CN115099526A - Mass distributed renewable energy power prediction method and related device - Google Patents
Mass distributed renewable energy power prediction method and related device Download PDFInfo
- Publication number
- CN115099526A CN115099526A CN202210900992.8A CN202210900992A CN115099526A CN 115099526 A CN115099526 A CN 115099526A CN 202210900992 A CN202210900992 A CN 202210900992A CN 115099526 A CN115099526 A CN 115099526A
- Authority
- CN
- China
- Prior art keywords
- cloud
- power
- renewable energy
- power prediction
- distributed renewable
- 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
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000011218 segmentation Effects 0.000 claims abstract description 22
- 238000013135 deep learning Methods 0.000 claims abstract description 13
- 230000008859 change Effects 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims abstract description 12
- 238000005516 engineering process Methods 0.000 claims abstract description 11
- 230000015654 memory Effects 0.000 claims description 26
- 238000013527 convolutional neural network Methods 0.000 claims description 17
- 238000013528 artificial neural network Methods 0.000 claims description 12
- 238000013256 Gubra-Amylin NASH model Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 12
- 238000012545 processing Methods 0.000 description 9
- 230000008901 benefit Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000002085 persistent effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
- G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Marketing (AREA)
- Medical Informatics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Remote Sensing (AREA)
- Astronomy & Astrophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method and a related device for predicting power of massive distributed renewable energy sources, wherein the method comprises the following steps: acquiring cloud picture images corresponding to a power region to be predicted, and identifying and classifying the cloud picture images through an identification and segmentation technology to obtain m different kinds of cloud layers and marking the cloud layers; establishing a power prediction model of mass distributed renewable energy sources based on a cloud picture by adopting a deep learning mode; according to m different types, dividing a plurality of distributed renewable energy power predictions to which the cloud layers belong into m Map tasks; performing power prediction on each Map task through a power prediction model to obtain m power prediction results; determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using a Reduce function to obtain the renewable energy power of the area to be predicted. The method improves accuracy of power prediction of the massive distributed renewable energy sources, and greatly shortens calculation time of the power prediction.
Description
Technical Field
The application relates to the technical field of renewable energy sources, in particular to a method and a related device for predicting power of massive distributed renewable energy sources.
Background
The result of power system power prediction is related to the specification of the power department production plan and the scheduling operation condition. Particularly, the power condition of the renewable energy is closely related to natural condition factors such as weather, temperature, humidity and the like, and has a strong nonlinear relation, and when the traditional power prediction mode is applied to the renewable energy, the error is large; meanwhile, due to the influence of factors such as solar radiation and weather, the output power of the massive distributed renewable energy sources has volatility and intermittency, and the safe, stable and economic operation of the power system can be influenced by the large-scale new energy grid-connected operation. The output power of the massive distributed renewable energy sources is predicted, the coordination of a conventional power supply and the massive distributed renewable energy source power generation is favorably arranged by a power system dispatching department, a dispatching plan is reasonably adjusted, the safe and stable operation of a power grid is ensured, the renewable energy sources are fully utilized, and the social benefit and the economic benefit of the massive distributed renewable energy source power generation are exerted. Therefore, it is very important to accurately predict the output power of the massive distributed renewable energy sources. How to efficiently, quickly and accurately predict the power of massive distributed renewable energy sources is a key problem of power prediction in the context of big data.
Disclosure of Invention
The application provides a massive distributed renewable energy power prediction method and a related device, which are used for improving accuracy and efficiency of massive distributed renewable energy power prediction.
In view of this, the first aspect of the present application provides a method for predicting power of massive distributed renewable energy sources, where the method includes:
acquiring cloud picture images corresponding to a power region to be predicted, and identifying and classifying the cloud picture images through an identification and segmentation technology to obtain m different kinds of cloud layers and marking the cloud layers;
establishing a power prediction model of mass distributed renewable energy sources based on a cloud picture by adopting a deep learning mode;
according to m different types, dividing a plurality of distributed renewable energy power predictions to which the cloud layers belong into m Map tasks;
performing power prediction on each Map task through the power prediction model to obtain m power prediction results;
and determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using the Reduce function to obtain the renewable energy power of the area to be predicted.
Optionally, the dividing, according to m different categories, the plurality of distributed renewable energy power predictions to which the cloud layer belongs into m Map tasks further includes:
acquiring the condition that m Map tasks occupy internal memory when the server runs respectively;
and carrying out Map task division again on the Map tasks occupying the memory size exceeding the preset memory threshold value of the server, and increasing the number of the servers so as to obtain n Map tasks.
Optionally, the determining, according to the cloud layer change condition, the corresponding weights of the m Map tasks, and combining the m power prediction results by using a Reduce function to obtain the renewable energy power of the area to be predicted specifically includes:
and comparing the cloud layer classification result obtained by classifying at the current moment with the historical cloud layer classification result, and determining the proportion of the current cloud layer type in the historical cloud layer type, thereby determining the corresponding weights of the m Map tasks, and adding the m power prediction results with the corresponding weights to obtain the renewable energy power of the area to be predicted.
Optionally, the method includes acquiring a cloud image corresponding to a power region to be predicted, and identifying and classifying the cloud image through an identification and segmentation technology to obtain m different types of cloud layers and perform labeling, and specifically includes:
inputting the cloud picture image as an original cloud picture image into a GAN model to produce a large number of new cloud picture images;
and inputting the original cloud picture image and the new cloud picture image into a segmentation model for segmentation, and then using the original cloud picture image and the new cloud picture image as input of a CNN neural network, and identifying and classifying the cloud picture images through the CNN neural network to obtain m different kinds of cloud layers and marking the cloud layers.
A second aspect of the present application provides a system for predicting power of mass distributed renewable energy sources, the system comprising:
the system comprises a classification unit, a prediction unit and a prediction unit, wherein the classification unit is used for acquiring cloud picture images corresponding to a region to be predicted of power, and identifying and classifying the cloud picture images through an identification and segmentation technology to obtain m different types of cloud layers and mark the cloud layers;
the system comprises an establishing unit, a power prediction unit and a power prediction unit, wherein the establishing unit is used for establishing a power prediction model of massive distributed renewable energy sources based on a cloud picture in a deep learning mode;
the system comprises a first dividing unit, a second dividing unit and a third dividing unit, wherein the first dividing unit is used for dividing a plurality of distributed renewable energy power predictions of cloud layers into m Map tasks according to m different types;
the prediction unit is used for carrying out power prediction on each Map task through the power prediction model to obtain m power prediction results;
and the calculating unit is used for determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using the Reduce function to obtain the renewable energy power of the area to be predicted.
Optionally, the method further comprises: a second dividing unit;
the second dividing unit is used for acquiring the memory occupied condition of the m Map tasks when the server runs respectively; and carrying out Map task division again on the Map tasks occupying the memory size exceeding the preset memory threshold of the server, and increasing the number of the servers so as to obtain n Map tasks.
Optionally, the computing unit is specifically configured to:
and comparing the cloud layer classification result obtained by classifying at the current moment with the historical cloud layer classification result, and determining the proportion of the current cloud layer type in the historical cloud layer type, thereby determining the corresponding weights of the m Map tasks, and adding the m power prediction results with the corresponding weights to obtain the renewable energy power of the area to be predicted.
Optionally, the classification unit is specifically configured to:
inputting the cloud picture image as an original cloud picture image into a GAN model to produce a large number of new cloud picture images;
and inputting the original cloud picture image and the new cloud picture image into a segmentation model for segmentation, and then using the original cloud picture image and the new cloud picture image as input of a CNN neural network, and identifying and classifying the cloud picture images through the CNN neural network to obtain m different kinds of cloud layers and marking the cloud layers.
A third aspect of the present application provides a mass distributed renewable energy power prediction device, the device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to perform the steps of the method for predicting power of mass distributed renewable energy sources according to the first aspect.
A fourth aspect of the present application provides a computer-readable storage medium for storing program code for executing the method for predicting mass distributed renewable energy power according to the first aspect.
According to the technical scheme, the method has the following advantages:
the application provides a method for predicting power of massive distributed renewable energy sources, which comprises the following steps: acquiring cloud picture images corresponding to a power region to be predicted, and identifying and classifying the cloud picture images through an identification and segmentation technology to obtain m different kinds of cloud layers and marking the cloud layers; establishing a power prediction model of mass distributed renewable energy sources based on a cloud picture by adopting a deep learning mode; according to m different types, dividing a plurality of distributed renewable energy power predictions to which the cloud layers belong into m Map tasks; performing power prediction on each Map task through a power prediction model to obtain m power prediction results; determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using a Reduce function to obtain the renewable energy power of the area to be predicted.
Compared with the prior art, the method for predicting the power of the massive distributed renewable energy sources based on the MapReduce and the cloud pictures is provided. And under a MapReduce parallel programming framework, power prediction is carried out on massive distributed renewable energy sources in the region by establishing a distributed renewable energy source power prediction model based on cloud map information. The method improves accuracy of prediction of the power of the massive distributed renewable energy sources, greatly shortens calculation time of power prediction, and improves efficiency.
Drawings
Fig. 1 is a schematic flowchart of an embodiment of a method for predicting power of mass distributed renewable energy provided in an embodiment of the present application;
fig. 2 is a schematic flow chart of an embodiment of a power prediction system for distributed renewable energy resources in the embodiment of the present application;
fig. 3 is a flow chart of cloud image preprocessing provided in an embodiment of the present application;
fig. 4 is a schematic diagram of a GAN network provided in an embodiment of the present application;
fig. 5 is a diagram of an FCN architecture provided in an embodiment of the present application;
fig. 6 is a flowchart of cloud image segmentation provided in an embodiment of the present application;
fig. 7 is a flow chart of a cloud-based mass distributed renewable energy power prediction model provided in an embodiment of the present application.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, 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 application.
Referring to fig. 1, a method for predicting power of a massive distributed renewable energy source provided in an embodiment of the present application includes:
it should be noted that the preprocessing of the cloud image includes two important steps of segmentation and identification of cloud regions and surface regions, and segmentation and identification of different cloud layer classes. In this embodiment, a GAN model, an FCN model, and a CNN model are combined to divide a cloud map into different areas, and the areas are corresponding to different weather types (e.g., clear sky, cloudy day, rainy day, and wind), and the process is shown in fig. 3.
1) The GAN model generates a cloud image dataset:
a Generative Adaptive Networks (GAN) is a deep learning model, and the main components of this network include: a generation model G (generator) and a discrimination model D (discriminator), as shown in fig. 4, in the training process, G will continuously improve the counterfeiting capability, and generate an image as false as possible to make D misjudge, and the goal of D is to distinguish the image generated by G from the real data as much as possible. Thus, G and D constitute a dynamic "gaming process".
The module takes a known cloud image as input, and can obtain more data sets after the GAN model is used for training so as to provide a foundation for the CNN model.
2) And (3) segmenting the model:
a satellite cloud graph segmentation model is designed based on a full convolution neural network (FCN), the structure diagram of the FCN is shown in figure 5, and the segmentation flow is shown in figure 6.
3) CNN neural network:
convolutional Neural Networks (CNN) is a feedforward type of Neural network, and is also one of the most widely used models in practical applications. The CNN mainly comprises a convolution layer, a pooling layer, a full-connection layer and a softmax classification output layer.
The module takes a cloud picture data set as input, and divides the cloud picture into five types of clear sky, cloudy sky, rainy day and wind through a CNN convolution layer, a pooling layer, a full connection layer and a softmax classification output layer, wherein the number of the cloud picture is A, B, C, D, E.
102, establishing a power prediction model of mass distributed renewable energy sources based on a cloud picture by adopting a deep learning mode;
it should be noted that, in the conventional method, feature extraction mainly depends on an extractor designed manually, professional knowledge and a complex parameter adjusting process are required, and each method is specific to a specific application and has poor generalization ability and robustness. Deep learning mainly includes data-driven feature extraction, deep and data set-specific feature representation can be obtained according to learning of a large number of samples, expression of the data set is more efficient and accurate, extracted abstract features are stronger in robustness and better in generalization capability, and the extracted abstract features can be end-to-end. Therefore, a deep learning mode is used for establishing a massive distributed renewable energy power prediction model based on the satellite cloud picture.
As shown in fig. 7, the CNN convolutional neural network is used to perform image processing on the geostationary satellite cloud images, so as to extract characteristic information related to the power prediction of the distributed renewable energy source. When the power of the massive distributed renewable energy sources is predicted, the extracted characteristic information and the traditional historical meteorological data are input into a Long Short-Term Memory network (LSTM) neural network together, and the accuracy of the prediction result can be effectively improved.
103, according to m different types, dividing a plurality of distributed renewable energy power predictions to which the cloud layers belong into m Map tasks;
further, the embodiment acquires the condition that m Map tasks occupy the memory when the server runs respectively; and carrying out Map task division again on the Map tasks occupying the memory size exceeding the preset memory threshold of the server, and increasing the number of the servers so as to obtain n Map tasks.
It should be noted that batch processing is suitable for tasks that require processing of large amounts of data. Whether the data set is processed directly from the persistent storage device or first loaded into memory, the batch processing system takes the amount of data into account during the design process and provides sufficient processing resources. Batch processing is often used to analyze historical data because it performs extremely well in dealing with large amounts of persistent data.
When mass data are required to be processed, a performance improvement mode for a certain single server is limited, such as adding a memory and improving the main frequency of a CPU; if a new server is added, the calculation speed is increased by adopting a parallel calculation mode, and the complexity and the development difficulty of the program are greatly increased. Google engineers find that the parallel computing mode can be described by a program model consisting of two processes of Map mapping and Reduce simplification, and the rest can be standardized as long as developers finish the two processes of Map and Reduce. Therefore, Google encapsulates the complexity of parallel computing on the basis of GFS, and realizes the computing framework of MapReduce.
The core idea of MapReduce is "divide and conquer". The Map task is 'minute', operation is carried out on some independent data, and the operation result is stored in a new list; the task of Reduce is "merge", which merges and generalizes the new list created by the Map function. In the invention, when data to be processed arrives, the Map divides the data to control and schedule the whole task, divides the data into a plurality of parts with approximate sizes, processes the data by combining a BP algorithm, combines the parts by Reduce, executes the task in parallel and returns the data, thereby realizing the data processing.
The invention provides a massive distributed renewable energy power prediction method based on MapReduce and cloud map information. For power prediction, distributed renewable energy sources in the same or similar areas in cloud layer conditions have similarity in processing conditions, and therefore, according to m different cloud layer types in step 101, a plurality of distributed renewable energy source power predictions to which a cloud layer belongs are divided into m Map tasks. In addition, in order to avoid that the occupation of the server is overlarge during the execution of the Map task due to single cloud layer type or large data quantity difference in the prediction region, the tasks exceeding 75% of the server threshold value in the m Map tasks are divided again, the number of the servers is increased, and finally the n Map tasks are obtained through division. Therefore, when the data is processed in a "division" manner, two steps are included: dividing mass distributed renewable energy sources into a plurality of Map tasks according to different current cloud layer types; secondly, observing the running condition of the server when each Map task is executed, and setting a threshold value for 75% of the memory of the server; if the Map data volume exceeds the threshold value, the task is decomposed into a plurality of Map tasks again, and the number of the servers is increased.
104, performing power prediction on each Map task through a power prediction model to obtain m power prediction results;
it should be noted that, by using the distributed renewable energy power prediction model based on deep learning and cloud maps, the accuracy of renewable energy power prediction is improved. Therefore, in the present embodiment, when power prediction is performed on each distributed renewable energy, power prediction is performed on each Map task by using the massive distributed renewable energy power prediction model based on the deep learning mode and the cloud Map established in step 102, so as to obtain n power prediction results.
And 105, determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using a Reduce function to obtain the renewable energy power of the area to be predicted.
Further, in this embodiment, step 105 is explained based on the n Map tasks obtained in step 103.
It should be noted that, in general, the satellite cloud map is updated generally once every 10 min. When the overall prediction of the small-magnitude distributed renewable energy power is performed, because the cloud Map time preprocessed in the step 101 is the current time, according to the Map task division principle in the step 103, under the current satellite cloud Map division result, the cloud layer type in each Map task is single, in order to further improve the prediction accuracy, the satellite cloud Map information updated every 10min in the previous 1h needs to be compared with the cloud Map division result at the current time, the proportion of the cloud layer of the type in each satellite cloud Map in the previous time is found, the average value of the proportions is taken to obtain the weight, the weights of n Map tasks are respectively set as a1 and a2 … … an, the weights are respectively multiplied by the Map task results, and the addition is performed, namely, the Reduce process is completed, and the prediction result of the marine distributed renewable energy power in the area is obtained.
The embodiment provides a massive distributed renewable energy power prediction method based on MapReduce and a satellite cloud picture, which comprises 101 and cloud picture preprocessing. And finding out cloud picture images corresponding to the areas needing power prediction, and identifying, classifying and labeling different types of cloud layers in the cloud pictures by applying a cloud picture identification technology. 102. And establishing a cloud-map-based massive distributed renewable energy power prediction model by adopting a deep learning mode. 103. And (3) dividing Map tasks: dividing a mass of distributed renewable energy into a plurality of Map tasks according to the difference of current cloud layer types; secondly, observing the running condition of the server when each Map task is executed, and setting a threshold value for 75% of the memory of the server; if the Map data volume exceeds the threshold value, the task is decomposed into a plurality of Map tasks again, and the number of the servers is increased. 104. And executing the Map task: and performing power prediction on each Map task by using a power prediction model, and storing a prediction result in a cloud. 105. And executing the Reduce task. And recording the change condition of the cloud layer area every 10min in the previous 1h, multiplying the Map task results by corresponding weights respectively according to the change condition, and combining the prediction results of the mass distributed renewable energy power of different areas by using a Reduce function to obtain the predicted value of the mass distributed renewable energy power in the area 1 h.
The above is a method for predicting power of mass distributed renewable energy sources provided in the embodiment of the present application, and the following is a system for predicting power of mass distributed renewable energy sources provided in the embodiment of the present application.
Referring to fig. 2, in an embodiment of the present application, a system for predicting power of mass distributed renewable energy sources includes:
the classification unit 201 is configured to obtain a cloud image corresponding to a region to be predicted of power, and identify and classify the cloud image by using an identification and segmentation technology to obtain m different types of cloud layers and mark the cloud layers;
the establishing unit 202 is used for establishing a power prediction model of the mass distributed renewable energy sources based on the cloud picture in a deep learning mode;
the first dividing unit 203 is configured to divide the power prediction of the distributed renewable energy sources to which the cloud layer belongs into m Map tasks according to m different categories;
the prediction unit 204 is configured to perform power prediction on each Map task through a power prediction model to obtain m power prediction results;
and the calculating unit 205 is configured to determine corresponding weights of the m Map tasks according to the cloud layer change condition, and merge the m power prediction results by using a Reduce function to obtain the renewable energy power of the area to be predicted.
Further, an embodiment of the present application further provides a device for predicting power of mass distributed renewable energy sources, where the device includes a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the method for predicting massive distributed renewable energy power according to the embodiment of the method according to the instructions in the program code.
Further, an embodiment of the present application also provides a computer-readable storage medium, which is used for storing program codes, where the program codes are used for executing the method for predicting the power of the distributed renewable energy sources in the mass.
It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described systems and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The terms "first," "second," "third," "fourth," and the like in the description of the present application and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged under appropriate circumstances such that the embodiments of the application described herein may be implemented, for example, in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that in the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" for describing an association relationship of associated objects, indicating that there may be three relationships, e.g., "a and/or B" may indicate: only A, only B and both A and B are present, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single item(s) or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b and c may be single or plural.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit 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 application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in 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 application. 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 above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present application.
Claims (10)
1. A method for predicting power of massive distributed renewable energy sources is characterized by comprising the following steps:
acquiring cloud picture images corresponding to a power region to be predicted, and identifying and classifying the cloud picture images through an identification and segmentation technology to obtain m different kinds of cloud layers and marking the cloud layers;
establishing a power prediction model of massive distributed renewable energy sources based on a cloud picture by adopting a deep learning mode;
according to m different types, dividing a plurality of distributed renewable energy power predictions to which the cloud layers belong into m Map tasks;
performing power prediction on each Map task through the power prediction model to obtain m power prediction results;
and determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using the Reduce function to obtain the renewable energy power of the area to be predicted.
2. The method for predicting the power of distributed renewable energy resources of claim 1, wherein the step of predicting the power of the distributed renewable energy resources to which the cloud layer belongs is divided into m Map tasks according to m different categories further comprises:
acquiring the condition that m Map tasks occupy internal memory when the server runs respectively;
and carrying out Map task division again on the Map tasks occupying the memory size exceeding the preset memory threshold value of the server, and increasing the number of the servers so as to obtain n Map tasks.
3. The method for predicting the power of the massively distributed renewable energy sources according to claim 1, wherein the determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using a Reduce function to obtain the renewable energy source power of the area to be predicted specifically comprises:
and comparing the cloud layer classification result obtained by classifying at the current moment with the historical cloud layer classification result, and determining the proportion of the current cloud layer type in the historical cloud layer type, thereby determining the corresponding weights of the m Map tasks, and adding the m power prediction results with the corresponding weights to obtain the renewable energy power of the area to be predicted.
4. The method for predicting the power of the mass distributed renewable energy sources according to claim 1, wherein cloud images corresponding to the power region to be predicted are obtained, the cloud images are identified and classified through an identification and segmentation technology, m different types of cloud layers are obtained and marked, and the method specifically comprises the following steps:
inputting the cloud picture image as an original cloud picture image into a GAN model to produce a large number of new cloud picture images;
and inputting the original cloud picture image and the new cloud picture image into a segmentation model for segmentation, and then using the original cloud picture image and the new cloud picture image as input of a CNN neural network, and identifying and classifying the cloud picture images through the CNN neural network to obtain m different kinds of cloud layers and marking the cloud layers.
5. A system for forecasting power of a massive distributed renewable energy source, comprising:
the system comprises a classification unit, a prediction unit and a prediction unit, wherein the classification unit is used for acquiring cloud picture images corresponding to a region to be predicted of power, and identifying and classifying the cloud picture images through an identification and segmentation technology to obtain m different types of cloud layers and mark the cloud layers;
the system comprises an establishing unit, a power prediction unit and a power prediction unit, wherein the establishing unit is used for establishing a power prediction model of massive distributed renewable energy sources based on a cloud picture in a deep learning mode;
the first dividing unit is used for dividing a plurality of distributed renewable energy power predictions to which the cloud layers belong into m Map tasks according to m different types;
the prediction unit is used for carrying out power prediction on each Map task through the power prediction model to obtain m power prediction results;
and the calculating unit is used for determining the corresponding weights of the m Map tasks according to the cloud layer change condition, and combining the m power prediction results by using the Reduce function to obtain the renewable energy power of the area to be predicted.
6. The system for forecasting distributed renewable energy mass power of claim 5, further comprising: a second dividing unit;
the second dividing unit is used for acquiring the memory occupied condition of the m Map tasks when the server runs respectively; and carrying out Map task division again on the Map tasks occupying the memory size exceeding the preset memory threshold value of the server, and increasing the number of the servers so as to obtain n Map tasks.
7. The system according to claim 5, wherein the computing unit is specifically configured to:
and comparing the cloud layer classification result obtained by classifying at the current moment with the historical cloud layer classification result, and determining the proportion of the current cloud layer type in the historical cloud layer type, thereby determining the corresponding weights of the m Map tasks, and adding the m power prediction results with the corresponding weights to obtain the renewable energy power of the area to be predicted.
8. The system according to claim 5, wherein the classification unit is specifically configured to:
inputting the cloud picture image as an original cloud picture image into a GAN model to produce a large number of new cloud picture images;
and inputting the original cloud picture image and the new cloud picture image into a segmentation model for segmentation, and then using the original cloud picture image and the new cloud picture image as input of a CNN neural network, and identifying and classifying the cloud picture images through the CNN neural network to obtain m different kinds of cloud layers and marking the cloud layers.
9. A mass distributed renewable energy power prediction device, the device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to execute the method for predicting distributed renewable energy mass power according to any one of claims 1 to 4 according to instructions in the program code.
10. A computer-readable storage medium for storing program code for performing the method for power prediction for mass distributed renewable energy sources of any of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210900992.8A CN115099526A (en) | 2022-07-28 | 2022-07-28 | Mass distributed renewable energy power prediction method and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210900992.8A CN115099526A (en) | 2022-07-28 | 2022-07-28 | Mass distributed renewable energy power prediction method and related device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115099526A true CN115099526A (en) | 2022-09-23 |
Family
ID=83300557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210900992.8A Pending CN115099526A (en) | 2022-07-28 | 2022-07-28 | Mass distributed renewable energy power prediction method and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115099526A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106022521A (en) * | 2016-05-19 | 2016-10-12 | 四川大学 | Hadoop framework-based short-term load prediction method for distributed BP neural network |
| CN106779154A (en) * | 2016-11-22 | 2017-05-31 | 浙江工业大学 | Area distribution formula photovoltaic power generation output forecasting method based on satellite cloud picture |
| CN111738327A (en) * | 2020-06-18 | 2020-10-02 | 河海大学常州校区 | Ultra-short-term irradiation prediction method based on typical cloud shielding irradiation difference |
| CN113128793A (en) * | 2021-05-19 | 2021-07-16 | 中国南方电网有限责任公司 | Photovoltaic power combination prediction method and system based on multi-source data fusion |
| CN113537561A (en) * | 2021-06-09 | 2021-10-22 | 中国电力科学研究院有限公司 | Ultra-short-term solar radiation prediction method and system based on foundation cloud picture |
| CN115186865A (en) * | 2022-05-27 | 2022-10-14 | 李波 | Mass distributed renewable energy power prediction method based on MapReduce and cloud picture |
-
2022
- 2022-07-28 CN CN202210900992.8A patent/CN115099526A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106022521A (en) * | 2016-05-19 | 2016-10-12 | 四川大学 | Hadoop framework-based short-term load prediction method for distributed BP neural network |
| CN106779154A (en) * | 2016-11-22 | 2017-05-31 | 浙江工业大学 | Area distribution formula photovoltaic power generation output forecasting method based on satellite cloud picture |
| CN111738327A (en) * | 2020-06-18 | 2020-10-02 | 河海大学常州校区 | Ultra-short-term irradiation prediction method based on typical cloud shielding irradiation difference |
| CN113128793A (en) * | 2021-05-19 | 2021-07-16 | 中国南方电网有限责任公司 | Photovoltaic power combination prediction method and system based on multi-source data fusion |
| CN113537561A (en) * | 2021-06-09 | 2021-10-22 | 中国电力科学研究院有限公司 | Ultra-short-term solar radiation prediction method and system based on foundation cloud picture |
| CN115186865A (en) * | 2022-05-27 | 2022-10-14 | 李波 | Mass distributed renewable energy power prediction method based on MapReduce and cloud picture |
Non-Patent Citations (1)
| Title |
|---|
| 王昊;师卫;李欢;: "Hadoop下基于贝叶斯网络的气象数据挖掘研究", 电子器件, no. 04, 20 August 2016 (2016-08-20), pages 841 - 846 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deng et al. | Multi-scale convolutional neural network with time-cognition for multi-step short-term load forecasting | |
| Ceci et al. | Predictive modeling of PV energy production: How to set up the learning task for a better prediction? | |
| CN112180471B (en) | Weather forecasting method, device, equipment and storage medium | |
| CN113128478B (en) | Model training method, pedestrian analysis method, device, equipment and storage medium | |
| CN114792156A (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| CN114493052B (en) | Multi-model fusion self-adaptive new energy power prediction method and system | |
| CN110992307A (en) | Insulator positioning and identifying method and device based on YOLO | |
| CN110889560A (en) | Express delivery sequence prediction method with deep interpretability | |
| CN116186548A (en) | Power load prediction model training method and power load prediction method | |
| CN116799796A (en) | Photovoltaic power generation power prediction method, device, equipment and medium | |
| Dutt et al. | Artificial intelligence and technology in weather forecasting and renewable energy systems: emerging techniques and worldwide studies | |
| Lin et al. | Integrated circuit board object detection and image augmentation fusion model based on YOLO | |
| CN113536944A (en) | Distribution line inspection data identification and analysis method based on image identification | |
| CN117391257A (en) | Road congestion condition prediction method and device | |
| Zhai et al. | Combining PSO-SVR and Random Forest Based Feature Selection for Day-ahead Peak Load Forecasting. | |
| CN115186865A (en) | Mass distributed renewable energy power prediction method based on MapReduce and cloud picture | |
| Sharma et al. | Deep Learning Based Prediction Of Weather Using Hybrid_stacked Bi-Long Short Term Memory | |
| CN111950753A (en) | Scenic spot passenger flow prediction method and device | |
| CN116739649A (en) | User response potential evaluation method and device | |
| CN115099526A (en) | Mass distributed renewable energy power prediction method and related device | |
| CN115687948A (en) | Power special transformer user unsupervised classification method based on load curve | |
| CN115687788A (en) | Intelligent business opportunity recommendation method and system | |
| Shaji et al. | Weather Prediction Using Machine Learning Algorithms | |
| CN114118410A (en) | Method, device and storage medium for extracting node feature of graph structure | |
| CN112434650A (en) | Multi-spectral image building change detection method and 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 |