution neural network.
The feature information matrix X represents the input features of the training objects, and the feature information matrix X of the nth training objectnThe system consists of net load active power and net load reactive power injected by a sampling moment node and a node voltage value of a power distribution network, and is expressed as follows:
in the formula (I), the compound is shown in the specification,a column vector representing the net load active power injected by each node at the sampling instant,a column vector representing the net load reactive power injected by each node at the sampling instant,and the column vector is formed by the node voltage values of the power distribution network at the sampling moment, and N is the number of the nodes of the power distribution network.
The topology information matrix A represents the connection relationship between the nodes, and the topology information matrix An of the nth training object is represented as:
in the formula (I), the compound is shown in the specification,topological information matrix A for the n-th training objectnRow i and column j elements of (1);
the label matrix Z represents the label values of the output features of the training objects, and the label matrix Z of the nth training objectnThe node voltage control value at the sampling moment is represented as:
in the formula (I), the compound is shown in the specification,representing the label value of the output characteristic of the ith node of the nth training object for the ith row element of the matrix; and N is the number of nodes of the power distribution network.
(3) The preprocessing of the training objects is to preprocess each element in a characteristic information matrix X and a label matrix Z of the training objects in a training set before model training; wherein
Feature information matrix X for the nth training objectnCarrying out pretreatment, wherein the pretreatment formula is as follows:
in the formula (I), the compound is shown in the specification,characteristic information matrix X of n training objects before preprocessingnThe kth input feature of the ith node,the characteristic information matrix X of the n-th training object after preprocessingnThe kth input feature of the ith node,is the mean value of all elements of the kth input feature in the feature information matrix X of the training object,the variance of all elements of the kth input feature in a feature information matrix X of a training object;
label matrix Z for nth training objectnCarrying out pretreatment, wherein the pretreatment formula is as follows:
in the formula (I), the compound is shown in the specification,label matrix Z for n training object before preprocessingnThe ith node in the tree outputs the tag value of the feature,label matrix Z for n training object after pretreatmentnThe tag value of the output characteristic of the ith node, muZIs the mean value, delta, of all elements of the output features in the label matrix Z of the training objectZAnd outputting the variance of all elements of the features in the label matrix Z of the training object.
(4) The distribution network agent model based on the graph convolution neural network is represented as follows:
in the formula, H(l+1)For power distribution networkOutputting a (l + 1) th hidden layer of the physical model; h (l)Is the output of the l layer hidden layer; a is a topological information matrix and B is a topological information matrix,IMis an M-order identity matrix;the array is a diagonal array and the array is a diagonal array,W(l)is the weight matrix of the l layer; y is an output matrix of the graph convolution neural network; h(k-1)The output of the last hidden layer; w(k)Is the weight of the output layer; σ (-) is a nonlinear activation function, expressed as:
σ(x)=LeakyReLU(x,α)=max(0,x)+α×min(0,x) (9)
wherein α is a negative slope; f (-) is the output layer function, F (-) 1 (-);
the power distribution network agent model based on the graph convolution neural network takes the minimum loss function as a target function and a batch gradient descent algorithm as a training mode, and the training process is expressed as follows:
in the formula (I), the compound is shown in the specification,the method is a loss function of a power distribution network proxy model based on a graph convolution neural network when a weight coefficient is omega;output matrix Y for the nth training objectnOutputting a fitting value of the characteristic by the ith node;label matrix Z for n training object after pretreatmentnThe ith node outputs the label value of the characteristic; n is a radical ofTThe number of the training objects is N, and the number of the nodes of the power distribution network is N; omega is the current weight coefficient of the distribution network proxy model based on the graph convolution neural network, and omegalThe first weight coefficient of the distribution network agent model based on the graph convolution neural network is gamma, and the gamma is the learning rate.
3) And issuing the trained power distribution network agent model to the edge computing devices at the edge side of the distribution network to serve as the power distribution network sub-agent models of all the edge computing devices.
4) According to each distribution network subagent model, on the basis of a predicted distributed power supply daily output curve and daily load output curve and voltage and net load power measurement information in the area governed by each edge computing device, on the edge side of a distribution network, setting local voltage control curve parameters of the distributed power supply in each area by taking the minimum node voltage deviation in the area governed by each edge computing device as a target function;
(1) the local voltage control curve of the distributed power supply is obtained by adopting the following formula:
in the formula, Vt,iThe voltage amplitude of node i, g (V), for a period of tt,i) For the expression of the distributed power supply in-situ voltage reactive power control strategy, g (V)t,i) With dead zone of regulation [ V ]i q,min,Vi q,max],Vi q,minAnd Vi q,maxThe reactive power control method is characterized in that the reactive power control method is a local voltage control curve parameter of the distributed power supply, and the reactive power generated by the distributed power supply in the regulation dead zone is 0.
(2) The minimum node voltage deviation in the area governed by each edge computing device is taken as an objective function and is expressed as follows:
Wherein f is the node voltage deviation, NtIs the number of time slices, NaCalculating the total number of nodes in the area covered by the device for the a-th edge; v t,iThe voltage amplitude of the node i in the period t;is the maximum voltage threshold value of the voltage of the power supply,as minimum voltage threshold, when Vt,iOut of desired voltage intervalThe objective function is used to reduce the voltage deviation.
(3) The setting of the local voltage control curve parameters of the distributed power supply in each area comprises the following steps:
(4.1) selecting upper and lower limits V of voltage regulation dead zone of distributed power supplyi q,minAnd Vi q,maxIs [0.9,1.1 ]]And the constraint conditions of the upper limit and the lower limit of the dead zone are expressed as follows:
(4.2) based on the upper and lower limits V of the dead zonei q,minAnd Vi q,maxAnd power distribution network measurement information in the area, and updating the characteristic information matrix X to obtain an updated characteristic information matrix X';
(4.3) inputting the updated characteristic information matrix X' into a power distribution network subagent model to obtain system voltage distribution in an area corresponding to the upper limit and the lower limit of the dead zone, and calculating the node voltage deviation, wherein the formula is as follows:
In the formula, NtIs the number of time slices, NaCalculating the total number of nodes in the area covered by the device for the a-th edge; vt,iThe voltage amplitude of the node i is t time period;in order to be the maximum voltage threshold value,at minimum voltage threshold, when Vt,iOut of desired voltage intervalThe objective function is used to reduce the voltage deviation;
and (4.4) scanning each dead zone upper and lower limit combination, and outputting the dead zone upper and lower limit combination with the minimum node voltage deviation as an optimal solution, namely the local voltage control curve parameters of the distributed power supply in the region.
5) Updating training objects based on the sub-agent model of the power distribution network to form an edge side sub-training set according to the local voltage control curves of the distributed power sources in the regions obtained in the step 4); respectively updating the weight coefficient of each edge side power distribution network sub-agent model by using a federal learning frame to obtain an updated power distribution network sub-agent model;
(1) the method for updating the training objects to form the edge side sub-training set based on the sub-agent model of the power distribution network comprises the following steps:
in-situ voltage control curve dead zone upper and lower limits V of distributed power supply in each zone based on settingi q,min/Vi q ,maxAnd measuring information of the distribution network in the region, and updating the characteristic information matrix X' to obtainTo updated feature information matrix X'm(ii) a The updated feature information matrix X'mInputting a distribution network sub-agent model to obtain system voltage distribution corresponding to the upper and lower limits of the dead zone, updating a tag matrix Z, and updating the updated tag matrix Z'm。
(2) The method for updating the weight coefficients of the subagent models of the distribution networks at the edge sides by using the federal learning framework respectively comprises the following steps:
(5.1) calculating the average loss of each edge side distribution network subagent model in the b-th iteration in the updating process of the weight coefficient by the following formula;
in the formula (f) g(omega) is an objective function of the g-th power distribution network sub-agent model, omega is a weight coefficient of the model,the sub-agent model of the distribution network corresponds to the n training object (x) when the weight coefficient is omegan,yn) The loss function of (a) is calculated,the number of training objects in the sub-training set of the g-th sub-agent model of the distribution network, DgA sub-training set of the g-th sub-agent model of the power distribution network is set; f (ω) is the overall objective function of the Federal learning framework, NGIs the total number of distributed power sources, NCThe sum of the number of the training objects in each sub-training set;
(5.2) each edge side power distribution network sub-agent model takes the minimum total objective function of the federal learning frame as a target, an edge side sub-training set is used for training, and the weight coefficient of each edge side power distribution network sub-agent model is adjusted to obtain each power distribution network sub-agent model before communication;
(5.3) uploading the updated self weight coefficients by each power distribution network sub-agent model through the following formula, carrying out weighted average on the updated self weight coefficients of each edge side power distribution network sub-agent model by the cloud, and issuing the weighted average weight coefficients to the edge side to finish the b-th iteration;
in the formula, Wg,bIs the weight coefficient, W, of the g-th sub-agent model of the distribution network before the b-th iteration b+1Weight coefficient, W, of the trained power distribution network proxy model of the cloud after the b-th iterationg,b+1And the weight coefficient is the weighted average of the g-th power distribution network sub-agent model after the b-th iteration.
6) And (5) re-setting the local voltage control curve parameters of the distributed power supplies in each region according to the updated power distribution network sub-agent model in the step 5), and outputting the local voltage control curve setting result.
In order to fully verify the advancement of the distributed power supply local voltage control method for cloud-edge cooperation and power grid privacy protection, in the embodiment, the following three schemes are adopted for comparative analysis:
scheme I: the reactive power output of the distributed power supply is not optimized, and the initial running state of the power distribution network is obtained;
scheme II: the local voltage control curve of the distributed power supply is set in a partitioning mode only by means of intra-area measurement information, and the output of the distributed power supply is optimized;
scheme III: by adopting the method, the local voltage control curve of the distributed power supply is set in a partitioning manner based on the intra-area measurement information and the interval cooperation method, and the output of the distributed power supply is optimized;
scheme IV: and the local voltage control curve of the distributed power supply is set in a partitioning manner by adopting a centralized optimization method, and the output of the distributed power supply is optimized.
The comparison of the optimization results of the scheme I, the scheme II, the scheme III and the scheme IV is shown in the table 3, the comparison of the distribution of the extreme system voltages of the scheme I and the scheme III is shown in the figure 4, the comparison of the distribution of the extreme system voltages of the scheme II and the scheme III is shown in the figure 5, and the comparison of the distribution of the extreme system voltages of the scheme III and the scheme IV is shown in the figure 6.
TABLE 3 comparison of simulation results under different control strategies
The computer hardware environment for executing the optimization calculation is Intel (R) core (TM) i7-9750H CPU, the dominant frequency is 2.60GHz, and the memory is 16 GB; the software environment is a Windows 10 operating system.
It can be seen from the comparison between schemes I and II that the connection of the distributed power supply causes the system voltage to fluctuate dramatically when the control means is not used. After the distributed power supply local voltage control method for cloud edge cooperation and power grid privacy protection is adopted to optimize the output of the distributed power supply, each distributed power supply adjusts reactive compensation in real time, and when the node voltage is low, the distributed power supply supports voltage by sending out reactive power; when the node voltage is high, the distributed power supply reduces the voltage by absorbing reactive power, so that the system voltage is maintained at a safe operation level.
Compared with the scheme II, the scheme III is based on a cloud-edge coordination framework, and a better voltage control effect is achieved.
As can be seen from the comparison of the schemes III and IV, the result obtained by optimizing the output of the distributed power supply in a partitioning manner by using the distributed power supply in-situ voltage control method for cloud edge cooperation and power grid privacy protection is slightly inferior to the result obtained by optimizing in a partitioning manner by using a centralized setting method; compared with a centralized setting method, the cloud-edge cooperation and power grid privacy protection distributed power supply on-site voltage control method effectively avoids the problem that user data privacy is revealed in the optimization process.
Claims (10)
1. A cloud edge cooperation and power grid privacy protection distributed power supply local voltage control method is characterized by comprising the following steps:
1) dividing the region under the jurisdiction of each edge computing device according to the selected active power distribution network, and inputting parameters of the power distribution network, including the network topology connection relation of the power distribution network, the access positions, the capacities and the parameters of the distributed power supplies; input distribution network NdHistorical power flow data of a typical day, and a predicted sunrise output curve and a predicted sunrise load output curve of the distributed power supply; setting a typical topological total number NsThe total sampling time length delta T and the sampling time interval delta T of the training object; setting parameters of a power distribution network proxy model based on a graph convolution neural network, wherein the parameters comprise a learning rate gamma, a regularization coefficient beta and times E of completely traversing a training set;
2) According to the parameters of the power distribution network and N in the step 1)dHistorical load flow data of a typical day, constructing a training set of a power distribution network agent model based on a graph convolution neural network at a cloud end, and determining the number of training objects in the training set, the composition of each training object and preprocessing the training objects; constructing a power distribution network proxy model based on a graph and convolution neural network at the cloud, and finishing the training of the power distribution network proxy model based on the graph and convolution neural network to obtain the trained power distribution network proxy model;
3) the trained power distribution network agent model is issued to edge computing devices on the edge side of the distribution network and serves as power distribution network sub-agent models of all the edge computing devices;
4) according to each distribution network subagent model, on the basis of a predicted distributed power supply daily output curve and daily load output curve and voltage and net load power measurement information in the area governed by each edge computing device, on the edge side of a distribution network, setting local voltage control curve parameters of the distributed power supply in each area by taking the minimum node voltage deviation in the area governed by each edge computing device as a target function;
5) updating the training objects based on the sub-agent model of the power distribution network to form an edge side sub-training set according to the local voltage control curves of the distributed power supplies in the regions obtained in the step 4); respectively updating the weight coefficients of the sub-agent models of the distribution network at each edge side by using a federal learning framework to obtain updated sub-agent models of the distribution network;
6) And (5) re-setting the local voltage control curve parameters of the distributed power supplies in each region according to the updated power distribution network sub-agent model in the step 5), and outputting the local voltage control curve setting result.
2. The distributed power supply local voltage control method for cloud-edge collaboration and power grid privacy protection as claimed in claim 1, wherein the power distribution network N in the step 1) isdHistorical trend data for a typical day, is:
in the distribution network NdIn the course of a typical day,a typical day distributed power supply does not participate in reactive power regulation,a typical daily distributed power supply is given as [1.0, 1.0 ]]Adjusting reactive power for the base curve of the dead zone to obtain distribution network NdHistorical trend data for a typical day.
3. The distributed power supply local voltage control method for cloud-edge coordination and power grid privacy protection according to claim 1, wherein in step 2),
the number of the training objects in the training set is as follows:
based on distribution network NdHistorical trend data of a typical day, and considering NsA typical topological structure is used for constructing a training set of a power distribution network agent model based on a graph convolution neural network, and the training set comprisesA training subject;
the composition of each training object comprises:
Each training object consists of three matrixes, namely a characteristic information matrix X, a topological information matrix A and a label matrix Z, and the label matrix Z is fitted by taking the characteristic information matrix X and the topological information matrix A as input based on a power distribution network agent model of the graph convolution neural network;
the feature information matrix X represents the input features of the training objects, and the feature information matrix X of the n-th training objectnThe system consists of net load active power and net load reactive power injected by a sampling moment node and a node voltage value of a power distribution network, and is expressed as follows:
in the formula (I), the compound is shown in the specification,a column vector representing the net load active power injected by each node at the sampling instant,a column vector representing the net load reactive power injected by each node at the sampling instant,representing a column vector formed by node voltage values of the power distribution network at the sampling moment, wherein N is the number of nodes of the power distribution network;
the topological information matrix A represents the connection relation between nodes, and the topological information matrix A of the n-th training objectnExpressed as:
in the formula (I), the compound is shown in the specification,topological information matrix A for the n-th training objectnRow ith and column jth elements of (1);
the label matrix Z represents the label values of the output features of the training objects, and the label matrix Z of the nth training objectnThe node voltage control value at the sampling moment is represented as:
4. The method for cloud-edge cooperative and power grid privacy protection distributed power supply local voltage control as claimed in claim 1, wherein the preprocessing of the training objects in step 2) is to preprocess each element in a feature information matrix X and a tag matrix Z of the training objects in a training set before model training; wherein
Characteristic information matrix X for n-th training objectnCarrying out pretreatment, wherein the pretreatment formula is as follows:
in the formula (I), the compound is shown in the specification,characteristic information matrix X of n training objects before preprocessingnTo the ith nodeThe number of k input features is such that,the characteristic information matrix X of the n-th training object after preprocessingnThe kth input feature of the ith node,is the mean value of all elements of the kth input feature in the feature information matrix X of the training object,the variance of all elements of the kth input feature in a feature information matrix X of a training object;
label matrix Z for nth training objectnCarrying out pretreatment, wherein the pretreatment formula is as follows:
in the formula (I), the compound is shown in the specification,label matrix Z for n training object before preprocessing nThe ith node outputs the tag value of the feature,label matrix Z for n training object after pretreatmentnThe label value of the output characteristic of the ith node, piZIs the mean value, delta, of all elements of the output features in the label matrix Z of the training objectZAnd outputting the variance of all elements of the features in the label matrix Z of the training object.
5. The method for cloud-edge cooperative and grid privacy protection distributed power supply local voltage control according to claim 1, wherein the power distribution network proxy model based on the graph convolution neural network in the step 2) is expressed as:
in the formula, H(l+1)Outputting a (l + 1) th hidden layer of the power distribution network agent model; h(l)Is the output of the l-th hidden layer; a is a topological information matrix,IMis an M-order identity matrix;is a diagonal matrix and is characterized in that,W(l)is the weight matrix of the l layer; y is an output matrix of the graph convolution neural network; h(k-1)The output of the last hidden layer; w(k)Is the weight of the output layer; σ (-) is a nonlinear activation function, expressed as:
σ(x)=LeakyReLU(x,α)=max(0,x)+α×min(0,x) (9)
wherein α is a negative slope; f (-) is the output layer function, F (-) 1 (-);
the power distribution network agent model based on the graph convolution neural network takes the minimum loss function as a target function and a batch gradient descent algorithm as a training mode, and the training process is expressed as follows:
In the formula (I), the compound is shown in the specification,the method is a loss function of a power distribution network proxy model based on a graph convolution neural network when a weight coefficient is omega;output matrix Y for the nth training objectnOutputting a fitting value of the characteristic by the ith node;label matrix Z for n training object after pretreatmentnThe ith node outputs the label value of the characteristic; n is a radical ofTThe number of the training objects is N, and the number of the nodes of the power distribution network is N; omega is the current weight coefficient of the power distribution network proxy model based on the graph convolution neural network, and omega islThe first weight coefficient of the power distribution network proxy model based on the graph convolution neural network is gamma, and the gamma is the learning rate.
6. The method for cloud-edge cooperative and grid privacy protection distributed power supply local voltage control according to claim 1, wherein the local voltage control curve of the distributed power supply in step 4) is obtained by using the following formula:
in the formula, Vt,iThe voltage amplitude of node i, g (V), for a period of tt,i) For the expression of the distributed power supply in-situ voltage reactive power control strategy, g (V)t,i) With dead zone of regulation [ V ]i q,min,Vi q,max],Vi q,minAnd Vi q,maxMinimum and maximum voltage regulation dead zone values for the local reactive voltage control strategy, respectivelyAnd the local voltage control curve parameter of the distributed power supply is that the reactive power sent by the distributed power supply in the regulation dead zone is 0.
7. The distributed power supply local voltage control method for cloud-edge collaboration and power grid privacy protection according to claim 1, wherein the minimum node voltage deviation in the area governed by each edge computing device in step 4) is taken as an objective function and is expressed as:
wherein f is the node voltage deviation, NtIs the number of time slices, NaCalculating the total number of nodes in the area covered by the device for the a-th edge; vt,iThe voltage amplitude of the node i in the period t;is the maximum voltage threshold value of the voltage of the power supply,as minimum voltage threshold, when Vt,iOut of desired voltage intervalThe objective function is used to reduce the voltage deviation.
8. The method for cloud-edge cooperative and power grid privacy protection distributed power supply local voltage control according to claim 1, wherein the setting of the local voltage control curve parameters of the distributed power supplies in each region in step 4) comprises:
(4.1) selecting upper and lower limits V of voltage regulation dead zone of distributed power supplyi q,minAnd Vi q,maxIs [0.9, 1.1 ]]And the constraint conditions of the upper limit and the lower limit of the dead zone are expressed as follows:
(4.2) based on the upper and lower limits V of the dead zonei q,minAnd Vi q,maxAnd power distribution network measurement information in the area, and updating the characteristic information matrix X to obtain an updated characteristic information matrix X';
(4.3) inputting the updated characteristic information matrix X' into a power distribution network sub-agent model to obtain system voltage distribution in the area corresponding to the upper limit and the lower limit of the dead zone, and calculating the node voltage deviation, wherein the formula is as follows:
in the formula, NtIs the number of time slices, NaCalculating the total number of nodes in the area covered by the device for the a-th edge; vt,iThe voltage amplitude of the node i in the period t;is the maximum voltage threshold value of the voltage of the power supply,as minimum voltage threshold, when Vt,iOut of desired voltage intervalThe objective function is used to reduce the voltage deviation;
and (4.4) scanning each dead zone upper and lower limit combination, and outputting the dead zone upper and lower limit combination with the minimum node voltage deviation as an optimal solution, namely the local voltage control curve parameters of the distributed power supply in the region.
9. The method for cloud-edge cooperative and power grid privacy protection distributed power supply local voltage control as claimed in claim 1, wherein the step 5) of updating the training objects based on the power distribution network sub-agent model to form an edge side sub-training set comprises:
in-situ voltage control curve dead zone upper and lower limits V of distributed power supply in each region based on settingi q,min/Vi q,maxAnd power distribution network measurement information in the area, updating the characteristic information matrix X 'to obtain an updated characteristic information matrix X' m(ii) a The updated feature information matrix X'mInputting a distribution network sub-agent model to obtain system voltage distribution corresponding to the upper and lower limits of the dead zone, updating a tag matrix Z, and updating the updated tag matrix Z'm。
10. The method for cloud-edge coordination and power grid privacy protection distributed power supply local voltage control according to claim 1, wherein the step 5) of updating the weight coefficients of the sub-agent models of the edge-side power distribution network by using a federal learning framework respectively comprises the following steps:
(5.1) calculating the average loss of each edge side distribution network subagent model in the b-th iteration in the updating process of the weight coefficient by the following formula;
in the formula (f)g(omega) is an objective function of the g-th power distribution network sub-agent model, omega is a weight coefficient of the model,the sub-agent model of the power distribution network corresponds to the nth training object (x) when the weight coefficient is omegan,yn) Is used to determine the loss function of (c),is the g thNumber of training objects in sub-training set of sub-agent model of power distribution network, DgA sub-training set of the g-th sub-agent model of the power distribution network; f (ω) is the overall objective function of the Federal learning framework, NGIs the total number of distributed power sources, NCThe sum of the number of the training objects in each sub-training set;
(5.2) each edge side power distribution network sub-agent model takes the minimum total objective function of the federal learning frame as a target, an edge side sub-training set is used for training, and the weight coefficient of each edge side power distribution network sub-agent model is adjusted to obtain each power distribution network sub-agent model before communication;
(5.3) uploading the updated self weight coefficients by each power distribution network sub-agent model through the following formula, carrying out weighted average on the updated self weight coefficients of each edge side power distribution network sub-agent model by the cloud, and issuing the weighted average weight coefficients to the edge side to finish the b-th iteration;
in the formula, Wg,bThe weight coefficient, W, of the g-th sub-agent model of the power distribution network before the b-th iterationb+1Weight coefficient, W, of the trained power distribution network proxy model of the cloud after the b-th iterationg,b+1And the weight coefficient is the weighted average of the g-th power distribution network sub-agent model after the b-th iteration.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110839904.3A CN113471982B (en) | 2021-07-22 | 2021-07-22 | Cloud edge cooperation and power grid privacy protection distributed power supply in-situ voltage control method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110839904.3A CN113471982B (en) | 2021-07-22 | 2021-07-22 | Cloud edge cooperation and power grid privacy protection distributed power supply in-situ voltage control method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113471982A CN113471982A (en) | 2021-10-01 |
CN113471982B true CN113471982B (en) | 2022-06-28 |
Family
ID=77882203
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110839904.3A Active CN113471982B (en) | 2021-07-22 | 2021-07-22 | Cloud edge cooperation and power grid privacy protection distributed power supply in-situ voltage control method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113471982B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114024322B (en) * | 2021-11-15 | 2023-08-25 | 云南电网有限责任公司电力科学研究院 | Reactive voltage self-learning control system and method for power distribution network |
CN114511179B (en) * | 2021-12-28 | 2023-09-12 | 江苏东晔电气设备有限公司 | Intelligent planning method and system for processing abnormality of power distribution cabinet |
CN114938372B (en) * | 2022-05-20 | 2023-04-18 | 天津大学 | Federal learning-based micro-grid group request dynamic migration scheduling method and device |
CN116667369B (en) * | 2023-08-01 | 2023-10-31 | 国网山西省电力公司晋城供电公司 | Distributed photovoltaic voltage control method based on graph convolution neural network |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111064633A (en) * | 2019-11-28 | 2020-04-24 | 国网甘肃省电力公司电力科学研究院 | Cloud-edge cooperative power information communication equipment automated testing resource allocation method |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109728603A (en) * | 2018-12-26 | 2019-05-07 | 天津大学 | Active power distribution network distributed electrical source partition voltage control strategy setting method on the spot |
CN111697599A (en) * | 2020-07-14 | 2020-09-22 | 国网信息通信产业集团有限公司 | Low-voltage distribution network control method and low-voltage distribution network control system |
CN111934325B (en) * | 2020-07-27 | 2023-08-29 | 北京奥德威特电力科技股份有限公司 | Cloud edge cooperative voltage reactive intelligent control system and control method |
CN112688422A (en) * | 2020-11-13 | 2021-04-20 | 广西电网有限责任公司电力科学研究院 | Intelligent low-voltage branch monitoring method based on cloud edge-end cooperation |
CN112653154B (en) * | 2020-11-30 | 2023-06-27 | 国网浙江省电力有限公司嘉兴供电公司 | Distributed photovoltaic power distribution network reactive power optimization control method based on edge calculation |
-
2021
- 2021-07-22 CN CN202110839904.3A patent/CN113471982B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111064633A (en) * | 2019-11-28 | 2020-04-24 | 国网甘肃省电力公司电力科学研究院 | Cloud-edge cooperative power information communication equipment automated testing resource allocation method |
Also Published As
Publication number | Publication date |
---|---|
CN113471982A (en) | 2021-10-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113471982B (en) | Cloud edge cooperation and power grid privacy protection distributed power supply in-situ voltage control method | |
Gao et al. | Day-ahead power forecasting in a large-scale photovoltaic plant based on weather classification using LSTM | |
Wu et al. | Multi-timescale forecast of solar irradiance based on multi-task learning and echo state network approaches | |
CN112564098B (en) | High-proportion photovoltaic power distribution network voltage prediction method based on time convolution neural network | |
CN110738344B (en) | Distributed reactive power optimization method and device for load prediction of power system | |
CN112465664B (en) | AVC intelligent control method based on artificial neural network and deep reinforcement learning | |
CN111222674A (en) | Short-term photovoltaic power generation capacity prediction method based on long-term and short-term memory neural network | |
CN105631483A (en) | Method and device for predicting short-term power load | |
CN113363998B (en) | Power distribution network voltage control method based on multi-agent deep reinforcement learning | |
CN105913151A (en) | Photovoltaic power station power generation amount predication method based on adaptive mutation particle swarm and BP network | |
CN113422371B (en) | Distributed power supply local voltage control method based on graph convolution neural network | |
CN107609667B (en) | Heat supply load prediction method and system based on Box _ cox transformation and UFCNN | |
CN111561732B (en) | Heat exchange station heat supply adjusting method and system based on artificial intelligence | |
CN104636985A (en) | Method for predicting radio disturbance of electric transmission line by using improved BP (back propagation) neural network | |
CN113489015B (en) | Multi-time-scale reactive voltage control method for power distribution network based on reinforcement learning | |
CN110070228B (en) | BP neural network wind speed prediction method for neuron branch evolution | |
CN109615124B (en) | SCADA master station load prediction method based on deep learning | |
CN106920014A (en) | A kind of short-term load forecasting method and device | |
CN111384726A (en) | High-permeability photovoltaic power distribution network partition voltage regulation method | |
CN111799808B (en) | Voltage distributed control method and system based on multi-agent deep reinforcement learning | |
CN109242136A (en) | A kind of micro-capacitance sensor wind power Chaos-Genetic-BP neural network prediction technique | |
CN115313403A (en) | Real-time voltage regulation and control method based on deep reinforcement learning algorithm | |
Pinson | Introducing distributed learning approaches in wind power forecasting | |
CN117154690A (en) | Photovoltaic power generation power prediction method and system based on neural network | |
CN113872213B (en) | Autonomous optimization control method and device for power distribution network voltage |
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 |