CN115544860A - Output modeling method of intermittent distributed power supply in complex operation scene - Google Patents
Output modeling method of intermittent distributed power supply in complex operation scene Download PDFInfo
- Publication number
- CN115544860A CN115544860A CN202211082099.5A CN202211082099A CN115544860A CN 115544860 A CN115544860 A CN 115544860A CN 202211082099 A CN202211082099 A CN 202211082099A CN 115544860 A CN115544860 A CN 115544860A
- Authority
- CN
- China
- Prior art keywords
- data
- output
- distributed power
- curve
- power supply
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid distribution networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides an output modeling method, an output modeling device, electronic equipment and an output modeling medium of an intermittent distributed power supply in a complex operation scene, wherein the output modeling method comprises the following steps: acquiring first data of the intermittent distributed power supply, wherein the first data is used for representing at least one of historical measured data, noise data and category labels; processing the first data through a deep neural generation confrontation network to obtain an output curve of the first data; determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model; and adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation. The method adopts a deep generation countermeasure network DGAN on the structure of the generation countermeasure network, and improves the feature discrimination capability of the GAN; the category label is added when the confrontation network training is generated, so that the training effect is improved; the upper limit and the lower limit of the output force corresponding to each class label are calculated by combining a Gaussian mixture model GMM, so that the simulation operation precision of the power distribution network is improved.
Description
Technical Field
The invention relates to the technical field of power distribution and computers, in particular to an output modeling method and device of an intermittent distributed power supply in a complex operation scene, electronic equipment and a medium.
Background
In order to improve the disaster tolerance capability and the reliability level of a power distribution system and power distribution equipment, a complex operation scene of the power distribution network needs to be simulated. However, due to the constraint of modeling complexity, the conventional simulation model often ignores the correlation between the influence of environmental factors and the operation dynamics of multiple devices, so that the self-organization critical characteristics and the chain reaction mechanism of the system under a complex operation scene cannot be accurately described, the evaluation and prediction accuracy of the system situation is influenced, and the comprehensiveness of the analysis and reasoning of the fault mechanism is weakened. Therefore, the method for generating the complex operation scene of the power distribution network based on the countermeasure generation network is provided, and the complete knowledge of the operation scene set of the power distribution network is formed by simulating the uncertainty of the distributed power supply, so that the situation prediction and deduction application of the power distribution network is supported. One important factor in the simulation of a complex scene is that the output power of an intermittent distributed power source (DG, such as wind power generation and a photovoltaic cell) is greatly influenced by the weather environment, has obvious uncertainty, randomness and fluctuation, and can influence the normal operation of a power system.
The accurate simulation of the operation scene of the intermittent distributed power supply is researched a lot, so that a plurality of theoretical and practical achievements are obtained, a researcher takes the minimum total cost of a power supply company as an optimization target, provides a simulation model of output data of the complex scene of the intermittent distributed power supply under a market condition, and solves the problem by adopting a heuristic method; the method comprises the following steps that a researcher takes maximum active power output as a target function, takes output of an intermittent distributed power supply, thermal stability limit of a line and the like as constraints, forms a mathematical model, and then solves the model by using a linear programming method; according to the randomness of the output of the wind turbine generator and the uncertainty of the load, researchers apply the opportunity constraint planning to the output data simulation of the DWG complex scene, and adopt the random power flow to judge whether the planning scheme violates the node voltage constraint and the branch power transmission constraint; a researcher provides a concept of equivalent network loss micro-increment rate, the optimal configuration position of the intermittent distributed power supply is obtained by adopting the concept index calculation, the minimum network loss of a system after the intermittent distributed power supply is connected into a power distribution network can be ensured, 3 indexes of voltage, network loss and environmental benefit for evaluating the benefits of the intermittent distributed power supply are considered, and output data of a complex scene of the intermittent distributed power supply is converted into a multi-target nonlinear programming problem in an analog mode.
In the existing modeling method for the uncertainty of the output of the intermittent distributed power supply in the prior art, a probability modeling method needs to accurately describe uncertainty factors, and an actual project can only obtain partial statistical information, so that accurate probability distribution is difficult to obtain, and the time sequence characteristic of the output of the intermittent distributed power supply is ignored; the robust optimization method is too conservative and cannot well balance the economy and safety of the planning scheme; the traditional artificial intelligence method scene generation technology is difficult to capture the nonlinear characteristics of high-dimensional data, has certain limitations and can deviate from the actual scene. The existing modeling method does not consider any prior information, so that the training direction of the generated confrontation network is easy to lose, and the training effect is poor and unstable.
Disclosure of Invention
The embodiment of the invention mainly aims to provide an output modeling method, an output modeling device, electronic equipment and an output modeling medium of an intermittent distributed power supply in a complex operation scene, and improve the precision of the simulation operation of a power distribution network including the output of the intermittent distributed power supply.
One aspect of the present invention provides an output modeling method for an intermittent distributed power supply in a complex operation scene, which is characterized by comprising:
responding to a modeling request, and acquiring first data of the intermittent distributed power supply, wherein the first data is used for representing at least one of historical measured data, noise data and a category label;
performing processing on the first data through a deep neural generation countermeasure network to obtain an output curve of the first data;
determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model;
and adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation.
According to the output modeling method of the intermittent distributed power supply in the complex operation scene, the first data comprises the following steps:
the category label data is label information generated when the historical measured data is classified according to scenes;
each piece of historical measured data has corresponding category label data, and the scene of the category label data comprises at least one of time, place, unit identification number and characteristic data of a distributed power supply.
According to the output modeling method of the intermittent distributed power supply under the complex operation scene, the deep neural generation countermeasure network comprises the following steps:
creating a greatly minimized game model based on a generated countermeasure network, wherein the generated countermeasure network comprises a generator and a discriminator;
setting an up-sampling neural network at the generator and setting a down-sampling neural network at the discriminator;
training the generated confrontation network to obtain the deep neural generated confrontation network;
the extremely minimized game model is as follows:
where E represents an expected value, D [ G (z) ] is a probability that generated data G (z) is judged to be true in D, D (x) represents a probability that true data x is judged to be true in D, the distribution of noise data z is z to Pz, px is a true distribution of data x, D is a discriminator, and G is a generator.
According to the output modeling method of the intermittent distributed power supply under the complex operation scene, the first data is processed through a deep neural generation countermeasure network to obtain an output curve of the first data, and the method comprises the following steps:
inputting the category labels to the generator and the discriminator respectively;
performing per-unit processing on the noise data through the generator, constructing a noise data vector by using the noise data subjected to per-unit processing, inserting the class label into the noise data vector to form a generator cross-domain vector, and inputting the generator cross-domain vector into the generator to obtain generated data;
and performing per-unit processing on the historical actual measurement data through a discriminator, constructing an actual measurement data vector by using the historical actual measurement data subjected to per-unit processing, splicing the actual measurement data vector with the category label to form a discriminator cross-domain vector, and inputting the discriminator cross-domain vector and the generated data into the discriminator to be processed to obtain the output curve.
According to the output modeling method of the intermittent distributed power supply under the complex operation scene, the upper limit and the lower limit of the output curve are determined by adopting a mixed Gaussian model, and the method comprises the following steps:
generating a plurality of output curves by using the deep neural generation countermeasure network, and clustering the output curves by combining a Gaussian mixture model to generate a plurality of curve output line sets;
and selecting the upper limit and the lower limit of the maximum output curve according to the curve output line set.
According to the output modeling method of the intermittent distributed power supply under the complex operation scene, clustering a plurality of output curves by combining a Gaussian mixture model to generate a plurality of curve output line sets, and the method comprises the following steps:
approximating any continuous probability distribution by a combination of a plurality of gaussian functions, the probability distribution function being:
wherein alpha is s Represents the weight of the s-th Gaussian branch model, M is the number of the Gaussian branch models,
as a parameter of the Gaussian mixture model, μ s Is taken as the average value of the values,
representing the variance.
According to the output modeling method of the intermittent distributed power supply in the complex operation scene, the configuration of the deep neural generation countermeasure network is adjusted according to the output curve and the upper limit and the lower limit of the output curve, and output data simulation is obtained, and the method comprises the following steps:
and calculating the upper limit and the lower limit of the maximum output curve corresponding to each class label by combining a Gaussian mixture model, and performing output data simulation on the complex scene of the intermittent distributed power supply by combining the output curve and the upper limit and the lower limit of the output.
Another aspect of the embodiments of the present invention provides an output modeling apparatus for an intermittent distributed power source in a complex operation scenario, including:
the data acquisition module is used for acquiring first data of the intermittent distributed power supply according to the modeling request, wherein the first data is used for representing at least one of historical measured data, noise data and category labels;
the generation countermeasure module is used for executing processing on the first data through a deep neural generation countermeasure network to obtain an output curve of the first data;
the mixed Gaussian module is used for determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model;
and the output simulation module is used for adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation.
Another aspect of the embodiments of the present invention provides an electronic device, which includes a processor and a memory;
the memory is used for storing programs;
the processor executes the program to implement the method as described in the foregoing.
The embodiment of the invention also discloses a computer program product or a computer program, which comprises computer instructions, and the computer instructions are stored in a computer readable storage medium. The computer instructions may be read by a processor of a computer device from a computer-readable storage medium, and executed by the processor, causing the computer device to perform the methods described above.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a system diagram of an embodiment of the present invention.
FIG. 2 is a schematic flow chart of a method according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a deep neural generation countermeasure network execution generation flow according to an embodiment of the present invention.
Fig. 4 is a process diagram of a GAN network according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of a data processing flow of the deep neural generation countermeasure network according to an embodiment of the present invention.
FIG. 6 is a schematic diagram of a processing flow of a Gaussian mixture model according to an embodiment of the present invention.
Fig. 7 is a diagram of an output modeling analysis apparatus of an intermittent distributed power supply in a complex operating scenario according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. In the following description, suffixes such as "module", "part", or "unit" used to indicate elements are used only for facilitating the explanation of the present invention, and have no peculiar meaning by themselves. Thus, "module", "component" or "unit" may be used mixedly. "first", "second", etc. are used for the purpose of distinguishing technical features only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or implicitly indicating the precedence of the indicated technical features. In the following description, the method steps are labeled continuously for convenience of examination and understanding, and the implementation order of the steps is adjusted without affecting the technical effect achieved by the technical scheme of the present invention by combining the whole technical scheme of the present invention and the logical relationship between the steps. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
As shown in fig. 1, an embodiment of the present invention provides a system schematic diagram, which includes a client 100 and a server 200, in some embodiments, the server 200 obtains first data of an intermittent distributed power source in response to a modeling request of the client 100, where the first data is at least one of historical measured data, noise data, and a category label; the server 200 executes processing on the first data through a deep neural generation countermeasure network to obtain an output curve of the first data; the server 200 determines the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model; the server 200 adjusts the configuration of the deep neural generative confrontation network according to the output curve and the upper limit and the lower limit of the output curve, and output data simulation is obtained.
The client may be a PC, a web page, an intelligent mobile terminal, or the like.
Referring to fig. 2, the method for modeling the output of the intermittent distributed power source in a complex operation scenario is illustrated, including but not limited to steps S100 to S400:
and S100, responding to a modeling request, and acquiring first data of the intermittent distributed power source, wherein the first data is used for representing at least one of historical measured data, noise data and class labels.
In some embodiments, the class label data is label information generated when the historical measured data is classified according to scenes, each piece of historical measured data has corresponding class label data, and the scenes of the class label data include at least one of time, place, unit identification number and characteristic data of the distributed power supply.
Illustratively, each measured data item has a corresponding category label representing a specified scenario, such as a specific month, address, set number, wind speed, irradiance, etc., where wind speed and irradiance represent characteristic data of wind power generation and solar power generation, respectively.
In the traditional data generation process, any prior information is not considered, so that the training direction is easy to lose, and the training effect is poor and unstable. According to the technical scheme of the embodiment, the auxiliary category information is added and is used as a part of data generation and judgment input, so that data of a specified category can be generated, and the category judgment is used as a basis during data judgment, so that the training direction for generating the countermeasure network can be iterated along the same direction, and the training effect is improved.
S200, processing the first data through a deep neural generation confrontation network to obtain an output curve of the first data.
Illustratively, referring to fig. 3, a flow diagram illustrating the generation of a deep neural generating confrontation network is illustrated, which includes, but is not limited to:
s210, establishing a maximum and minimum game model based on a generated countermeasure network, wherein the generated countermeasure network comprises a generator and a discriminator;
in some embodiments, the GAN is composed of a generator G and a discriminator D, G is a data generator, and can map simple feature distribution to a high-dimensional space to generate distribution that approximates real data as much as possible, D receives the output data G (z) synthesized by G and the real data x sampled from historical data at the same time, and maximizes the difference between the two as much as possible, and G and D play continuously, and finally reach a state of "nash equilibrium", and can summarize a very-minimized game model of the whole GAN network by the following formula:
s220, setting an up-sampling neural network on the generator and setting a down-sampling neural network on the discriminator;
in some embodiments, an upsampling neural network is provided in G for generating high resolution output samples for time aligning the output samples; a down-sampling neural network layer is added in the D so as to carry out down-sampling on the input historical measured data and the simulation test data, and the down-sampling neural network layer is used for compressing the input signal and facilitating the subsequent characteristic judgment.
And S230, training the generation countermeasure network to obtain a deep neural generation countermeasure network.
In some embodiments, in the training process, as iteration is performed, G improves the similarity between generated data and real data by adjusting the weight of the network, and D also improves discrimination capability by learning, and after repeated iteration, the obtained DGAN can be used for simulating an output scene when training is completed.
In some embodiments, referring to the process flow diagram for generating a countermeasure network shown in fig. 4, historical measured data, noise data, and category label data are imported into an improved deep neural generation countermeasure network, and an output curve corresponding to the historical measured data is generated by the improved deep neural generation countermeasure network.
The Deep Generation Adaptive Networks (DGANs) of this embodiment is an improvement on GAN, and introduces a deep neural network into a structure of GAN, and improves data quality generated by GAN based on a strong feature extraction capability of the deep neural network. The measured data has rich characteristics, and the characteristics of the DGAN are just suitable for extracting dynamic characteristics so as to construct an accurate output curve. A Generative Adaptive Network (GAN) is a deep learning model that implicitly learns the probability distribution of data by training a set of real data to fit the model, and is used to evaluate the overall effectiveness of the generated model.
Illustratively, referring to fig. 5, it illustrates a schematic diagram of the deep neural generation countermeasure network data processing flow, including but not limited to steps S240 to S260:
s240, inputting the category label to the generator and the discriminator respectively.
In some embodiments, the DGAN adds category label data to the training process, which may be the particular month, address, set number, wind speed, irradiance, etc., and uses this information as part of the G and D inputs, so that data specifying a category may be generated and a category decision may be based on when the data is determined. The cost function of the modified conditional DGAN is adjusted as follows:
wherein y is the auxiliary tag data information.
S250, performing per-unit processing on the noise data through a generator, constructing a noise data vector by using the noise data subjected to per-unit processing, inserting a class label into the noise data vector to form a generator cross-domain vector, and inputting the generator cross-domain vector into the generator to obtain generated data;
and S260, performing per-unit processing on the historical measured data through the discriminator, constructing a measured data vector by using the per-unit processed historical measured data, splicing the measured data vector with the class label to form a discriminator cross-domain vector, and inputting the discriminator cross-domain vector and the generated data into the discriminator to be processed to obtain an output curve.
In some embodiments, the specific input data processing is:
(1) Input data processing of the generator G: performing per-unit processing on the noise data, constructing a noise data vector by using the noise data subjected to per-unit processing, inserting a class label y into the noise data vector to form a G cross-domain vector, inputting the G cross-domain vector into a generator G to obtain generated data G (z), wherein the G (z) is also expressed in a vector form;
(2) Input data processing to the discriminator D: performing per-unit processing on the historical actual measurement data, constructing an actual measurement data vector by using the historical actual measurement data after the per-unit processing, splicing the actual measurement data vector with the category label y to form a D cross-domain vector, and inputting the D cross-domain vector and the generated data G (z) into a discriminator D for processing.
And S300, determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model.
In some embodiments, reference is made to FIG. 6, which illustrates a Gaussian mixture model process flow diagram. Including but not limited to steps S310 to S320:
s310, generating a plurality of output curves by using a deep neural generation countermeasure network, and clustering the output curves by combining a Gaussian mixture model to generate a plurality of curve output line sets;
in some embodiments, after obtaining the DGAN model, the information in the class labels is fixed and the DGAN model is used to generate a large number of output curves. The generated output curves are clustered by combining a Gaussian Mixture Model (GMM) to generate a plurality of output line sets.
And S320, selecting the upper limit and the lower limit of the maximum output curve according to the curve output line set.
In some embodiments, the GMM effectively approximates any continuous probability distribution by a combination of multiple Gaussian functions, the probability distribution function of which is as follows:
approximating any continuous probability distribution by a combination of multiple gaussian functions, the probability distribution function being:
wherein alpha is s Representing the weight of the s-th Gaussian partial model, M is the number of Gaussian partial models,
is a parameter of the Gaussian mixture model, mu s Is taken as the mean value of the average value,
represents the variance
S400, adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation.
In some embodiments, the upper and lower force curve boundaries in this case are obtained by taking the upper and lower limits of the maximum force curve among the different force curve sets. The simulation of the complex operation scene can be completed by combining the simulation of the faults of the power distribution network.
Through the technical scheme of the embodiment of the invention, the method at least has the following beneficial effects: the deep generation countermeasure network DGAN is adopted on the structure of the generated countermeasure network, the neural network is added in both the generator G and the discriminator D, and the data quality generated by the GAN is improved and the feature discrimination capability of the GAN is improved based on the stronger feature extraction capability of the neural network; the class label is added when the confrontation network training is generated, the class label is used as a part of data generation and judgment input, and the training direction of the confrontation network can be iterated along the same direction through the guidance of the class label, so that the training effect is improved; after the output curve is obtained, calculating the upper limit and the lower limit of the output corresponding to each class label by combining a Gaussian mixture model GMM, and simulating the output data of the DG complex scene by combining the output curve and the upper limit and the lower limit of the output.
Fig. 7 is a diagram of an output modeling analysis device of an intermittent distributed power source in a complex operation scenario according to an embodiment of the present invention. The device comprises a data acquisition module 710, a generation countermeasure module 720, a Gaussian mixture module 730 and an output simulation module 740.
The device comprises: the data acquisition module is used for acquiring first data of the intermittent distributed power supply according to the modeling request, wherein the first data is used for representing at least one of historical measured data, noise data and category labels; the generation countermeasure module is used for executing processing on the first data through a deep neural generation countermeasure network to obtain an output curve of the first data; the mixed Gaussian module is used for determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model; and the output simulation module is used for adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation.
Illustratively, under the cooperation of a data acquisition module, a generation countermeasure module, a Gaussian mixture module and an output simulation module in the device, the device of the embodiment can realize the output modeling method of any intermittent distributed power supply in a complex operation scene, namely acquiring first data of the intermittent distributed power supply, wherein the first data is used for representing at least one of historical measured data, noise data and a category label; generating the first data into an antagonistic network through a deep nerve to execute processing to obtain an output curve of the first data; determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model; and adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation. Through the technical scheme of the embodiment of the invention, the method has at least the following beneficial effects: the deep generation countermeasure network DGAN is adopted on the structure of the generated countermeasure network, the neural network is added in the generator G and the discriminator D, and the data quality generated by the GAN is improved and the feature discrimination capability of the GAN is improved based on the stronger feature extraction capability of the neural network; the class labels are added when the confrontation network training is generated, the class labels are used as a part of data generation and judgment input, and the training direction of the confrontation network can be iterated along the same direction through class label guidance, so that the training effect is improved; after the output curve is obtained, calculating the upper limit and the lower limit of the output corresponding to each class label by combining a Gaussian mixture model GMM, and performing output data simulation on the DG complex scene by combining the output curve and the upper limit and the lower limit of the output.
The embodiment of the invention also provides electronic equipment, which comprises a processor and a memory;
the memory stores a program;
the processor executes a program to execute the output modeling method of the intermittent distributed power supply under the complex operation scene; the electronic device has a function of loading and operating a software system for modeling output of the intermittent distributed power supply in a complex operation scene, for example, a Personal Computer (PC), a mobile phone, a smart phone, a Personal Digital Assistant (PDA), a wearable device, a Pocket PC (PPC), a tablet PC, and the like.
The embodiment of the invention also provides a computer-readable storage medium, which stores a program, and the program is executed by a processor to implement the output modeling method of the intermittent distributed power supply in the complex operation scene.
In alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flow charts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed, and in which sub-operations described as part of larger operations are performed independently.
The embodiment of the invention also discloses a computer program product or a computer program, which comprises computer instructions, and the computer instructions are stored in a computer readable storage medium. The computer instructions can be read by a processor of the computer device from a computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the output modeling method of the intermittent distributed power source under the complex operation scene.
Furthermore, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the described functions and/or features may be integrated in a single physical device and/or software module, or one or more functions and/or features may be implemented in a separate physical device or software module. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary for an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be understood within the ordinary skill of an engineer, given the nature, function, and internal relationship of the modules. Accordingly, those skilled in the art can, using ordinary skill, practice the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative of and not intended to limit the scope of the invention, which is to be determined from the appended claims along with their full scope of equivalents.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. An output modeling method of an intermittent distributed power supply in a complex operation scene is characterized by comprising the following steps:
responding to a modeling request, and acquiring first data of the intermittent distributed power supply, wherein the first data is used for representing at least one of historical measured data, noise data and a category label;
performing processing on the first data through a deep neural generation countermeasure network to obtain an output curve of the first data;
determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model;
and adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation.
2. The intermittent distributed power supply output modeling method under the complex operation scenario of claim 1, wherein the first data comprises:
the category label data is label information generated when the historical measured data is classified according to scenes;
each piece of historical measured data has corresponding category label data, and the scene of the category label data comprises at least one of time, place, unit identification number and characteristic data of a distributed power supply.
3. The intermittent distributed power supply output modeling method under the complex operation scene of claim 1, wherein the deep neural generation countermeasure network comprises:
creating a maximum minimization game model based on a generated countermeasure network, wherein the generated countermeasure network comprises a generator and a discriminator;
setting an up-sampling neural network on the generator and setting a down-sampling neural network on the discriminator;
training the generated confrontation network to obtain the deep neural generated confrontation network;
the extremely minimized game model is as follows:
where E represents an expected value, D [ G (z) ] is a probability that generated data G (z) is judged to be true in D, D (x) represents a probability that true data x is judged to be true in D, the distribution of noise data z is z to Pz, px is a true distribution of data x, D is a discriminator, and G is a generator.
4. The output modeling method of the intermittent distributed power source in the complex operation scene according to claim 3, wherein the processing the first data through the deep neural generation countermeasure network to obtain the output curve of the first data includes:
inputting the category labels to the generator and the discriminator respectively;
performing per-unit processing on the noise data through the generator, constructing a noise data vector by using the noise data subjected to per-unit processing, inserting the class label into the noise data vector to form a generator cross-domain vector, and inputting the generator cross-domain vector into the generator to obtain generated data;
performing per-unit processing on the historical measured data through a discriminator, constructing a measured data vector by using the per-unit processed historical measured data, splicing the measured data vector with the class label to form a discriminator cross-domain vector, and inputting the discriminator cross-domain vector and the generated data into the discriminator to be processed to obtain the output curve.
5. The output modeling method of the intermittent distributed power source under the complex operation scene according to claim 1, wherein the determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model comprises:
generating a plurality of output curves by using the deep neural generation countermeasure network, and clustering the output curves by combining a Gaussian mixture model to generate a plurality of curve output line sets;
and selecting the upper limit and the lower limit of the maximum output curve according to the curve output line set.
6. The intermittent distributed power supply output modeling method under the complex operation scene of claim 5, wherein the clustering the output curves by combining a Gaussian mixture model to generate a plurality of curve output line sets comprises:
approximating any continuous probability distribution by a combination of a plurality of gaussian functions, the probability distribution function being:
wherein alpha is s Representing the weight of the s-th Gaussian partial model, M is the number of Gaussian partial models,
is a parameter of the Gaussian mixture model, mu s Is taken as the mean value of the average value,
representing the variance.
7. The output modeling method of the intermittent distributed power supply in the complex operation scene according to claim 5, wherein the adjusting the configuration of the deep neural generative confrontation network according to the output curve and the upper limit and the lower limit of the output curve to obtain an output data simulation comprises:
and calculating the upper limit and the lower limit of the maximum output curve corresponding to each class label by combining a Gaussian mixture model, and performing output data simulation on the complex scene of the intermittent distributed power supply by combining the output curve and the upper limit and the lower limit of the output.
8. An output modeling device of an intermittent distributed power supply in a complex operation scene is characterized by comprising:
the data acquisition module is used for acquiring first data of the intermittent distributed power supply according to the modeling request, wherein the first data is used for representing at least one of historical measured data, noise data and category labels;
the generation countermeasure module is used for executing processing on the first data through a deep neural generation countermeasure network to obtain an output curve of the first data;
the mixed Gaussian module is used for determining the upper limit and the lower limit of the output curve by adopting a mixed Gaussian model;
and the output simulation module is used for adjusting the configuration of the deep neural generation countermeasure network according to the output curve and the upper limit and the lower limit of the output curve to obtain output data simulation.
9. An electronic device comprising a processor and a memory;
the memory is used for storing programs;
the processor executes the program to realize the output modeling method of the intermittent distributed power source in any one of claims 1-7 under a complex operation scene.
10. A computer-readable storage medium, wherein the storage medium stores a program, which is executed by a processor to implement the method for modeling an output of the intermittent distributed power source according to any one of claims 1 to 7 in a complex operating scenario.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211082099.5A CN115544860A (en) | 2022-09-06 | 2022-09-06 | Output modeling method of intermittent distributed power supply in complex operation scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211082099.5A CN115544860A (en) | 2022-09-06 | 2022-09-06 | Output modeling method of intermittent distributed power supply in complex operation scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115544860A true CN115544860A (en) | 2022-12-30 |
Family
ID=84725093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211082099.5A Pending CN115544860A (en) | 2022-09-06 | 2022-09-06 | Output modeling method of intermittent distributed power supply in complex operation scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115544860A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116842409A (en) * | 2023-08-28 | 2023-10-03 | 南方电网数字电网研究院有限公司 | New energy power generation scene generation method and device, computer equipment and storage medium |
-
2022
- 2022-09-06 CN CN202211082099.5A patent/CN115544860A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116842409A (en) * | 2023-08-28 | 2023-10-03 | 南方电网数字电网研究院有限公司 | New energy power generation scene generation method and device, computer equipment and storage medium |
| CN116842409B (en) * | 2023-08-28 | 2024-02-20 | 南方电网数字电网研究院有限公司 | New energy power generation scene generation method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107133695A (en) | A kind of wind power forecasting method and system | |
| CN112101684B (en) | Plug-in hybrid electric vehicle real-time energy management method and system | |
| Park et al. | Comparing scenario reduction methods for stochastic transmission planning | |
| CN110807554A (en) | Generation method and system based on wind power/photovoltaic classical scene set | |
| CN112241836B (en) | Virtual load leading parameter identification method based on incremental learning | |
| CN110210670A (en) | A kind of prediction technique based on power-system short-term load | |
| CN115544860A (en) | Output modeling method of intermittent distributed power supply in complex operation scene | |
| Li et al. | GMM-HMM-based medium-and long-term multi-wind farm correlated power output time series generation method | |
| Abdelaziz et al. | Convolutional Neural Network With Genetic Algorithm for Predicting Energy Consumption in Public Buildings | |
| CN117674421A (en) | Digital twin power grid data transmission system | |
| CN116933037A (en) | Photovoltaic output prediction method based on multi-model fusion and related device | |
| CN116010831A (en) | Combined clustering scene reduction method and system based on potential decision result | |
| de Santi et al. | Field-level simulation-based inference with galaxy catalogs: the impact of systematic effects | |
| CN116070458A (en) | New wind power plant scene generation method based on RAC-GAN | |
| CN115907228A (en) | Short-term power load prediction analysis method based on PSO-LSSVM | |
| Li et al. | Quantifying the impact of building load forecasts on optimizing energy storage systems | |
| CN114169118A (en) | Power distribution network topological structure identification method considering distributed power supply output correlation | |
| CN114528679A (en) | Multi-mode fault early warning method and system for numerical control system | |
| CN107102543A (en) | The forming method and device of a kind of energy router anti-interference controller | |
| Cho et al. | Application of Parallel ANN-PSO to Hourly Solar PV Estimation | |
| CN113011597A (en) | Deep learning method and device for regression task | |
| CN111339646A (en) | Temperature data enhancement method for full-automatic control | |
| CN117878932B (en) | Method and device for evaluating available transmission capacity of multi-region power system | |
| Mehdary et al. | Hyperparameter Optimization with Genetic Algorithms and XGBoost: A Step Forward in Smart Grid Fraud Detection | |
| CN117892886B (en) | Ship motion extremely short-term probability forecasting method and system based on confidence interval |
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 |