CN116865343B - Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network - Google Patents
Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network Download PDFInfo
- Publication number
- CN116865343B CN116865343B CN202311117041.4A CN202311117041A CN116865343B CN 116865343 B CN116865343 B CN 116865343B CN 202311117041 A CN202311117041 A CN 202311117041A CN 116865343 B CN116865343 B CN 116865343B
- Authority
- CN
- China
- Prior art keywords
- distribution network
- distributed photovoltaic
- power distribution
- voltage
- photovoltaic power
- 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
- 238000009826 distribution Methods 0.000 title claims abstract description 176
- 238000000034 method Methods 0.000 title claims abstract description 66
- 230000003044 adaptive effect Effects 0.000 claims description 28
- 230000008569 process Effects 0.000 claims description 20
- 239000003795 chemical substances by application Substances 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 17
- 230000009467 reduction Effects 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 4
- 238000013480 data collection Methods 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 abstract description 12
- 238000010248 power generation Methods 0.000 abstract description 11
- 238000010521 absorption reaction Methods 0.000 abstract description 3
- 230000001105 regulatory effect Effects 0.000 abstract description 2
- 230000035945 sensitivity Effects 0.000 description 13
- 230000001276 controlling effect Effects 0.000 description 9
- 238000013499 data model Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004134 energy conservation Methods 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012826 global research Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0205—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system
- G05B13/024—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- 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
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- 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
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00032—Systems characterised by the controlled or operated power network elements or equipment, the power network elements or equipment not otherwise provided for
-
- 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
-
- 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)
- Power Engineering (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides a model-free self-adaptive control method, device and medium for a distributed photovoltaic power distribution network, which are suitable for the technical field of power distribution network regulation and control and can improve the operation safety and stability of the power distribution network and the absorption rate of distributed photovoltaic power generation. The method comprises the following steps: the distributed photovoltaic power distribution network is divided into a plurality of areas, the similarity between different nodes in the same area is higher than that between nodes in different areas, and then all nodes in any area can be regulated and controlled based on a similar monitoring mode, such as a Gaussian agent model and/or an MFA control model.
Description
Technical Field
The invention belongs to the technical field of power distribution network regulation and control, and particularly relates to a model-free self-adaptive control method, device and medium for a distributed photovoltaic power distribution network.
Background
Compared with the traditional fossil energy, the distributed power supply (Distributed Generation, DG) adopts renewable energy sources to generate power, has the characteristics of environmental friendliness and the like, and has become a global research hotspot. Distributed photovoltaic is one of important forms of DG, and large-scale access and full-scale absorption are current development trends. However, the distributed photovoltaic access voltage level is low, the monomer capacity is small, the quantity is large, the data acquisition is insufficient, the distributed photovoltaic access voltage level has the characteristics of strong randomness, intermittence and fluctuation, the existing distribution network has a single regulation and control mode for the distributed photovoltaic, the voltage fluctuation of the distribution network is more obvious along with the continuous improvement of the permeability of the distribution network, the electric energy quality is continuously deteriorated, and a plurality of challenges are brought to the safe, economic and stable operation of the distribution network.
Aiming at the actual situation that the distributed photovoltaic in the existing low-voltage distribution area lacks effective regulation, the control mode of the distributed photovoltaic at present mainly comprises on-site control and centralized monitoring mode by accessing a power distribution network regulation system. The centralized monitoring mode is to connect a plurality of distributed photovoltaic power stations to a central control system, and perform centralized monitoring and management on the power stations through the central control system. Although centralized control can improve the regulation and control efficiency of the distributed photovoltaic power distribution network, the problems of high system complexity, insufficient regulation and control precision and the like exist at the same time. The existing centralized regulation and control mode is difficult to meet the regulation and control management requirements of the distributed photovoltaic power distribution network in aspects of various regulation and control resources, data acquisition and the like.
Disclosure of Invention
The invention provides a model-free self-adaptive control method, a model-free self-adaptive control device and a model-free self-adaptive control medium for a distributed photovoltaic power distribution network, which can effectively improve the operation safety and stability of the power distribution network, can also improve the absorption rate of distributed photovoltaic power generation, and achieve the purposes of energy conservation and emission reduction.
Aiming at the problems, the invention adopts the following technical scheme:
in a first aspect, a model-free adaptive control method for a distributed photovoltaic power distribution network is provided, including:
Dividing the distributed photovoltaic power distribution network into a plurality of areas, wherein the similarity between different nodes in the same area is higher than the similarity between nodes in different areas;
inputting historical operation data of a first area into a Gaussian agent model, and determining parameters of the Gaussian agent model, wherein the first area is any one area of a plurality of areas;
inputting current operation data of the first area into a Gaussian agent model, and determining a voltage predicted value of the first area;
the output power of each node of the first region is controlled based on the voltage prediction value.
Optionally, dividing the distributed photovoltaic power distribution network into a plurality of regions includes:
determining voltage regulation capability of the distributed photovoltaic power distribution network and electrical distances between different nodes in the distributed photovoltaic power distribution network, wherein the voltage regulation capability comprises integral voltage regulation capability of the distributed photovoltaic power distribution network and voltage regulation capability of each node in the distributed photovoltaic power distribution network, and the electrical distances represent connection compactness between the different nodes in the distributed photovoltaic power distribution network;
based on voltage regulation capability and electrical distance, the distributed photovoltaic power distribution network is divided into a plurality of areas.
Further, based on the voltage regulation capability and the electrical distance, dividing the distributed photovoltaic power distribution network into a plurality of areas includes:
Based on the voltage regulation capability and the electric distance, acquiring a clustering result of each node in the distributed photovoltaic power distribution network;
based on clustering results, the distributed photovoltaic power distribution network is divided into a plurality of areas, and the connection closeness between different nodes in the same area is greater than that between nodes in different areas.
Optionally, controlling the output power of each node of the first region based on the voltage prediction value includes:
based on the voltage forecast, the output power of the generator and/or the load of the utility network is adjusted.
In a second aspect, a model-free adaptive control method for a distributed photovoltaic power distribution network is provided, including:
dividing the distributed photovoltaic power distribution network into a plurality of areas, wherein the similarity between different nodes in the same area is higher than the similarity between nodes in different areas;
training a model-free adaptive MFA control model based on historical operation data and a dynamic linearization data model of a second region, wherein the second region is any one of a plurality of regions, and the model-free adaptive MFA control model is used for learning dynamic characteristics of the distributed photovoltaic power distribution network;
inputting current operation data of the second area into an MFA control model, and acquiring a predicted value of the distributed photovoltaic power distribution network, wherein the predicted value corresponds to the dynamic characteristic of the distributed photovoltaic power distribution network;
Based on the predicted value, the output voltages of the nodes in the second region are controlled.
Optionally, dividing the distributed photovoltaic power distribution network into a plurality of regions includes:
acquiring voltage regulation capability of the distributed photovoltaic power distribution network and electrical distances among different nodes in the distributed photovoltaic power distribution network, wherein the voltage regulation capability comprises integral voltage regulation capability of the distributed photovoltaic power distribution network and voltage regulation capability of each node in the distributed photovoltaic power distribution network, and the electrical distances represent connection compactness among the different nodes in the distributed photovoltaic power distribution network;
based on voltage regulation capability and electrical distance, the distributed photovoltaic power distribution network is divided into a plurality of areas.
Further, based on the voltage regulation capability and the electrical distance, dividing the distributed photovoltaic power distribution network into a plurality of areas includes:
determining a clustering result of each node in the distributed photovoltaic power distribution network based on the voltage regulation capability and the electric distance;
based on clustering results, the distributed photovoltaic power distribution network is divided into a plurality of areas, and the connection closeness between different nodes in the same area is greater than that between nodes in different areas.
Optionally, controlling the output voltage of each node in the second region based on the predicted value includes:
Acquiring deviation between a current value and a predicted value of a distributed photovoltaic power distribution network;
the output voltages of the nodes in the second region are adjusted based on the deviation between the current value and the predicted value.
In a third aspect, a model-free adaptive control device for a distributed photovoltaic power distribution network is provided, including: the device comprises a dividing module, a determining module and a control module; wherein,
the division module is used for dividing the distributed photovoltaic power distribution network into a plurality of areas, and the similarity between different nodes in the same area is higher than the similarity between the nodes in different areas;
the determining module is used for inputting the historical operation data of the first area into the Gaussian proxy model, determining parameters of the Gaussian proxy model, and enabling the first area to be any one of the areas;
the determining module is also used for inputting the current operation data of the first area into the Gaussian agent model and determining the voltage predicted value of the first area;
and the control module is used for controlling the output power of each node of the first area based on the voltage predicted value.
Optionally, the determining module is further configured to determine a voltage regulation capability of the distributed photovoltaic power distribution network and an electrical distance between different nodes in the distributed photovoltaic power distribution network, where the voltage regulation capability includes an overall voltage regulation capability of the distributed photovoltaic power distribution network and a voltage regulation capability of each node in the distributed photovoltaic power distribution network, and the electrical distance represents connection compactness between the different nodes in the distributed photovoltaic power distribution network;
The division module is further used for dividing the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electric distance.
Further, the determining module is further used for determining a clustering result of each node in the distributed photovoltaic power distribution network based on the voltage regulation capability and the electric distance;
the division module is further used for dividing the distributed photovoltaic power distribution network into a plurality of areas based on the clustering result, and the connection closeness between different nodes in the same area is greater than the connection closeness between the nodes in different areas.
Optionally, the control module is further configured to adjust the output power of the generator and/or the load of the utility network based on the voltage prediction value.
In a fourth aspect, a model-free adaptive control device for a distributed photovoltaic power distribution network is provided, including: the device comprises a dividing module, a training module, an acquisition module and a control module; wherein,
the division module is used for dividing the distributed photovoltaic power distribution network into a plurality of areas, and the similarity between different nodes in the same area is higher than the similarity between the nodes in different areas;
the training module is used for training a model-free self-adaptive MFA control model based on historical operation data and a dynamic linearization data model of a second area, wherein the second area is any area in the plurality of areas, and the model-free self-adaptive MFA control model is used for learning dynamic characteristics of the distributed photovoltaic power distribution network;
The acquisition module is used for inputting the current operation data of the second area into an MFA control model, and acquiring a predicted value of the distributed photovoltaic power distribution network, wherein the predicted value corresponds to the dynamic characteristic of the distributed photovoltaic power distribution network;
and the control module is used for controlling the output voltage of each node in the second area based on the predicted value.
Optionally, the acquiring module is further configured to acquire voltage adjustment capability of the distributed photovoltaic power distribution network and electrical distances between different nodes in the distributed photovoltaic power distribution network, where the voltage adjustment capability includes an overall voltage adjustment capability of the distributed photovoltaic power distribution network and voltage adjustment capability of each node in the distributed photovoltaic power distribution network, and the electrical distances represent connection compactness between the different nodes in the distributed photovoltaic power distribution network;
the division module is further used for dividing the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electric distance.
Further, the acquisition module is further used for acquiring a clustering result of each node in the distributed photovoltaic power distribution network based on the voltage regulation capability and the electric distance;
the division module is further used for dividing the distributed photovoltaic power distribution network into a plurality of areas based on the clustering result, and the connection closeness between different nodes in the same area is greater than the connection closeness between the nodes in different areas.
Optionally, the obtaining module is further configured to obtain a deviation between a current value and a predicted value of the distributed photovoltaic power distribution network;
the control module is also used for adjusting the output voltage of each node in the second area based on the deviation between the current value and the predicted value.
A fifth aspect provides a model-free adaptive control device for a distributed photovoltaic power distribution network, comprising: the processor may be configured to perform the steps of,
the processor is coupled with the memory;
the processor is configured to read and execute a program or instructions stored in the memory, so that the device executes the model-free adaptive control method of the distributed photovoltaic power distribution network according to the first aspect or the second aspect.
In a sixth aspect, a computer readable storage medium is provided, in which a program or an instruction is stored, which when read and executed by a computer, causes the computer to perform the model-free adaptive control method of a distributed photovoltaic power distribution network according to the first or second aspect.
The model-free self-adaptive control method, device and medium for the distributed photovoltaic power distribution network can divide the distributed power distribution network into a plurality of areas, and the similarity between different nodes in the same area is higher than that between nodes in different areas, so that all nodes in any area can be controlled by adopting a similar monitoring mechanism, such as adjusting output voltage, output power, power grid load and the like, thereby overcoming the problems of high system complexity and low regulation precision in a centralized monitoring mode of the distributed photovoltaic power distribution network, improving the operation safety and stability of the distributed photovoltaic power distribution network, effectively eliminating the negative effects caused by the randomness, intermittence and fluctuation of the distributed photovoltaic power distribution network, improving the consumption rate of the distributed photovoltaic power generation, and further improving the new energy utilization rate to achieve the purposes of energy conservation and emission reduction.
In addition, in the process, accurate monitoring of the distributed photovoltaic power distribution network can be realized without establishing an accurate system model, namely, a model-free monitoring method is provided, and compared with a system model method, the method has the following advantages:
1. the process of developing and identifying the system model is omitted: the method without the system model does not need to build an accurate mathematical model in advance, and avoids complex system model development and recognition processes.
2. The adaptability is strong: the system model-free method can learn the dynamic characteristics of the system in real time or off-line, train and adjust the controller based on the learned dynamic characteristics, has better adaptability and robustness, and is suitable for complex and uncertain systems.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a model-free adaptive control method of a distributed photovoltaic power distribution network according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of another model-free adaptive control method for a distributed photovoltaic power distribution network according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a model-free adaptive control device of a distributed photovoltaic power distribution network according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of another model-free adaptive control device for a distributed photovoltaic power distribution network according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a model-free adaptive control device for a distributed photovoltaic power distribution network according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Firstly, with reference to fig. 1 and fig. 2, a model-free adaptive control method for a distributed photovoltaic power distribution network provided by the embodiment of the invention is described in detail.
Exemplary, fig. 1 is a model-free adaptive control method of a distributed photovoltaic power distribution network according to an embodiment of the present invention. As shown in fig. 1, the method includes:
s101, dividing the distributed photovoltaic power distribution network into a plurality of areas.
The similarity between different nodes in the same area is higher than the similarity between nodes in different areas.
Optionally, dividing the distributed photovoltaic power distribution network into a plurality of regions includes:
step 1, determining voltage regulation capability of a distributed photovoltaic power distribution network and electrical distances among different nodes in the distributed photovoltaic power distribution network, wherein the voltage regulation capability comprises integral voltage regulation capability of the distributed photovoltaic power distribution network and voltage regulation capability of each node in the distributed photovoltaic power distribution network, and the electrical distances represent connection compactness among the different nodes in the distributed photovoltaic power distribution network;
and 2, dividing the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electric distance.
Further, based on the voltage regulation capability and the electrical distance, dividing the distributed photovoltaic power distribution network into a plurality of areas includes:
Based on the voltage regulation capability and the electric distance, acquiring a clustering result of each node in the distributed photovoltaic power distribution network;
based on clustering results, the distributed photovoltaic power distribution network is divided into a plurality of areas, and the connection closeness between different nodes in the same area is greater than that between nodes in different areas.
S102, inputting historical operation data of a first area into a Gaussian agent model, and determining parameters of the Gaussian agent model, wherein the first area is any area in a plurality of areas;
s103, inputting current operation data of the first area into a Gaussian agent model, and determining a voltage predicted value of the first area;
s104, controlling the output power of each node of the first area based on the voltage predicted value.
Optionally, controlling the output power of each node of the first region based on the voltage prediction value includes:
based on the voltage forecast, the output power of the generator and/or the load of the utility network is adjusted.
The following is a detailed description by way of one example.
In a distributed photovoltaic power generation system, the voltage and current of a photovoltaic module are related to factors such as illumination intensity and temperature, and the change of the factors can influence parameters such as output power and voltage of the photovoltaic power generation system. In addition, parameters such as output power and voltage of the photovoltaic power generation system can influence parameters such as voltage of the power distribution network, so that stable operation of the power distribution network is affected. Therefore, in the distributed photovoltaic power generation system, how to ensure that parameters such as voltage of the power distribution network are not out of limit so as to ensure safe operation of the power distribution network is a very important problem.
Based on the data driving technology, the real-time voltage, current, power and other parameters of the distributed photovoltaic power generation system can be used as input data, and the control of the system can be realized by using a model-free self-adaptive control method.
Firstly, collecting real-time data, wherein the real-time data of a distributed photovoltaic power generation system, including parameters such as illumination intensity, temperature, voltage, current and power, are collected to be used as input of a control method;
secondly, preprocessing input data, including data cleaning, feature extraction and other operations, so as to obtain data suitable for control;
then, using Gaussian agent modeling, and sending the input data into the Gaussian agent model for processing and analysis to obtain corresponding control parameters; and then, according to the control parameters output by the Gaussian model, voltage regulation is carried out on the distributed photovoltaic power generation system, and parameters such as output power and voltage of the photovoltaic module are regulated so as to ensure that the parameters such as voltage of the power distribution network are not out of limit.
The method is specifically realized as follows:
gaussian proxy modeling
The existing voltage control algorithm must use the original tide model to conduct rewarding calculation in the training process. To address this problem, a gaussian proxy model may be employed that produces the same input-output relationship as the power flow equation:
（1）
（2）
（3）
（4）
（5）
Wherein a Gaussian process is expressed asBy mean function->Covariance->It is shown that the goal of the Gaussian process is to give a new input +.>Is predicted under the condition of (1). Wherein (1)>Representing training points->And test point->N-dimensional covariance vector between>Representation->Is a variance of (2);the representation is used for representing->And->A kernel function of the correlation between them. />Representing the variance scale>Is the variance of the signal and,is a super parameter set, < >>Represents a covariance matrix (Covariance Matrix). When the super parameter set is fixed and the optimal position of the pseudo set is determined, the mean and covariance of the predicted values can be obtained according to the formula (1) and the formula (2). Since it is deduced from the data, no physical power flow model is required.
Voltage regulation problem is solved in subregion of distributed photovoltaic distribution network
The aim is to divide the entire distributed photovoltaic power distribution network into several areas so that the centralized voltage regulation problem is divided into several sub-problems that can be solved in a distributed manner. While considering the regional voltage regulation capability, determining the electrical distance based on the voltage sensitivity and aggregating nodes with similar properties based on the electrical distance. Wherein the electrical distance is defined based on a voltage-active power sensitivity and a voltage-reactive power sensitivity matrix The following are provided:
（6）
wherein,for the electrical distance between node i and node j, < >>For the voltage-active power sensitivity between node i and node j, +.>Is the voltage-reactive power sensitivity between node i and node j.
Equation (6) defines the electrical distanceIt is based on voltage-active power sensitivity (+)>) And voltage-reactive power sensitivity (+)>) Calculated. By decomposing the voltage regulation problem into distributed sub-problems, the electrical distance is used to measure the connection closeness between nodes. Wherein->And->Respectively, the electrical distance defined based on the voltage-active power sensitivity and the voltage-reactive power sensitivity, and is specifically shown as the following formula:
（7）
equation (7) gives a method of calculating the electrical distance based on the voltage-active power sensitivity and the voltage-reactive power sensitivity. These sensitivities represent the degree to which the voltage at node i is sensitive to the injected active and reactive power at node j. Wherein,and->The sensitivity of the voltage at node i to the injected active power and the injected reactive power at node j is shown, respectively. Electric distance d ij The smaller the electrical connection between nodes i and j will be, the tighter.
Further, the voltage regulation capability is defined as:
（8）
Equation (8) defines the wholeVoltage regulation capability (c) of network and region voltage regulation capability (c) of each region (cluster k) k ) Relationship between them. It is based on the maximum reactive power (Q i ) And active power (P i ) The reduction and the voltage deviation between nodes (Δvj) are calculated. Where c is the voltage regulation capability of the whole network, c k Voltage regulation capability for region k;voltage deviation of node with maximum voltage violation value Q i And P i The maximum reactive power and active power curtailment that can be used by node i for voltage regulation, respectively. clip function restriction->The value of (2) is [0,1 ]]Is not limited in terms of the range of (a).
The performance index of the cluster is defined based on the modularity index:
(9)
equation (9) defines a cluster performance index (ρ) based on a modularity index, which measures the closeness of the clustering result, wherein the closer the electrical connections between nodes, the tighter the connections within the cluster, and the looser the connections between clusters. The index can be used to search for the best cluster while maximizing defined performance. Wherein the method comprises the steps ofRepresenting the sum of the weights of all nodes; />Representing the weight of node j. For performance indicators, the larger the value, the tighter the electrical connection within the cluster and the looser the connection between clusters. The algorithm may be used to search for the best cluster while maximizing defined performance.
The Gaussian agent model is used for voltage prediction and power control, and the method is concretely as follows:
and (3) data collection: data is collected including time, load level, weather conditions, and corresponding voltage measurements (or power measurements). The input variable is noted as x and the voltage measurement is noted as y.
Feature selection: suitable input features of the collected data are selected, such as time stamp t, seasonal index s, load level l, temperature t and humidity h. The input eigenvector is noted as x= [ t, s, l, t, h ].
Training a Gaussian agent model: using the collected data, parameters of a mean function m (x) and a covariance function k (x, x') of the gaussian process are estimated. The Gaussian process is GP (m (x), k (x, x').
Model prediction: given new input features(e.g., current time of day operational data), voltage predictions are made using a gaussian proxy model. The predicted voltage value is +.>The estimated variance is->. I.e. calculate condition distribution +.>。
And (3) power control: based on predicted voltage valuesAnd performing power control operation. The specific manner of operation will depend on the application scenario, for example, the output power of the generator may be adjusted or the load of the grid may be controlled to regulate the output power of all nodes in the area.
Fig. 2 is a schematic diagram of another model-free adaptive control method for a distributed photovoltaic power distribution network according to an embodiment of the present invention. As shown in fig. 2, the method includes:
S201, dividing the distributed photovoltaic power distribution network into a plurality of areas, wherein the similarity between different nodes in the same area is higher than the similarity between the nodes in different areas.
Specific implementation may refer to S101, and will not be described herein.
S202, training a model-free self-adaptive MFA control model based on historical operation data and a dynamic linearization data model of a second area, wherein the second area is any one of a plurality of areas, and the model-free self-adaptive MFA control model is used for learning dynamic characteristics of a distributed photovoltaic power distribution network.
And S203, inputting the current operation data of the second area into an MFA control model, and acquiring a predicted value of the distributed photovoltaic power distribution network, wherein the predicted value corresponds to the dynamic characteristic of the distributed photovoltaic power distribution network.
S204, controlling the output voltage of each node in the second area based on the predicted value.
Optionally, controlling the output voltage of each node in the second region based on the predicted value includes:
acquiring deviation between a current value and a predicted value of a distributed photovoltaic power distribution network;
the output voltages of the nodes in the second region are adjusted based on the deviation between the current value and the predicted value.
The following is a detailed description by way of one example.
Collecting and processing data: firstly, operation data of the distributed photovoltaic system are required to be collected, and the data are processed, including operations such as data cleaning, sampling and preprocessing. These processing operations may improve data quality and availability, making subsequent data-driven modeling more accurate and reliable.
Establishing a dynamic linearization data model: for the complex discrete time nonlinear characteristics of a distributed photovoltaic system, an equivalent dynamic linearization data model needs to be built at each dynamic operating point. The model may be obtained through a series of mathematical operations and data processing techniques such as interpolation, regression, least squares, and the like. The dynamic linearization data model is as follows:
（10）
（11）
（12）
where k is a time step or discrete time index, representing the current point in time or time step, u (k) is a control input or manipulated variable, representing the control input at time step k, u (k-1) is the control input at time step (k-1),for parameter estimation values, representing parameter estimation at time step k, y (k) is the state variable or system output, representing the state variable or system output at time step k,/>A desired state variable or system output, representing the desired state variable or system output at time step (k+1), - >Predicted values for the state variables or the system outputs, representing the predicted values for the state variables or the system outputs at time step k, r being the control gain for adjusting the influence of the changes of the control inputs on the model, +.>For the state variable or the estimated value of the system output, it is indicated at time step k that θ is a parameter for balancing the trade-off between the control input variation and the state variable estimation error, α is a parameter for balancing the speed and stability of the parameter estimation, b is a parameter indicating a constant, Δu (k-1) is the control input variation, and it is indicated between time step (k-1) and time step k.
The derivation process is as follows:
firstly, according to the Taylor expansion of a discrete nonlinear system and combining a differential median theorem, a new dynamic linearization model is given:
（13）
wherein the method comprises the steps ofAnd->Is a time-varying parameter of a distributed photovoltaic system, < >>With this time-varying parameter, equation (13) can be expressed as follows:
（14）
wherein,usingAbout->Around->The taylor series expansion and differential median theorem of (c) can be obtained:
（15）
wherein,，/> let->The sign of (2) is known and without loss of generality, let +. >The following formula must exist:
（16）
wherein B and B are two positive constants. Assume that=0, we define:
（17）
（18）
assume thatIs->Estimate of->Is->Substituting equations (17) and (18) into equation (15) to obtain:
（19）
wherein,，/>including the estimation error of the model and possible disturbances of the system. Definition:
（20）
wherein the method comprises the steps ofIs an estimate of d (k),>is the output of the approximation model (16).
Then, equation (19) can be rewritten as:
（21）
by introducing the idea of in-mold control according to equations (19) - (21), we can define two cost functions as follows:
（22）
（23）
wherein alpha is>0，θ>0 is a weighting coefficient, r>0 is the influence coefficient of model estimation error, and the introduced termIn order to reduce the effects of model estimation errors and disturbances that may be present in the system by using an in-mold control architecture. Then, the partial derivatives of (22) and (23) are made equal to 0, resulting in equations (10) and (11). From (16) and (17), it is known that +.>So that +.sup.f should be satisfied in the estimation algorithm (11) of β (k)>. Thus, +.>Is set to 1/b and an update algorithm of formula (12) is proposed. To this end, a new MFA control algorithm is derived consisting of (10), (11) and (12).
Training an MFA controller using on-line or off-line learning
On the basis of the dynamic linearization data model, the MFA controller can be trained in an online or offline learning mode. The MFA controller is a controller based on fuzzy logic and has good robustness and adaptability. The goal of the training is to learn the dynamics of the system to obtain a more accurate and reliable control strategy, thereby achieving control of the adjustable quantity and node voltage. The model predictive controller uses the following formula:
objective function:
（24）
the objective function is used for model predictive controllers. It is a cost function that by minimizing it, a more accurate and reliable control strategy can be obtained. The objective function includes the active power variation delta P k Reactive power variation delta Q k And a voltage variation DeltaV k Where Q, R and S are weight matrices.
Power balance constraint:
（25）
these constraints are used to ensure power balance of the power system. The left side is the sum of the active power and reactive power at the current moment and the variation thereof, and the right side is zero, indicating the power balance.
Inequality constraint:
（26）
these constraints are used to limit the active power P i Reactive power Q i Voltage V i Current I i Is a range of values. They ensure the stability and safety of the power system.
Relationship between control variable and state variable:
（27）
these equations describe the control variables (ΔP, ΔQ, ΔV) and the state variables (P i 、Q i 、V i ) Relationship between them. By calculating the difference between the control variable and the state variable, the deviation between the adjustment quantity and the expected value can be determined, thereby realizing the control of the adjustable quantity and the node voltage. Wherein P, Q, V respectively represent the active, reactive and voltage values at the current moment,，/>，/>and respectively representing expected active, reactive and voltage values, wherein N is the number of nodes.
Evaluating and optimizing controller performance: finally, performance evaluation and optimization are required to be carried out on the MFA controller obtained through training so as to verify the effectiveness and feasibility of the MFA controller. The evaluation can be performed by means of simulation, actual test and the like, and the optimization needs to be adjusted and optimized according to actual conditions and control targets.
According to the model-free self-adaptive control method for the distributed photovoltaic power distribution network, the distributed power distribution network can be divided into a plurality of areas, and the similarity between different nodes in the same area is higher than that between nodes in different areas, so that all nodes in any area can be adjusted by adopting a similar monitoring mechanism for all nodes in any area, such as adjusting output voltage, output power, power grid load and the like, the problems of high system complexity and low regulation precision in a centralized monitoring mode of the distributed photovoltaic power distribution network are solved, the operation safety and stability of the power distribution network are improved, the negative effects caused by the randomness, intermittence and fluctuation of the distributed photovoltaic power are effectively eliminated, the consumption rate of the distributed photovoltaic power generation is improved, the new energy utilization rate is improved, and the purposes of energy conservation and emission reduction are achieved.
In addition, in the process, accurate monitoring of the distributed photovoltaic power distribution network can be realized without establishing an accurate system model, namely, a model-free monitoring method is provided, and compared with a system model method, the method has the following advantages:
1. the process of developing and identifying the system model is omitted: the method without the system model does not need to build an accurate mathematical model in advance, and avoids complex system model development and recognition processes.
2. The adaptability is strong: the system model-free method can learn the dynamic characteristics of the system in real time or off-line, train and adjust the controller based on the learned dynamic characteristics, has better adaptability and robustness, and is suitable for complex and uncertain systems.
The model-free self-adaptive control method of the distributed photovoltaic power distribution network provided by the embodiment of the invention is described in detail above with reference to fig. 1 and 2, and the model-free self-adaptive control device of the distributed photovoltaic power distribution network provided by the embodiment of the invention is described below with reference to fig. 3 to 5.
Fig. 3 is a schematic structural diagram of a model-free adaptive control device of a distributed photovoltaic power distribution network according to an embodiment of the present invention. The device can execute the model-free self-adaptive control method of the distributed photovoltaic power distribution network shown in the figure 1.
As shown in fig. 3, the apparatus 300 includes: a dividing module 301, a determining module 302 and a control module 303; wherein,
the dividing module 301 is configured to divide the distributed photovoltaic power distribution network into a plurality of areas, where the similarity between different nodes in the same area is higher than the similarity between nodes in different areas;
a determining module 302, configured to input historical operating data of a first area into the gaussian proxy model, determine parameters of the gaussian proxy model, and the first area is any one area of the plurality of areas;
the determining module 302 is further configured to input current operation data of the first area into a gaussian proxy model, and determine a voltage prediction value of the first area;
and a control module 303, configured to control output power of each node of the first area based on the voltage prediction value.
Optionally, the determining module 302 is further configured to determine a voltage regulation capability of the distributed photovoltaic power distribution network and an electrical distance between different nodes in the distributed photovoltaic power distribution network, where the voltage regulation capability includes an overall voltage regulation capability of the distributed photovoltaic power distribution network and a voltage regulation capability of each node in the distributed photovoltaic power distribution network, and the electrical distance represents connection compactness between the different nodes in the distributed photovoltaic power distribution network;
The dividing module 301 is further configured to divide the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electrical distance.
Further, the determining module 302 is further configured to determine a clustering result of each node in the distributed photovoltaic power distribution network based on the voltage adjustment capability and the electrical distance;
the division module 301 is further configured to divide the distributed photovoltaic power distribution network into a plurality of areas based on the clustering result, where the connection closeness between different nodes in the same area is greater than the connection closeness between nodes in different areas.
Optionally, the control module 303 is further configured to adjust the output power of the generator and/or the load of the utility network based on the voltage prediction value.
Fig. 4 is a schematic structural diagram of another model-free adaptive control device for a distributed photovoltaic power distribution network according to an embodiment of the present invention. The device can execute the model-free self-adaptive control method of the distributed photovoltaic power distribution network shown in the figure 2.
As shown in fig. 4, the apparatus 400 includes:
a dividing module 401, a training module 402, an obtaining module 403 and a control module 404; wherein,
the dividing module 401 is configured to divide the distributed photovoltaic power distribution network into a plurality of areas, where the similarity between different nodes in the same area is higher than the similarity between nodes in different areas;
The training module 402 is configured to train a model-free adaptive MFA control model based on historical operation data and a dynamic linearization data model of a second area, where the second area is any one of the plurality of areas, and the model-free adaptive MFA control model is configured to learn dynamic characteristics of the distributed photovoltaic power distribution network;
the obtaining module 403 is configured to input current operation data of the second area into the MFA control model, and obtain a predicted value of the distributed photovoltaic power distribution network, where the predicted value corresponds to a dynamic characteristic of the distributed photovoltaic power distribution network;
a control module 404 for controlling the output voltages of the nodes in the second region based on the predicted values.
Optionally, the obtaining module 403 is further configured to obtain a voltage adjustment capability of the distributed photovoltaic power distribution network and an electrical distance between different nodes in the distributed photovoltaic power distribution network, where the voltage adjustment capability includes an overall voltage adjustment capability of the distributed photovoltaic power distribution network and a voltage adjustment capability of each node in the distributed photovoltaic power distribution network, and the electrical distance represents connection compactness between the different nodes in the distributed photovoltaic power distribution network;
the dividing module 401 is further configured to divide the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electrical distance.
Further, the obtaining module 403 is further configured to obtain a clustering result of each node in the distributed photovoltaic power distribution network based on the voltage adjustment capability and the electrical distance;
the division module 401 is further configured to divide the distributed photovoltaic power distribution network into a plurality of areas based on the clustering result, where the connection closeness between different nodes in the same area is greater than the connection closeness between nodes in different areas.
Optionally, the obtaining module 403 is further configured to obtain a deviation between a current value and a predicted value of the distributed photovoltaic power distribution network;
the control module 404 is further configured to adjust the output voltage of each node in the second region based on the deviation between the current value and the predicted value.
Fig. 5 is a schematic structural diagram of a model-free adaptive control device of a distributed photovoltaic power distribution network according to an embodiment of the present invention.
As shown in fig. 5, the apparatus 500 includes: a processor 501, the processor 501 coupled to the memory 502;
the processor 501 is configured to read and execute a program or instructions stored in the memory 502, so that the apparatus 500 performs a model-free adaptive control method of the distributed photovoltaic power distribution network as shown in fig. 1 or fig. 2.
Optionally, the apparatus 500 may further comprise a transceiver 503 for the apparatus 500 to communicate with other apparatuses.
It should be noted that, for convenience of description, fig. 3 to 5 only show main components of the model-free adaptive control device of the distributed photovoltaic power distribution network. In practical applications, the model-free adaptive control device of the distributed photovoltaic power distribution network may further include a part or component not shown in the figure.
The embodiment of the invention also provides a computer readable storage medium, which stores a program or instructions, and when the program or instructions are read and executed by a computer, the computer is caused to execute the model-free adaptive control method of the distributed photovoltaic power distribution network.
Although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (4)
1. The model-free self-adaptive control method for the distributed photovoltaic power distribution network is characterized by comprising the following steps of:
dividing the distributed photovoltaic power distribution network into a plurality of areas, wherein the similarity between different nodes in the same area is higher than the similarity between nodes in different areas;
Inputting historical operation data of a first region into a Gaussian proxy model, and determining parameters of the Gaussian proxy model, wherein the first region is any region in the plurality of regions; using the gaussian proxy model, the same input-output relationship as the power flow equation is generated:
（1），
（2），
（3），
（4），
（5），
wherein a Gaussian process is expressed asBy mean function->And covariance function->Representing, the goal of the Gaussian process is to give a new input/>Prediction +.>；
In the formulae (1) to (5):representing training points->And test point->A covariance function between; />Representation->Is a variance of (2); />Representing a variance scale; />Is the signal variance; />Is a super parameter set; />Representing a covariance matrix;
inputting current operation data of the first area into the Gaussian agent model, and determining a voltage predicted value of the first area;
controlling output power of each node of the first region based on the voltage predicted value;
the dividing the distributed photovoltaic power distribution network into a plurality of areas includes:
determining a voltage regulation capability of the distributed photovoltaic power distribution network and an electrical distance between different nodes in the distributed photovoltaic power distribution network, wherein the voltage regulation capability comprises an overall voltage regulation capability of the distributed photovoltaic power distribution network and a voltage regulation capability of each node in the distributed photovoltaic power distribution network, and the electrical distance represents connection compactness between the different nodes in the distributed photovoltaic power distribution network;
Dividing the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electrical distance;
the dividing the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electrical distance includes:
based on the voltage regulation capability and the electrical distance, acquiring a clustering result of each node in the distributed photovoltaic power distribution network; the performance index of the cluster is defined based on the modularity index:
(6)，
in formula (6):representing the sum of the weights of all nodes; />A weight representing node j;
dividing the distributed photovoltaic power distribution network into a plurality of areas based on the clustering result, wherein the connection closeness between different nodes in the same area is greater than that between nodes in different areas;
the voltage regulation capability is defined as:
(7)，
wherein the above formula (7) defines a relationship between the voltage regulation capability of the entire network and the area voltage regulation capability of each area, which is calculated based on the maximum reactive power and the active power reduction of the nodes and the voltage deviation between the nodes; in the formula (7): c is the voltage regulation capability of the whole network, c k Voltage regulation capability for region k; deltaV j Voltage deviation for a node having a maximum voltage violation value; q (Q) i And P i Maximum reactive power and active power reduction for voltage regulation by node i, respectively, the clip function limit takes on values of [0,1]Is defined by the range of (2);
and carrying out voltage prediction and power control by using the Gaussian agent model, wherein the voltage prediction and power control are specifically as follows:
and (3) data collection: collecting data comprising time, load level, weather conditions and corresponding voltage measurements; recording an input variable as x and a voltage measured value as y;
feature selection: selecting proper input characteristics of the collected data, including a time stamp t, a seasonal index s, a load level l, a temperature t and a humidity h, and recording an input characteristic vector as x= [ t, s, l, t, h ];
training a Gaussian agent model: using the collected data, estimating parameters of a mean function m (x) and a covariance function k (x, x') of the gaussian process; the Gaussian process is GP (m (x), k (x, x'));
model prediction: given new input featuresPerforming voltage prediction by using a Gaussian agent model; the predicted voltage value is +.>The estimated variance is->I.e. calculate condition distribution +.>；
Power control: based on predicted voltage valuesAnd performing power control operation, and adjusting the output power of the generator and/or controlling the load of a commercial power network to regulate and control the output power of all nodes in the area.
2. The utility model provides a no model self-adaptation controlling means of distributed photovoltaic distribution network which characterized in that includes: the device comprises a dividing module, a determining module and a control module; wherein,
the dividing module is used for dividing the distributed photovoltaic power distribution network into a plurality of areas, and the similarity between different nodes in the same area is higher than the similarity between the nodes in different areas;
the determining module is used for inputting historical operation data of a first area into the Gaussian proxy model, and determining parameters of the Gaussian proxy model, wherein the first area is any area in the plurality of areas; using the gaussian proxy model, the same input-output relationship as the power flow equation is generated:
（1），
（2），
（3），
（4），
（5），
wherein one is highThe process is expressed asBy mean function->And covariance function->It is shown that the goal of the Gaussian process is to give a new input +.>Is predicted under the condition of (1)；
In the formulae (1) to (5):representing training points->And test point->A covariance function between; />Representation->Is a variance of (2); />Representing a variance scale; />Is the signal variance; />Is prepared from radix Ginseng RubraA number set; />Representing a covariance matrix; the determining module is further configured to input current operation data of the first area into the gaussian proxy model, and determine a voltage prediction value of the first area;
The control module is used for controlling the output power of each node of the first area based on the voltage predicted value;
the determining module is further configured to determine a voltage regulation capability of the distributed photovoltaic power distribution network and an electrical distance between different nodes in the distributed photovoltaic power distribution network, where the voltage regulation capability includes an overall voltage regulation capability of the distributed photovoltaic power distribution network and a voltage regulation capability of each node in the distributed photovoltaic power distribution network, and the electrical distance represents a connection compactness between the different nodes in the distributed photovoltaic power distribution network;
the dividing module is further configured to divide the distributed photovoltaic power distribution network into a plurality of areas based on the voltage regulation capability and the electrical distance;
the determining module is further configured to determine a clustering result of each node in the distributed photovoltaic power distribution network based on the voltage regulation capability and the electrical distance; the performance index of the cluster is defined based on the modularity index:
(6)，
in formula (6):representing the sum of the weights of all nodes; />A weight representing node j;
the dividing module is further configured to divide the distributed photovoltaic power distribution network into a plurality of areas based on the clustering result, where the connection closeness between different nodes in the same area is greater than the connection closeness between nodes in different areas;
The voltage regulation capability is defined as:
(7)，
wherein the above formula (7) defines a relationship between the voltage regulation capability of the entire network and the area voltage regulation capability of each area, which is calculated based on the maximum reactive power and the active power reduction of the nodes and the voltage deviation between the nodes; in the formula (7): c is the voltage regulation capability of the whole network, c k Voltage regulation capability for region k; deltaV j Voltage deviation for a node having a maximum voltage violation value; q (Q) i And P i Maximum reactive power and active power reduction for voltage regulation by node i, respectively, the clip function limit takes on values of [0,1]Is defined by the range of (2);
and carrying out voltage prediction and power control by using the Gaussian agent model, wherein the voltage prediction and power control are specifically as follows:
and (3) data collection: collecting data comprising time, load level, weather conditions and corresponding voltage measurements; recording an input variable as x and a voltage measured value as y;
feature selection: selecting proper input characteristics of the collected data, including a time stamp t, a seasonal index s, a load level l, a temperature t and a humidity h, and recording an input characteristic vector as x= [ t, s, l, t, h ];
training a Gaussian agent model: using the collected data, estimating parameters of a mean function m (x) and a covariance function k (x, x') of the gaussian process; the Gaussian process is GP (m (x), k (x, x'));
Model prediction: given new input featuresPerforming voltage prediction by using a Gaussian agent model; the predicted voltage value is +.>The estimated variance is->I.e. calculate condition distribution +.>；
And (3) power control: based on predicted voltage valuesAnd performing power control operation, and adjusting the output power of the generator and/or controlling the load of a commercial power network to regulate and control the output power of all nodes in the area.
3. The utility model provides a no model self-adaptation controlling means of distributed photovoltaic distribution network which characterized in that includes: the processor may be configured to perform the steps of,
the processor is coupled with the memory;
wherein the processor is configured to read and execute the program or instructions stored in the memory, so that the apparatus performs the model-free adaptive control method of the distributed photovoltaic power distribution network according to claim 1.
4. A computer-readable storage medium, characterized in that a program or instructions is stored, which, when read and executed by a computer, cause the computer to perform the model-free adaptive control method of a distributed photovoltaic power distribution network according to claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311117041.4A CN116865343B (en) | 2023-09-01 | 2023-09-01 | Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311117041.4A CN116865343B (en) | 2023-09-01 | 2023-09-01 | Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116865343A CN116865343A (en) | 2023-10-10 |
| CN116865343B true CN116865343B (en) | 2024-03-29 |
Family
ID=88221880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311117041.4A Active CN116865343B (en) | 2023-09-01 | 2023-09-01 | Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116865343B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117277355B (en) * | 2023-11-09 | 2024-03-22 | 一能电气有限公司 | Intelligent monitoring data power transmission method and system |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102497125A (en) * | 2011-11-30 | 2012-06-13 | 东北大学 | Photovoltaic inversion control device and model free control method based on field programmable gata array (FPGA) |
| CN104615094A (en) * | 2014-11-24 | 2015-05-13 | 国网辽宁省电力有限公司锦州供电公司 | City-class high-density multipoint distributed photovoltaic cluster monitoring method |
| CN105023070A (en) * | 2015-08-12 | 2015-11-04 | 河海大学常州校区 | Output power prediction method of photovoltaic system |
| CN107196333A (en) * | 2017-06-07 | 2017-09-22 | 天津大学 | Distributed photovoltaic assemblage classification method based on modularization index |
| CN109193765A (en) * | 2018-09-17 | 2019-01-11 | 中国农业大学 | A kind of distributed photovoltaic cluster regulation method and device |
| CN110380450A (en) * | 2019-08-13 | 2019-10-25 | 南方电网科学研究院有限责任公司 | A kind of photovoltaic control method, device, equipment and computer readable storage medium |
| CN111327040A (en) * | 2020-03-25 | 2020-06-23 | 上海电力大学 | Data-driven direct-current micro-grid power and voltage control method and device |
| CN111384726A (en) * | 2020-01-21 | 2020-07-07 | 国网安徽省电力有限公司六安供电公司 | High-permeability photovoltaic power distribution network partition voltage regulation method |
| CN111682594A (en) * | 2020-06-15 | 2020-09-18 | 天津大学 | Data-driven model-free adaptive voltage control method for flexible substation of power distribution network |
| US11070056B1 (en) * | 2020-03-13 | 2021-07-20 | Dalian University Of Technology | Short-term interval prediction method for photovoltaic power output |
| CN113381454A (en) * | 2021-06-02 | 2021-09-10 | 国网河北省电力有限公司 | New energy joint debugging method combining ultra-short-term prediction and regional control deviation |
| CN114709874A (en) * | 2022-04-28 | 2022-07-05 | 全球能源互联网集团有限公司 | Distribution network cluster division method considering 5G base station and distributed photovoltaic access |
| CN116365591A (en) * | 2023-05-25 | 2023-06-30 | 北京智盟信通科技有限公司 | Distributed light Fu Qun group control method, device and storage medium |
-
2023
- 2023-09-01 CN CN202311117041.4A patent/CN116865343B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102497125A (en) * | 2011-11-30 | 2012-06-13 | 东北大学 | Photovoltaic inversion control device and model free control method based on field programmable gata array (FPGA) |
| CN104615094A (en) * | 2014-11-24 | 2015-05-13 | 国网辽宁省电力有限公司锦州供电公司 | City-class high-density multipoint distributed photovoltaic cluster monitoring method |
| CN105023070A (en) * | 2015-08-12 | 2015-11-04 | 河海大学常州校区 | Output power prediction method of photovoltaic system |
| CN107196333A (en) * | 2017-06-07 | 2017-09-22 | 天津大学 | Distributed photovoltaic assemblage classification method based on modularization index |
| CN109193765A (en) * | 2018-09-17 | 2019-01-11 | 中国农业大学 | A kind of distributed photovoltaic cluster regulation method and device |
| CN110380450A (en) * | 2019-08-13 | 2019-10-25 | 南方电网科学研究院有限责任公司 | A kind of photovoltaic control method, device, equipment and computer readable storage medium |
| CN111384726A (en) * | 2020-01-21 | 2020-07-07 | 国网安徽省电力有限公司六安供电公司 | High-permeability photovoltaic power distribution network partition voltage regulation method |
| US11070056B1 (en) * | 2020-03-13 | 2021-07-20 | Dalian University Of Technology | Short-term interval prediction method for photovoltaic power output |
| CN111327040A (en) * | 2020-03-25 | 2020-06-23 | 上海电力大学 | Data-driven direct-current micro-grid power and voltage control method and device |
| CN111682594A (en) * | 2020-06-15 | 2020-09-18 | 天津大学 | Data-driven model-free adaptive voltage control method for flexible substation of power distribution network |
| CN113381454A (en) * | 2021-06-02 | 2021-09-10 | 国网河北省电力有限公司 | New energy joint debugging method combining ultra-short-term prediction and regional control deviation |
| CN114709874A (en) * | 2022-04-28 | 2022-07-05 | 全球能源互联网集团有限公司 | Distribution network cluster division method considering 5G base station and distributed photovoltaic access |
| CN116365591A (en) * | 2023-05-25 | 2023-06-30 | 北京智盟信通科技有限公司 | Distributed light Fu Qun group control method, device and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| short-term prediction model to forecast power of photovoltaic based on MFA-Elman;XinYu Ma 等;Energy Reports;20220731;第8卷(第4期);第495-507页 * |
| 风电接入下基于AP聚类的无功功率―电压控制分区方法;周琼;志皓;丰颖;孙景文;;电力系统自动化(13);第19-27、158页 * |
| 高比例新能源接入的配电网集群划分及电压控制;阎怀东 等;电力需求侧管理;20210731;第23卷(第4期);第8-13页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116865343A (en) | 2023-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wen et al. | Real-time identification of power fluctuations based on LSTM recurrent neural network: A case study on Singapore power system | |
| Ge et al. | A hybrid model for short-term PV output forecasting based on PCA-GWO-GRNN | |
| Zhang et al. | Short-term wind speed interval prediction based on artificial intelligence methods and error probability distribution | |
| CN105046374A (en) | Power interval predication method based on nucleus limit learning machine model | |
| Zhou et al. | Comparison of machine-learning models for predicting short-term building heating load using operational parameters | |
| Wang et al. | Prediction of early stabilization time of electrolytic capacitor based on ARIMA-Bi_LSTM hybrid model | |
| CN116865343B (en) | Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network | |
| CN112630659A (en) | Lithium battery SOC estimation method based on improved BP-EKF algorithm | |
| Tian et al. | An adaptive online sequential extreme learning machine for short-term wind speed prediction based on improved artificial bee colony algorithm | |
| CN116914751B (en) | Intelligent power distribution control system | |
| CN113919545A (en) | Photovoltaic power generation power prediction method and system with integration of multiple data models | |
| CN113612237A (en) | Method for positioning resonance-induced subsynchronous oscillation source in offshore wind farm | |
| CN111815039A (en) | Weekly scale wind power probability prediction method and system based on weather classification | |
| CN115995810A (en) | Wind power prediction method and system considering weather fluctuation self-adaptive matching | |
| Alharbi et al. | Short-term wind speed and temperature forecasting model based on gated recurrent unit neural networks | |
| Sun et al. | Interval forecasting for wind speed using a combination model based on multiobjective artificial hummingbird algorithm | |
| Peng et al. | Echo State Network ship motion modeling prediction based on Kalman filter | |
| CN114117852A (en) | Regional heat load rolling prediction method based on finite difference working domain division | |
| Chi et al. | Comparison of two multi-step ahead forecasting mechanisms for wind speed based on machine learning models | |
| CN116826728A (en) | Power distribution network state structure estimation method and system under condition of few measurement samples | |
| Sun et al. | Prediction-based manufacturing center self-adaptive demand side energy optimization in cyber physical systems | |
| CN116862050A (en) | Time sequence network-based daily prediction method, system, storage medium and equipment for carbon emission factors | |
| Ding et al. | A statistical upscaling approach of region wind power forecasting based on combination model | |
| CN115775038A (en) | Short-term load prediction method based on IGWO optimization LSSVM | |
| Ren et al. | Particle swarm optimization based predictive control of proton exchange membrane fuel cell (pemfc) |
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 |