CN114374233A - Method and system for adjusting micro-grid power output based on virtual generator - Google Patents
Method and system for adjusting micro-grid power output based on virtual generator Download PDFInfo
- Publication number
- CN114374233A CN114374233A CN202210279547.4A CN202210279547A CN114374233A CN 114374233 A CN114374233 A CN 114374233A CN 202210279547 A CN202210279547 A CN 202210279547A CN 114374233 A CN114374233 A CN 114374233A
- Authority
- CN
- China
- Prior art keywords
- power output
- power
- load
- value
- microgrid
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
- H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/22—The renewable source being solar energy
- H02J2300/24—The renewable source being solar energy of photovoltaic origin
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
Abstract
The invention discloses a method and a system for adjusting the power output of a microgrid based on a virtual generator, wherein the method comprises the following steps: acquiring a load parameter value of each power output node on the microgrid; resolving an energy output structure of the microgrid based on the load parameter value; inputting the load parameter values associated with the energy output structure and the energy output structure into the twin network model to obtain load regulation parameter values of each power output node; sending the load regulation parameter values to each power output node; each power output node adjusts a respective power output value based on the load adjustment parameter value. The invention adopts a twin network model to match load regulation parameter values for each function output node based on an energy output structure, thereby realizing self-adaptive regulation of the regulation parameters of each function node and enabling each power output node to meet the requirement of optimal power output of a micro-grid.
Description
Technical Field
The invention relates to the technical field of power control, in particular to a method and a system for adjusting micro-grid power output based on a virtual generator.
Background
A microgrid generally refers to a local electrical grid generating and distributing electrical power in an area remote from a large electrical energy production center, having autonomous operability and being located near a consumption area, with limited losses inherent to long-distance distribution grids. With the continuous improvement of the permeability of distributed power supplies such as photovoltaic power, wind power and the like in a power system, the voltage frequency and amplitude of a power grid fluctuate violently, and the power output value of the whole microgrid is unstable, a virtual generator technology is adopted at present, a distributed power supply is provided with energy storage units, and the characteristics of a synchronous motor are simulated through inverter control to increase the inertia of the system, so that the frequency supporting capacity of the system is enhanced. The existing micro-grid system causes the mismatch of power output along with the instability of a consumption end and a production end, so that the power matching between a virtual generator and the micro-grid cannot be well realized.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a method and a system for adjusting the power output of a microgrid based on a virtual generator, which are used for realizing the self-adaptive adjustment of adjustment parameters of each functional node, so that each power output node can meet the requirement of the optimal power output of the microgrid.
In order to solve the above problem, the present invention provides a method for adjusting a microgrid power output based on a virtual generator, the method comprising:
acquiring a load parameter value of each power output node on the microgrid;
resolving an energy output structure of the microgrid based on the load parameter value;
inputting a load parameter value associated with an energy output structure and the energy output structure into a twin network model to obtain a load regulation parameter value of each power output node, training the twin network by using a microgrid public data set, using a parameter trained in the public data set in a transfer learning manner, initializing the twin network, and retraining on the energy output structure data set to obtain a trained twin network;
sending the load regulation parameter values to each power output node;
each power output node adjusts a respective power output value based on the load adjustment parameter value.
Each of the power output nodes includes: one or more regenerative power sources, one or more generators, and an energy storage system; the load parameter values include: load voltage, load current, load power, load frequency.
The energy output structure for resolving the microgrid based on the load parameter values comprises:
acquiring hardware basic structures of all power output nodes of the micro-grid;
and filtering power output nodes without energy output based on the corresponding relation between the load parameter values and the hardware basic structure, and generating an energy output structure.
Inputting the load parameter value and the energy output structure into the twin network model to obtain the load adjusting parameter value of each power output node comprises:
training a twin network by using a microgrid public data set, initializing the twin network by using parameters trained in the public data set in a transfer learning manner, and performing retraining on an energy output structure data set to obtain a trained twin network;
simultaneously sending the load parameter value in the optimal power output state and the load parameter value in the energy output structure into a twin network;
a first sub-network in the twin network receives a load parameter value in a power output optimal state and outputs a first characteristic value with a dimension space; a second sub-network in the twin network receives the load parameter value under the energy output structure and outputs a second characteristic value with a dimensional space;
calculating the distance between the first characteristic value and the second characteristic value, comparing the similarity degree of the first characteristic value and the second characteristic value, and taking the load parameter value with the highest similarity degree in the optimal power output state as an identification result;
and acquiring the load adjusting parameter value of each corresponding power output node under the identification result.
The calculating of the distance between the first characteristic value and the second characteristic value, and the comparing of the similarity between the first characteristic value and the second characteristic value comprises:
and calculating Euclidean distances of the first characteristic value and the second characteristic value to compare the similarity degree of the first characteristic value and the second characteristic value.
The calculating the euclidean distance between the first feature value and the second feature value to compare the similarity between the first feature value and the second feature value includes:
the euclidean distance of the produced feature values is expressed using a contrast loss function.
The load regulation parameters include: the control value of the motor output power of each generator in more than one generator, the control value of the regenerated electricity output power of each regenerated power supply in more than one regenerated power supply and the control value of the energy storage output power of the energy storage system.
The adjusting, by the respective power output node, the respective power output value based on the load adjustment parameter value comprises:
each generator of the more than one generator outputs power to the micro-grid based on the corresponding motor output power control value;
each of the one or more regenerative power sources outputs power to the microgrid based on the corresponding regenerative power output power control value;
and the energy storage system outputs power to the micro-grid based on the corresponding energy storage output power control value.
Each of the one or more generators outputs power to the microgrid in a droop control mode.
Correspondingly, the invention also provides a virtual generator system, which comprises:
the virtual controller is used for acquiring the load parameter value of each power output node on the microgrid; resolving an energy output structure of the microgrid based on the load parameter value; inputting the load parameter values associated with the energy output structure and the energy output structure into the twin network model to obtain load regulation parameter values of each power output node; the load adjusting parameter values are sent to each power output node, the twin network model trains the twin network by using a microgrid public data set, then the parameters trained in the public data set are used in a transfer learning mode, the twin network is initialized, and retraining is carried out on an energy output structure data set to obtain the trained twin network;
each power output node for adjusting a respective power output value based on the load regulation parameter value, the each power output node comprising: more than one regenerative power source, more than one generator, and an energy storage system.
According to the invention, the energy output structure acting on the microgrid is known by acquiring the load parameter values of each power output node on the microgrid, and then the twin network model is adopted to match the load regulation parameter values for each functional output node based on the energy output structure, so that the regulation parameters of each functional node are adaptively regulated, each power output node can meet the requirement of the optimal power output of the microgrid, and the power matching between the virtual generator and the microgrid is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of a virtual generator system according to an embodiment of the present invention;
fig. 2 is a flowchart of a method for adjusting the power output of a microgrid based on a virtual generator according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Specifically, fig. 1 shows a schematic structural diagram of a virtual generator system in an embodiment of the present invention, which specifically includes:
the virtual controller is used for acquiring the load parameter value of each power output node on the microgrid; resolving an energy output structure of the microgrid based on the load parameter value; inputting the load parameter values associated with the energy output structure and the energy output structure into the twin network model to obtain load regulation parameter values of each power output node; sending the load regulation parameter values to each power output node; the twin network model trains a twin network by using a microgrid public data set, then uses parameters trained in the public data set in a transfer learning mode to initialize the twin network, and retrains on an energy output structure data set to obtain the trained twin network;
each power output node for adjusting a respective power output value based on the load regulation parameter value, the each power output node comprising: more than one regenerative power source, more than one generator, and an energy storage system.
Specifically, more than one generator forms a generating set, and this generating set includes: the generator comprises a first generator, a second generator and an Nth generator, wherein N is a natural number which is larger than 0. The generator set may output electrical power to the microgrid, the electrical power relating to voltage values, current values, frequencies, and the like, and thus the generator set may apply voltages and frequencies to the microgrid. Each generator in the generator set realizes electric power output by adopting a droop control mode on frequency, the voltage value, the current value and the frequency of the output of the generator depend on the rotating speed of a shaft of the generator set, and when the electric power output by the generator set changes in a sounding mode, the rotating speed of the shaft of the generator set is self-adaptive, so that the frequency related to the electric power is also adjusted.
Specifically, the one or more regenerative power sources herein include: the first regenerative power source, the second regenerative power source to the Nth regenerative power source, wherein N is a natural number greater than 0. Since the renewable power source is subject to variable weather, it belongs to an unstable power source, which is an intermittent renewable energy source, which may include photovoltaic panels, wind turbines, water turbines or thermodynamic machines, etc. One end of the renewable energy source can be connected with the energy storage system, and the other end of the renewable energy source is connected with the inverter, and the renewable power source can deliver certain electric power to the energy storage system or the inverter.
Specifically, the energy storage system generally includes an electrochemical cell or a capacitor, etc., which can absorb the electric energy generated by the regenerative power source, and also can adaptively deliver a certain electric power to the microgrid based on the inverter.
Specifically, the inverter may convert the electrical power provided by the power generation stack, the energy storage system, the regenerative power source into an AC current and an AC voltage, which are then injected into the microgrid.
Based on the hardware structure, the virtual controller can accurately and adaptively adjust the power output value of each node only by obtaining the load parameter value of the energy storage system, the load parameter value of the generator set and the load parameter value of the regenerative power supply, so that the stability of voltage, current, frequency and the like transmitted to the microgrid is guaranteed. The load parameter values referred to herein include: load voltage, load current, load power, load frequency, etc., may reflect the power transmission characteristics of the various power output nodes.
Specifically, the analyzing the energy output structure of the microgrid based on the load parameter value includes: acquiring hardware basic structures of all power output nodes of the micro-grid; and filtering power output nodes without energy output based on the corresponding relation between the load parameter values and the hardware basic structure, and generating an energy output structure.
Specifically, the inputting the load parameter value and the energy output structure into the twin network model to obtain the load adjustment parameter value of each power output node of each microgrid includes: training a twin network by using a microgrid public data set, initializing the twin network by using parameters trained in the public data set in a transfer learning manner, and performing retraining on an energy output structure data set to obtain a trained twin network; simultaneously sending the load parameter value in the optimal power output state and the load parameter value in the energy output structure into a twin network; a first sub-network in the twin network receives a load parameter value in a power output optimal state and outputs a first characteristic value with a dimension space; a second sub-network in the twin network receives the load parameter value under the energy output structure and outputs a second characteristic value with a dimensional space; calculating the distance between the first characteristic value and the second characteristic value, comparing the similarity degree of the first characteristic value and the second characteristic value, and taking the load parameter value with the highest similarity degree in the optimal power output state as an identification result; and acquiring the load regulation parameter values of each power output node of each micro-grid corresponding to the identification result.
Specifically, the calculating a distance between the first characteristic value and the second characteristic value, and the comparing the similarity between the first characteristic value and the second characteristic value includes: and calculating Euclidean distances of the first characteristic value and the second characteristic value to compare the similarity degree of the first characteristic value and the second characteristic value.
Specifically, the calculating the euclidean distance between the first characteristic value and the second characteristic value to compare the similarity between the first characteristic value and the second characteristic value includes: the euclidean distance of the produced feature values is expressed using a contrast loss function.
The twin network in this embodiment is composed of two symmetrical neural networks, has the same weight and architecture, two different inputs are respectively input to the two branch networks to obtain characteristic values, and the similarity between the two characteristic values is calculated to determine whether the two inputs are the same.
The twin network training aims to make the distance of the generated characteristic values between the same species as close as possible and the distance of the generated characteristic values between different species as far as possible.
In the embodiment of the invention, a contrast Loss function (contrast Loss) is adopted to calculate the Euclidean distance, and the specific expression is as follows:
wherein d =
D represents the Euclidean distance of two sample features,
in order to be a sample of the sample,
is a sample; n represents the same batch of data, y is a label indicating whether two samples are matched, y =1 represents that the two samples are similar or matched, y =0 represents that the two samples are not matched, margin is a set threshold, and L is a loss function value that is obtained, the loss function value is mainly used in dimensionality reduction, namely, the two samples are still similar in a characteristic space after dimensionality reduction (characteristic extraction); and after dimensionality reduction, two samples which are originally dissimilar are still dissimilar in the feature space.
Specifically, the load regulation parameter includes: the control value of the motor output power of each generator in more than one generator, the control value of the regenerated electricity output power of each regenerated power supply in more than one regenerated power supply and the control value of the energy storage output power of the energy storage system.
Specifically, the adjusting, by the power output nodes, the respective power output values based on the load adjustment parameter values includes: each generator of the more than one generator outputs power to the micro-grid based on the corresponding motor output power control value; each of the one or more regenerative power sources outputs power to the microgrid based on the corresponding regenerative power output power control value; and the energy storage system outputs power to the micro-grid based on the corresponding energy storage output power control value.
Specifically, the virtual controller may control power input between the regenerative power source and the energy storage system, the inverter may transmit power on the regenerative power source and the energy storage system to the microgrid, in this process, the energy storage system may consume power input by the regenerative power source or input power to the microgrid, the series of load parameters may be obtained by the virtual controller, and the virtual controller may make corresponding output power adjustment for the energy storage system in combination with the need of power adjustment.
In particular, the generators in the electrical machine set adopt the characteristic of droop control, which can connect various generators in parallel in a virtual generator system, which will deliver the same frequency of voltage and current.
In particular, the virtual controller enables output power regulation of the regenerative power source, which may cause the regenerative power source to output a maximum power or saturated power, which may be lower than the maximum power that the regenerative power source is capable of delivering.
In particular, the virtual controller may implement a limit on the state of charge of the energy storage system, which may keep its state of charge not more than 100%, or below some threshold below 100%, in order to avoid degradation of the energy storage system.
According to the virtual generator system, the load parameter values of all power output nodes on the micro-grid are obtained, so that an energy output structure acting on the micro-grid is known, then a twin network model is adopted to match load adjusting parameter values for all the function output nodes based on the energy output structure, so that the adjusting parameters of all the function nodes are adjusted in a self-adaptive mode, all the power output nodes can meet the requirement of optimal power output of the micro-grid, and the power matching between the virtual generator and the micro-grid is realized.
Specifically, fig. 2 shows a flowchart of a method for adjusting the power output of the microgrid based on a virtual generator in the embodiment of the present invention, including the following steps:
s201, obtaining a load parameter value of each power output node on the microgrid;
based on the schematic diagram of the virtual generator system shown in fig. 1, each power output node includes: one or more regenerative power sources, one or more generators, and an energy storage system; the load parameter values include: load voltage, load current, load power, load frequency.
The distributed energy supply system is caused to be combined with the application state of renewable energy sources to realize self-adaptive energy supply output by combining with a micro-grid.
S202, resolving an energy output structure of the microgrid based on the load parameter values;
the energy output structure for resolving the microgrid based on the load parameter values comprises the following steps: acquiring hardware basic structures of all power output nodes of the micro-grid; and filtering power output nodes without energy output based on the corresponding relation between the load parameter values and the hardware basic structure, and generating an energy output structure.
Referring to fig. 1, when N is 4, there is a hardware infrastructure formed by four regenerative power sources, four generators, and an energy storage system, and when a load parameter value on a regenerative energy source is obtained, it can be monitored which regenerative energy sources can cooperate with the self-adaptation to complete power output to the microgrid, for example, the load parameter values of a first regenerative power source and a second regenerative power source can satisfy power output to the microgrid, the power output to the microgrid can be monitored by a third regenerative power source and a fourth regenerative power source department, when a load parameter value on the energy storage system is obtained, it can be monitored whether the energy storage system can cooperate with the self-adaptation to complete power output to the microgrid, and when a load parameter value on the generator is monitored, it can be known the power generation parameters of the generators and the power generation number of the generators, thereby preparing for adjusting the functional output of the generators, and whether the generators need to be shut down or started, wherein the first generator set and the second generator set are set to participate in power generation, and the third generator and the fourth generator are not participated in power generation.
When the value of N is 4, a hardware basic structure based on the value exists, the hardware contact structure is expressed in a netlist form, the netlist is used for describing the mutual connection relation of all power output nodes, when it is monitored that some power output nodes do not carry out power transmission on the micro-grid, the power output nodes which do not participate in the netlist need to be deleted as blank point rows, and the content of the initial netlist is adjusted to obtain the energy supply netlist reflecting the state of the energy output structure in real time.
By filtering the non-functional output nodes, partial data interference can be reduced, and redundancy in the data processing process can be reduced, so that the twin network can be matched with corresponding load adjustment parameter values more accurately.
S203, inputting the load parameter values associated with the energy output structure and the energy output structure into the twin network model to obtain load regulation parameter values of each power output node;
it should be noted that inputting the load parameter value and the energy output structure into the twin network model to obtain the load adjustment parameter value of each power output node of each microgrid includes: training a twin network by using a microgrid public data set, initializing the twin network by using parameters trained in the public data set in a transfer learning manner, and performing retraining on an energy output structure data set to obtain a trained twin network; simultaneously sending the load parameter value in the optimal power output state and the load parameter value in the energy output structure into a twin network; a first sub-network in the twin network receives a load parameter value in a power output optimal state and outputs a first characteristic value with a dimension space; a second sub-network in the twin network receives the load parameter value under the energy output structure and outputs a second characteristic value with a dimensional space; calculating the distance between the first characteristic value and the second characteristic value, comparing the similarity degree of the first characteristic value and the second characteristic value, and taking the load parameter value with the highest similarity degree in the optimal power output state as an identification result; and acquiring the load regulation parameter values of each power output node of each micro-grid corresponding to the identification result.
In the specific implementation process, the energy output structure is output to the twin network model in the form of the energy supply netlist, namely a version number is given to the twin network model, the twin network model can train the twin network for the public data set in the micro-grid state based on the energy supply netlist, and then the load parameter value in the optimal power output state and the load parameter value of the energy output structure are simultaneously sent to the twin network.
It should be noted that, the calculating a distance between the first characteristic value and the second characteristic value, and the comparing the similarity between the first characteristic value and the second characteristic value includes: and calculating Euclidean distances of the first characteristic value and the second characteristic value to compare the similarity degree of the first characteristic value and the second characteristic value.
It should be noted that the calculating the euclidean distance between the first feature value and the second feature value to compare the similarity between the first feature value and the second feature value includes: the euclidean distance of the produced feature values is expressed using a contrast loss function.
The twin network in this embodiment is composed of two symmetrical neural networks, has the same weight and architecture, two different inputs are respectively input to the two branch networks to obtain characteristic values, and the similarity between the two characteristic values is calculated to determine whether the two inputs are the same.
The twin network training aims to make the distance of the generated characteristic values between the same species as close as possible and the distance of the generated characteristic values between different species as far as possible.
In the embodiment of the invention, a contrast Loss function (contrast Loss) is adopted to calculate the Euclidean distance, and the specific expression is as follows:
wherein
D represents the Euclidean distance of two sample features,
in order to be a sample of the sample,
is a sample; n represents the same batch of data, y is a label indicating whether two samples are matched, y =1 represents that the two samples are similar or matched, y =0 represents that the two samples are not matched, margin is a set threshold, and L is a loss function value that is obtained, the loss function value is mainly used in dimensionality reduction, namely, the two samples are still similar in a characteristic space after dimensionality reduction (characteristic extraction); and after dimensionality reduction, two samples which are originally dissimilar are still dissimilar in the feature space.
It should be noted that the load parameter values in the optimal power output state are associated with load regulation parameter values, and these load regulation parameter values trigger the next power output regulation process of the virtual generator system.
S204, sending the load regulation parameter values to each power output node;
it should be noted that the load regulation parameters include: the control value of the motor output power of each generator in more than one generator, the control value of the regenerated electricity output power of each regenerated power supply in more than one regenerated power supply and the control value of the energy storage output power of the energy storage system.
It should be noted that, if the power output adjustment of the power output node is not involved, the default setting is set to be a null value, and when the corresponding function output node receives the null value, the function output adjustment process is not performed, and the original mode is maintained.
And S205, each power output node adjusts each power output value based on the load adjusting parameter value.
It should be noted that the adjusting of the respective power output value by the respective power output node based on the load adjustment parameter value includes: each generator of the more than one generator outputs power to the micro-grid based on the corresponding motor output power control value; each of the one or more regenerative power sources outputs power to the microgrid based on the corresponding regenerative power output power control value; and the energy storage system outputs power to the micro-grid based on the corresponding energy storage output power control value.
Each of the one or more generators outputs power to the microgrid in a droop control mode. After the motor output power control value is generated, the virtual controller thereby controls the genset for power output based on an electromechanical equation, which is implemented specifically for a mechanical differential equation modeling the dynamics of the rotor of the genset.
To sum up, the embodiment of the invention obtains the load parameter values of each power output node on the microgrid, so as to know the energy output structure acting on the microgrid, and then matches the load regulation parameter values for each function output node by adopting a twin network model based on the energy output structure, so as to realize the self-adaptive regulation of the regulation parameters of each function node, so that each power output node can meet the requirement of the optimal power output of the microgrid, and the power matching between the virtual generator and the microgrid is realized.
The above embodiments of the present invention are described in detail, and the principle and the implementation of the present invention are described herein by using specific embodiments, and the description of the above embodiments is only used to help understanding the method of the present invention and the core idea thereof; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A method of regulating microgrid power output based on a virtual generator, the method comprising:
acquiring a load parameter value of each power output node on the microgrid;
resolving an energy output structure of the microgrid based on the load parameter value;
inputting a load parameter value associated with an energy output structure and the energy output structure into a twin network model to obtain a load regulation parameter value of each power output node, training the twin network by using a microgrid public data set, using a parameter trained in the public data set in a transfer learning manner, initializing the twin network, and retraining on the energy output structure data set to obtain a trained twin network;
sending the load regulation parameter values to each power output node;
each power output node adjusts a respective power output value based on the load adjustment parameter value.
2. The method of regulating microgrid power output based on a virtual generator of claim 1, wherein the respective power output nodes include: one or more regenerative power sources, one or more generators, and an energy storage system; the load parameter values include: load voltage, load current, load power, load frequency.
3. The method of claim 2, wherein resolving the energy output structure of the microgrid based on the load parameter values comprises:
acquiring hardware basic structures of all power output nodes of the micro-grid;
and filtering power output nodes without energy output based on the corresponding relation between the load parameter values and the hardware basic structure, and generating an energy output structure.
4. The method of claim 3, wherein inputting the load parameter values and the energy output structure into the twin network model to derive the load regulation parameter values for each power output node comprises:
simultaneously sending the load parameter value in the optimal power output state and the load parameter value in the energy output structure into a twin network;
a first sub-network in the twin network receives a load parameter value in a power output optimal state and outputs a first characteristic value with a dimension space; a second sub-network in the twin network receives the load parameter value under the energy output structure and outputs a second characteristic value with a dimensional space;
calculating the distance between the first characteristic value and the second characteristic value, comparing the similarity degree of the first characteristic value and the second characteristic value, and taking the load parameter value with the highest similarity degree in the optimal power output state as an identification result;
and acquiring the load adjusting parameter value of each corresponding power output node under the identification result.
5. The method of regulating microgrid power output based on a virtual generator of claim 4, wherein the step of calculating a distance between the first characteristic value and the second characteristic value, and the step of comparing the similarity between the first characteristic value and the second characteristic value comprises:
and calculating Euclidean distances of the first characteristic value and the second characteristic value to compare the similarity degree of the first characteristic value and the second characteristic value.
6. The method of regulating microgrid power output based on a virtual generator of claim 5, wherein the calculating of Euclidean distances of the first characteristic value and the second characteristic value to compare degrees of similarity of the first characteristic value and the second characteristic value comprises:
the euclidean distance of the produced feature values is expressed using a contrast loss function.
7. The virtual generator-based method of regulating microgrid power output of any of claims 1 to 6, wherein said load regulation parameters comprise: the control value of the motor output power of each generator in more than one generator, the control value of the regenerated electricity output power of each regenerated power supply in more than one regenerated power supply and the control value of the energy storage output power of the energy storage system.
8. The method of regulating microgrid power output based on a virtual generator of claim 7, wherein the regulation of the respective power output value by the respective power output node based on the load regulation parameter value comprises:
each generator of the more than one generator outputs power to the micro-grid based on the corresponding motor output power control value;
each of the one or more regenerative power sources outputs power to the microgrid based on the corresponding regenerative power output power control value;
and the energy storage system outputs power to the micro-grid based on the corresponding energy storage output power control value.
9. The method of claim 8, wherein each of the one or more generators outputs power to the microgrid in a droop control mode.
10. A virtual generator system, comprising:
the virtual controller is used for acquiring the load parameter value of each power output node on the microgrid; resolving an energy output structure of the microgrid based on the load parameter value; inputting the load parameter values associated with the energy output structure and the energy output structure into the twin network model to obtain load regulation parameter values of each power output node; the load adjusting parameter values are sent to each power output node, the twin network model trains the twin network by using a microgrid public data set, then the parameters trained in the public data set are used in a transfer learning mode, the twin network is initialized, and retraining is carried out on an energy output structure data set to obtain the trained twin network;
each power output node for adjusting a respective power output value based on the load regulation parameter value, the each power output node comprising: more than one regenerative power source, more than one generator, and an energy storage system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210279547.4A CN114374233B (en) | 2022-03-22 | 2022-03-22 | Method and system for adjusting micro-grid power output based on virtual generator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210279547.4A CN114374233B (en) | 2022-03-22 | 2022-03-22 | Method and system for adjusting micro-grid power output based on virtual generator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114374233A true CN114374233A (en) | 2022-04-19 |
| CN114374233B CN114374233B (en) | 2022-06-17 |
Family
ID=81146510
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210279547.4A Active CN114374233B (en) | 2022-03-22 | 2022-03-22 | Method and system for adjusting micro-grid power output based on virtual generator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114374233B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114726002A (en) * | 2022-06-09 | 2022-07-08 | 长沙电机厂集团长瑞有限公司 | Virtual control-based multi-generator-set output control method and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103633739A (en) * | 2013-11-28 | 2014-03-12 | 中国科学院广州能源研究所 | Microgrid energy management system and method |
| US20170091791A1 (en) * | 2015-09-25 | 2017-03-30 | General Electric Company | Digital power plant system and method |
| US20190131795A1 (en) * | 2016-06-21 | 2019-05-02 | General Electric Company | System and method for controlling a power generating unit |
| US20200014212A1 (en) * | 2018-07-05 | 2020-01-09 | Abb Schweiz Ag | Technologies for provisioning power controllers for grid services |
| CN111722540A (en) * | 2020-06-19 | 2020-09-29 | 中国电力科学研究院有限公司 | Energy Internet digital twin simulation system and method |
| CN112332444A (en) * | 2020-09-14 | 2021-02-05 | 华北电力大学（保定） | Microgrid energy management system based on digital twins |
| CN113221456A (en) * | 2021-05-11 | 2021-08-06 | 上海交通大学 | Digital twin modeling and multi-agent coordination control method for smart microgrid |
| CN113807705A (en) * | 2021-09-18 | 2021-12-17 | 国网冀北电力有限公司智能配电网中心 | Digital twin operation driven power distribution network planning method and device and terminal |
-
2022
- 2022-03-22 CN CN202210279547.4A patent/CN114374233B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103633739A (en) * | 2013-11-28 | 2014-03-12 | 中国科学院广州能源研究所 | Microgrid energy management system and method |
| US20170091791A1 (en) * | 2015-09-25 | 2017-03-30 | General Electric Company | Digital power plant system and method |
| US20190131795A1 (en) * | 2016-06-21 | 2019-05-02 | General Electric Company | System and method for controlling a power generating unit |
| US20200014212A1 (en) * | 2018-07-05 | 2020-01-09 | Abb Schweiz Ag | Technologies for provisioning power controllers for grid services |
| CN111722540A (en) * | 2020-06-19 | 2020-09-29 | 中国电力科学研究院有限公司 | Energy Internet digital twin simulation system and method |
| CN112332444A (en) * | 2020-09-14 | 2021-02-05 | 华北电力大学（保定） | Microgrid energy management system based on digital twins |
| CN113221456A (en) * | 2021-05-11 | 2021-08-06 | 上海交通大学 | Digital twin modeling and multi-agent coordination control method for smart microgrid |
| CN113807705A (en) * | 2021-09-18 | 2021-12-17 | 国网冀北电力有限公司智能配电网中心 | Digital twin operation driven power distribution network planning method and device and terminal |
Non-Patent Citations (3)
| Title |
|---|
| NAJMEH BAZMOHAMMADI: ""Microgrid Digital Twins: Concepts, Applications, and Future Trends"", 《IEEE ACCESS》 * |
| 张宏伟: "" 基于孪生神经网络的两阶段目标跟踪方法"", 《红外与激光工程》 * |
| 潘春鹏: ""交直流微网数字孪生及其应用研究"", 《万方在线》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114726002A (en) * | 2022-06-09 | 2022-07-08 | 长沙电机厂集团长瑞有限公司 | Virtual control-based multi-generator-set output control method and system |
| CN114726002B (en) * | 2022-06-09 | 2022-08-23 | 长沙电机厂集团长瑞有限公司 | Virtual control-based multi-generator-set output control method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114374233B (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Abdullah et al. | Towards green energy for smart cities: Particle swarm optimization based MPPT approach | |
| Zhang et al. | Fully distributed coordination of multiple DFIGs in a microgrid for load sharing | |
| Mishra et al. | Biogeography based optimal state feedback controller for frequency regulation of a smart microgrid | |
| Syahputra et al. | Control of Synchronous Generator in Wind Power Systems Using Neuro-Fuzzy Approach | |
| Mensou et al. | An efficient nonlinear Backstepping controller approach of a wind power generation system based on a DFIG | |
| CN114374233B (en) | Method and system for adjusting micro-grid power output based on virtual generator | |
| Nguyen et al. | Cluster-based predictive PCC voltage control of large-scale offshore wind farm | |
| CN109301882A (en) | A kind of electric system distributing economic load dispatching method based on ADMM | |
| Bakeer et al. | Resilient virtual synchronous generator approach using DC-link capacitor energy for frequency support of interconnected renewable power systems | |
| Abbes et al. | An aggregate model of PMSG-based, grid connected wind farm: Investigation of LVRT capabilities | |
| CN106611966B (en) | Multi-inverter type exchanges micro-capacitance sensor distribution economy Automatic Generation Control algorithm | |
| CN109560557A (en) | A kind of micro-capacitance sensor frequency intelligence control system and its frequency stabilization control algolithm | |
| Nguyen | A strategy development for optimal generating power of small wind-diesel-solar hybrid microgrid system | |
| CN107591843A (en) | A kind of double-fed wind field idle generating optimization method in system recovery procedure | |
| Bhaduri et al. | Supervisory controller for power management of microgrid using hybrid technique | |
| Saleh | The analysis and development of a power flow-based controller for microgrid systems | |
| Kouadria et al. | Sliding mode control of the active and reactive power of DFIG for variable-speed wind energy conversion system | |
| Shayeghi et al. | Management of Autonomous Microgrids Using Multi-Agent Based Online Optimized NF-PID Controller | |
| Huynh et al. | An Improved Artificial Bee Colony Algorithm-based Configuration Optimization of a Remote Microgrid Considering Renewable Energy Systems | |
| Lian et al. | Cost allocation strategies for wide-area control of power systems using Nash bargaining solution | |
| Yilin et al. | A distributed control based on finite-time consensus algorithm for the cluster of DC microgrids | |
| Dong et al. | Surrogate-assisted cooperation control of network-connected doubly fed induction generator wind farm with maximized reactive power capacity | |
| Gowri et al. | ANFIS based MPPT and droop controller for power sharing in interline power system | |
| Veeramani et al. | Evaluation of wind energy parameter optimization of A DFIG controller based on cultural algorithms | |
| Pandey et al. | Frequency control of an autonomous hybrid generation system |
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 |