CN117674197B - Frequency adjustment method, storage medium and equipment using virtual power plant active support - Google Patents
Frequency adjustment method, storage medium and equipment using virtual power plant active support Download PDFInfo
- Publication number
- CN117674197B CN117674197B CN202410133128.9A CN202410133128A CN117674197B CN 117674197 B CN117674197 B CN 117674197B CN 202410133128 A CN202410133128 A CN 202410133128A CN 117674197 B CN117674197 B CN 117674197B
- Authority
- CN
- China
- Prior art keywords
- power
- vpp
- region
- aggregation node
- adjustment
- 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
- 238000000034 method Methods 0.000 title claims abstract description 86
- 238000003860 storage Methods 0.000 title claims abstract description 11
- 230000002776 aggregation Effects 0.000 claims abstract description 110
- 238000004220 aggregation Methods 0.000 claims abstract description 110
- 230000033228 biological regulation Effects 0.000 claims abstract description 35
- 238000012725 vapour phase polymerization Methods 0.000 claims description 148
- 238000009826 distribution Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 18
- 230000007812 deficiency Effects 0.000 claims description 15
- 239000013598 vector Substances 0.000 claims description 15
- 238000007476 Maximum Likelihood Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 12
- 230000001105 regulatory effect Effects 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000004146 energy storage Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 238000010248 power generation Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 241000764238 Isis Species 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 230000002452 interceptive effect Effects 0.000 claims description 2
- 230000003827 upregulation Effects 0.000 claims description 2
- 230000008901 benefit Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 9
- 125000003275 alpha amino acid group Chemical group 0.000 description 7
- 238000004590 computer program Methods 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 150000001875 compounds Chemical class 0.000 description 6
- 230000003993 interaction Effects 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 238000004088 simulation Methods 0.000 description 4
- 238000012733 comparative method Methods 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007430 reference method Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 150000001413 amino acids Chemical class 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000002803 fossil fuel Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000013076 uncertainty analysis Methods 0.000 description 1
Classifications
-
- 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
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a frequency adjusting method, a storage medium and equipment for active support of a virtual power plant, wherein operation data of distributed resources in each region VPP are input into a pre-trained GCN-BiLSTM combination model for prediction, and a predicted power value output by an aggregation node in each region VPP is obtained; calculating an optimal power adjustment task born by the aggregation node under each region VPP; calculating the actual regulation task of the aggregation node under each region VPP; and calculating the adjusted system frequency variation. The advantages are that: the method can promote the effective management and utilization of distributed resources so as to better reversely support the frequency adjustment of the main power grid; the accuracy of the power prediction of the aggregation node under the VPP is improved; the reliability of the VPP participating in frequency regulation is effectively improved; the accuracy of responding to the AGC command of the main power grid is effectively improved.
Description
Technical Field
The invention relates to a frequency adjusting method, a storage medium and equipment using active support of a virtual power plant, and belongs to the technical field of operation and control of power systems.
Background
Due to the intermittent nature and uncertainty of renewable energy sources, negative effects are brought to the supply and demand balance of the power grid, and the frequency of the system fluctuates. Traditionally, to keep the system frequency stable, the output of fossil fuel generators is regulated. However, because the slope capability of the main network cannot meet the requirement of frequency adjustment, and meanwhile, renewable energy sources are continuously connected with the power grid, the capacity ratio of the traditional unit is gradually reduced, and the frequency adjustment capability of the main network is gradually insufficient. Considering that there are massive distributed flexible resources on the distribution side that can be mined, it is necessary to utilize some controllable resources that can support the frequency adjustment of the main network. In particular, a flexible controllable virtual power plant (virtual power plants, VPP) is built by aggregating mass distributed resources, and a novel paradigm is provided for power grid frequency adjustment as a novel management and control technology.
There are few methods related to frequency adjustment by using VPP, and these methods mainly use VPP to participate in market to provide adjustment resources to ensure stability of system frequency. However, there are two main difficulties with this approach: 1) VPP makes frequency adjustments entirely from the point of view of maximizing its own economic benefits, without from the point of view of confidence margin. Because the VPP ignores the influence of actual adjustable resources while pursuing own benefits, the risk of system frequency adjustment is greatly increased; 2) The VPP ignores the topological relevance of the network when performing frequency adjustment, and the frequency adjustment of the system is affected by the power loss of the transmission line.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a frequency adjusting method, a storage medium and equipment for active support by utilizing a virtual power plant.
In order to solve the technical problems, the invention provides a frequency adjustment method utilizing active support of a virtual power plant, which is applied to a frequency adjustment control architecture of a multi-region VPP, wherein the frequency adjustment control architecture of the multi-region VPP comprises a main power grid, a control center and each region VPP, and each region VPP comprises a plurality of distributed resources;
the method comprises the following steps:
acquiring operation data of distributed resources in each region VPP, a planned power value of an aggregation node under the region VPP, and an adjusting power value and a frequency deviation coefficient required by a main power grid automatic power generation control instruction; the operation data comprise wind power, maximum output value of photovoltaic and interactive power value of an energy storage system at each moment, and VPP represents a virtual power plant;
inputting the operation data of the distributed resources in each region VPP into a pre-trained GCN-BiLSTM combination model for prediction to obtain a predicted power value output by an aggregation node under each region VPP;
based on the predicted power values, calculating effective power values output by aggregation nodes under the VPPs of all the areas, and uploading the effective power values to a control center so that the control center calculates the effective power values according to the adjustment power values required by the automatic power generation control instruction of the main power grid to obtain the optimal power adjustment tasks born by the aggregation nodes under the VPPs of all the areas;
transmitting the output power value of the aggregation node under each region VPP to a control center, so that the control center calculates power adjustment deviation caused by network loss according to the output power value of the aggregation node under each region VPP, and performs power compensation calculation according to the power adjustment deviation to obtain the actual adjustment task of the aggregation node under each region VPP;
and transmitting the actual adjustment task of the aggregation node under each region VPP to implement power adjustment, and uploading the actual adjustment power of the aggregation node under each region VPP to a main power grid so that the main power grid calculates the adjusted power grid frequency variation according to the actual adjustment power of the aggregation node under each region VPP.
Further, in the frequency adjustment control architecture of the multi-area VPP, the main power grid is used for sending an AGC instruction to be adjusted to the control center, the control center is used for receiving the adjustment instruction of the main power grid, simultaneously coordinating the output power of the aggregation node under each area VPP and feeding back adjustment information to the main power grid, and each area VPP is used for collecting the operation information of the internal distributed resource to predict the output power of the aggregation node, and simultaneously uploading the calculated adjustment capacity of the aggregation node to the control center.
Further, the training of the GCN-BiLSTM combination model comprises the following steps:
acquiring a constructed GCN-BiLSTM combination model, wherein the GCN-BiLSTM combination model comprises a GCN network layer, a BiLSTM network layer and a combination model output layer which are sequentially connected; the GCN network layer comprises an input layer, a hidden layer and a GCN output layer which are sequentially connected; the BiLSTM network layer comprises two unidirectional LSTM neural networks;
acquiring historical operation data of distributed resources in each region VPP;
carrying out normalization processing on the historical operation data, and dividing a data set after normalization processing into a training set and a testing set;
and updating parameters of the GCN-BiLSTM combination model by using the training set, the testing set and the adaptive motion estimation optimization algorithm to obtain optimal parameters, and substituting the optimal parameters into the GCN-BiLSTM combination model to obtain a trained GCN-BiLSTM combination model.
Further, the obtaining the predicted power value output by the aggregation node under each region VPP includes:
normalizing the operation data of the distributed resources in each region VPP;
extracting spatial feature vectors of the normalized data by using a GCN network layer of a GCN-BiLSTM combination model;
inputting the space feature vector into a BiLSTM network layer of a GCN-BiLSTM combination model for time feature extraction to obtain a space-time feature vector;
and inputting the space-time feature vector to the combined model output layer to obtain the predicted power value output by the aggregation node under each region VPP.
Further, the formula for calculating the effective power value output by the aggregation node under each region VPP based on the predicted power value is as follows:
；
in the method, in the process of the invention,is the effective value of the aggregation node under the region VPP, < >>Predicted power value output for aggregation node under region VPP, +.>A lower limit for predicting a bias power deficiency;
lower limit of predicted deviation power deficitThe determination process of (1) is as follows:
deviation between the historical predicted power value output by the aggregation node under the regional VPP and the corresponding historical true valueDenoted as->Is used for the normal distribution of the (c),μ AN 、/>respectively the mean and the variance of the prediction bias;
inputting deviation data between a determined set of historical predicted power values and corresponding historical true valuesX i Obtaining the maximum likelihood estimation of the mean value and the variance of normal distribution, wherein the expression is as follows:
；
；
in the method, in the process of the invention,is thatμ AN Is used for the maximum likelihood estimation of (a),nfor predicting the number of samples of the deviation data; />Is->Maximum likelihood estimator of (a);
combining the maximum likelihood estimation of the mean value and the variance of the normal distribution, and obtaining a prediction deviation at a confidence level of 1-αThe following power deficiency is expressed as:
；
in the method, in the process of the invention,G(. Cndot.) isIs the inverse of the probability density function of (c),G(α 1 ) For the lower limit of the predicted bias power deficiency output by the aggregation node under VPP,G(α 2 ) For the upper limit of the predicted deviation power deficiency output by the aggregation node under the VPP, the inverse function adopts a symmetric probability area, and the symmetric probability area is expressed as:α 1 -α 2 =1-α，α 1 =α/2，α 2 =1-α/2，αindicating a level of significance.
Further, the calculating, by the control center according to the adjustment power value and the effective power value required by the main power grid automatic power generation control instruction, the optimal power adjustment task borne by the aggregation node under each regional VPP includes:
calculating the adjustment capability of the aggregation nodes under the VPPs of different areas according to the effective power values output by the aggregation nodes under the VPPs of different areas and the planned power values of the aggregation nodes under the VPPs of corresponding areas;
combining the adjustment capability of the aggregation node under different regions VPP to obtain the adjustment capability of the aggregation node under the multi-region VPP;
determining a power regulation task born by the multi-region VPP according to the aggregation node regulation capability under the multi-region VPP;
and determining the optimal regulation task distributed to each aggregation node according to the regulation capability of the aggregation node under the multi-region VPP and the power regulation task born by the multi-region VPP.
Further, the calculation formula of the optimal adjustment task allocated to each aggregation node is as follows:
；
in the method, in the process of the invention,is thattTime of day allocation to the firstiOptimal regulation task of individual aggregation nodes, +.>Is thattThe power regulation task assumed by the time multizone VPP,/->Is thattTime of day (time)iThe up-regulation capability of the individual aggregation nodes,is thattUp-regulation capability of aggregation node under time multi-region, < ->Is thattTime of day (time)iDown-regulation capability of individual aggregation nodes, < >>Is thattDown regulation capability of AN under time multiple area, < ->Representing the set of all aggregation nodes in the distribution network, +.>，N a Is the number of aggregation nodes.
Further, in the calculationtTime of day allocation to the firstiOptimal tuning tasks for individual aggregation nodesAfter that, pair->Power supplementing is carried out, and the power supplementing is determined as the firstiThe actual regulation task of the aggregation nodes comprises the following steps:
representing the power regulation deviation caused by the network loss asCombining the adjustment capability of each aggregation node, according to a proportion distribution method, obtaining power compensation by using the following formula:
；
in the method, in the process of the invention,is the firstiPower compensation borne by the aggregation nodes;
combining the optimal adjustment tasks allocated to each aggregation node, the actual adjustment tasks of each aggregation node are obtained by using the following formula:
；
in the method, in the process of the invention,is the firstiActual adjustment tasks of the individual aggregation nodes;
further, the calculation formula of the adjusted power grid frequency variation is as follows:
；
in the method, in the process of the invention,βas a coefficient of frequency deviation (f-co) of the frequency deviation,for the adjusted grid frequency variation, +.>The multi-zone VPP power adjustment is obtained by summing the actual adjustment tasks of the aggregation nodes corresponding to each VPP.
A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform the method.
A computer device, comprising,
one or more processors, memory, and one or more programs, wherein one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing the method.
The invention has the beneficial effects that:
(1) The frequency regulation control architecture of the multi-region VPP can promote the effective management and utilization of distributed resources so as to better reversely support the frequency regulation of a main power grid;
(2) According to the invention, by gathering the advantages of GCN and BiLSTM, a novel GCN-BiLSTM combination model for extracting functional space-time characteristics is constructed, and the accuracy of power prediction of the aggregation node under VPP is greatly improved;
(3) Aiming at the problems that the current VPP does not start from the perspective of confidence margin and omits the influence of actual adjustable resources, the invention uses interval estimation to determine the effective power value output by AN from the perspective of confidence margin, thereby effectively improving the reliability of the VPP participating in frequency adjustment;
(4) The invention can respond to the AGC command of the main network with optimized power distribution;
(5) Aiming at the problem that the current VPP ignores the topological relevance of the network when performing frequency adjustment, the frequency adjustment of the system is affected by the power loss of a transmission line, the invention can correct the adjustment task of the AN under each region VPP based on the power compensation design scheme of proportional distribution, and effectively improves the accuracy of responding to the AGC instruction of the main power grid.
Drawings
FIG. 1 is a flow chart of a method for frequency adjustment using active support of a virtual power plant according to an embodiment of the present invention;
FIG. 2 is a topology diagram of a test system provided by an embodiment of the present invention;
fig. 3 is a diagram of a frequency adjustment control architecture of a multi-zone VPP provided by an embodiment of the invention;
FIG. 4 is a graph showing a comparison of AN AN power prediction curve for 10 consecutive days for different models provided by AN embodiment of the present invention;
FIG. 5 is a box plot of power prediction bias for aggregated nodes for different models provided by an embodiment of the present invention;
FIG. 6 is a graph of the system frequency adjustment results provided by an embodiment of the present invention;
FIG. 7 is a graph of comparative results of system frequency adjustment with confidence margin considered provided by an embodiment of the present invention;
fig. 8 is a graph of a comparison of system frequency adjustment taking into account transmission line power loss, according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
Embodiment 1, as shown in fig. 1, the present embodiment provides a frequency adjustment method using active support of a virtual power plant, including:
(1) Acquiring operation data of distributed resources in the VPP, a planned power value of AN AN under the VPP, and AN adjustment power value and a frequency deviation coefficient required by a main network AGC instruction; the operation data comprise wind power and photovoltaic maximum output values at all times and interaction power values of the energy storage system;
(2) Building a frequency adjustment control framework of the multi-region VPP; the control architecture comprises a main power grid, a control center and each region VPP, and in addition, each region VPP comprises a large amount of distributed resources such as wind power, photovoltaic and energy storage systems;
(3) Each regional VPP inputs the operation data of the internal distributed resources into a pre-constructed new GCN-BiLSTM combination model extracted by the functional space-time characteristics to predict, and a predicted power value output by AN AN is obtained;
(4) From the point of confidence margin, each region VPP calculates according to the predicted power value output by the AN by using a preset AN power deficiency measuring and calculating method based on interval estimation to obtain the effective power value output by the AN, and the effective power value is uploaded to a control center through information interaction;
(5) The control center calculates by adopting a control strategy based on scheduling, which is set up in advance, according to the adjustment power value required by the AGC instruction issued by the main power network and the effective power value output by the AN, so as to obtain the optimal power adjustment task borne by the AN under each region VPP;
(6) From the point of view of topological relevance of the network, the control center calculates according to the determined AN output power value under each regional VPP by utilizing a Matpower flow program to obtain power adjustment deviation caused by network loss, calculates power compensation according to a power compensation design scheme based on proportional distribution, obtains actual adjustment tasks of the AN under each regional VPP, sends the actual adjustment tasks to the AN under each regional VPP through information interaction to implement power adjustment, and then uploads multi-regional VPP adjustment information to a main power grid;
(7) And the main power grid calculates according to the determined power regulation information of the multi-region VPP by utilizing the pre-proposed relation of active power and frequency, and the regulated power grid frequency variation is obtained.
The method comprises the following specific steps:
step 1: and acquiring operation data of distributed resources in the VPP, a planned power value of the AN under the VPP, AN adjustment power value and a frequency deviation coefficient required by a main network AGC instruction, wherein the operation data comprises wind power and photovoltaic maximum output values at all times and AN interaction power value of AN energy storage system.
In this embodiment, the topology of the test system shown in fig. 2 is used. The topological connection relation of each node is shown in detail in the figure. The regional test system consists of 9 nodes, and comprises AN AN formed by aggregation of a plurality of photovoltaics, wind power and energy storage, such as nodes 2, 5 and 6 in the figure.
Step 2: a frequency-regulated control architecture of a multi-zone VPP is built, comprising a main grid, a control center and each zone VPP, and furthermore each zone VPP contains a large amount of distributed resources inside, such as wind power, photovoltaic and energy storage systems.
In the present embodiment, a frequency adjustment control architecture diagram of the multi-zone VPP shown in fig. 3 is employed.
The main power grid is responsible for issuing AGC instructions required to be regulated to the control center, the control center is responsible for receiving the regulating instructions of the main power grid, simultaneously coordinating the output power of the AN under each region VPP and feeding back regulating information to the main power grid, each region VPP is responsible for collecting the running information of internal distributed resources to predict the output power of the AN, and simultaneously uploading the calculated AN regulating capacity to the control center.
Step 3: and each region VPP inputs the operation data of the internal distributed resources into a pre-constructed GCN-BiLSTM combination model extracted by the novel functional space-time characteristics to predict, so as to obtain a predicted power value output by the AN.
Step 3.1: normalizing the operation data of the distributed resource, wherein the normalization process is expressed as:
（1）；
in the formula (1), the components are as follows,Xthe original data is represented by a representation of the original data,and->Representing the maximum and minimum values in the original data respectively,xrepresenting normalized data.
Step 3.2: the normalized dataset is divided into a training set and a test set.
Step 3.3: and (3) inputting the training set obtained in the step (3.2) into the GCN-BiLSTM combination model to be used as the input of the GCN-BiLSTM model.
Step 3.4: constructing a GCN network layer:
the GCN network layer is composed of an input layer, a hidden layer and an output layer, the sequence data input in the step 3.3 is input into the GCN network for spatial feature extraction, and the spatial feature vector is obtained by the following formula:
；
wherein, in the formula (2),is->Laplace transform matrix of>As an adjacency matrixAAnd unit matrixI N The specific expression of the sum is shown in the formula (5),H l() is at the first nodelThe feature matrix of the layer is used,W l() is the firstlThe weight matrix of the layer is used to determine,σis an activation function; in the formula (3), ∈>Is->Is a degree matrix of (2).
Step 3.5: building a BiLSTM network layer:
the BiLSTM network layer is composed of two traditional unidirectional LSTM neural networks, the spatial feature vector output in the step 3.4 is input into the BiLSTM network layer for time feature extraction, and the space-time feature vector is obtained by using the following formula:
；
wherein, in the formula (6),h t is thattThe output of the bit cell at time instant BiLSTM,fas a function of the LSTM outputs for stitching the two directions; in the formula (7) of the present invention,is thattOutputting the reverse LSTM unit at the moment; in the formula (8), ∈>Is thattThe output of the forward LSTM cell at time instant,x t is thattSpatial characteristic information input at moment;Lis the operation of the LSTM unit.
Step 3.6: the output layer is composed of a full connection layer, the spatial feature vector output in the step 3.5 is input into the output layer, and the predicted power value output by the AN is obtained by the following formula:
（9）；
in the formula (9), the amino acid sequence of the compound,the output vector is represented as such,wa weight vector representing the output layer is presented,brepresenting the offset vector of the output layer.
Step 3.7: and updating the GCN-BiLSTM network parameters by adopting a self-adaptive motion estimation optimization algorithm, and bringing the optimal parameters into a GCN-BiLSTM network model for training to obtain a trained AN output power prediction model.
Step 3.8: and predicting by using the trained GCN-BILSTM combination model to obtain AN AN output power predicted value.
Step 3.9: the predicted power value of the AN outputted by the GCN-BiLSTM network in the step 3.6And obtaining AN actual AN output power predicted value after the inverse normalization processing.
In this embodiment, the construction, training and testing of the prediction model is completed by the software MATLAB. To verify the validity of the proposed predictive model, three reference models, GCN, biLSTM and Multi-layer perceptron (MLP), were established for comparison. Figure 4 compares consecutive 10 day AN power prediction curves for different models. As can be seen from fig. 4, these 4 models can predict the trend of the node power as a whole, wherein the model predicts the best and has smaller deviation. For example, in fig. 4 (a) corresponding to aggregate node 2 power prediction, the MLP model prediction capability is comparable, but the error is larger in some time periods; the overall prediction error of the GCN model and the BiLSTM model is large compared to the above models. In addition, in the application of the above model, the power prediction effect of aggregation nodes 5 and 6 corresponding to (b) and (c) in fig. 4 is also substantially the same as that of aggregation node 2. In conclusion, the AN power value predicted by the GCN-BiLSTM combination model provided by the method has better fitting degree with a true value and higher prediction precision. Furthermore, comparison of the models shows that the proposed combined model has good power prediction performance at different ANs.
Step 4: from the point of confidence margin, each region VPP calculates according to the predicted power value output by the AN by using a preset AN power deficiency measuring and calculating method based on interval estimation, obtains the effective power value output by the AN, and uploads the effective power value to a control center through information interaction.
Step 4.1: deviation between the historical predicted power value output by the aggregation node under the regional VPP and the corresponding historical true valueDenoted as->Is used for the normal distribution of the (c),μ AN 、/>respectively the mean and the variance of the prediction bias;
step 4.2: inputting a determined set of AN output predictive deviation dataX i According to the maximum likelihood estimation method, the following formula is used to obtainMaximum likelihood estimation of mean and variance of normal distribution:
（10）；
（11）；
in the formula (10),is thatμ AN Is used for the maximum likelihood estimation of (a),nfor predicting the number of samples of the deviation data; in the formula (11), ∈>Is->Maximum likelihood estimator of (a);
step 4.3: combining the maximum likelihood estimates of the mean and variance of the prediction bias output by AN in step 4.2, and obtaining the prediction bias at a confidence level of 1 according to the interval estimation by using the following formulaαThe following power deficiency is expressed as:
（12）；
in the formula (12) of the present invention,G(. Cndot.) isIs the inverse of the probability density function of (c),G(α 1 ) For the lower limit of the predicted bias power deficiency output by the aggregation node under VPP,G(α 2 ) For the upper limit of the predicted deviation power deficiency output by the aggregation node under the VPP, the inverse function adopts a symmetric probability area, and the symmetric probability area is expressed as:α 1 -α 2 =1-α，α 1 =α/2，α 2 =1-α/2，αrepresenting a level of significance;
step 4.4: a lower limit of the predicted deviation power deficiency outputted according to the AN in the step 4.3G(α 1 ) The effective power output by AN is obtained using the following equation:
（13）；
in the formula (13), the amino acid sequence of the compound,、/>the effective value and the predicted value of the AN output power are respectively.
In this embodiment, the validity of the proposed predictive model is further verified in connection with model uncertainty analysis. FIG. 5 provides a box plot of power prediction bias for the aggregate node for different models. It can be seen that the model predictive bias data distribution herein is most stable and the median is closer to 0 than the other models. That is, according to the simulation result, the model has the lowest fluctuation degree of the prediction deviation data and higher prediction precision.
Step 5: and the control center calculates by adopting a control strategy based on scheduling, which is set up in advance, according to the adjustment power value required by the AGC instruction issued by the main power network and the effective power value output by the AN, so as to obtain the optimal power adjustment task borne by the AN under each region VPP.
Step 5.1: obtaining time according to the effective power value output by the AN and the planned power value of the AN under the VPP by using the following formulatAN regulatory capability under the area of (a):
；
wherein, in the formula (14),is the firstiUp-regulation capability of individual AN, +.>Is the firstiMaximum power value output by individual AN, +.>Represent the firstiPlanned power value of individual AN->Representing a set of all ANs in the distribution network; in formula (15), ∈>Is the firstiDown regulation capability of individual AN, +.>Is the firstiThe minimum power value outputted by the AN, wherein the maximum power value outputted by the AN means the effective power value outputted by the AN +.>。
Step 5.2: combining the step (5.1) timetThe regional lower AN adjustment capability of (2) is obtained using the following equation:
；
wherein, in the formula (16),for the upward adjustment capability of AN in multiple zones,N a is the number of ANs; in the formula (17), ∈>Is the downward tuning capability of AN in multiple zones.
Step 5.3: combining the adjustment capability of the AN under the multi-zone in the step 5.2, the power adjustment task borne by the multi-zone VPP is obtained by using the following formula:
（18）；
in the formula (18), the amino acid sequence of the compound,the power regulation task assumed for a multizone VPP,/->And adjusting the power value required by the AGC command issued by the main power network.
Step 5.4: based on the multi-zone lower AN adjustment capability in step 5.2 and the power adjustment task undertaken by the multi-zone VPP in step 5.3, AN optimal adjustment task assigned to each AN is obtained using the following equation:
（19）；
in the formula (19), the amino acid sequence of the compound,to be allocated to the firstiOptimal tuning tasks for the ANs.
Step 6: from the point of view of topological relevance of the network, the control center calculates according to the determined AN output power value under each regional VPP by utilizing a Matpower flow program to obtain power adjustment deviation caused by network loss, calculates power compensation according to a power compensation design scheme based on proportional distribution, obtains actual adjustment tasks of the AN under each regional VPP, transmits the actual adjustment tasks to the AN under each regional VPP through information interaction to implement power adjustment, and then uploads multi-regional VPP adjustment information to a main power grid.
Step 6.1: representing the power regulation deviation caused by the network loss asCombining the adjustment capability of each AN in step 5.1, according to the proportional allocation method, power compensation is obtained by using the following formula:
（20）；
in the formula (20), the amino acid sequence of the compound,is the firstiPower compensation assumed by the individual ANs.
Step 6.2: combining the optimal adjustment tasks allocated to each AN in the step 5.4, and obtaining the actual adjustment tasks of each AN by using the following formula:
（21）；
in the formula (21), the amino acid sequence of the amino acid,is the firstiThe actual tuning task of the AN.
Step 7: and the main power grid calculates according to the determined power regulation information of the multi-region VPP by utilizing the pre-proposed relation of active power and frequency, and the regulated power grid frequency variation is obtained.
According to the multi-region VPP power regulation information, the regulated power grid frequency variation is obtained by using the following formula:
（22）；
in the formula (22), the amino acid sequence of the compound,βas a coefficient of frequency deviation (f-co) of the frequency deviation,for the adjusted system frequency variation, +.>The amount of power adjustment for the multi-zone VPP.
In this embodiment, fig. 6 provides the frequency deviation of the system using 15min sampling time, and the frequency adjustment results of the proposed method and the reference method. As can be seen in connection with fig. 6, the participation of the multi-zone VPP in frequency regulation can suppress system frequency fluctuations. For example, when the system frequency of occurrence is low, except at 16:45 and 19:00, the multi-zone VPP can reverse grid charging to support system frequency regulation, returning to around the target value of 50 Hz. When the frequency of system occurrence is high, except at the frequency of 3:30 and 9: in 30 few time points, the multi-zone VPP is able to reduce discharge to the grid to provide system frequency regulation and return to around the target value. Furthermore, the multi-zone VPP is able to adjust all frequencies to within the safe range although the frequencies cannot be adjusted to around the target value in the above-mentioned time points. That is, according to the simulation result, the method proposed herein can effectively solve the frequency fluctuation problem (including the problems of higher frequency and lower frequency) of the system. Notably, we observe that the proposed method is close to the frequency adjustment result compared to the reference method in this example.
In the present embodiment, the effectiveness of the proposed frequency adjustment method considering the confidence margin is verified by the comparison result of fig. 7. The comparison simulation is carried out by adopting an operation result (comparison method 1) without considering the confidence margin and an operation result (comparison method 2) with consideration of the actual power adjustment and the proposed frequency adjustment method with consideration of the confidence margin. As shown in fig. 7, it can be clearly observed that the proposed frequency adjustment method taking into account the confidence margin can adjust the frequency to around the target value under the effective power adjustment, compared to the comparative method 2, whereas the comparative method 1, while being able to adjust the frequency to around the target value, does not take into account the confidence margin so that the adjustment capability exceeds its actual adjustment capability, which greatly increases the risk of the system frequency adjustment. Thus, the reliability of the proposed frequency adjustment method taking the confidence margin into account can be ensured by the simulation result.
In this embodiment, fig. 8 provides a comparison of the system frequency adjustment taking into account transmission line power loss. The comparison method 1 is an idealized frequency adjustment method without considering network loss, and the comparison method 2 is an actual frequency adjustment method without adopting power compensation and considering network loss. From comparing the frequency adjustment results of fig. 8, it can be clearly observed that the method proposed herein can achieve a similar frequency adjustment effect to that of the comparison method 1 in consideration of network loss, while the comparison method 2 can adjust the frequency to be within the safe range, but the frequency curve thereof has more fluctuation phenomenon. Thus, the proposed method is more advantageous in terms of the effectiveness of frequency adjustment than the two comparative methods. Furthermore, the effectiveness of the power compensation design based on proportional distribution in consideration of the power loss of the transmission line is also fully verified.
(1) The frequency adjustment control framework of the multi-region VPP provided by the invention can promote the effective management and utilization of distributed resources so as to better reversely support the frequency adjustment of the main power grid.
(2) According to the invention, by gathering the advantages of GCN and BiLSTM, a novel GCN-BiLSTM combination model for extracting functional space-time characteristics is constructed, and the accuracy of power prediction of the aggregation node under VPP is greatly improved.
(3) Aiming at the problems that the current VPP does not start from the perspective of confidence margin and neglects the influence of actual adjustable resources, the invention provides a method for AN power deficiency measurement and calculation based on interval estimation to determine the effective power value of AN output from the perspective of confidence margin, thereby effectively improving the reliability of the VPP participating in frequency adjustment.
(4) The present invention proposes a scheduling-based control strategy enabling a multi-zone VPP to respond to AGC instructions of a main network with an optimized power allocation.
(5) Aiming at the problem that the current VPP ignores the topological relevance of the network when performing frequency adjustment, and the frequency adjustment of the system is affected by the power loss of a transmission line, the invention provides a power compensation design scheme based on proportional distribution so as to correct the adjustment task of AN under each region VPP, and effectively improves the accuracy of responding to the AGC instruction of a main power grid.
Embodiment 2, which is based on the same inventive concept as the other embodiments, introduces a computer-readable storage medium storing one or more programs, which include instructions, which when executed by a computing device, cause the computing device to perform the method.
Embodiment 3, which is based on the same inventive concept as the other embodiments, introduces a computer device comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing any of the methods.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (8)
1. The frequency regulation method utilizing the active support of the virtual power plant is characterized by being applied to a frequency regulation control framework of a multi-region VPP, wherein the frequency regulation control framework of the multi-region VPP comprises a main power grid, a control center and each region VPP, and each region VPP comprises a plurality of distributed resources;
the method comprises the following steps:
acquiring operation data of distributed resources in each region VPP, a planned power value of an aggregation node under the region VPP, and an adjusting power value and a frequency deviation coefficient required by a main power grid automatic power generation control instruction; the operation data comprise wind power, maximum output value of photovoltaic and interactive power value of an energy storage system at each moment, and VPP represents a virtual power plant;
inputting the operation data of the distributed resources in each region VPP into a pre-trained GCN-BiLSTM combination model for prediction to obtain a predicted power value output by an aggregation node under each region VPP;
based on the predicted power values, calculating effective power values output by aggregation nodes under the VPPs of all the areas, and uploading the effective power values to a control center so that the control center calculates the effective power values according to the adjustment power values required by the automatic power generation control instruction of the main power grid to obtain the optimal power adjustment tasks born by the aggregation nodes under the VPPs of all the areas;
transmitting the output power value of the aggregation node under each region VPP to a control center, so that the control center calculates power adjustment deviation caused by network loss according to the output power value of the aggregation node under each region VPP, and performs power compensation calculation according to the power adjustment deviation to obtain the actual adjustment task of the aggregation node under each region VPP;
transmitting the actual adjustment task of the aggregation node under each region VPP to implement power adjustment, and uploading the actual adjustment power of the aggregation node under each region VPP to a main power grid so that the main power grid calculates and obtains the adjusted power grid frequency variation according to the actual adjustment power of the aggregation node under each region VPP;
the formula for calculating the effective power value output by the aggregation node under each region VPP based on the predicted power value is as follows:
；
in the method, in the process of the invention,for the value of the active power of the aggregation node under the regional VPP,/for the aggregation node>Predicted power value output for aggregation node under region VPP, +.>A lower limit for predicting a bias power deficiency;
lower limit of predicted deviation power deficitThe determination process of (1) is as follows:
deviation between the historical predicted power value output by the aggregation node under the regional VPP and the corresponding historical true valueRepresentation ofIs->Is used for the normal distribution of the (c),μ AN 、/>respectively the mean and the variance of the prediction bias;
inputting deviation data between a determined set of historical predicted power values and corresponding historical true valuesX i Obtaining the maximum likelihood estimation of the mean value and the variance of normal distribution, wherein the expression is as follows:
；
；
in the method, in the process of the invention,is thatμ AN Is used for the maximum likelihood estimation of (a),nfor predicting the number of samples of the deviation data; />Is->Maximum likelihood estimator of (a);
combining the maximum likelihood estimation of the mean value and the variance of the normal distribution, and obtaining a prediction deviation at a confidence level of 1-αThe following power deficiency is expressed as:
；
in the method, in the process of the invention, G(. Cndot.) isIs the inverse of the probability density function of (c),G(α 1 ) For the lower limit of the predicted bias power deficiency output by the aggregation node under the region VPP,G(α 2 ) For the upper limit of the predicted deviation power deficiency output by the aggregation node under the region VPP, the inverse function adopts a symmetric probability region, and the symmetric probability region is expressed as:α 1 -α 2 =1-α，α 1 =α/2，α 2 =1-α/2，αrepresenting a level of significance;
the step of calculating the effective power value by the control center according to the adjustment power value required by the main power grid automatic power generation control instruction to obtain the optimal power adjustment task born by the aggregation node under each region VPP comprises the following steps:
according to the effective power value output by the aggregation node under the different area VPPs and the planned power value of the aggregation node under the corresponding area VPPs, the adjustment capability of the aggregation node under the different area VPPs is calculated, and the method specifically comprises the following steps:
；
wherein, in the formula (14),is the firstiUp-regulation capability of individual aggregation nodes, +.>Is the firstiMaximum power value output by each aggregation node, +.>Represent the firstiThe planned power values of the individual aggregation nodes,representing a set of all aggregation nodes in the distribution network,N a is the number of aggregation nodes; formula (VI)(15) In (I)>Is the firstiDown-regulation capability of individual aggregation nodes, < >>Is the firstiThe minimum power value output by the aggregation node refers to the effective power value output by the aggregation node +.>；
Combining the adjustment capability of the aggregation node under different regions VPP to obtain the adjustment capability of the aggregation node under the multi-region VPP;
determining a power regulation task born by the multi-region VPP according to the aggregation node regulation capability under the multi-region VPP;
and determining the optimal power regulation task distributed to each aggregation node according to the aggregation node regulation capability under the multi-region VPP and the power regulation task born by the multi-region VPP.
2. The method for adjusting frequency by utilizing active support of virtual power plant according to claim 1, wherein in the frequency adjustment control architecture of the multi-zone VPP, the main power grid is used for issuing AGC command to be adjusted to the control center, the control center is used for receiving the adjustment command of the main power grid, simultaneously coordinating the output power of the aggregation node under each zone VPP and feeding back the adjustment information to the main power grid, and each zone VPP is used for collecting the operation information of the internal distributed resource to predict the output power of the aggregation node, and simultaneously uploading the calculated adjustment capability of the aggregation node to the control center.
3. The method for frequency adjustment with virtual power plant active support of claim 1, wherein the training of the GCN-BiLSTM combination model comprises:
acquiring a constructed GCN-BiLSTM combination model, wherein the GCN-BiLSTM combination model comprises a GCN network layer, a BiLSTM network layer and a combination model output layer which are sequentially connected; the GCN network layer comprises an input layer, a hidden layer and a GCN output layer which are sequentially connected; the BiLSTM network layer comprises two unidirectional LSTM neural networks;
acquiring historical operation data of distributed resources in each region VPP;
carrying out normalization processing on the historical operation data, and dividing a data set after normalization processing into a training set and a testing set;
and updating parameters of the GCN-BiLSTM combination model by using the training set, the testing set and the adaptive motion estimation optimization algorithm to obtain optimal parameters, and substituting the optimal parameters into the GCN-BiLSTM combination model to obtain a trained GCN-BiLSTM combination model.
4. A method for adjusting frequency by active support of a virtual power plant according to claim 3, wherein obtaining the predicted power value output by the aggregation node in each region VPP comprises:
normalizing the operation data of the distributed resources in each region VPP;
extracting spatial feature vectors of the normalized data by using a GCN network layer of a GCN-BiLSTM combination model;
inputting the space feature vector into a BiLSTM network layer of a GCN-BiLSTM combination model for time feature extraction to obtain a space-time feature vector;
and inputting the space-time feature vector to the combined model output layer to obtain the predicted power value output by the aggregation node under each region VPP.
5. The method for adjusting frequency with active support of virtual power plant according to claim 1, wherein the calculation formula of the optimal power adjustment task allocated to each aggregation node is:
；
in the method, in the process of the invention,is thattTime of day allocation to the firstiOptimal power regulation task for individual aggregation nodes, +.>Is thattThe power regulation task assumed by the time multizone VPP,/->Is thattTime of day (time)iThe up-regulation capability of the individual aggregation nodes,is thattUp-regulation capability of aggregation node under time multi-region, < ->Is thattTime of day (time)iDown-regulation capability of individual aggregation nodes, < >>Is thattThe downward adjustment capability of aggregation nodes under multiple areas at a time.
6. The method for adjusting frequency by using active support of virtual power plant according to claim 5, wherein the frequency is calculated bytTime of day allocation to the firstiOptimal power adjustment tasks for individual aggregation nodesAfter that, pair->Power compensation is carried out, and the determination is the firstiThe actual regulation task of the aggregation nodes comprises the following steps:
representing the power regulation deviation caused by the network loss asCombining the adjustment capability of each aggregation node, according to a proportion distribution method, obtaining power compensation by using the following formula:
；
in the method, in the process of the invention,is the firstiPower compensation borne by the aggregation nodes;
combining the optimal power adjustment tasks allocated to each aggregation node, the actual adjustment tasks of each aggregation node are obtained by using the following formula:
；
in the method, in the process of the invention,is the firstiActual adjustment tasks of the individual aggregation nodes;
the calculation formula of the regulated power grid frequency variation is as follows:
；
in the method, in the process of the invention,βas a coefficient of frequency deviation (f-co) of the frequency deviation,for the adjusted grid frequency variation, +.>The multi-zone VPP power adjustment is obtained by summing the actual adjustment tasks of the aggregation nodes corresponding to each VPP.
7. A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform the method of any of claims 1-6.
8. A computer device, comprising,
one or more processors, memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing the method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410133128.9A CN117674197B (en) | 2024-01-31 | 2024-01-31 | Frequency adjustment method, storage medium and equipment using virtual power plant active support |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410133128.9A CN117674197B (en) | 2024-01-31 | 2024-01-31 | Frequency adjustment method, storage medium and equipment using virtual power plant active support |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117674197A CN117674197A (en) | 2024-03-08 |
| CN117674197B true CN117674197B (en) | 2024-04-16 |
Family
ID=90064537
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410133128.9A Active CN117674197B (en) | 2024-01-31 | 2024-01-31 | Frequency adjustment method, storage medium and equipment using virtual power plant active support |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117674197B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115313361A (en) * | 2022-07-25 | 2022-11-08 | 天津大学 | Joint optimization control method and device for large-scale adjustable resources |
| CN116760103A (en) * | 2023-06-19 | 2023-09-15 | 云南电网有限责任公司昭通供电局 | Self-adaptive prediction energy management method for virtual power plant on-line optimization power dispatching |
| CN117318027A (en) * | 2023-09-25 | 2023-12-29 | 国网江西省电力有限公司供电服务管理中心 | EMD-based real-time day-ahead frequency regulation and control method for virtual power plant |
-
2024
- 2024-01-31 CN CN202410133128.9A patent/CN117674197B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115313361A (en) * | 2022-07-25 | 2022-11-08 | 天津大学 | Joint optimization control method and device for large-scale adjustable resources |
| CN116760103A (en) * | 2023-06-19 | 2023-09-15 | 云南电网有限责任公司昭通供电局 | Self-adaptive prediction energy management method for virtual power plant on-line optimization power dispatching |
| CN117318027A (en) * | 2023-09-25 | 2023-12-29 | 国网江西省电力有限公司供电服务管理中心 | EMD-based real-time day-ahead frequency regulation and control method for virtual power plant |
Non-Patent Citations (1)
| Title |
|---|
| A Cyber-Physical Collaboration Based Control Method for Frequency Regulation With VPP;Jinrui Guo等;IEEE SYSTEMS JOURNAL;20231207;第1-12页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117674197A (en) | 2024-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10504195B2 (en) | System and method for decentralized energy production | |
| Li et al. | Coordinated load frequency control of multi-area integrated energy system using multi-agent deep reinforcement learning | |
| CN111555281B (en) | Method and device for simulating flexible resource allocation of power system | |
| CN113054688B (en) | Renewable energy cluster output distributed coordination control method considering uncertainty | |
| CN111682536B (en) | Random-robust optimization operation method for virtual power plant participating in dual market before day | |
| CN111245024A (en) | Comprehensive energy system robust optimization operation method based on model predictive control | |
| CN115275983A (en) | Photovoltaic power fluctuation stabilizing control method based on distributed resource clustering | |
| CN115795992A (en) | Park energy Internet online scheduling method based on virtual deduction of operation situation | |
| CN109976155B (en) | Method and system for randomly optimizing and controlling interior of virtual power plant participating in gas-electricity market | |
| CN115775046A (en) | Virtual power plant optimal scheduling method and system, electronic equipment and storage medium | |
| CN117239740B (en) | Optimal configuration and flexibility improvement method and system for virtual power plant system | |
| CN117767433A (en) | Real-time county energy internet scheduling method and system based on digital twin | |
| Naug et al. | Data driven methods for energy reduction in large buildings | |
| CN117674197B (en) | Frequency adjustment method, storage medium and equipment using virtual power plant active support | |
| CN112039058A (en) | Unit combination method, system, medium and device based on wind power prediction interval | |
| CN111049195A (en) | AGC parameter optimization method and device | |
| Gong | Aggregated load forecast and control for creating alternative power system resources using thermostatically controlled loads | |
| CN115051374A (en) | Control method and device for electric heating equipment to participate in electric power peak regulation and storage medium | |
| CN115347583A (en) | Energy internet power instruction distribution method and system based on multiple intelligent agents | |
| CN115115276A (en) | Virtual power plant scheduling method and system considering uncertainty and privacy protection | |
| CN114819573A (en) | Day-ahead scheduling method and device of new energy system and computer equipment | |
| CN114552634A (en) | Distributed power supply large-scale grid-connected multi-target dynamic cluster division method and device | |
| CN110717694A (en) | Energy storage configuration random decision method and device based on new energy consumption expected value | |
| CN113779862B (en) | Power electronic flexible switch access planning method, device, equipment and storage medium | |
| Kousounadis-Knousen et al. | A New Co-Optimized Hybrid Model Based on Multi-Objective Optimization for Probabilistic Wind Power Forecasting in a Spatiotemporal Framework |
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 |