CN111143776A - Electric quantity load prediction method and device - Google Patents
Electric quantity load prediction method and device Download PDFInfo
- Publication number
- CN111143776A CN111143776A CN201911380885.1A CN201911380885A CN111143776A CN 111143776 A CN111143776 A CN 111143776A CN 201911380885 A CN201911380885 A CN 201911380885A CN 111143776 A CN111143776 A CN 111143776A
- Authority
- CN
- China
- Prior art keywords
- data
- test
- external factor
- electric quantity
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000013528 artificial neural network Methods 0.000 claims abstract description 49
- 238000012545 processing Methods 0.000 claims abstract description 36
- 238000012360 testing method Methods 0.000 claims description 93
- 238000012549 training Methods 0.000 claims description 57
- 230000000737 periodic effect Effects 0.000 claims description 40
- 238000004590 computer program Methods 0.000 claims description 17
- 230000003203 everyday effect Effects 0.000 claims description 4
- 238000013277 forecasting method Methods 0.000 claims 2
- 230000005611 electricity Effects 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 3
- 238000005457 optimization Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 22
- 238000010586 diagram Methods 0.000 description 17
- 238000010248 power generation Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- Mathematical Analysis (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Pure & Applied Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Physics (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention is suitable for the technical field of energy, and provides an electric quantity load prediction method and a device, wherein the method comprises the following steps: acquiring time sequence data corresponding to a moment to be predicted and external factor data corresponding to the time sequence data; processing the time series data by adopting a residual error attention network to acquire semantic data characteristics of the time series data; processing the external factor data by adopting a self-encoder to acquire external factor characteristics of the external factor data; acquiring a combined feature according to the semantic data feature and the external factor feature; and processing the combination characteristics by adopting a neural network to obtain an electric quantity load prediction result at the moment to be predicted. The combined use of the two prediction methods makes up the defect of poor accuracy of single load prediction, solves the problem of large prediction deviation, and the accurate prediction result is beneficial to scheduling optimization of subsequent work; the method has simple overall operation, ensures the accuracy, convenience and rapidness of prediction and saves processing resources.
Description
Technical Field
The invention belongs to the technical field of energy, and particularly relates to an electric quantity load prediction method and device.
Background
The photovoltaic is a short-term solar photovoltaic power generation system, is a novel power generation system which utilizes the photovoltaic effect of a solar cell semiconductor material and can directly convert solar radiation into electric energy, and meanwhile, based on the classification of the solar photovoltaic power generation system, the photovoltaic is centralized, such as a large northwest ground photovoltaic power generation system; one is distributed, such as a roof photovoltaic power generation system of a factory and commercial enterprise factory. At present, electric quantity supply users are divided into industries, businesses, residents, offices and the like, electric quantity loads, load magnitude and load characteristics of different users are different, and accuracy difficulty of load prediction is different. There are many methods for predicting the load of the photovoltaic power, such as exponential smoothing, differential integrated Moving Average autoregressive model (autoregressive integrated Moving Average model), etc., but there are current situations that the load prediction accuracy is poor and the prediction deviation is large. In view of the above problem, a new method for solving the problem of poor accuracy of photovoltaic power load prediction is needed.
Disclosure of Invention
In view of this, embodiments of the present invention provide an electric quantity load prediction method, an apparatus, a terminal device, and a computer readable storage medium, so as to solve the technical problem in the prior art that an electric quantity load, especially a photovoltaic electric quantity load, cannot be accurately predicted.
In a first aspect of the embodiments of the present invention, a method for predicting an electric load is provided, including:
acquiring time sequence data corresponding to a moment to be predicted and external factor data corresponding to the time sequence data;
processing the time series data by adopting a residual attention network to acquire semantic data characteristics of the time series data;
processing the external factor data by adopting an auto-encoder to acquire external factor characteristics of the external factor data;
acquiring a combined feature according to the semantic data feature and the external factor feature;
and processing the combined characteristics by adopting a neural network to obtain an electric quantity load prediction result at the moment to be predicted.
In a second aspect of the embodiments of the present invention, there is provided an electric quantity load prediction apparatus, including:
the data acquisition module is used for acquiring time sequence data corresponding to the time to be predicted and external factor data corresponding to the time sequence data;
the first characteristic acquisition module is used for processing the time series data by adopting a residual attention network so as to acquire semantic data characteristics of the time series data;
the second characteristic acquisition module is used for processing the external factor data by adopting a self-encoder so as to acquire the external factor characteristics of the external factor data;
the combined feature acquisition module is used for acquiring combined features according to the semantic data features and the external factor features;
and the result acquisition module is used for processing the combined characteristics by adopting a neural network to acquire the electric quantity load prediction result at the moment to be predicted.
In a third aspect of the embodiments of the present invention, a terminal device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the power load prediction method are implemented.
In a fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided, where a computer program is stored, and the computer program, when executed by a processor, implements the method steps of the electric quantity load prediction method.
The electric quantity load prediction method provided by the embodiment of the invention has the beneficial effects that at least: the embodiment of the invention overcomes the defect of poor accuracy of single load prediction by matching two prediction methods, solves the problem of large prediction deviation, and simultaneously, the accurate prediction result is beneficial to scheduling optimization of subsequent work; the method has simple overall operation, ensures the accuracy, convenience and rapidness of prediction, improves the overall processing speed and saves processing resources.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed for the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a first schematic flow chart illustrating an implementation process of a power load prediction method according to an embodiment of the present invention;
fig. 2 is a schematic flow chart illustrating an implementation process of acquiring time-series data corresponding to a time to be predicted and external factor data corresponding to the time-series data in the electric quantity load prediction method according to the embodiment of the present invention;
fig. 3 is a schematic flow chart illustrating an implementation process of obtaining a preset rule in the electric quantity load prediction method according to the embodiment of the present invention;
fig. 4 is a schematic flow chart illustrating an implementation process of obtaining semantic data features in the electric quantity load prediction method according to the embodiment of the present invention;
fig. 5 is a schematic flow chart illustrating an implementation process of the electric quantity load prediction method according to the embodiment of the present invention;
fig. 6 is a schematic flow chart illustrating an implementation process of training an initial residual attention network, an initial self-encoder, and an initial neural network in the electric quantity load prediction method according to the embodiment of the present invention;
fig. 7 is a schematic flow chart of an implementation process of testing the trained residual attention network, the trained self-encoder, and the trained neural network by using the test data in the electric quantity load prediction method according to the embodiment of the present invention;
fig. 8 is a first schematic diagram of an electrical load prediction apparatus according to an embodiment of the present invention;
fig. 9 is a schematic diagram of a data acquisition module in the electric quantity load prediction apparatus according to the embodiment of the present invention;
fig. 10 is a schematic diagram of a rule determination unit in the electric load prediction apparatus according to an embodiment of the present invention;
fig. 11 is a schematic diagram of a first characteristic obtaining module in the electric quantity load prediction apparatus according to the embodiment of the present invention;
fig. 12 is a second schematic diagram of an electrical load prediction apparatus according to an embodiment of the present invention;
FIG. 13 is a diagram illustrating a training module of the electrical load prediction apparatus according to an embodiment of the present invention;
FIG. 14 is a schematic diagram of a test unit in the electrical load prediction apparatus according to an embodiment of the present invention;
fig. 15 is a schematic diagram of a terminal device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention, are within the scope of the invention. Unless otherwise specified, the technical means used in the examples are conventional means well known to those skilled in the art.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the present application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification of the present application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
As used in this specification and the appended claims, the term "if" may be interpreted contextually as "when", "upon" or "in response to a determination" or "in response to a detection". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
In order to explain the technical means of the present invention, the following description will be given by way of specific examples.
Referring to fig. 1, a first implementation flow diagram of the electric quantity load prediction method according to the embodiment of the present invention is shown, where the method may include:
step S10: and acquiring time sequence data corresponding to the time to be predicted and external factor data corresponding to the time sequence data.
Further, in order to acquire time-series data corresponding to a time to be predicted and external factor data corresponding to the time-series data, it is necessary to acquire time-series electric quantity load data. Fig. 2 is a schematic diagram of an implementation flow of acquiring time-series data corresponding to a time to be predicted and external factor data corresponding to the time-series data in the electric quantity load prediction method according to the embodiment of the present invention, in this embodiment, electric quantity load data of a time series is acquired; according to a preset rule, extracting time series data corresponding to the moment to be predicted from the electric quantity load data of the time series, wherein the time series data comprise trend data, periodic data and proximity data; and acquiring external factor data according to the time sequence data, wherein the external factor data at least comprises temperature data and weather data. One way of acquiring time-series data corresponding to a time to be predicted and external factor data corresponding to the time-series data may include the steps of:
step S101: and acquiring the electric quantity load data of the time sequence.
And preprocessing the initial time sequence data corresponding to the time to be predicted to obtain the electric quantity load average value and the electric quantity load value variance of the initial time sequence data corresponding to the time to be predicted. The initial time series data corresponding to the time to be predicted may be simply referred to as initial time series data.
The method for acquiring the average value and the variance of the electric quantity load value comprises the following steps:
wherein x isn,iRepresenting the electric quantity load value at the ith moment of the nth day,
characterizing the average value of the electrical load,
and characterizing the electric quantity load value variance.
And determining abnormal data in the initial time sequence data according to the electric quantity load average value and the electric quantity load value variance. And judging whether the difference value between the electric quantity load value of the initial time sequence data and the electric quantity load average value is larger than the electric quantity load standard deviation of a preset multiple.
The 3 sigma criterion is also called Layida criterion, which is that a group of detection data is assumed to only contain random errors, the detection data is calculated to obtain standard deviation, an interval is determined according to a certain probability, and if the error exceeding the interval is considered to be not the random error but a coarse error, the data containing the error is rejected. And 3 sigma is suitable when there are more groups of data.
The 3 sigma principle is as follows: the probability of the numerical distribution in (μ - σ, μ + σ) is 0.6827; the probability of the numerical distribution in (μ -3 σ, μ +3 σ) is 0.9545; the probability of the numerical distribution in (μ -3 σ, μ +3 σ) is 0.9973; it can be considered that the values are almost entirely concentrated in the interval of (mu-3 sigma, mu +3 sigma) vi > 3 sigma, and the possibility of exceeding this range is only less than 0.3%.
The 3 sigma criterion is based on equal-precision repeated measurement of normal distribution, so that the interference or noise of singular data is difficult to meet the normal distribution. If the absolute value nui of the residual error of a certain measured value in a group of measured data is larger than 3 sigma, the measured value is a bad value and should be removed.
The 3 sigma criterion exists because the probability of falling outside 3 sigma is only 0.27% for a normally distributed random error, which is usually taken as the limit error, and it is less likely to occur in a limited number of measurements. The 3 sigma criterion is the most common and simplest gross error criterion and is generally applied when the number of measurements is sufficiently large (n ≧ 30) or when n > 10 is used for coarse discrimination.
A normal distribution has two parameters, mu and sigma2The first parameter mu is the mean of the random variables following a normal distribution, and the second parameter sigma is the distribution of the continuous random variables2Is the variance of this random variable, so a normal distribution is denoted as N (μ, σ)2)。
μ is a position parameter of the normal distribution, describing the central tendency position of the normal distribution. The probability law is that the probability of taking values adjacent to μ is high, while the probability of taking values further away from μ is lower. The normal distribution is completely symmetrical about the axis of symmetry X ═ μ. The expectation, mean, median, and mode of normal distribution are the same and equal to μ.
The sigma describes the degree of dispersion of data distribution of normally distributed data, the larger the sigma is, the more dispersed the data distribution is, and the smaller the sigma is, the more concentrated the data distribution is. Also known as a normally distributed shape parameter, the larger σ, the flatter the curve, whereas the smaller σ, the thinner the curve.
And if the difference value between the electric quantity load value of the initial time sequence data and the electric quantity load average value is larger than the electric quantity load standard deviation of a preset multiple, determining the initial time sequence data corresponding to the moment to be predicted as abnormal data. And correcting the abnormal data to obtain corrected initial time series data, namely time series data corresponding to the time to be predicted.
The correction method comprises the following steps:
wherein the content of the first and second substances,
representing the corrected value of the electric quantity load at the ith moment of the nth day,
representing the electric quantity load values of m similar unit time ith moments before and after the electric quantity load value,
and representing the electric quantity load values of p similar unit time ith moments before and after.
When m is 1 and p is 2, the formula is:
wherein the content of the first and second substances,
representing the electric quantity load correction value, x, at the ith moment of the nth dayn±1I represents the electric quantity load value at the ith moment of 2 similar days before and after the electric quantity load value,
and characterizing the electric quantity load values of the i-th time of the front and back 2 similar days.
Day of the same kind: for example, a certain time t currently being the second friday (day) of the month, and two days (days) of the respective times of the first friday (day) and the third friday (day) adjacent to the month; similar day: such as the two days (days) at a time t currently being the second friday of the month, the corresponding times of the second thursday (day) of the month and the saturday (day).
It should be understood that the same kind of unit time and the similar unit time may be any time herein, and a plurality of unit times may be acquired at the same time, which is not limited herein.
Step S102: and according to a preset rule, extracting time series data corresponding to the moment to be predicted from the electric quantity load data of the time series, wherein the time series data comprise trend data, periodic data and proximity data.
Further, in order to acquire the preset rule, explicit trend data, periodic data, and proximity data are required. Fig. 3 is a schematic diagram of a flow of implementing a preset rule obtaining in the electric quantity load prediction method according to the embodiment of the present invention, where in this embodiment, the trend data is electric quantity load data of a preset number of weeks at the same time every week before the time to be predicted; the periodic data is electric quantity load data of the same time every day on preset days before the time to be predicted; the proximity data is electric quantity load data of the same moment every hour in preset hours before the moment to be predicted. One way to obtain the preset rule may include the following steps:
step S1021: the trend data is the electric quantity load data of the preset week number at the same time every week before the time to be predicted.
Step S1022: the periodic data is electric quantity load data of the same time every day on preset days before the time to be predicted.
Step S1023: the proximity data is electric quantity load data of the same moment every hour in preset hours before the moment to be predicted.
The time series data contains three types of data: trend data: selecting the data of the last 7 weeks; periodic data: selecting data for the last 21 days; proximity data: the last 24 hours of data were selected.
It should be understood that the time series data may also be added with other preset data, not limited to trend data, periodic data and proximity data, but may also be reduced by one or all of them to replace other preset data, which is not limited herein.
It should be understood that the trending data may be selected for 4 weeks or 20 weeks; the periodic data can be selected from 10 days or 200 days; the proximity data may be selected to be 10 hours or 18 hours, but is not limited thereto.
After obtaining the preset rule, the following steps can be performed:
step S103: and acquiring external factor data according to the time sequence data, wherein the external factor data at least comprises temperature data and weather data.
It should be understood that the external factor data includes not only temperature and weather, but also humidity, wind speed, visibility, pressure, etc., without limitation.
Referring to fig. 1, further, after acquiring time-series data corresponding to a time to be predicted and external factor data corresponding to the time-series data, the following steps may be performed:
step S30: and processing the time sequence data by adopting a residual attention network to acquire semantic data characteristics of the time sequence data.
Further, in order to obtain semantic data features, a residual attention network is required to process the time series data. Please refer to fig. 4, which is a schematic diagram illustrating a flow of implementing semantic data feature acquisition in the electric quantity load prediction method according to the embodiment of the present invention, in the embodiment, a first residual attention network is used to process the trend data, so as to acquire a trend data feature corresponding to the trend data; processing the periodic data by adopting a second residual attention network to obtain periodic data characteristics corresponding to the periodic data; and processing the proximity data by adopting a third residual attention network to acquire the proximity data characteristic corresponding to the proximity data. One way to obtain semantic data features may include the steps of:
step S301: and processing the trend data by adopting a first residual attention network to obtain trend data characteristics corresponding to the trend data.
Step S302: and processing the periodic data by adopting a second residual attention network to obtain periodic data characteristics corresponding to the periodic data.
Step S303: and processing the proximity data by adopting a third residual attention network to acquire the proximity data characteristic corresponding to the proximity data.
Referring to fig. 1, further, after obtaining the semantic data features, the following steps may be performed:
step S40: and processing the external factor data by adopting an auto-encoder to acquire the external factor characteristics of the external factor data.
Referring to fig. 1, further, after obtaining the external factor characteristics, the following steps may be performed:
step S50: and acquiring a combined feature according to the semantic data feature and the external factor feature.
And splicing and combining the extracted highly abstract trend data features, periodic data features, proximity data features and external factor features to obtain combined features.
Referring to fig. 1, further, after obtaining the combination feature, the following steps may be performed:
step S60: and processing the combined characteristics by adopting a neural network to obtain an electric quantity load prediction result at the moment to be predicted.
Referring to fig. 5, a schematic flow chart of an implementation of the electric quantity load prediction method according to the embodiment of the present invention is shown, and further, before the step of obtaining the semantic data features of the time series data, the following steps may be performed:
step S20: and training the initial residual attention network, the initial self-encoder and the initial neural network to obtain the residual attention network, the self-encoder and the neural network which meet preset requirements.
Further, in order to obtain semantic data features, a residual attention network is required to process the time series data. Please refer to fig. 6, which is a schematic diagram illustrating an implementation process of training an initial residual attention network, an initial self-encoder, and an initial neural network in the electric quantity load prediction method according to an embodiment of the present invention, in this embodiment, a manner of obtaining the residual attention network, the initial self-encoder, and the neural network that satisfy preset requirements may include the following steps:
step S201: the time series data and the external factor data are segmented to obtain training data and test data, the training data comprise trend training data, periodic training data, proximity training data and external factor training data, and the test data comprise trend test data, periodic test data, proximity test data and external factor test data.
Step S202: and training the initial residual attention network and the initial self-encoder by adopting the training data to obtain the combination characteristics of the trained residual attention network, the trained self-encoder and the training data, wherein the training combination characteristics comprise semantic data characteristics and external factor characteristics.
The residual attention network in the embodiment is applied to feature extraction of power time series data, and more abstract semantic data features are obtained. The method comprises the following specific steps: respectively using trend data, periodic data and proximity data of training data to train a residual attention network, using external characteristic data to train a self-encoder, after the residual attention network and the self-encoder are trained, inputting the trend data, the periodic data and the proximity data of the training data into the residual attention network to obtain highly abstract trend data characteristics, periodic data characteristics and proximity data characteristics, and inputting the external factor characteristics into the self-encoder to obtain abstract external factor characteristics.
Step S203: and training the initial neural network by adopting the combined features of the training data to obtain the trained neural network.
Step S204: and testing the trained residual attention network, the trained self-encoder and the trained neural network by adopting the test data so as to determine that the residual attention network, the self-encoder and the neural network which meet the preset requirements are used for electric quantity load prediction.
Further, in order to determine that the residual attention network, the self-encoder and the neural network which meet the preset requirements are used for electric quantity load prediction, the trained residual attention network, the trained self-encoder and the trained neural network need to be tested by using the test data. Please refer to fig. 7, which is a schematic diagram illustrating an implementation process of testing the trained residual attention network, the trained self-encoder, and the trained neural network by using the test data in the electric quantity load prediction method according to the embodiment of the present invention, in this embodiment, the test data is input into the trained residual attention network and the trained self-encoder to obtain a combination feature of the test data; inputting the combined features of the test data into the trained neural network to obtain a test prediction result; obtaining a test index according to the test prediction result and the test data; judging whether the test index meets a preset requirement or not; if the test index meets a preset requirement, determining that the trained residual attention network, the trained self-encoder and the trained neural network are respectively a residual attention network, a self-encoder and a neural network for electric quantity load prediction; and if the test index does not meet the preset requirement, returning to the step of training the initial residual error attention network and the initial self-encoder by adopting the training data. One way to determine that a residual attention network, a self-encoder and a neural network that meet preset requirements are used for electric load prediction may include the steps of:
step S2041: and inputting the test data into the trained residual attention network and the trained self-encoder to obtain the combined features of the test data.
Step S2042: and inputting the combined features of the test data into the trained neural network to obtain a test prediction result.
Step S2043: and obtaining a test index according to the test prediction result and the test data.
The test index comprises a root mean square error and a relative error rate, and the root mean square error of the test index is obtained in the following mode:
wherein, yiCharacterizing the test values in the time series data,
predictive values characterizing the time series data;
the relative error rate obtaining mode of the test indexes is as follows:
the resulting combined features are used together with the time series data to train the neural network. The method comprises the steps of using test set data to test an algorithm model, using a trained residual attention network to extract trend characteristics, periodic characteristics and proximity characteristics of the test set data respectively to obtain highly abstract trend characteristics, periodic characteristics and proximity characteristics respectively, using a trained self-encoder to extract external factor characteristics such as temperature and weather to obtain abstract external factor characteristics, combining the characteristics to obtain combined characteristics, and inputting the combined characteristics into a trained neural network to obtain a prediction result.
Step S2044: and judging whether the test index meets a preset requirement.
Step S2045: and if the test index meets the preset requirement, determining that the trained residual attention network, the trained self-encoder and the trained neural network are respectively a residual attention network, a self-encoder and a neural network for electric quantity load prediction.
Step S2046: and if the test index does not meet the preset requirement, returning to the step of training the initial residual error attention network and the initial self-encoder by adopting the training data.
It should be understood that the above-mentioned letters and/or symbols are only used for clearly explaining the meaning of specific parameters of the device or steps, and other letters or symbols can be used for representing the device or steps, which is not limited herein.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
The electric quantity load prediction method provided by the embodiment of the invention has the beneficial effects that at least: the embodiment overcomes the defect of poor accuracy of single load prediction by using two prediction methods in a matching way, solves the problem of large prediction deviation, and simultaneously, the accurate prediction result is beneficial to scheduling optimization of subsequent work; the method has simple overall operation, ensures the accuracy, convenience and rapidness of prediction, improves the overall processing speed and saves processing resources.
Fig. 8 is a first schematic diagram of the electric quantity load prediction device provided in the embodiment of the present invention, and for convenience of description, only the parts related to the embodiment of the present application are shown.
Referring to fig. 8, the electric quantity load prediction apparatus includes a data obtaining module 71, a first characteristic obtaining module 73, a second characteristic obtaining module 74, a combined characteristic obtaining module 75, and a result obtaining module 76. The data obtaining module 71 is configured to obtain time series data corresponding to a time to be predicted and external factor data corresponding to the time series data; the first feature obtaining module 73 is configured to process the time series data by using a residual attention network to obtain semantic data features of the time series data; the second characteristic obtaining module 74 is configured to process the external factor data by using a self-encoder to obtain external factor characteristics of the external factor data; the combined feature obtaining module 75 is configured to obtain a combined feature according to the semantic data feature and the external factor feature; the result obtaining module 76 is configured to process the combined features by using a neural network, and obtain a prediction result of the electric quantity load at the time to be predicted.
Referring to fig. 9, the data acquisition module 71 further includes a basic data determination unit 711, a rule determination unit 712, and an external factor determination unit 713. Wherein the basic data determination 711 is used for acquiring time series electric quantity load data; the rule determining unit 712 is configured to extract time series data corresponding to a time to be predicted from the time series of electric quantity load data according to a preset rule, where the time series data includes trend data, periodic data, and proximity data; the external factor determination unit 713 is configured to acquire external factor data including at least temperature data and weather data based on the time-series data.
Referring to fig. 10, the rule determination unit 712 further includes a trend data preset sub-unit 7121, a periodic data preset sub-unit 7122 and a proximity data preset sub-unit 7123. The trend data presetting subunit 7121 is configured to preset the electric load data of the week at the same time before the time to be predicted; the periodic data presetting subunit 7122 is used for presetting electric quantity load data of the same time every day for preset days before the time to be predicted; the proximity data presetting subunit 7123 is used for presetting electric quantity load data of the same moment in each hour in preset hours before the moment to be predicted.
Referring to fig. 11, the first feature acquisition module 73 further includes a trend feature acquisition unit 731, a period feature acquisition unit 732, and an adjacent feature acquisition unit 733. The trend feature acquiring unit 731 is configured to process the trend data by using a first residual attention network, and acquire a trend data feature corresponding to the trend data; the periodic feature obtaining unit 732 is configured to process the periodic data by using a second residual attention network, and obtain a periodic data feature corresponding to the periodic data; the proximity feature obtaining unit 733 is configured to process the proximity data by using a third residual attention network, and obtain a proximity data feature corresponding to the proximity data.
Further, please refer to fig. 12, which is a schematic diagram of a power load prediction apparatus according to an embodiment of the present invention. The electric quantity load prediction device comprises a training module 72, which is used for training the initial residual attention network, the initial self-encoder and the initial neural network to obtain the residual attention network, the self-encoder and the neural network which meet preset requirements.
Referring to fig. 13, the training module 72 further includes a segmentation unit 721, a first training unit 722, a second training unit 723, and a testing unit 724. The segmentation unit 721 is configured to segment the time series data and the external factor data to obtain training data and test data, where the training data includes trend training data, periodic training data, proximity training data, and external factor training data, and the test data includes trend test data, periodic test data, proximity test data, and external factor test data; the first training unit 722 is configured to train the initial residual attention network and the initial self-encoder with the training data to obtain a combined feature of the trained residual attention network, the trained self-encoder, and the training data, where the training combined feature includes a semantic data feature and an external factor feature; the second training unit 723 is configured to train the initial neural network by using the combined features of the training data to obtain a trained neural network; the testing unit 724 is configured to test the trained residual attention network, the trained self-encoder, and the trained neural network by using the test data, so as to determine that the residual attention network, the self-encoder, and the neural network that meet preset requirements are used for electric quantity load prediction.
Referring to fig. 14, further, the test unit 724 includes a first test subunit 7241, a second test subunit 7242, an index determination subunit 7243, a judgment subunit 7244, a determination subunit 7245, and a return subunit 7246. Wherein, the first testing subunit 7241 is configured to input the test data into the trained residual attention network and the trained self-encoder, and obtain a combined feature of the test data; the second testing subunit 7242 is configured to input the combined features of the test data into the trained neural network, and obtain a test prediction result; the index determination subunit 7243 is configured to obtain a test index according to the test prediction result and the test data; the judgment subunit 7244 is configured to judge whether the test index meets a preset requirement; the determining subunit 7245 is configured to determine that the trained residual attention network, the trained self-encoder, and the trained neural network are respectively a residual attention network, a self-encoder, and a neural network for electric quantity load prediction if the test index meets a preset requirement; the returning subunit 7246 is configured to, if the test indicator does not meet the preset requirement, return to the step of training the initial residual attention network and the initial self-encoder by using the training data.
Fig. 15 is a schematic diagram of a terminal device according to an embodiment of the present invention. As shown in fig. 15, the terminal device 8 includes a memory 81, a processor 80, and a computer program 82 stored in the memory 81 and executable on the processor 80, and the processor 80 implements the steps of the electrical load prediction method when executing the computer program 82, such as steps S10 to S60 shown in fig. 1-7.
The terminal device 8 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The terminal device may include, but is not limited to, the processor 80 and the memory 81. Those skilled in the art will appreciate that fig. 15 is merely an example of a terminal device 8 and does not constitute a limitation of terminal device 8 and may include more or fewer components than shown, or some components may be combined, or different components, e.g., the terminal device may also include input-output devices, network access devices, buses, etc.
The Processor 80 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The storage 81 may be an internal storage unit of the terminal device 8, such as a hard disk or a memory of the terminal device 8. The memory 81 may also be an external storage device of the terminal device 8, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) and the like provided on the terminal device 8. Further, the memory 81 may also include both an internal storage unit and an external storage device of the terminal device 8. The memory 81 is used for storing the computer program and other programs and data required by the terminal device. The memory 81 may also be used to temporarily store data that has been output or is to be output.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow in the method of the embodiments described above can be realized by a computer program, which can be stored in a computer-readable storage medium and can realize the steps of the embodiments of the methods described above when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
Specifically, the present application further provides a computer-readable storage medium, which may be a computer-readable storage medium contained in the memory in the foregoing embodiments; or it may be a separate computer-readable storage medium not incorporated into the terminal device. The computer readable storage medium stores one or more computer programs:
a computer-readable storage medium comprising a computer program stored thereon, which, when executed by a processor, performs the steps of the method for predicting electrical load.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described embodiments of the apparatus/terminal device are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A method for predicting a power load, comprising:
acquiring time sequence data corresponding to a moment to be predicted and external factor data corresponding to the time sequence data;
processing the time series data by adopting a residual attention network to acquire semantic data characteristics of the time series data;
processing the external factor data by adopting an auto-encoder to acquire external factor characteristics of the external factor data;
acquiring a combined feature according to the semantic data feature and the external factor feature;
and processing the combined characteristics by adopting a neural network to obtain an electric quantity load prediction result at the moment to be predicted.
2. The electric quantity load prediction method according to claim 1, wherein the acquiring of the time-series data corresponding to the time to be predicted and the external factor data corresponding to the time-series data includes:
acquiring time-series electric quantity load data;
according to a preset rule, extracting time series data corresponding to the moment to be predicted from the electric quantity load data of the time series, wherein the time series data comprise trend data, periodic data and proximity data;
and acquiring external factor data according to the time sequence data, wherein the external factor data at least comprises temperature data and weather data.
3. The electric quantity load prediction method according to claim 2, wherein in the step of extracting time series data corresponding to a time to be predicted from the electric quantity load data of the time series according to a preset rule, the time series data includes trend data, periodic data and proximity data, and the trend data is the electric quantity load data of the same time every week in a preset number of weeks before the time to be predicted;
the periodic data is electric quantity load data of the same time every day on preset days before the time to be predicted;
the proximity data is electric quantity load data of the same moment every hour in preset hours before the moment to be predicted.
4. The electrical load forecasting method of claim 1, wherein the semantic data features include trending data features, periodic data features, and proximity data features;
the processing the time series data by using the residual attention network to acquire semantic data features of the time series data comprises:
processing the trend data by adopting a first residual attention network to obtain trend data characteristics corresponding to the trend data;
processing the periodic data by adopting a second residual attention network to obtain periodic data characteristics corresponding to the periodic data;
and processing the proximity data by adopting a third residual attention network to acquire the proximity data characteristic corresponding to the proximity data.
5. The electric load forecasting method according to claim 1, wherein before the step of processing the time-series data by using a residual attention network to obtain semantic data features of the time-series data, the method further comprises:
training an initial residual attention network, an initial self-encoder and an initial neural network to obtain the residual attention network, the self-encoder and the neural network which meet preset requirements, wherein the training comprises the following steps:
segmenting the time series data and the external factor data to obtain training data and test data, wherein the training data comprises trend training data, periodic training data, proximity training data and external factor training data, and the test data comprises trend test data, periodic test data, proximity test data and external factor test data;
training the initial residual attention network and the initial self-encoder by using the training data to obtain a combination feature of the trained residual attention network, the trained self-encoder and the training data, wherein the training combination feature comprises a semantic data feature and an external factor feature;
training the initial neural network by adopting the combined features of the training data to obtain a trained neural network;
and testing the trained residual attention network, the trained self-encoder and the trained neural network by adopting the test data so as to determine that the residual attention network, the self-encoder and the neural network which meet the preset requirements are used for electric quantity load prediction.
6. The method of claim 5, wherein the testing the trained residual attention network, the trained self-encoder, and the trained neural network with the test data to determine that the residual attention network, the self-encoder, and the neural network meet preset requirements for the battery load prediction comprises:
inputting the test data into the trained residual attention network and the trained self-encoder to obtain the combined features of the test data;
inputting the combined features of the test data into the trained neural network to obtain a test prediction result;
obtaining a test index according to the test prediction result and the test data;
judging whether the test index meets a preset requirement or not;
if the test index meets a preset requirement, determining that the trained residual attention network, the trained self-encoder and the trained neural network are respectively a residual attention network, a self-encoder and a neural network for electric quantity load prediction;
and if the test index does not meet the preset requirement, returning to the step of training the initial residual error attention network and the initial self-encoder by adopting the training data.
7. The electrical load prediction method according to claim 6, wherein in the obtaining of the test indicators according to the test prediction results and the test data, the test indicators include root mean square errors and relative error rates, and the root mean square errors of the test indicators are obtained by:
wherein, yiCharacterizing the test values in the time series data,
predictive values characterizing the time series data;
the relative error rate obtaining mode of the test indexes is as follows:
8. a device for predicting a load of electricity, comprising:
the data acquisition module is used for acquiring time sequence data corresponding to the time to be predicted and external factor data corresponding to the time sequence data;
the first characteristic acquisition module is used for processing the time series data by adopting a residual attention network so as to acquire semantic data characteristics of the time series data;
the second characteristic acquisition module is used for processing the external factor data by adopting a self-encoder so as to acquire the external factor characteristics of the external factor data;
the combined feature acquisition module is used for acquiring combined features according to the semantic data features and the external factor features;
and the result acquisition module is used for processing the combined characteristics by adopting a neural network to acquire the electric quantity load prediction result at the moment to be predicted.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911380885.1A CN111143776B (en) | 2019-12-27 | 2019-12-27 | Electric quantity load prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911380885.1A CN111143776B (en) | 2019-12-27 | 2019-12-27 | Electric quantity load prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111143776A true CN111143776A (en) | 2020-05-12 |
| CN111143776B CN111143776B (en) | 2022-06-28 |
Family
ID=70521168
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911380885.1A Active CN111143776B (en) | 2019-12-27 | 2019-12-27 | Electric quantity load prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111143776B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020165816A1 (en) * | 2001-05-02 | 2002-11-07 | Barz Graydon Lee | Method for stochastically modeling electricity prices |
| CN106056233A (en) * | 2016-05-11 | 2016-10-26 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Power load prediction method |
| US20180106869A1 (en) * | 2015-05-08 | 2018-04-19 | Volvo Truck Corporation | A method for monitoring the status of a plurality of battery cells in a battery pack |
| US20180240202A1 (en) * | 2015-08-19 | 2018-08-23 | China Electric Power Research Institute Company Limited | Method of predicting distribution network operation reliability |
| US20190130273A1 (en) * | 2017-10-27 | 2019-05-02 | Salesforce.Com, Inc. | Sequence-to-sequence prediction using a neural network model |
| CN110163447A (en) * | 2019-05-29 | 2019-08-23 | 电子科技大学 | Long term power load forecasting method based on residual GM grey forecasting model |
| CN110222882A (en) * | 2019-05-21 | 2019-09-10 | 国家电网公司西南分部 | A kind of prediction technique and device of electric system Mid-long Term Load |
| CN110445126A (en) * | 2019-06-25 | 2019-11-12 | 中国电力科学研究院有限公司 | A kind of non-intrusion type load decomposition method and system |
-
2019
- 2019-12-27 CN CN201911380885.1A patent/CN111143776B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020165816A1 (en) * | 2001-05-02 | 2002-11-07 | Barz Graydon Lee | Method for stochastically modeling electricity prices |
| US20180106869A1 (en) * | 2015-05-08 | 2018-04-19 | Volvo Truck Corporation | A method for monitoring the status of a plurality of battery cells in a battery pack |
| US20180240202A1 (en) * | 2015-08-19 | 2018-08-23 | China Electric Power Research Institute Company Limited | Method of predicting distribution network operation reliability |
| CN106056233A (en) * | 2016-05-11 | 2016-10-26 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | Power load prediction method |
| US20190130273A1 (en) * | 2017-10-27 | 2019-05-02 | Salesforce.Com, Inc. | Sequence-to-sequence prediction using a neural network model |
| CN110222882A (en) * | 2019-05-21 | 2019-09-10 | 国家电网公司西南分部 | A kind of prediction technique and device of electric system Mid-long Term Load |
| CN110163447A (en) * | 2019-05-29 | 2019-08-23 | 电子科技大学 | Long term power load forecasting method based on residual GM grey forecasting model |
| CN110445126A (en) * | 2019-06-25 | 2019-11-12 | 中国电力科学研究院有限公司 | A kind of non-intrusion type load decomposition method and system |
Non-Patent Citations (6)
| Title |
|---|
| TONGGUANG YANG等: "LSTM-Attention-Embedding Model-Based Day-Ahead Prediction of Photovoltaic Power Output Using Bayesian Optimization", 《IEEE ACCESS》 * |
| 周晖等: "夏季最大负荷发生时间的预测", 《现代电力》 * |
| 朱晓明: "BP-灰度模型的电力负荷预测模型研究", 《科技通报》 * |
| 牛勇等: "改进灰色模型在中长期电力负荷预测中的应用", 《东北电力大学学报(自然科学版)》 * |
| 王喜平等: "夏季短期电力负荷ARIMA-SVR组合预测模型", 《黑龙江电力》 * |
| 陈巧玲等: "基于残差修正的组合模型在电力负荷预测中的应用", 《技术与市场》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111143776B (en) | 2022-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110991761B (en) | Heat supply load prediction method and device | |
| CN113125851B (en) | Power consumption statistical method, device, equipment and storage medium | |
| CN115905927A (en) | Method and device for identifying abnormal electricity consumption user, electronic equipment and storage medium | |
| CN113516275A (en) | Power distribution network ultra-short term load prediction method and device and terminal equipment | |
| CN113723861A (en) | Abnormal electricity consumption behavior detection method and device, computer equipment and storage medium | |
| CN114912720A (en) | Memory network-based power load prediction method, device, terminal and storage medium | |
| CN112801315A (en) | State diagnosis method and device for power secondary equipment and terminal | |
| CN116523140A (en) | Method and device for detecting electricity theft, electronic equipment and storage medium | |
| CN113391256B (en) | Electric energy meter metering fault analysis method and system of field operation terminal | |
| CN111091420A (en) | Method and device for predicting power price | |
| CN108280608B (en) | Product life analysis method and terminal equipment | |
| CN113592192A (en) | Short-term power load prediction method and device and terminal equipment | |
| CN111143776B (en) | Electric quantity load prediction method and device | |
| CN111523083A (en) | Method and device for determining power load declaration data | |
| CN111127114A (en) | Method and device for determining power load declaration data | |
| CN111144634A (en) | Method and device for predicting power price | |
| CN113376564B (en) | Smart electric meter metering correction method and device based on data analysis and terminal | |
| CN112395179B (en) | Model training method, disk prediction method, device and electronic equipment | |
| CN115473216A (en) | Method and system for improving line loss calculation of power grid | |
| CN113590608A (en) | Data stream processing-based user electricity consumption information collecting and correcting method | |
| CN112614005B (en) | Method and device for processing reworking state of enterprise | |
| CN113592218A (en) | Photovoltaic user baseline load estimation method and device and terminal equipment | |
| CN108269004B (en) | Product life analysis method and terminal equipment | |
| CN113505943A (en) | Method, system, equipment and medium for predicting short-term load of power grid | |
| CN112531629B (en) | Method and device for automatically setting protection setting value of power distribution network and terminal equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |