CN112734096A - Urban saturation load prediction method and system - Google Patents
Urban saturation load prediction method and system Download PDFInfo
- Publication number
- CN112734096A CN112734096A CN202011639715.3A CN202011639715A CN112734096A CN 112734096 A CN112734096 A CN 112734096A CN 202011639715 A CN202011639715 A CN 202011639715A CN 112734096 A CN112734096 A CN 112734096A
- Authority
- CN
- China
- Prior art keywords
- urban
- sequence
- load
- data
- load prediction
- 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 56
- 229920006395 saturated elastomer Polymers 0.000 claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 40
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000000638 solvent extraction Methods 0.000 claims abstract description 5
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 5
- 230000003068 static effect Effects 0.000 claims description 5
- 230000005855 radiation Effects 0.000 claims description 4
- 238000005192 partition Methods 0.000 claims description 3
- 238000001556 precipitation Methods 0.000 claims description 2
- 238000001744 unit root test Methods 0.000 claims description 2
- 230000001131 transforming effect Effects 0.000 claims 1
- 238000012163 sequencing technique Methods 0.000 abstract description 2
- 230000005611 electricity Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 5
- 230000008859 change Effects 0.000 description 5
- 230000007774 longterm Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013277 forecasting method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- 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)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Probability & Statistics with Applications (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method and a system for predicting urban saturation load, comprising the following steps: acquiring historical load data of substations in each area after urban partitioning and meteorological data of corresponding time periods to construct a training sample set; performing smoothness processing on the training sample set by adopting a linear clustering algorithm, and training a time sequence-based saturated load prediction model by using the processed training sample set; and predicting meteorological data of the transformer substation in the region to be tested in the time period to be tested by adopting the trained saturated load prediction model, and obtaining a total urban saturated load prediction result according to the transformer substation saturated load prediction result of each region in the same time period. The method can predict the total amount of the urban loads, and simultaneously can acquire the spatial distribution of the regional loads, thereby providing a basis for the optimal sequencing of the urban distribution network planning project.
Description
Technical Field
The invention relates to the technical field of saturated load prediction, in particular to a method and a system for predicting urban saturated load.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The saturated load prediction in the load prediction of the power system is currently applied to an urban power load saturation analysis method, and a saturation value is obtained based on a logistic curve equation and historical data; or applying an intelligent algorithm to prediction, applying ant colony algorithm, cellular automata and other theory building models, analyzing the application of a neural network model in medium and long term load prediction, and analyzing the saturation density conditions under different land properties by an S curve method. The typical analysis method of the regional saturated load mainly comprises a saturated load prediction method based on the average electricity of people, a saturated load prediction method based on the urban load density and a prediction method based on the spatial saturated load density.
The prediction method based on the per-capita electricity quantity needs to determine population bearing capacity matched with urban resources and environments according to national macro planning, urban general planning and various special plans, determine per-capita saturated electricity quantity suitable for the city by referring to per-capita electricity quantity of a typical developed country in foreign countries, obtain the size of urban saturated load, and predict the approximate arrival time of the urban electricity demand entering saturation; the idea of adopting the method of using the electricity per person to predict the saturation load is roughly as follows: and multiplying the total population of the saturated year by the per-person saturated power consumption to obtain the saturated scale of the power consumption of the whole society in the area.
The saturated load scale of the urban load density can be obtained by the ratio of the saturated scale of the electricity consumption of the whole society to the maximum load utilization hours; based on urban development positioning, referring to the load density situation similar to the urban positioning, deducing to obtain the load density when the local area load increases to represent the saturation situation, and calculating to obtain the scale of the urban saturated load by combining the urban planning land area determined by urban overall planning.
The prediction method based on the space saturation load density comprises the following steps: multiplying the saturated load density of the saturated year by the power utilization area to obtain the saturated scale of the power consumption of the whole society in the area; the maximum load saturation scale is obtained by the ratio of the total social electricity consumption saturation scale to the maximum load.
The inventor believes that the traditional method has certain limitations as the complexity of the influence factors of load prediction increases, and is difficult to solve the problem of complex nonlinearity, so that the load prediction is inaccurate. At present, a scheme for realizing prediction by constructing a mapping relation between an input variable and an output variable through a neural network model and a space load prediction method based on a double-layer Bayes classification model exist, and each cell is reasonably classified according to the input characteristic thereof by utilizing a classification model finished based on sample training, so that a class label, namely a load density index, is obtained. However, in practical application, the above method must face a problem of how to collect and process a large amount of sample data, and many factors affecting the power load, such as weather factors, climate factors, historical loads, economic factors, industrial structures, power utilization structures, and the like, and many corresponding factor indexes, such as average temperature, average humidity, days of holidays, GDP acceleration, industrial structures, power utilization structures, saturated load density, and the like; the prediction accuracy of the prediction model is affected by the number, quality, distribution and training effect of the samples on the model.
In addition, when urban load is predicted, if modeling prediction is directly carried out on the basis of the overall load curve, the modeling error is small due to high load level, but random error is difficult to control; if modeling prediction is directly carried out based on the substation sequence, and finally the urban load prediction result is obtained by summation, although the random error tends to be 0, the modeling is difficult and the prediction error is increased due to the low load level of each sequence.
Disclosure of Invention
In order to solve the problems, the invention provides a method and a system for predicting urban saturated load, which are used for deconstructing urban total load by using historical load data of each regional substation, performing smoothness processing on load sequences of each substation forming the urban total load by a linear clustering method, improving data linearity, performing optimal ARIMA prediction on the sequences subjected to linear clustering processing, and adding load prediction results of each region to obtain a predicted value of the urban total load.
In order to achieve the purpose, the invention adopts the following technical scheme:
in a first aspect, the present invention provides a method for predicting an urban saturation load, including:
acquiring historical load data of substations in each area after urban partitioning and meteorological data of corresponding time periods to construct a training sample set;
performing smoothness processing on the training sample set by adopting a linear clustering algorithm, and training a time sequence-based saturated load prediction model by using the processed training sample set;
and predicting meteorological data of the transformer substation in the region to be tested in the time period to be tested by adopting the trained saturated load prediction model, and obtaining a total urban saturated load prediction result according to the transformer substation saturated load prediction result of each region in the same time period.
In a second aspect, the present invention provides a system for predicting an urban saturation load, including:
the data acquisition module is used for acquiring historical load data of the transformer substation in each area after the urban partition and meteorological data of a corresponding time period to construct a training sample set;
the training module is used for performing smoothness processing on the training sample set by adopting a linear clustering algorithm and training a time sequence-based saturation load prediction model by using the processed training sample set;
and the prediction module is used for predicting the meteorological data of the transformer substation in the region to be tested in the time period to be tested by adopting the trained saturated load prediction model, and obtaining the prediction result of the total saturated load of the city according to the prediction result of the saturated load of the transformer substation in each region in the same time period.
In a third aspect, the present invention provides an electronic device comprising a memory and a processor, and computer instructions stored on the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method of the first aspect is performed.
In a fourth aspect, the present invention provides a computer readable storage medium for storing computer instructions which, when executed by a processor, perform the method of the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
according to the urban load forecasting method, the urban total load is deconstructed by fully utilizing the load data of the historical transformer substation through a data-driven linear clustering forecasting method, and the internal composition and the change rule of the urban load are deeply researched, so that the accuracy of urban medium-term and long-term load forecasting is improved.
The linear clustering prediction method of the invention carries out linear clustering analysis prediction on each substation load subsequence forming the urban total load, deconstructs the urban total load by using the load data of the historical substations, divides the city into a plurality of regions, analyzes to obtain the saturated load density of each functional land block, predicts the load of each region, predicts the total amount of the urban load, and simultaneously can acquire the spatial distribution of the regional load, thereby providing a basis for the optimal sequencing of the urban distribution network planning project.
The linear clustering prediction method provided by the invention searches and combines data sequence pairs with linear complementary characteristics by traversing all data sequences, thereby improving the overall linearity of data; on the basis, performing optimal ARIMA model modeling and prediction on each sequence subjected to linear clustering preprocessing, and summing prediction results to obtain a final prediction value;
according to the method, on one hand, the original data sequence can be smoothed, and the data dimension is reduced, so that the modeling difficulty is reduced; on the other hand, the information contained in a large number of data sequences is fully utilized, the final prediction result is obtained through respective prediction and summation, the random error of prediction is effectively reduced, and therefore the prediction precision is improved. When urban power load prediction is carried out, linear clustering analysis prediction is carried out on each substation load subsequence forming urban total load, and the internal composition of the urban load can be deeply analyzed while the final prediction precision is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
Fig. 1 is a flowchart of a method for predicting an urban saturation load according to embodiment 1 of the present invention;
fig. 2 is a flowchart of the linear clustering process and ARIMA method provided in embodiment 1 of the present invention.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, and it should be understood that the terms "comprises" and "comprising", and any variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example 1
The power system load prediction analyzes the change rule of the power load and the influence of relevant factors on the power load by researching historical data, so that the prediction of future loads is realized; the medium-and-long-term load prediction is the basis for realizing the planning work of the power system, however, the power demand acceleration starts to slow down when the current part of large cities enter the later stage of the urbanization stage, and the industrial structure enters the adjustment period, so that the medium-and-long-term load does not increase rapidly in a single-side rising manner any more, but shows a relatively obvious fluctuation trend, and the difficulty of the medium-and-long-term load prediction is increased.
Therefore, the urban saturated load prediction method is provided, as shown in fig. 1, load data of a historical substation is fully utilized to deconstruct urban total loads through a linear clustering DLC prediction method based on data driving, the internal composition and the change rule of the urban loads are deeply researched, and therefore the accuracy of urban medium and long-term load prediction is improved. The method specifically comprises the following steps:
s1: acquiring historical load data of substations in each area after urban partitioning and meteorological data of corresponding time periods to construct a training sample set;
s2: performing smoothness processing on the training sample set by adopting a linear clustering algorithm, and training a time sequence-based saturated load prediction model by using the processed training sample set;
s3: and predicting meteorological data of the transformer substation in the region to be tested in the time period to be tested by adopting the trained saturated load prediction model, and obtaining a total urban saturated load prediction result according to the transformer substation saturated load prediction result of each region in the same time period.
In step S1, the city may be divided into a plurality of areas according to functions, such as residential areas, administrative office areas, industrial areas, etc., and since each area has different functions, the load data change rules are different due to different weather changes in the same time period, so that the load is predicted independently for each area, and the total load of the city can be predicted and the spatial distribution of the area load can be clarified.
In the present embodiment, the meteorological data includes, but is not limited to, short wave radiation, long wave radiation, air temperature, humidity, wind direction, wind speed, air pressure, cloud cover, precipitation rate, etc.; and a training sample set is constructed by the historical load data of each regional substation and the meteorological data of the corresponding time period.
In the step S2, a linear clustering algorithm is adopted to perform smoothness processing on the training sample set with linear smoothness as a standard, so that the load sets of a plurality of substations or regions are smoother, and the accuracy of the model is further improved; in the embodiment, subsequences in the training sample set are clustered, and the sum of the subsequences in each category has better linear characteristic and more obvious correlation than the sum of all subsequences in the data set;
let ytT is the time series of the system power load, ytIs formed by yk.tN, N being the number of subsequences, T being the number of time samples, typically in units of years;
yk.tthe load sequence of a transformer substation can be realized, and the load sequence of a region can also be realized, namely:
linear clustering is actually an optimization problem, with the goal of finding the best cluster that provides the best global linearity, expressed as:
wherein S isi,tIs an existing cluster i, i ═ 1, 2., M ≦ N;
is the corresponding linear fit sequence; f. ofRMSIs a root mean square calculation:
x is an n-dimensional vector, x ═ x1,x2,...xn)。
In this embodiment, as shown in fig. 2, the performing smoothness processing on the training sample set by using a linear clustering algorithm specifically includes:
(1) establishing a linear fitting model of each data sequence of the training sample set, and calculating fitting root mean square errors one by one; the method specifically comprises the following steps:
for each subsequence yk.tEstablishing least square linear fitting model, calculating the root mean square value of corresponding linear fitting residual error, and enabling ukN as a measure of the linearity of each original subsequence.
(2) Selecting a data sequence with the maximum fitting root mean square error, namely a sequence with the strongest fluctuation; the method specifically comprises the following steps:
finding the maximum RMS value u in step (2)kmaxThe subsequence of (1), denoted as ykmax.tI.e. the most fluctuating subsequence, ykmax.tIs the main optimization objective in the iteration.
(3) Summing the maximum root-mean-square error sequence and the rest sequences one by one, establishing a new linear fitting model for each sum sequence, and recalculating the fitting root-mean-square error of each sum sequence; the method specifically comprises the following steps:
for summing up sequence Yj,t＝ykmax.t+yj.tJ 1,2, N, j ≠ k, a new linear fitting model is established, and the correspondence is calculatedThe root mean square value of the fitting residual of (1), denoted as uj(ii) a The purpose is to determine whether there are other subsequences that can be summed to ykmax.tTo improve the linear fit.
(4) If the root mean square error of the sum sequence is smaller than the two addend sequences, selecting the sum sequence with the minimum fitting root mean square error, replacing the two addend sequences, adding the sum sequence into the data set, and continuing the step (1); otherwise, stopping iteration, and carrying out optimal ARIMA model modeling and prediction; the method specifically comprises the following steps:
finding u in step (3)jIs the minimum value of (d), is denoted as ujmin；
If ujmin<ukmaxPresence of subsequence yjmin.tCan be summed to ykmax.tTo improve the linear fit, yjmin.tAnd ykmax.tInstead of it and yjmin.tReturning to the step (1);
if ujmin≥ukmaxAnd then, iteration is stopped, the subsequences cannot become smoother through summation, and after linear clustering pretreatment, the smoothness of the subsequences is improved, and the number of samples is reduced.
In step S2, the time series-based saturated load prediction model is a time series model ARIMA-based saturated load prediction model; ARIMA performs well in characterizing and predicting time series, and is used to predict the aggregate load for each cluster and to analyze the error of load prediction.
The ARIMA model is:
wherein epsilontIs white noise;
and θ is a coefficient.
The ARIMA model consists of two parts: an auto-regressive module (AR) and a moving average Module (MA), the AR part describing the recorded characteristics of the past system state, and the MA part reflecting the influence of noise in the current system state;
the auto regression section (AR) is:
the moving average part (MA) is:
yt＝εt-θ1εt-1-θ2εt-2-...-θqεt-q
p and q are the corresponding orders of the two parts, and since the ARIMA model is only suitable for the lexicographic order, the partitioning preprocessing is required if the sequence is not lexicographic, wherein d represents the difference order.
Establishing an ARIMA optimization model as follows: s i,t1, 2.. said, M, predicting their respective future values, and summing them to obtain a final total load prediction; the method specifically comprises the following steps:
(1) removing trend items in each data sequence of the sample to be detected through difference, and converting the trend items into static sequences; the method specifically comprises the following steps:
for the preprocessed sequence Si,tA unit root test is performed to check if it is lexicographically ordered and any non-lexicographically ordered sequences will be differentially converted to lexicographically ordered sequences.
(2) Constructing ARIMA models under different p values and q values for each static sequence, and specifically:
establishing a plurality of ARIMA (p, d, q) models for each lexicographic sequence, combining different p and q parameters; due to sequence length constraints, p and q are constrained to a relatively low order, avoiding overfitting: p is 0,1, 2; q is 0, 1.
(3) Selecting an optimal ARIMA model according to the information content AIC criterion of the Chichi pool, which specifically comprises the following steps:
finding out the optimal one for each lexicographic sequence by utilizing Akaike information criterion in all ARIMA models established in the step (2);
AIC＝2n+Tln(fRSS/T)
where n is the number of model parameters, T is the sequence length, fRSSIs the sum of the squares of the residuals, reflecting the model accuracy;the method is a standard for measuring the modeling effect, not only considering the fitting precision, but also considering the complexity of the established model;
models with smaller AIC values are more optimal, so the optimal ARIMA model is chosen:
wherein the content of the first and second substances,
is Si,tARIMA fit values of (a).
(4) Based on the optimal ARIMA model, predicting each subsequence respectively, specifically:
predicting each preprocessing sequence S based on the corresponding optimal ARIMA model selected in step (3)i,tThe predicted result is recorded as S i,t+τ1, 2.. gtat, where Δ T is a prediction time period, Δ T cannot be too large due to limitations of the ARIMA model.
(5) Adding the prediction results of all the subsequences to obtain the prediction result of the whole system, which specifically comprises the following steps:
summing all ARIMA predictors, i.e.:
in the embodiment, the error analysis modeling error and the random error of the saturated load prediction model are divided into two parts: the modeling error refers to the deviation between the established model and the real rule, and can be quantified by the error between the fitting value and the real value of the sample; in general, the higher the load level, the smoother the load sequence and the smaller the modeling error. The random error is the degree of deviation of the true value of the load from the original change rule due to various uncertain factors, can be regarded as white Gaussian noise, and is quantified by subtracting the modeling error from the total prediction error.
When urban load is predicted, if modeling prediction is directly carried out on the basis of a total load curve, the modeling error is small due to high load level, but random error is difficult to control; if modeling prediction is carried out respectively directly on the basis of the substation subsequences, and finally the urban load prediction results are obtained through summation, the random error can tend to be 0 in average value; however, because each sequence has a low load level, modeling is difficult, and modeling errors are increased, in this embodiment, a linear clustering prediction method is used to perform linear smoothing processing on the substation load subsequences, and on this basis, modeling prediction is performed respectively and final results are obtained by summing up, so that random errors can be controlled at a low level while the modeling errors are reduced, and the overall prediction accuracy is improved.
Example 2
The embodiment provides a city saturation load prediction system, including:
the data acquisition module is used for acquiring historical load data of the transformer substation in each area after the urban partition and meteorological data of a corresponding time period to construct a training sample set;
the training module is used for performing smoothness processing on the training sample set by adopting a linear clustering algorithm and training a time sequence-based saturation load prediction model by using the processed training sample set;
and the prediction module is used for predicting the meteorological data of the transformer substation in the region to be tested in the time period to be tested by adopting the trained saturated load prediction model, and obtaining the prediction result of the total saturated load of the city according to the prediction result of the saturated load of the transformer substation in each region in the same time period.
It should be noted that the above modules correspond to steps S1 to S3 in embodiment 1, and the above modules are the same as the examples and application scenarios realized by the corresponding steps, but are not limited to the disclosure in embodiment 1. It should be noted that the modules described above as part of a system may be implemented in a computer system such as a set of computer-executable instructions.
In further embodiments, there is also provided:
an electronic device comprising a memory and a processor and computer instructions stored on the memory and executed on the processor, the computer instructions when executed by the processor performing the method of embodiment 1. For brevity, no further description is provided herein.
It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general purpose processors, digital signal processors DSP, application specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and may provide instructions and data to the processor, and a portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.
A computer readable storage medium storing computer instructions which, when executed by a processor, perform the method described in embodiment 1.
The method in embodiment 1 may be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, among other storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor. To avoid repetition, it is not described in detail here.
Those of ordinary skill in the art will appreciate that the various illustrative elements, i.e., 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 application.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A method for predicting urban saturation load is characterized by comprising the following steps:
acquiring historical load data of substations in each area after urban partitioning and meteorological data of corresponding time periods to construct a training sample set;
performing smoothness processing on the training sample set by adopting a linear clustering algorithm, and training a time sequence-based saturated load prediction model by using the processed training sample set;
and predicting meteorological data of the transformer substation in the region to be tested in the time period to be tested by adopting the trained saturated load prediction model, and obtaining a total urban saturated load prediction result according to the transformer substation saturated load prediction result of each region in the same time period.
2. The method of claim 1, wherein the performing smoothness processing on the training sample set by using the linear clustering algorithm comprises:
constructing a linear fitting model for each data sequence of the training sample set, and calculating a fitting root mean square error;
summing the data sequence with the maximum fitting root mean square error and other data sequences one by one, constructing a new linear fitting model for each sum sequence, and calculating the fitting root mean square error of each sum sequence;
judging whether the fitting root mean square error of the sum sequence is smaller than the two addend sequences, if so, replacing the two addend sequences with the minimum fitting root mean square error sum sequence, and continuing iteration until the fitting root mean square error of the sum sequence is smaller than the two addend sequences; if not, the iteration stops.
3. The urban saturated load prediction method according to claim 1, wherein the prediction process of the time-series-based saturated load prediction model comprises:
converting the sample data sequence to be tested into a static sequence through difference;
constructing ARIMA models with different orders for the static sequence;
and respectively predicting each data sequence of the sample to be tested based on the ARIMA model, and adding the prediction results of all the data sequences to obtain the prediction result of the sample to be tested.
4. The method of claim 3, wherein the transforming the sample data sequence to be tested into a static sequence by difference comprises:
and performing unit root test on the sample data sequence to be tested, judging whether the sample data sequence is in a lexicographic order, and converting the data sequence which is not in the lexicographic order into the lexicographic order through difference.
5. The method as claimed in claim 3, wherein the ARIMA model is selected according to the red pool information quantity.
6. The urban saturated load prediction method according to claim 1, wherein the time series-based saturated load prediction model comprises an auto-regression module and a moving average module, wherein the auto-regression module records historical characteristics, and the moving average module reflects the influence of current noise.
7. The method of claim 1, wherein the meteorological data includes but is not limited to short wave radiation, long wave radiation, air temperature, humidity, wind direction, wind speed, air pressure, cloud cover, precipitation rate.
8. An urban saturation load prediction system, comprising:
the data acquisition module is used for acquiring historical load data of the transformer substation in each area after the urban partition and meteorological data of a corresponding time period to construct a training sample set;
the training module is used for performing smoothness processing on the training sample set by adopting a linear clustering algorithm and training a time sequence-based saturation load prediction model by using the processed training sample set;
and the prediction module is used for predicting the meteorological data of the transformer substation in the region to be tested in the time period to be tested by adopting the trained saturated load prediction model, and obtaining the prediction result of the total saturated load of the city according to the prediction result of the saturated load of the transformer substation in each region in the same time period.
9. An electronic device comprising a memory and a processor and computer instructions stored on the memory and executed on the processor, the computer instructions when executed by the processor performing the method of any of claims 1-7.
10. A computer-readable storage medium storing computer instructions which, when executed by a processor, perform the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011639715.3A CN112734096A (en) | 2020-12-31 | 2020-12-31 | Urban saturation load prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011639715.3A CN112734096A (en) | 2020-12-31 | 2020-12-31 | Urban saturation load prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112734096A true CN112734096A (en) | 2021-04-30 |
Family
ID=75609015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011639715.3A Pending CN112734096A (en) | 2020-12-31 | 2020-12-31 | Urban saturation load prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112734096A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117955095A (en) * | 2024-01-11 | 2024-04-30 | 国网湖北省电力有限公司信息通信公司 | Power load prediction method, device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704966A (en) * | 2017-10-17 | 2018-02-16 | 华南理工大学 | A kind of Energy Load forecasting system and method based on weather big data |
| CN107895211A (en) * | 2017-11-27 | 2018-04-10 | 上海积成能源科技有限公司 | A kind of long-medium term power load forecasting method and system based on big data |
| CN109508835A (en) * | 2019-01-01 | 2019-03-22 | 中南大学 | A kind of wisdom Power Network Short-Term Electric Load Forecasting method of integrated environment feedback |
-
2020
- 2020-12-31 CN CN202011639715.3A patent/CN112734096A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704966A (en) * | 2017-10-17 | 2018-02-16 | 华南理工大学 | A kind of Energy Load forecasting system and method based on weather big data |
| CN107895211A (en) * | 2017-11-27 | 2018-04-10 | 上海积成能源科技有限公司 | A kind of long-medium term power load forecasting method and system based on big data |
| CN109508835A (en) * | 2019-01-01 | 2019-03-22 | 中南大学 | A kind of wisdom Power Network Short-Term Electric Load Forecasting method of integrated environment feedback |
Non-Patent Citations (2)
| Title |
|---|
| 李震 等: "基于数据驱动的线性聚类ARIMA 长期电力负荷预测", 《科学技术与工程》 * |
| 程潜善 等: "一种应用大数据技术的中长期负荷预测方法", 《武汉大学学报（工学版）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117955095A (en) * | 2024-01-11 | 2024-04-30 | 国网湖北省电力有限公司信息通信公司 | Power load prediction method, device, electronic equipment and storage medium |
| CN117955095B (en) * | 2024-01-11 | 2024-06-25 | 国网湖北省电力有限公司信息通信公司 | Power load prediction method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | LASSO and LSTM integrated temporal model for short-term solar intensity forecasting | |
| CN106951976B (en) | Bus passenger flow prediction method based on mode classification | |
| CN109919353B (en) | Distributed photovoltaic prediction method of ARIMA model based on spatial correlation | |
| CN110570651A (en) | Road network traffic situation prediction method and system based on deep learning | |
| CN112465243B (en) | Air quality forecasting method and system | |
| CN110648014B (en) | Regional wind power prediction method and system based on space-time quantile regression | |
| CN109242265B (en) | Urban water demand combined prediction method based on least square sum of errors | |
| CN111242395B (en) | Method and device for constructing prediction model for OD (origin-destination) data | |
| CN112348290B (en) | River water quality prediction method, river water quality prediction device, storage medium and storage device | |
| CN113554466A (en) | Short-term power consumption prediction model construction method, prediction method and device | |
| CN109637196A (en) | En-route sector traffic probability density prediction technique | |
| CN112419711B (en) | Closed parking lot parking demand prediction method based on improved GMDH algorithm | |
| CN115983448B (en) | Multi-energy load prediction method based on space-time diagram neural network | |
| CN109686402A (en) | Based on key protein matter recognition methods in dynamic weighting interactive network | |
| CN117977568A (en) | Power load prediction method based on nested LSTM and quantile calculation | |
| CN115099511A (en) | Photovoltaic power probability estimation method and system based on optimized copula | |
| CN112734096A (en) | Urban saturation load prediction method and system | |
| CN107316096A (en) | A kind of track traffic one-ticket pass passenger amount of entering the station Forecasting Methodology | |
| CN113537596A (en) | Short-time passenger flow prediction method for new line station of urban rail transit | |
| Pham et al. | Research on applying machine learning models to predict the electricity generation capacity of rooftop solar energy systems on buildings | |
| Nizar et al. | Forecasting of temperature by using LSTM and bidirectional LSTM approach: case study in Semarang, Indonesia | |
| CN116913098A (en) | Short-time traffic flow prediction method integrating air quality and vehicle flow data | |
| CN116992305A (en) | Weather forecast method and system based on big data | |
| CN115829418A (en) | Power consumer load characteristic portrait construction method and system suitable for load management | |
| Li et al. | The research of urban rail transit sectional passenger flow prediction method |
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 |