CN116542362A - Load prediction method and device, electronic equipment and storage medium - Google Patents
Load prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116542362A CN116542362A CN202310305393.6A CN202310305393A CN116542362A CN 116542362 A CN116542362 A CN 116542362A CN 202310305393 A CN202310305393 A CN 202310305393A CN 116542362 A CN116542362 A CN 116542362A
- Authority
- CN
- China
- Prior art keywords
- sequence
- load
- determined
- load prediction
- current
- 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 66
- 238000012545 processing Methods 0.000 claims description 58
- 238000004590 computer program Methods 0.000 claims description 16
- 238000013507 mapping Methods 0.000 claims description 13
- 239000004973 liquid crystal related substance Substances 0.000 claims description 7
- 230000000694 effects Effects 0.000 abstract description 6
- 238000000354 decomposition reaction Methods 0.000 description 23
- 239000013598 vector Substances 0.000 description 11
- 238000004891 communication Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000010606 normalization Methods 0.000 description 5
- 230000007246 mechanism Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 230000000295 complement effect Effects 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 241000288105 Grus Species 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- 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
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a load prediction method, a load prediction device, electronic equipment and a storage medium, wherein the load prediction method comprises the following steps: acquiring a historical load sequence, and determining at least one modal component to be determined corresponding to the historical load sequence; invoking load prediction models corresponding to the modal components to be determined to determine a predicted sequence to be determined of the corresponding modal components to be determined based on the load prediction models; a target load prediction sequence corresponding to a load prediction period is determined based on at least one prediction sequence to be determined. The method solves the problems that the traditional load prediction method needs to consider multidimensional load influence factors or the accuracy of a load prediction result is low, and achieves the effect of accurately predicting load prediction data of a load prediction period by time data and air-phase data of historical load data and the load prediction period on the premise that other load influence factors do not need to be considered.
Description
Technical Field
The present invention relates to the field of power load prediction technologies, and in particular, to a load prediction method, a load prediction device, an electronic device, and a storage medium.
Background
The load prediction plays a key role in the aspect of stable operation of the power system, and the power generation plan, the power consumption scheduling, the market transaction and the like can be more reasonably assisted by the power system according to the load prediction result.
The traditional load prediction method comprises a regression analysis method, a time sequence method, a gray model method and the like, but the accuracy of a prediction result is lower when the load prediction is carried out by the prediction method. In addition, a machine learning prediction method, such as a neural network method, a support vector machine, a fuzzy theory and the like, can be adopted, but the prediction method needs to consider multidimensional load influence factors when carrying out load prediction, and the prediction process is more complex and takes longer time.
In order to solve the above problems, an improvement in the load prediction method is required.
Disclosure of Invention
The invention provides a load prediction method, a load prediction device, electronic equipment and a storage medium, which are used for solving the problems that the traditional load prediction method needs to consider multidimensional load influence factors or the accuracy of a load prediction result is low.
In a first aspect, an embodiment of the present invention provides a load prediction method, including:
acquiring a historical load sequence, and determining at least one modal component to be determined corresponding to the historical load sequence; the historical load sequence comprises load data corresponding to a plurality of historical moments;
Invoking load prediction models corresponding to the modal components to be determined to determine a predicted sequence to be determined of the corresponding modal components to be determined based on the load prediction models; wherein the load prediction model is a model constructed in advance;
determining a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined; wherein the load prediction period is determined based on a load prediction demand.
In a second aspect, an embodiment of the present invention further provides a load prediction apparatus, including:
the system comprises a modal component determining module, a modal component determining module and a processing module, wherein the modal component determining module is used for acquiring a historical load sequence and determining at least one modal component to be determined corresponding to the historical load sequence; the historical load sequence comprises load data corresponding to a plurality of historical moments;
the prediction sequence determining module is used for retrieving load prediction models corresponding to the modal components to be determined so as to determine the prediction sequences to be determined of the corresponding modal components to be determined based on the load prediction models; wherein the load prediction model is a model constructed in advance;
a load sequence determining module, configured to determine a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined; wherein the load prediction period is determined based on a load prediction demand.
In a third aspect, an embodiment of the present invention further provides an electronic device, including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the load prediction method of any one of the embodiments of the present invention.
In a fourth aspect, an embodiment of the present invention further provides a computer readable storage medium, where computer instructions are stored, where the computer instructions are configured to cause a processor to execute the load prediction method according to any one of the embodiments of the present invention.
According to the technical scheme, the historical load sequence is obtained, at least one modal component to be determined corresponding to the historical load sequence is determined, at least one group of Gaussian white noise is added to the historical load sequence to obtain corresponding sequences to be used, modal decomposition is carried out on each sequence to be used respectively, and the modal component to be determined corresponding to each sequence to be used is obtained. Further, a load prediction model corresponding to each modal component to be determined is called to determine a predicted sequence to be determined of the corresponding modal component to be determined based on each load prediction model, model call parameters corresponding to each modal component to be determined can be determined through the target mapping table, and the load prediction model corresponding to each modal component to be determined is called based on the model call parameters to perform load prediction on the corresponding modal component to be determined based on each load prediction model, and at least one predicted sequence to be determined is determined. Further, a target load prediction sequence corresponding to the load prediction period is determined based on at least one prediction sequence to be determined, the prediction sequences to be used are obtained by superposing the prediction sequences to be determined, and the target load prediction sequence corresponding to the load prediction period is obtained based on the ratio of the prediction sequences to be used to the number of sequences of the sequences to be used. The method solves the problems that the traditional load prediction method needs to consider multidimensional load influence factors or the accuracy of a load prediction result is low, and achieves the effect of accurately predicting load prediction data of a load prediction period by time data and air-phase data of historical load data and the load prediction period on the premise that other load influence factors do not need to be considered.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a load prediction method according to a first embodiment of the present invention;
FIG. 2 is a schematic diagram of a BiGRU neural network layer according to a first embodiment of the invention;
FIG. 3 is a flowchart of a load prediction method according to a second embodiment of the present invention;
fig. 4 is a schematic structural view of a load predicting device according to a third embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device implementing a load prediction method according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the sequences so used may be interchanged where appropriate so that embodiments of the invention described herein may be practiced otherwise than as shown or described.
Before the technical scheme is elaborated, an application scene of the technical scheme is simply introduced so as to more clearly understand the technical scheme. The rapid development of distributed power supplies and novel loads further increases the uncertainty and volatility on both sides of the power system source load, and based on the uncertainty and volatility, higher requirements are put on the precision of load prediction. In the technical scheme, in order to more accurately predict the load of the power system, historical load data is decomposed to obtain intrinsic mode components under different frequencies, and then frequency distribution and time sequence characteristics of a load sequence are obtained. Furthermore, the association relation between time sequences is accurately acquired based on a pre-trained neural network model, and the accuracy of prediction is increased by combining an attention mechanism, so that the effect of more accurately predicting the load of the power system is achieved.
Example 1
Fig. 1 is a flowchart of a load prediction method according to an embodiment of the present invention, where the load prediction method may be applicable to predicting load data of a target area in a load prediction period according to historical load prediction data of the target area collected by a power system, time data of the load prediction period, and acquired weather data of the target area in the load prediction period, so as to assist a worker of the power system in planning a power generation plan, a power consumption schedule, and a power market transaction in the target area based on a prediction result, where the method may be performed by a load prediction device, which may be implemented in the form of hardware and/or software, and where the load prediction device may be configured in a computing device that may perform the load prediction method.
As shown in fig. 1, the method includes:
s110, acquiring a historical load sequence, and determining at least one modal component to be determined, which corresponds to the historical load sequence.
When load prediction is performed on a target area needing load prediction, a historical load sequence in the target area needs to be acquired, wherein the historical load sequence comprises load data corresponding to a plurality of historical moments. The to-be-determined modal component may be understood as a characteristic component obtained after performing modal decomposition on the historical load sequence, and in the technical solution, at least one to-be-determined modal component refers to a modal component under different frequencies corresponding to the historical load sequence after performing modal decomposition, for example, complementary set empirical mode decomposition (Complementary Ensemble Empirical Mode Decomposition, CEEMD) may be adopted when decomposing the historical load sequence.
For example, if the load of the next cycle of the target area needs to be predicted, analysis processing is needed according to the historical load prediction of the previous cycle, so as to obtain the association relationship between the historical load data in the historical load sequence in the previous cycle, and based on the association relationship, the load of the next cycle needs to be predicted is predicted. The period referred to herein may be set by user, for example, may be set to 1 day, 2 days, or one week.
Specifically, a historical load sequence in a target area is collected, modal decomposition is carried out on the historical load sequence, and at least one group of Gaussian noise is added to the historical load sequence, so that modal decomposition is carried out on the historical load sequence under the treatment of each group of Gaussian noise on the basis, and a to-be-determined modal component under at least one frequency is obtained.
S120, invoking load prediction models corresponding to the modal components to be determined to determine a prediction sequence to be determined of the corresponding modal components to be determined based on the load prediction models.
In the technical scheme, the load prediction model is a pre-constructed model, and comprises a bi-directional gating cyclic neural network layer (BiGRU) and an attention mechanism layer (attention). In practical application, because frequencies corresponding to the components to be determined are different, different load prediction models are required to be adopted when processing the modal components to be determined. By prediction sequence to be determined is understood load prediction data at the respective frequency based on the prediction of the modal component to be determined. Further, taking one of the modal components to be determined as an example, after determining a load prediction model corresponding to the modal component to be determined, analyzing the corresponding modal component to be determined based on the load prediction model to determine a predicted sequence to be determined of the historical load sequence under the frequency corresponding to the modal component to be determined.
In order to more clearly describe the technical scheme, a load prediction model used in the technical scheme is first briefly described. In practical application, the load prediction model firstly performs feature extraction on information of a modal component deep layer to be determined through a bidirectional gating cyclic neural network layer; then, the attention mechanism layer is utilized to distribute corresponding weights to the deep information of the modal components to be determined; and finally, the characteristic information with different weights is put into a softmax function layer for characteristic analysis, so as to obtain a corresponding predicted sequence to be determined of the modal component to be determined.
In practical application, the method for retrieving the load prediction model corresponding to each modal component to be determined comprises the following steps: determining a model calling parameter and at least one model calling parameter according to the mode components to be determined and the current mode components to be determined based on the target mapping table; invoking a current load prediction model corresponding to the current modal component to be determined based on the at least one model invocation parameter; acquiring at least one associated data corresponding to a load prediction period; and processing at least one associated data based on the current load prediction model to obtain a prediction sequence to be determined, which corresponds to the current modal component to be determined.
It will be appreciated that different modal components to be determined need to be processed using different load prediction models, and in order to determine a current load prediction model corresponding to a current modal component to be determined, the current load prediction model corresponding to the current modal component to be determined may be found in the target mapping table.
Wherein the target mapping table comprises at least one modal component and at least one model call parameter corresponding to each modal component. That is, the target mapping table is used for recording model call parameters corresponding to different modal components, so as to determine a load prediction model corresponding to the modal component to be determined based on the model call parameters. The model calling parameters refer to frequencies corresponding to the modal components to be determined.
In this technical solution, at least one of the associated data includes to-be-predicted time data and to-be-predicted weather data, where the to-be-predicted time data refers to time information that needs to be subjected to load prediction, for example, a current date is 3 months and 1 day, and load data of the target area is required to be predicted for 3 months and 2 days, and the to-be-predicted time data is 3 months and 2 days. The weather data to be predicted refers to weather data corresponding to a period in which load prediction is required, for example, if the temperature of 3 months and 2 days is predicted to be 10-18 ℃, the temperature data in the weather data to be predicted is 10-18 ℃. The weather data to be predicted can be obtained through a weather platform.
Specifically, taking one of the modal components to be determined as a current modal component to be determined as an example, searching at least one model calling parameter, such as a frequency parameter, corresponding to the current modal component to be determined from a target mapping table, determining a current load prediction model corresponding to the current modal component to be determined according to the at least one model calling parameter, and simultaneously acquiring time data to be predicted and weather data to be predicted in a load prediction period. Further, the current modal component to be determined, the time data to be predicted and the weather data to be predicted are input into a current load prediction model to obtain a predicted sequence to be determined, which corresponds to the current modal component to be determined.
Specifically, the load prediction model needs to be built in advance before the prediction processing is performed on the history load sequence based on the load prediction model. It can be understood that the whole structure of the BiGRU network layer consists of two GRU network layers with opposite directions, namely an input layer, an output layer and a hidden layer, wherein the hidden layer comprises a forward layer and a reverse layer, and the special structure of the BiGRU also determines that the output is jointly determined by the states of the GRUs with the two directions. Wherein, the structure of the BiGRU network layer is shown in figure 2,
The data processing mode of the GRU neural network unit is as follows
Wherein x is t Input variable for time t, h t-1 The result is output for the last unit hidden layer,is x t And h t-1 And (h) is a complex result of t Outputting the result for the hidden layer of the unit, W Z ，W r ，W，U Z ，U r U is a trainable parameter matrix, l is an identity matrix,>representing a composite relationship, σ represents a sigmoid activation function.
Wherein the input variables refer to the data such as load, time, weather (temperature, humidity, wind speed) and the like corresponding to the time, and therefore x t Is a multidimensional sequence consisting of a historical load sequence, a historical time sequence, a historical temperature sequence, a historical humidity sequence and a historical wind speed sequence.
By outputting GRU units of a forward layer and a reverse layerAnd->And carrying out weighted summation to obtain the output of the BiGRU at the time t as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,for the output weight of the forward layer, +.>Output weight for reverse layer, b t Is the offset.
Further, the correlation problem between the input sequence and the output result is handled by the attention mechanism layer. The principle is that enough weight values are distributed to key information through probability distribution, so that the accuracy of BiGRU network model prediction is improved. The specific implementation flow is as follows:
firstly, establishing a full connection layer, outputting the state of a BiGRU network to h t Converted into corresponding weight e t：
e t ＝w t tanh(W t h t +b t )
In the formula e t Representing a state output vector h t Weights, w t And W is t Finger weight coefficient matrix, b t Is an offset, h t Represents hidden layer state output vector of BiGRU network at t moment, tanh is hyperbolic tangent function, h (W t h t +b t ) Representing the output vector h for the state t Is a function of the transformation of the above.
Next, an attention weighting coefficient a is generated by a softmax function t The corresponding formula is as follows:
wherein a is t Representing a state output vector h t Attention weighting coefficient of e t Representing a state output vector h t Weights of e j State output vector h representing arbitrary time j The value of j is 1 to n, and n is the number of state output vectors).
Finally, obtaining a feature vector corresponding to the BiGRU output sequence:
v＝Σa t h t
wherein v represents a feature vector, a t Representing a state output vector h t Attention weight coefficient of (h) t And the hidden layer state output vector of the BiGRU network at the time t is represented.
S130, determining a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined.
Based on the analysis, at least one group of Gaussian noise is added to the historical load sequence, and modal decomposition is carried out on the historical load sequence after the noise is added, so that modal components to be determined under different frequencies can be obtained. Further, a load prediction model corresponding to each modal component to be determined is called to perform load prediction on the corresponding modal component to be determined, so that at least one predicted sequence to be determined can be obtained. Optionally, determining, based on the at least one predicted sequence to be determined, a target load prediction sequence corresponding to the load prediction period, includes: superposing at least one predicted sequence to be determined to obtain a corresponding predicted sequence to be used; and obtaining a target load prediction sequence corresponding to the load prediction period based on the ratio of at least one predicted sequence to be used to the number of sequences of the sequences to be used.
The target load prediction sequence refers to a prediction result of load information of a load prediction period based on a historical load sequence, and a worker of a power system can plan a power generation plan, a power consumption schedule, a power market transaction and the like of a target area in the load prediction period according to the target load prediction sequence. It is understood that at least one load prediction data is included in the target load prediction sequence.
In practical application, in order to obtain a target load prediction sequence corresponding to a load prediction period, respectively performing superposition processing on corresponding load prediction data in at least one obtained prediction sequence to be determined, and forming a prediction sequence to be used based on each superposed load prediction data. Further, an average value of each load prediction data in the prediction sequence to be used is solved, and a load sequence formed by the solved average load prediction data is used as a target load prediction sequence of the load prediction period.
According to the technical scheme, the historical load sequence is obtained, at least one modal component to be determined corresponding to the historical load sequence is determined, at least one group of Gaussian white noise is added to the historical load sequence to obtain corresponding sequences to be used, modal decomposition is carried out on each sequence to be used respectively, and the modal component to be determined corresponding to each sequence to be used is obtained. Further, a load prediction model corresponding to each modal component to be determined is called to determine a predicted sequence to be determined of the corresponding modal component to be determined based on each load prediction model, model call parameters corresponding to each modal component to be determined can be determined through the target mapping table, and the load prediction model corresponding to each modal component to be determined is called based on the model call parameters to perform load prediction on the corresponding modal component to be determined based on each load prediction model, and at least one predicted sequence to be determined is determined. Further, a target load prediction sequence corresponding to the load prediction period is determined based on at least one prediction sequence to be determined, the prediction sequences to be used are obtained by superposing the prediction sequences to be determined, and the target load prediction sequence corresponding to the load prediction period is obtained based on the ratio of the prediction sequences to be used to the number of sequences of the sequences to be used. The method solves the problems that the traditional load prediction method needs to consider multidimensional load influence factors or the accuracy of a load prediction result is low, and achieves the effect of accurately predicting load prediction data of a load prediction period by time data and air-phase data of historical load data and the load prediction period on the premise that other load influence factors do not need to be considered.
Example two
Fig. 3 is a flowchart of a load prediction method according to a second embodiment of the present invention, and optionally, at least one modal component to be determined corresponding to a historical load sequence is determined to be refined.
As shown in fig. 3, the method includes:
s210, acquiring a historical load sequence, and carrying out normalization processing on the historical load sequence to obtain a sequence to be processed.
The historical load sequence comprises at least one historical load data, normalization processing is carried out on each historical load data, and the normalized historical load data form a sequence to be processed. That is, included in the sequence to be processed is normalized data of each of the history load data in the history load sequence.
In order to facilitate the prediction of load data of a target area in a load prediction period, when a historical load sequence is processed, at least one historical load data in the historical load sequence may be normalized respectively.
Specifically, when normalization processing is performed on each history load data, the processing can be performed based on the following formula:
wherein P represents historical load data to be normalized, P min Representing the smallest historical load data in the historical load sequence, P max Representing the largest historical load data, P, in the historical load sequence i Normalized data normalized by P is shown.
S220, carrying out interpolation processing on the missing load data in the sequence to be processed, so as to determine the sequence to be processed after interpolation as a target processing sequence.
Specifically, at least one historical load data in the historical load sequence is normalized respectively, and a sequence to be processed can be obtained based on the normalized data. Furthermore, if the missing load data exists in the sequence to be processed, an interpolation algorithm can be adopted to perform interpolation processing on the sequence to be processed, so as to obtain a target processing sequence.
When the difference processing is performed based on the interpolation algorithm, the interpolation processing may be performed on the historical load sequence missing in the historical load sequence before normalization, or the interpolation processing may be performed on the exact missing load data in the sequence to be processed after the sequence to be processed is obtained. The interpolation algorithm used in the technical scheme can be a cubic spline interpolation algorithm, a nearest neighbor interpolation algorithm, a bilinear interpolation algorithm, a bicubic interpolation algorithm and the like.
S230, decomposing the target processing sequence to obtain at least one modal component to be determined.
Specifically, when the target processing sequence is decomposed, the technical scheme adopts a CEEMD decomposition method, wherein the CEEMD decomposition method is proposed on the traditional empirical mode decomposition (Empirical Mode Decomposition, EMD). Specifically, a pair of Gaussian white noise is added into a target processing sequence, and modal decomposition is carried out on the sequence added with the noise, so that at least one modal component to be determined is obtained.
In practical application, the target processing sequence is decomposed to obtain at least one modal component to be determined, including: carrying out Gaussian white noise treatment on the target treatment sequence to obtain a corresponding sequence to be used; for each sequence to be used, determining an upper envelope curve and a lower envelope curve corresponding to the current sequence to be used; based on the current sequence to be used, the upper envelope and the lower envelope, a current modal component to be determined corresponding to the current sequence to be used is determined.
The Gaussian white noise comprises positive sequence noise or negative sequence noise respectively. The sequence to be used can be understood as a load sequence after white noise is added to the target processing sequence.
Specifically, a pair of Gaussian white noise is added to the target processing sequence, namely, positive sequence noise and negative sequence noise are added to the target processing sequence, so that the problems of modal aliasing, residual auxiliary noise and the like are easy to occur when the target processing sequence is directly decomposed, and further the decomposed modal components to be determined can embody the distribution characteristics of the target processing sequence on the time scale.
In practical application, a pair of white gaussian noise ω (t) is added to the target processing sequence x (t), so that it is possible to obtain:
wherein x (t) represents a target processing sequence,representing the positive load sequence after addition of Gaussian white noise, ">Indicating the negative load sequence after adding Gaussian white noise, ">Representing positive sequence noise in Gaussian white noise, < ->Representing negative sequence noise among gaussian white noise.
It is thus clear that 2 load sequences can be obtained by adding a pair of gaussian white noise to the target processing sequence.
Further, taking adding one pair of Gaussian white noise as an example, performing EMD decomposition on the obtained 2 load sequences (i.e. sequences to be used) respectively, and marking the j-th to-be-determined modal component of the i-th sequence as IMF ij . On the basis, taking one current sequence to be used as an example, an upper envelope curve and a lower envelope curve corresponding to the current sequence to be used are determined, so that the current modal component to be determined is obtained based on the current sequence to be used and the corresponding upper envelope curve and lower envelope curve.
Optionally, determining the upper envelope and the lower envelope corresponding to the currently to-be-used sequence includes: determining at least one set of extremum data in a currently to-be-used sequence; and carrying out interpolation processing in at least one group of extreme value data to obtain an upper envelope curve and a lower envelope curve corresponding to the current sequence to be used.
Wherein the extremum data comprises local maximum data and local minimum data.
Specifically, determining local maximum values and local minimum values in the current sequence to be used, and performing interpolation processing in the current sequence to be used by using a cubic spline interpolation function to form an upper envelope curve and a lower envelope curve corresponding to the current sequence to be used.
On the basis, determining a current to-be-determined modal component corresponding to the current to-be-used sequence based on the current to-be-used sequence, the upper envelope and the lower envelope, wherein the method comprises the following steps: determining a mean value sequence corresponding to the upper envelope curve and the corresponding lower envelope curve at the same moment; determining a current to-be-used component sequence corresponding to the current to-be-used sequence based on the difference value of the current to-be-used sequence and the mean value sequence; obtaining a current residual component sequence corresponding to the current sequence to be used based on the difference value between the current sequence to be used and the current component sequence to be used; and determining a current to-be-determined modal component corresponding to the current to-be-used sequence based on the current to-be-used component sequence and the current residual component sequence.
In the present technical solution, the average value sequence refers to a sequence formed by an average value corresponding to an upper envelope curve and a lower envelope curve corresponding to a currently to-be-used sequence at the same time. The current sequence of components to be used may be understood as a sequence component determined based on the difference between the sequence to be used and the mean sequence. The current residual component sequence may be understood as a mean sequence formed based on the difference between the current sequence to be used and the current sequence of components to be used.
Specifically, the current sequence to be used includes at least one normalized historical load data, and each historical load data corresponds to one historical moment, so when determining the average value sequence corresponding to the sequence to be used, it is necessary to subtract the normalized data corresponding to the upper envelope line and the lower envelope line corresponding to the same moment, and form the average value sequence based on at least one difference result.
Specifically, the average sequence corresponding to the current sequence to be used may be determined based on the following formula:
m(t)＝[e max (t)+e min (t)]/2
wherein m (t) represents a mean sequence, e max Represents the upper envelope, e min And (t) represents the lower envelope.
It is understood that when the mean processing is performed based on the upper envelope and the lower envelope, the mean processing is performed on the data at the same time.
Further, the current component sequence to be used is determined based on the following formula:
h 1 (t)＝y(t)-m(t)
wherein h is 1 (t) represents a current component sequence to be used, y (t) represents a current sequence to be used, and m (t) represents a mean sequence.
On the basis, the current component sequence h to be used is determined 1 (t) whether the modality constraints, i.e., IMF constraints, are satisfied, if so, h 1 (t) recording as a first to-be-determined modal component; if not, h is 1 (t) as a new currentAnd (3) the sequence to be used and the residual component sequence corresponding to the new current sequence to be used are redetermined until the residual component sequence meets the IMF constraint condition, and the corresponding modal component to be determined is obtained.
Wherein the IMF constraints include:
(1) The difference value between the number of extreme points and the number of zero crossing points of the current component sequence to be used is less than or equal to 1;
(2) At any moment, the average value of the upper envelope corresponding to the local maximum of the current component sequence to be used and the lower envelope of the corresponding local minimum is 0.
Further, after obtaining the modal component to be determined, based on the difference value between the current sequence to be used and the current sequence of components to be used, a current residual component sequence is obtained, which can be determined specifically by the following formula:
r(t)＝y(t)-IMF 1
wherein r (t) represents the current residual component sequence, IMF 1 Representing the current sequence of components to be used, y (t) representing the current sequence to be used.
After the current residual component sequence is obtained, the current residual component sequence is used as a new current sequence to be used, and the corresponding current residual component sequence is repeatedly obtained until the current residual component sequence meets the standard deviation condition, so that EMD decomposition is completed.
Finally, the sequence to be used currently can be expressed after EMD decomposition:
Wherein y (t) represents the currently-to-be-used sequence, IMF i (i=1, 2, …, m) is the modal component to be determined for the different frequencies, r (t) represents the current residual component.
And by analogy, after EMD decomposition is carried out on all the sequences to be used, obtaining corresponding modal components to be determined, and taking the average value of at least one modal component to be determined as a CEEMD decomposition result. Specifically, CEEMD decomposition results were as follows:
wherein, IMF j Representing the modal component to be determined at the jth frequency, IMF ij Represents the modal component to be determined at the j-th frequency of the i-th sequence to be used, and n represents the logarithm of the sequence to be used (in this case n=1, i.e. only 1 sequence to be used).
Wherein r represents the residual component sequence, r i Representing the residual component of the i-th sequence to be used, n represents the logarithm of the sequence to be used (in this case n=1, i.e. only 1 sequence to be used).
S240, retrieving load prediction models corresponding to the modal components to be determined to determine a prediction sequence to be determined of the corresponding modal components to be determined based on the load prediction models.
S250, determining a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined.
According to the technical scheme, the historical load sequence is obtained, at least one modal component to be determined corresponding to the historical load sequence is determined, at least one group of Gaussian white noise is added to the historical load sequence to obtain corresponding sequences to be used, modal decomposition is carried out on each sequence to be used respectively, and the modal component to be determined corresponding to each sequence to be used is obtained. Further, a load prediction model corresponding to each modal component to be determined is called to determine a predicted sequence to be determined of the corresponding modal component to be determined based on each load prediction model, model call parameters corresponding to each modal component to be determined can be determined through the target mapping table, and the load prediction model corresponding to each modal component to be determined is called based on the model call parameters to perform load prediction on the corresponding modal component to be determined based on each load prediction model, and at least one predicted sequence to be determined is determined. Further, a target load prediction sequence corresponding to the load prediction period is determined based on at least one prediction sequence to be determined, the prediction sequences to be used are obtained by superposing the prediction sequences to be determined, and the target load prediction sequence corresponding to the load prediction period is obtained based on the ratio of the prediction sequences to be used to the number of sequences of the sequences to be used. The method solves the problems that the traditional load prediction method needs to consider multidimensional load influence factors or the accuracy of a load prediction result is low, and achieves the effect of accurately predicting load prediction data of a load prediction period by time data and air-phase data of historical load data and the load prediction period on the premise that other load influence factors do not need to be considered.
Example III
Fig. 4 is a schematic structural diagram of a load prediction device according to a third embodiment of the present invention. As shown in fig. 4, the apparatus includes: a modal component determination module 310, a prediction sequence determination module 320, and a load sequence determination module 330.
The modal component determining module 310 is configured to acquire a historical load sequence and determine at least one modal component to be determined corresponding to the historical load sequence; the historical load sequence comprises load data corresponding to a plurality of historical moments;
a prediction sequence determining module 320, configured to invoke a load prediction model corresponding to each of the modal components to be determined, so as to determine a prediction sequence to be determined of the corresponding modal component to be determined based on each of the load prediction models; the load prediction model is a model constructed in advance;
a load sequence determining module 330, configured to determine a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined; wherein the load prediction period is determined based on the load prediction demand.
According to the technical scheme, the historical load sequence is obtained, at least one modal component to be determined corresponding to the historical load sequence is determined, at least one group of Gaussian white noise is added to the historical load sequence to obtain corresponding sequences to be used, modal decomposition is carried out on each sequence to be used respectively, and the modal component to be determined corresponding to each sequence to be used is obtained. Further, a load prediction model corresponding to each modal component to be determined is called to determine a predicted sequence to be determined of the corresponding modal component to be determined based on each load prediction model, model call parameters corresponding to each modal component to be determined can be determined through the target mapping table, and the load prediction model corresponding to each modal component to be determined is called based on the model call parameters to perform load prediction on the corresponding modal component to be determined based on each load prediction model, and at least one predicted sequence to be determined is determined. Further, a target load prediction sequence corresponding to the load prediction period is determined based on at least one prediction sequence to be determined, the prediction sequences to be used are obtained by superposing the prediction sequences to be determined, and the target load prediction sequence corresponding to the load prediction period is obtained based on the ratio of the prediction sequences to be used to the number of sequences of the sequences to be used. The method solves the problems that the traditional load prediction method needs to consider multidimensional load influence factors or the accuracy of a load prediction result is low, and achieves the effect of accurately predicting load prediction data of a load prediction period by time data and air-phase data of historical load data and the load prediction period on the premise that other load influence factors do not need to be considered.
Optionally, the mode component determining module includes: the to-be-processed sequence determination submodule is used for carrying out normalization processing on the historical load sequence to obtain a to-be-processed sequence;
the target processing sequence determining submodule is used for carrying out interpolation processing on the missing load data in the sequence to be processed so as to determine the interpolated sequence to be processed as a target processing sequence;
and the modal component determination submodule is used for carrying out decomposition processing on the target processing sequence to obtain at least one modal component to be determined.
Optionally, the mode component determining submodule includes: the sequence to be used determining unit is used for carrying out Gaussian white noise processing on the target processing sequence to obtain a corresponding sequence to be used; the Gaussian white noise respectively comprises positive sequence noise or negative sequence noise;
an envelope determining unit configured to determine, for each sequence to be used, an upper envelope and a lower envelope corresponding to a current sequence to be used;
and the modal component determining unit is used for determining a current modal component to be determined corresponding to the current sequence to be used based on the current sequence to be used, the upper envelope and the lower envelope.
Optionally, the envelope determining unit includes: an extremum data determining subunit, configured to determine at least one set of extremum data in a currently to-be-used sequence; the extremum data comprises local maximum value data and local minimum value data;
And the envelope determining subunit is used for carrying out interpolation processing on at least one group of extreme value data to obtain an upper envelope and a lower envelope corresponding to the current sequence to be used.
Optionally, the modality component determination unit includes: the average value sequence determining subunit is used for determining an average value sequence corresponding to the upper envelope line and the corresponding lower envelope line at the same time;
the to-be-used component determining subunit is used for determining a current to-be-used component sequence corresponding to the current to-be-used sequence based on the difference value between the current to-be-used sequence and the average value sequence;
a residual component determining subunit, configured to obtain a current residual component sequence corresponding to the current sequence to be used based on a difference value between the current sequence to be used and the current component sequence to be used;
and the modal component determining subunit is used for determining a current modal component to be determined corresponding to the current sequence to be used based on the current sequence of components to be used and the current residual component sequence.
Optionally, the prediction sequence determining module includes: the parameter determination submodule is used for determining parameters for determining at least one model call with the current modal component to be determined based on the target mapping table for each modal component to be determined; the target mapping table comprises at least one modal component and at least one model calling parameter to be matched corresponding to each modal component;
The model determining submodule is used for calling a current load prediction model corresponding to a current modal component to be determined based on at least one model calling parameter;
a data acquisition sub-module for acquiring at least one associated data corresponding to a load prediction period; the at least one associated data comprises time data to be predicted and weather data to be predicted;
and the sequence prediction sub-module is used for processing at least one associated data based on the current load prediction model to obtain a predicted sequence to be determined, which corresponds to the current modal component to be determined.
Optionally, the load sequence determining module includes: the prediction sequence determination submodule is used for carrying out superposition processing on at least one prediction sequence to be determined to obtain a corresponding prediction sequence to be used;
and the load sequence determination submodule is used for obtaining a target load prediction sequence corresponding to the load prediction period based on the ratio of at least one predicted sequence to be used to the number of sequences of the sequences to be used.
The load prediction device provided by the embodiment of the invention can execute the load prediction method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 5 shows a schematic structural diagram of the electronic device 10 of the embodiment of the present invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 5, the electronic device 10 includes at least one processor 11, and a memory, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, in which the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and sequences necessary for the operation of the electronic device 10 may also be stored. The processor 11, the ROM 12 and the RAM 13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.
Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/sequences with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the load prediction method.
In some embodiments, the load prediction method may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM 12 and/or the communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the load prediction method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the load prediction method in any other suitable way (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive sequences and instructions from, and transmit sequences and instructions to, a storage system, at least one input device, and at least one output device.
The computer program for implementing the load prediction method of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable sequence processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a sequence server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system may be interconnected by any form or medium of digital sequential communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A load prediction method, comprising:
acquiring a historical load sequence, and determining at least one modal component to be determined corresponding to the historical load sequence; the historical load sequence comprises load data corresponding to a plurality of historical moments;
invoking load prediction models corresponding to the modal components to be determined to determine a predicted sequence to be determined of the corresponding modal components to be determined based on the load prediction models; wherein the load prediction model is a model constructed in advance;
Determining a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined; wherein the load prediction period is determined based on a load prediction demand.
2. The method according to claim 1, wherein said determining at least one modal component to be determined corresponding to the historical load sequence comprises:
normalizing the historical load sequence to obtain a sequence to be processed;
carrying out interpolation processing on the missing load data in the sequence to be processed so as to determine the sequence to be processed after interpolation as a target processing sequence;
and decomposing the target processing sequence to obtain at least one modal component to be determined.
3. The method according to claim 2, wherein the decomposing the target processing sequence to obtain at least one modal component to be determined comprises:
carrying out Gaussian white noise treatment on the target treatment sequence to obtain a corresponding sequence to be used; wherein the Gaussian white noise respectively comprises positive sequence noise or negative sequence noise;
for each sequence to be used, determining an upper envelope curve and a lower envelope curve corresponding to the current sequence to be used;
And determining a current to-be-determined modal component corresponding to the current to-be-used sequence based on the current to-be-used sequence, the upper envelope and the lower envelope.
4. A method according to claim 3, wherein said determining an upper envelope and a lower envelope corresponding to a currently to be used sequence comprises:
determining at least one set of extremum data in the currently to-be-used sequence; wherein the extremum data comprises local maximum data and local minimum data;
and carrying out interpolation processing in the at least one group of extreme value data to obtain an upper envelope curve and a lower envelope curve corresponding to the current sequence to be used.
5. A method according to claim 3, wherein said determining a current to-be-determined modal component corresponding to the current to-be-used sequence based on the current to-be-used sequence, the upper envelope and the lower envelope comprises:
determining a mean value sequence corresponding to the upper envelope curve and the corresponding lower envelope curve at the same moment;
determining a current to-be-used component sequence corresponding to the current to-be-used sequence based on the difference value of the current to-be-used sequence and the mean value sequence;
Obtaining a current residual component sequence corresponding to the current sequence to be used based on the difference value between the current sequence to be used and the current component sequence to be used;
and determining a current to-be-determined modal component corresponding to the current to-be-used sequence based on the current to-be-used component sequence and the current residual component sequence.
6. The method of claim 1, wherein the invoking the load prediction model corresponding to each modal component to be determined comprises:
determining, for each modal component to be determined, a model call parameter corresponding to the current modal component to be determined based on a target mapping table; the target mapping table comprises at least one modal component and at least one model calling parameter to be matched corresponding to each modal component;
based on the at least one model calling parameter, calling a current load prediction model corresponding to the current modal component to be determined;
acquiring at least one associated data corresponding to the load prediction period; wherein the at least one associated data comprises time data to be predicted and weather data to be predicted;
And processing the at least one associated data based on the current load prediction model to obtain a prediction sequence to be determined corresponding to the current modal component to be determined.
7. The method according to claim 1, wherein the determining a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined comprises:
superposing the at least one predicted sequence to be determined to obtain a corresponding predicted sequence to be used;
and obtaining a target load prediction sequence corresponding to the load prediction period based on the ratio of at least one predicted sequence to be used to the number of sequences of the sequences to be used.
8. A load predicting apparatus, comprising:
the system comprises a modal component determining module, a modal component determining module and a processing module, wherein the modal component determining module is used for acquiring a historical load sequence and determining at least one modal component to be determined corresponding to the historical load sequence; the historical load sequence comprises load data corresponding to a plurality of historical moments;
the prediction sequence determining module is used for retrieving load prediction models corresponding to the modal components to be determined so as to determine the prediction sequences to be determined of the corresponding modal components to be determined based on the load prediction models; wherein the load prediction model is a model constructed in advance;
A load sequence determining module, configured to determine a target load prediction sequence corresponding to a load prediction period based on at least one prediction sequence to be determined; wherein the load prediction period is determined based on a load prediction demand.
9. An electronic device, the electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the load prediction method of any one of claims 1-7.
10. A computer readable storage medium storing computer instructions for causing a processor to perform the load prediction method of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310305393.6A CN116542362A (en) | 2023-03-24 | 2023-03-24 | Load prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310305393.6A CN116542362A (en) | 2023-03-24 | 2023-03-24 | Load prediction method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116542362A true CN116542362A (en) | 2023-08-04 |
Family
ID=87454982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310305393.6A Pending CN116542362A (en) | 2023-03-24 | 2023-03-24 | Load prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116542362A (en) |
-
2023
- 2023-03-24 CN CN202310305393.6A patent/CN116542362A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112541124B (en) | Method, apparatus, device, medium and program product for generating a multitasking model | |
| CN114723966B (en) | Multi-task recognition method, training method, device, electronic equipment and storage medium | |
| CN117032938B (en) | Operator parallel scheduling method and device, electronic equipment and storage medium | |
| US20240135698A1 (en) | Image classification method, model training method, device, storage medium, and computer program | |
| CN116307215A (en) | Load prediction method, device, equipment and storage medium of power system | |
| CN117332896A (en) | New energy small time scale power prediction method and system for multilayer integrated learning | |
| CN117332897A (en) | Artificial intelligent driven new energy small time scale power interpolation integrated prediction method | |
| CN116151446A (en) | Ultra-short-term wind power prediction method and device based on CEEMD and CNN-LSTM models | |
| CN116542362A (en) | Load prediction method and device, electronic equipment and storage medium | |
| CN113642654B (en) | Image feature fusion method and device, electronic equipment and storage medium | |
| CN112559727B (en) | Method, apparatus, device, storage medium, and program for outputting information | |
| CN114399101A (en) | TCN-BIGRU-based gas load prediction method and device | |
| CN115456167B (en) | Lightweight model training method, image processing device and electronic equipment | |
| CN112632846B (en) | Power transmission section limit probability assessment method of power system and electronic equipment | |
| CN117094452B (en) | Drought state prediction method, and training method and device of drought state prediction model | |
| CN114331379B (en) | Method for outputting task to be handled, model training method and device | |
| CN116933896B (en) | Super-parameter determination and semantic conversion method, device, equipment and medium | |
| CN116188875B (en) | Image classification method, device, electronic equipment, medium and product | |
| CN117934137A (en) | Bad asset recovery prediction method, device and equipment based on model fusion | |
| CN117454313A (en) | Multi-source heterogeneous data fusion method and device, electronic equipment and storage medium | |
| CN117634437A (en) | Method and device for describing current running state of equipment and electronic equipment | |
| CN115578583B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN116012873B (en) | Pedestrian re-identification method and device, electronic equipment and storage medium | |
| CN117878905A (en) | Power grid load prediction method, device, equipment and medium based on white noise signals | |
| CN116960957A (en) | Vehicle charging load prediction method and device, electronic equipment and storage medium |
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 |