CN117713083A - Power system short-term power load prediction system and method based on data management platform - Google Patents
Power system short-term power load prediction system and method based on data management platform Download PDFInfo
- Publication number
- CN117713083A CN117713083A CN202311777620.1A CN202311777620A CN117713083A CN 117713083 A CN117713083 A CN 117713083A CN 202311777620 A CN202311777620 A CN 202311777620A CN 117713083 A CN117713083 A CN 117713083A
- Authority
- CN
- China
- Prior art keywords
- power load
- time sequence
- short
- power
- time
- 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
- 238000013523 data management Methods 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000013598 vector Substances 0.000 claims description 214
- 238000005259 measurement Methods 0.000 claims description 30
- 238000004458 analytical method Methods 0.000 claims description 22
- 238000012937 correction Methods 0.000 claims description 19
- 230000011218 segmentation Effects 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 abstract description 9
- 238000012545 processing Methods 0.000 abstract description 6
- 238000012300 Sequence Analysis Methods 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 9
- 230000007774 longterm Effects 0.000 description 6
- 238000010248 power generation Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000001932 seasonal effect Effects 0.000 description 3
- 238000012731 temporal analysis Methods 0.000 description 3
- 238000000700 time series analysis Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013179 statistical model Methods 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Water Supply & Treatment (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Public Health (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A power system short-term power load prediction system and a power system short-term power load prediction method based on a data management platform are capable of monitoring and collecting power load values of a power system in real time, and introducing a data processing and analyzing algorithm into the data management platform to conduct time sequence analysis of the power load values, so that short-term power load prediction of the power system is conducted based on characteristic information of power load time sequence data, corresponding power scheduling optimization is conducted to adapt to dynamic changes of the power system, and therefore operation efficiency and reliability of the power system are improved.
Description
Technical Field
The present disclosure relates to the field of intelligent prediction technologies, and more particularly, to a system and method for predicting short-term power load of a power system based on a data management platform.
Background
Short-term power load prediction of a power system is of great importance for power system operation and scheduling. Accurately predicting the electrical load may help the utility to rationally schedule power generation, optimize power scheduling, and improve reliability and economy of the power system.
In the past, traditional power load prediction methods are mainly based on statistical models and time series analysis, and are usually used for modeling and prediction by using historical load data, but nonlinear characteristics and time-varying changes of the power load cannot be captured, and prediction accuracy is limited. In particular, because the electrical load is affected by a variety of factors, such as climate, holidays, economic development, etc., it has complex non-linear, stochastic and time-varying characteristics that present significant challenges for electrical load prediction.
Accordingly, an optimized power system short-term power load prediction system is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a power system short-term power load prediction system and a power system short-term power load prediction method based on a data management platform, which can be used for carrying out time sequence analysis on power load values by monitoring and collecting the power load values of a power system in real time and introducing a data processing and analyzing algorithm into the data management platform, so that the short-term power load prediction of the power system is carried out based on characteristic information of the power load time sequence data, and corresponding power dispatching optimization is carried out to adapt to dynamic changes of the power system, thereby improving the running efficiency and reliability of the power system.
In a first aspect, a power system short-term power load prediction system based on a data management platform is provided, comprising:
the power load data acquisition module is used for acquiring power load values of the power system at a plurality of preset time points in a preset time period;
the power load local time sequence characteristic analysis module is used for carrying out local time sequence characteristic analysis on the power load time sequence input vectors after arranging the power load values of the plurality of preset time points into the power load time sequence input vectors according to the time dimension so as to obtain a sequence of the power load local time sequence characteristic vectors;
the power load short-time fluctuation semantic measurement module is used for carrying out power load short-time fluctuation semantic measurement analysis on every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors so as to obtain power load short-time fluctuation feature vectors;
the power load multi-dimensional characteristic fusion module is used for fusing the power load short-time fluctuation characteristic vector and the global power load time sequence characteristic vector to obtain a power load multi-dimensional time sequence characteristic vector as a power load multi-dimensional time sequence characteristic after splicing the sequences of the power load local time sequence characteristic vectors to obtain the global power load time sequence characteristic vector;
and the short-time power load prediction module is used for determining a short-time power load predicted value based on the power load multi-dimensional multi-scale time sequence characteristic.
In a second aspect, a method for predicting short-term power load of a power system based on a data management platform is provided, which includes:
acquiring power load values of a power system at a plurality of preset time points in a preset time period;
after the power load values at the plurality of preset time points are arranged into power load time sequence input vectors according to the time dimension, carrying out local time sequence feature analysis on the power load time sequence input vectors to obtain a sequence of power load local time sequence feature vectors;
carrying out power load short-time fluctuation semantic metric analysis on every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain power load short-time fluctuation feature vectors;
after the sequence of the power load local time sequence feature vector is spliced to obtain a global power load time sequence feature vector, the power load short-time fluctuation feature vector and the global power load time sequence feature vector are fused to obtain a power load multi-dimensional multi-scale time sequence feature vector as a power load multi-dimensional multi-scale time sequence feature;
a short-term power load predictor is determined based on the power load multi-dimensional multi-scale timing characteristic.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of a power system short-term power load prediction system based on a data management platform according to an embodiment of the present application.
Fig. 2 is a flowchart of a method for predicting short-term power load of a power system based on a data management platform according to an embodiment of the present application.
Fig. 3 is a schematic architecture diagram of a method for predicting short-term power load of a power system based on a data management platform according to an embodiment of the present application.
Fig. 4 is a schematic view of a scenario of a power system short-term power load prediction system based on a data management platform according to an embodiment of the present application.
Detailed Description
The following description of the technical solutions in the embodiments of the present application will be made with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Unless defined otherwise, all technical and scientific terms used in the examples of this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In the description of the embodiments of the present application, unless otherwise indicated and defined, the term "connected" should be construed broadly, and for example, may be an electrical connection, may be a communication between two elements, may be a direct connection, or may be an indirect connection via an intermediary, and it will be understood by those skilled in the art that the specific meaning of the term may be understood according to the specific circumstances.
It should be noted that, the term "first\second\third" in the embodiments of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that the embodiments of the present application described herein may be implemented in sequences other than those illustrated or described herein.
Short-term power load prediction of a power system is of great importance for power system operation and scheduling. The power company needs to make a power generation plan according to the prediction condition of the power load, accurate load prediction can help the company reasonably arrange the running time and load distribution of the power generation equipment so as to meet the power consumption requirement of users, and through predicting the load, the power company can avoid excessive or insufficient power generation, improve the utilization rate of the power generation equipment and reduce the fuel cost and the operation cost.
The power load prediction can help the power company to optimize power scheduling, namely, power resources are reasonably allocated according to a load prediction result, and the power company can reasonably arrange load distribution of a power transmission line through predicting loads, so that line overload and power transmission loss are avoided. In addition, the load prediction can also guide the power company to adjust the running state of the generator set in different time periods so as to realize the optimal economic benefit and system stability.
Accurate load prediction can help the electric company to discover potential electric power system problems in time and take corresponding measures to avoid or mitigate the influence of system faults. For example, if it is predicted that the load for a certain period of time will exceed the system capacity, the utility company may take measures in advance, such as starting a standby genset or adjusting load distribution, to ensure stable operation of the system.
The accurate load prediction can help the electric power company optimize energy purchasing and power generation cost, and through predicting the load, the electric power company can accurately predict the future power demand, so that the time and the quantity of purchasing power are reasonably arranged. In addition, the load prediction can also guide the electric company to adjust the running state of the generator set in different time periods so as to realize the optimal economic benefit and energy utilization rate.
Traditional power load prediction methods are mainly based on statistical models and time series analysis, the methods are generally used for modeling and predicting historical load data, a moving average method is used for calculating the average value of the historical load data by using a sliding window, and then the average value is used as a predicted value of future load, and the method is simple and easy to use, but cannot capture the nonlinear and time-varying characteristics of the load. The seasonal decomposition method decomposes the load data into trend, season and random components, and then models and predicts these components, respectively, and future load trend and seasonal variation can be predicted by modeling the season and trend components. However, this approach has difficulty dealing with nonlinear characteristics of the load and complex seasonal variations. An autoregressive moving average model predicts future load using past load data based on time series analysis, combines the characteristics of Autoregressive (AR) and Moving Average (MA), and captures the autocorrelation and moving average properties of the load data.
However, conventional methods often have a simple linear relationship based on a linear model or assuming the load, and cannot accurately capture the complex nonlinear characteristics of the load, which results in limited prediction accuracy, especially in the face of complex load variation conditions. The power load is affected by various external factors, such as weather, holidays, economic development, etc., and it is often difficult for conventional methods to effectively incorporate these factors into the predictive model, resulting in predicted results that are disturbed by these factors. The power load has obvious time variability, namely the load changes along with the time, and the traditional method often cannot accurately model and predict the time variability characteristics of the load, so that the prediction result deviates from the actual situation.
Traditional power load prediction methods have certain limitations in capturing complex nonlinear characteristics, processing external factors and time-varying properties. In order to improve prediction precision and accuracy, modern power load prediction methods can better cope with these challenges by means of machine learning, artificial intelligence, big data analysis and other technologies.
In one embodiment of the present application, fig. 1 is a block diagram of a data management platform-based power system short-term power load prediction system according to an embodiment of the present application. As shown in fig. 1, a power system short-term power load prediction system based on a data management platform includes: a power load data acquisition module 110, configured to acquire power load values of a power system at a plurality of predetermined time points within a predetermined time period; the power load local time sequence feature analysis module 120 is configured to perform local time sequence feature analysis on the power load time sequence input vectors after the power load values at the plurality of predetermined time points are arranged into the power load time sequence input vectors according to a time dimension, so as to obtain a sequence of power load local time sequence feature vectors; the power load short-term fluctuation semantic measurement module 130 is configured to perform power load short-term fluctuation semantic measurement analysis on every two adjacent power load local time sequence feature vectors in the sequence of power load local time sequence feature vectors to obtain power load short-term fluctuation feature vectors; the power load multi-dimensional feature fusion module 140 is configured to fuse the power load short-time fluctuation feature vector and the global power load time sequence feature vector to obtain a power load multi-dimensional time sequence feature vector as a power load multi-dimensional time sequence feature after the sequence of the power load local time sequence feature vector is spliced to obtain the global power load time sequence feature vector; a short-term power load prediction module 150 for determining a short-term power load prediction value based on the power load multi-dimensional multi-scale timing characteristics.
According to the technical scheme, the power system short-term power load prediction system based on the data management platform is provided, the power load value of the power system can be collected through real-time monitoring, a data processing and analyzing algorithm is introduced into the data management platform to conduct time sequence analysis of the power load value, short-term power load prediction of the power system is conducted based on characteristic information of the power load time sequence data, so that corresponding power dispatching optimization is conducted to adapt to dynamic changes of the power system, and therefore operation efficiency and reliability of the power system are improved.
Specifically, in the technical solution of the present application, first, power load values of a power system at a plurality of predetermined time points within a predetermined period of time are acquired. Then, it is considered that the power load value of the power system has a dynamic change rule of time sequence in the time dimension, that is, there is a certain correlation and rule between time sequence data of the power load. Therefore, it is necessary to arrange the power load values at the plurality of predetermined time points in the time dimension as power load timing input vectors, thereby integrating timing distribution information of the power load values in the time dimension. Then, considering that the power load value has time sequence fluctuation, the power load value may show different change modes and trends under different time spans, so that in order to better reveal the time sequence rule and the change trend of the load data, in the technical scheme of the application, vector segmentation is further carried out on the power load time sequence input vector so as to obtain a sequence of the power load local time sequence input vector.
In one specific embodiment of the present application, the power load local time sequence feature analysis module includes: the power load time sequence vector segmentation unit is used for carrying out vector segmentation on the power load time sequence input vectors after the power load values of the plurality of preset time points are arranged into the power load time sequence input vectors according to the time dimension so as to obtain a sequence of power load local time sequence input vectors; and the power load local time sequence characteristic extraction unit is used for enabling the sequence of the power load local time sequence input vectors to pass through a power load time sequence characteristic extractor based on a one-dimensional convolution layer to obtain the sequence of the power load local time sequence characteristic vectors.
And then, the sequence of the power load local time sequence input vectors is subjected to feature mining in a power load time sequence feature extractor based on a one-dimensional convolution layer so as to extract local time sequence feature information of the power load value in each local time period respectively, thereby obtaining the sequence of the power load local time sequence feature vectors. By extracting the local time sequence characteristics of the power load data, the time sequence change mode, the periodic change and other characteristic information of the load can be better captured, so that the accuracy of power load prediction is improved.
It should be appreciated that there is typically a short time fluctuation in the electrical load data, i.e., a sudden and transient change in the load that occurs in a short time. These short-term fluctuations may be caused by various factors, such as weather changes, people's electricity usage behavior, etc. The accurate measurement of the short-time fluctuation of the power load has important significance for the power load prediction and the system scheduling. Therefore, in the technical scheme of the application, the power load short-time fluctuation semantic measurement coefficient between every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors is further calculated to obtain the power load short-time fluctuation feature vector composed of a plurality of power load short-time fluctuation semantic measurement coefficients. And calculating the short-time fluctuation semantic measurement coefficient of the power load between two adjacent power load local time sequence feature vectors, so that the short-time fluctuation degree of the load data can be quantified. These metric coefficients may reflect the temporal variations and fluctuations of the load data, thereby providing information about the temporal fluctuation characteristics of the load data. Thus, the characteristics of the load data can be more comprehensively described, and the accuracy of short-time load prediction is improved.
In a specific embodiment of the present application, the power load short-time fluctuation semantic measurement module is configured to: and calculating the power load short-time fluctuation semantic measurement coefficient between every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain the power load short-time fluctuation feature vector composed of a plurality of power load short-time fluctuation semantic measurement coefficients.
The power load short-time fluctuation semantic measurement module is used for: calculating power load short-time fluctuation semantic measurement coefficients between every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors according to the following coefficient formula to obtain the power load short-time fluctuation feature vectors composed of a plurality of power load short-time fluctuation semantic measurement coefficients; wherein, the coefficient formula is:
wherein p is (x) And q (x) The short-time fluctuation semantic measurement coefficient of the power load between every two adjacent power load local time sequence characteristic vectors in the sequence of the power load local time sequence characteristic vectors is N, wherein N is the scale of every two adjacent power load local time sequence characteristic vectors in the sequence of the power load local time sequence characteristic vectors, S i Is the power load short-time fluctuation semantic measurement coefficient of the power load short-time fluctuation feature vector.
In the power load data, there are not only short-term fluctuations and local timing characteristics, but also long-term trends and global characteristic information, which have an important role in analyzing long-term changes in power load. Therefore, in order to capture global features and long-term trends of load data, more comprehensive information is provided for load prediction, and in the technical scheme of the application, the sequences of the power load local time sequence feature vectors are further spliced to obtain global power load time sequence feature vectors. It should be appreciated that the global power load timing feature vector may provide feature information about overall timing trends, periodic variations, and long-term fluctuations of the power load data, which is beneficial to improving accuracy and reliability of predictions.
Further, the short-time fluctuation feature vector of the power load can reflect the short-time change and fluctuation of the power load data, and the global power load time sequence feature vector can reflect the long-term trend and global feature of the whole time domain of the power load data, so that the overall change condition and trend of the power load can be revealed. Therefore, in order to more comprehensively describe the time sequence characteristic of the load data, in the technical scheme of the application, the power load short-time fluctuation characteristic vector and the global power load time sequence characteristic vector are further fused to obtain the power load multi-dimensional multi-scale time sequence characteristic vector. It should be appreciated that by fusing the power load short-term fluctuation feature vector and the global power load timing feature vector, feature information about different scales and different dimensions of power load data may be combined to construct a more characterizable power load multi-dimensional, multi-scale timing feature vector that more fully characterizes the load data, including short-term fluctuation, long-term trend, and periodic variation, for more accurate short-term power load prediction.
In one embodiment of the present application, the short-term power load prediction module includes: the power load multi-dimensional characteristic correction unit is used for carrying out characteristic correction on the power load multi-dimensional time sequence characteristic vector to obtain a corrected power load multi-dimensional time sequence characteristic vector; and the power load decoding prediction unit is used for enabling the corrected power load multidimensional multi-scale time sequence characteristic vector to pass through a predictor based on a decoder to obtain a short-term power load predicted value.
More specifically, the electric load multidimensional feature correction unit includes: a feature correction subunit, configured to correct the power load short-time fluctuation feature vector and the global power load time sequence feature vector to obtain a correction feature vector; and the correction feature fusion subunit is used for fusing the correction feature vector with the power load multi-dimensional multi-scale time sequence feature vector to obtain the corrected power load multi-dimensional multi-scale time sequence feature vector.
In particular, in the above technical solution, the sequence of the power load local timing feature vectors expresses local time domain power load timing related features of the power load timing distribution in a local time domain determined by vector slicing in a global time domain, whereby the power load short time fluctuation feature vector expresses local time domain power load timing related semantic fluctuation features in the global time domain and the global power load timing feature vector expresses local time domain image semantic context related features in the global time domain. In this way, when the power load short-time fluctuation feature vector and the global power load time sequence feature vector are fused, feature correspondence sparsity is considered to be caused by different feature distribution modes based on local time domains of the power load short-time fluctuation feature vector and the global power load time sequence feature vector in global time domains, so that the expression effect of the power load multi-dimensional multi-scale time sequence feature vector can be influenced, and feature correspondence optimization is expected to be performed based on the time sequence significance and the time sequence criticality of feature expression of each of the power load short-time fluctuation feature vector and the global power load time sequence feature vector, so that the expression effect of the power load multi-dimensional multi-scale time sequence feature vector is improved.
Based on this, the applicant of the present application corrects the electric load short-time fluctuation feature vector and the global electric load timing feature vector, specifically expressed as: correcting the power load short-time fluctuation feature vector and the global power load time sequence feature vector by using the following optimization formula to obtain a correction feature vector, wherein the optimization formula is as follows:
wherein V is 1 Is the short-time fluctuation characteristic vector of the power load, and V 2 Is the global power load timing feature vector,representing the position-wise evolution of feature vectors, v 1max ―1 And v 2max ―1 Respectively the feature vectors V 1 And V 2 Reciprocal of maximum eigenvalue, alpha and beta are weight superparameters, V c Representing correction feature vectors>Indicating subtraction by position, +.; and fusing the correction feature vector with the power load multi-dimensional multi-scale time sequence feature vector to obtain the corrected power load multi-dimensional multi-scale time sequence feature vector.
Here, by the electric load short-time fluctuation feature vector V 1 And the global power load timing feature vector V 2 To obtain a pre-segmented local group of eigenvalue sets from which the electric load short-time fluctuation eigenvector V is regressed 1 And the global power load timing feature vector V 2 In this way, the saliency distribution by position of the feature values can be promoted based on the idea of the furthest point sampling, so that sparse interaction control among feature vectors is performed through key features with the saliency distribution, and the correction of the feature vector V is realized c For the short-time fluctuation feature vector V of the electric load 1 And the global power load timing feature vector V 2 Is a reduction of the original manifold geometry of (a). Thus, the correction feature vector V is further used c Fusion with the power load multi-dimensional multi-scale time sequence feature vector can improve the expression effect of the power load multi-dimensional multi-scale time sequence feature vector, thereby improving the corrected power load multi-dimensional multi-scale time sequence feature vectorThe accuracy of the decoded regression of the timing feature vector is performed by the decoder. In this way, short-term power load prediction can be performed based on time sequence changes of power load data of the power system, so that corresponding power scheduling optimization is performed to adapt to dynamic changes of the power system, and therefore operation efficiency and reliability of the power system are improved.
And then, passing the corrected power load multi-dimensional multi-scale time sequence characteristic vector through a predictor based on a decoder to obtain a short-term power load predicted value. That is, the time sequence multidimensional characteristic information of the power load value is utilized to carry out decoding regression so as to predict the short-term power load of the power system, and corresponding power scheduling optimization is carried out to adapt to the dynamic change of the power system, so that the operation efficiency and the reliability of the power system are improved.
In a specific embodiment of the present application, the power load decoding prediction unit is configured to: performing a decoding regression on the corrected power load multi-dimensional multi-scale timing feature vector using the decoder-based predictor in a decoding formula to obtain the short-term power load predictor; wherein, the decoding formula is:wherein X represents the corrected power load multi-dimensional multi-scale time sequence characteristic vector, Y represents a short-term power load predicted value, W represents a weight matrix, B represents a bias vector, and +.>Representing a matrix multiplication.
In summary, the power system short-term power load prediction system 100 based on the data management platform according to the embodiments of the present application is illustrated, which can monitor and collect the power load value of the power system in real time, and introduce a data processing and analysis algorithm into the data management platform to perform time sequence analysis of the power load value, so as to perform short-term power load prediction of the power system based on the characteristic information of the power load time sequence data, so as to perform corresponding power scheduling optimization to adapt to dynamic changes of the power system, thereby improving the operation efficiency and reliability of the power system.
As described above, the data management platform-based power system short-term power load prediction system 100 according to the embodiment of the present application may be implemented in various terminal devices, such as a server or the like for data management platform-based power system short-term power load prediction. In one example, the data management platform based power system short-term power load prediction system 100 according to embodiments of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the data management platform based power system short-term power load prediction system 100 may be a software module in the operating system of the terminal device or may be an application developed for the terminal device; of course, the data management platform based power system short-term power load prediction system 100 may also be one of a number of hardware modules of the terminal device.
Alternatively, in another example, the data management platform based power system short-term power load prediction system 100 and the terminal device may be separate devices, and the data management platform based power system short-term power load prediction system 100 may be connected to the terminal device through a wired and/or wireless network and transmit the interaction information in a agreed data format.
In one embodiment of the present application, fig. 2 is a flowchart of a method for short-term power load prediction for a power system based on a data management platform according to an embodiment of the present application. Fig. 3 is a schematic architecture diagram of a method for predicting short-term power load of a power system based on a data management platform according to an embodiment of the present application. As shown in fig. 2 and fig. 3, a method for predicting short-term power load of a power system based on a data management platform according to an embodiment of the present application includes: 210, acquiring power load values of a power system at a plurality of preset time points in a preset time period; 220, after the power load values at the plurality of preset time points are arranged into power load time sequence input vectors according to the time dimension, performing local time sequence feature analysis on the power load time sequence input vectors to obtain a sequence of power load local time sequence feature vectors; 230, carrying out power load short-time fluctuation semantic metric analysis on every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain power load short-time fluctuation feature vectors; 240, after the sequence of the power load local time sequence feature vector is spliced to obtain a global power load time sequence feature vector, the power load short-time fluctuation feature vector and the global power load time sequence feature vector are fused to obtain a power load multi-dimensional multi-scale time sequence feature vector as a power load multi-dimensional multi-scale time sequence feature; a short-term power load predictor is determined based on the power load multi-dimensional multi-scale timing characteristics 250.
In the power system short-term power load prediction method based on the data management platform, after the power load values at the plurality of preset time points are arranged into power load time sequence input vectors according to a time dimension, performing local time sequence feature analysis on the power load time sequence input vectors to obtain a sequence of power load local time sequence feature vectors, including: after the power load values of the plurality of preset time points are arranged into the power load time sequence input vectors according to the time dimension, vector segmentation is carried out on the power load time sequence input vectors so as to obtain a sequence of power load local time sequence input vectors; and passing the sequence of the power load local time sequence input vectors through a power load time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the power load local time sequence feature vectors.
In the power system short-term power load prediction method based on the data management platform, power load short-term fluctuation semantic metric analysis is performed on every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain power load short-term fluctuation feature vectors, and the method comprises the following steps: and calculating the power load short-time fluctuation semantic measurement coefficient between every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain the power load short-time fluctuation feature vector composed of a plurality of power load short-time fluctuation semantic measurement coefficients.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described data management platform-based power system short-term power load prediction method have been described in detail in the above description of the data management platform-based power system short-term power load prediction system with reference to fig. 1, and thus, repetitive descriptions thereof will be omitted.
Fig. 4 is a schematic view of a scenario of a power system short-term power load prediction system based on a data management platform according to an embodiment of the present application. As shown in fig. 4, in this application scenario, first, power load values (e.g., C as illustrated in fig. 4) of a power system at a plurality of predetermined time points within a predetermined period of time are acquired; the obtained electrical load values are then input into a server (e.g., S as illustrated in fig. 4) deployed with a data management platform-based electrical system short-term electrical load prediction algorithm, wherein the server is capable of processing the electrical load values based on the data management platform' S electrical system short-term electrical load prediction algorithm to determine short-term electrical load prediction values.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (10)
1. A data management platform-based short-term power load prediction system for a power system, comprising:
the power load data acquisition module is used for acquiring power load values of the power system at a plurality of preset time points in a preset time period;
the power load local time sequence characteristic analysis module is used for carrying out local time sequence characteristic analysis on the power load time sequence input vectors after arranging the power load values of the plurality of preset time points into the power load time sequence input vectors according to the time dimension so as to obtain a sequence of the power load local time sequence characteristic vectors;
the power load short-time fluctuation semantic measurement module is used for carrying out power load short-time fluctuation semantic measurement analysis on every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors so as to obtain power load short-time fluctuation feature vectors;
the power load multi-dimensional characteristic fusion module is used for fusing the power load short-time fluctuation characteristic vector and the global power load time sequence characteristic vector to obtain a power load multi-dimensional time sequence characteristic vector as a power load multi-dimensional time sequence characteristic after splicing the sequences of the power load local time sequence characteristic vectors to obtain the global power load time sequence characteristic vector;
and the short-time power load prediction module is used for determining a short-time power load predicted value based on the power load multi-dimensional multi-scale time sequence characteristic.
2. The data management platform-based power system short-term power load prediction system of claim 1, wherein the power load local timing feature analysis module comprises:
the power load time sequence vector segmentation unit is used for carrying out vector segmentation on the power load time sequence input vectors after the power load values of the plurality of preset time points are arranged into the power load time sequence input vectors according to the time dimension so as to obtain a sequence of power load local time sequence input vectors;
and the power load local time sequence characteristic extraction unit is used for enabling the sequence of the power load local time sequence input vectors to pass through a power load time sequence characteristic extractor based on a one-dimensional convolution layer to obtain the sequence of the power load local time sequence characteristic vectors.
3. The data management platform-based power system short-term power load prediction system of claim 2, wherein the power load short-term fluctuation semantic measurement module is configured to: and calculating the power load short-time fluctuation semantic measurement coefficient between every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain the power load short-time fluctuation feature vector composed of a plurality of power load short-time fluctuation semantic measurement coefficients.
4. A data management platform based power system short term power load prediction system according to claim 3, wherein the power load short term fluctuation semantic measurement module is configured to: calculating power load short-time fluctuation semantic measurement coefficients between every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors according to the following coefficient formula to obtain the power load short-time fluctuation feature vectors composed of a plurality of power load short-time fluctuation semantic measurement coefficients;
wherein, the coefficient formula is:
wherein p is (x) And q (x) The short-time fluctuation semantic measurement coefficient of the power load between every two adjacent power load local time sequence characteristic vectors in the sequence of the power load local time sequence characteristic vectors is N, wherein N is the scale of every two adjacent power load local time sequence characteristic vectors in the sequence of the power load local time sequence characteristic vectors, S i Is the power load short-time fluctuation semantic measurement coefficient of the power load short-time fluctuation feature vector.
5. The data management platform-based power system short-term power load prediction system of claim 4, wherein the short-term power load prediction module comprises:
the power load multi-dimensional characteristic correction unit is used for carrying out characteristic correction on the power load multi-dimensional time sequence characteristic vector to obtain a corrected power load multi-dimensional time sequence characteristic vector;
and the power load decoding prediction unit is used for enabling the corrected power load multidimensional multi-scale time sequence characteristic vector to pass through a predictor based on a decoder to obtain a short-term power load predicted value.
6. The data management platform-based power system short-term power load prediction system according to claim 5, wherein the power load multidimensional feature correction unit comprises:
a feature correction subunit, configured to correct the power load short-time fluctuation feature vector and the global power load time sequence feature vector to obtain a correction feature vector;
and the correction feature fusion subunit is used for fusing the correction feature vector with the power load multi-dimensional multi-scale time sequence feature vector to obtain the corrected power load multi-dimensional multi-scale time sequence feature vector.
7. The data management platform-based power system short-term power load prediction system of claim 6, wherein the power load decoding prediction unit is configured to: performing a decoding regression on the corrected power load multi-dimensional multi-scale timing feature vector using the decoder-based predictor in a decoding formula to obtain the short-term power load predictor;
wherein, the decoding formula is:wherein X represents the corrected power load multi-dimensional multi-scale time sequence characteristic vector, Y represents a short-term power load predicted value, W represents a weight matrix, B represents a bias vector, and +.>Representing a matrix multiplication.
8. A method for predicting short-term power load of a power system based on a data management platform, comprising:
acquiring power load values of a power system at a plurality of preset time points in a preset time period;
after the power load values at the plurality of preset time points are arranged into power load time sequence input vectors according to the time dimension, carrying out local time sequence feature analysis on the power load time sequence input vectors to obtain a sequence of power load local time sequence feature vectors;
carrying out power load short-time fluctuation semantic metric analysis on every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain power load short-time fluctuation feature vectors;
after the sequence of the power load local time sequence feature vector is spliced to obtain a global power load time sequence feature vector, the power load short-time fluctuation feature vector and the global power load time sequence feature vector are fused to obtain a power load multi-dimensional multi-scale time sequence feature vector as a power load multi-dimensional multi-scale time sequence feature;
a short-term power load predictor is determined based on the power load multi-dimensional multi-scale timing characteristic.
9. The method for predicting short-term power load of a power system based on a data management platform according to claim 8, wherein after arranging the power load values at the plurality of predetermined time points into power load time sequence input vectors according to a time dimension, performing local time sequence feature analysis on the power load time sequence input vectors to obtain a sequence of power load local time sequence feature vectors, comprising:
after the power load values of the plurality of preset time points are arranged into the power load time sequence input vectors according to the time dimension, vector segmentation is carried out on the power load time sequence input vectors so as to obtain a sequence of power load local time sequence input vectors;
and passing the sequence of the power load local time sequence input vectors through a power load time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the power load local time sequence feature vectors.
10. The method for short-term power load prediction of a power system based on a data management platform according to claim 9, wherein performing a power load short-term fluctuation semantic metric analysis on every two adjacent power load local timing feature vectors in the sequence of power load local timing feature vectors to obtain a power load short-term fluctuation feature vector comprises: and calculating the power load short-time fluctuation semantic measurement coefficient between every two adjacent power load local time sequence feature vectors in the sequence of the power load local time sequence feature vectors to obtain the power load short-time fluctuation feature vector composed of a plurality of power load short-time fluctuation semantic measurement coefficients.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311777620.1A CN117713083A (en) | 2023-12-22 | 2023-12-22 | Power system short-term power load prediction system and method based on data management platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311777620.1A CN117713083A (en) | 2023-12-22 | 2023-12-22 | Power system short-term power load prediction system and method based on data management platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117713083A true CN117713083A (en) | 2024-03-15 |
Family
ID=90147788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311777620.1A Pending CN117713083A (en) | 2023-12-22 | 2023-12-22 | Power system short-term power load prediction system and method based on data management platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117713083A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117977587A (en) * | 2024-04-02 | 2024-05-03 | 南京鼎研电力科技有限公司 | Power load prediction system and method based on deep neural network |
-
2023
- 2023-12-22 CN CN202311777620.1A patent/CN117713083A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117977587A (en) * | 2024-04-02 | 2024-05-03 | 南京鼎研电力科技有限公司 | Power load prediction system and method based on deep neural network |
| CN117977587B (en) * | 2024-04-02 | 2024-06-07 | 南京鼎研电力科技有限公司 | Power load prediction system and method based on deep neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2020101900A4 (en) | A method, device and equipment for detecting abnormal electric meter | |
| CN111680841B (en) | Short-term load prediction method, system and terminal equipment based on principal component analysis | |
| CN117713083A (en) | Power system short-term power load prediction system and method based on data management platform | |
| CN110579734B (en) | Method for analyzing electric energy quality of single-phase electric energy meter and single-phase electric energy meter | |
| CN113554361B (en) | Comprehensive energy system data processing and calculating method and processing system | |
| CN116701481B (en) | Electric heating electricity consumption fluctuation analysis system and method | |
| US20230307908A1 (en) | Extraction of instantaneous renewable generation from net load measurements of power distribution systems | |
| CN117113159A (en) | Deep learning-based power consumer side load classification method and system | |
| CN117977587B (en) | Power load prediction system and method based on deep neural network | |
| CN114358445A (en) | Business process residual time prediction model recommendation method and system | |
| Brusokas et al. | HeatFlex: Machine learning based data-driven flexibility prediction for individual heat pumps | |
| CN111091223A (en) | Distribution transformer short-term load prediction method based on Internet of things intelligent sensing technology | |
| Sulaiman et al. | Smart meter data analysis issues: a data analytics perspective | |
| CN117236380A (en) | Power system fault prediction method, system, electronic equipment and medium | |
| CN117151276A (en) | Intelligent management system of electricity selling platform | |
| Zhuang et al. | Power load data completion method considering low rank property | |
| CN115130788A (en) | Big data analysis-based power utilization prediction system | |
| CN106845763B (en) | Power grid reliability analysis method and device | |
| Wang | An enhanced Markov chain Monte Carlo-integrated cross-entropy method with a partially collapsed Gibbs sampler for probabilistic spinning reserve adequacy evaluation of generating systems | |
| CN113808071A (en) | Non-invasive load monitoring method and system based on deep learning | |
| CN114004408A (en) | User power load prediction method based on data analysis | |
| LU500121B1 (en) | A method, device and equipment for electric meter detecting | |
| CN117154723B (en) | Platform short-term load prediction method and system based on multi-source data and model fusion | |
| CN117639107B (en) | Power quality assessment method and system for power distribution network | |
| CN116957306B (en) | User side response potential evaluation method and system based on resource collaborative interaction |
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 |