CN113919599A - Medium-and-long-term load prediction method - Google Patents
Medium-and-long-term load prediction method Download PDFInfo
- Publication number
- CN113919599A CN113919599A CN202111419900.6A CN202111419900A CN113919599A CN 113919599 A CN113919599 A CN 113919599A CN 202111419900 A CN202111419900 A CN 202111419900A CN 113919599 A CN113919599 A CN 113919599A
- Authority
- CN
- China
- Prior art keywords
- medium
- long term
- economic factors
- load
- model
- 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 41
- 230000007774 longterm Effects 0.000 claims abstract description 44
- 230000001364 causal effect Effects 0.000 claims abstract description 37
- 238000012360 testing method Methods 0.000 claims abstract description 21
- 238000010998 test method Methods 0.000 claims abstract description 9
- 230000006870 function Effects 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 7
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 6
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000002596 correlated effect Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 238000013112 stability test Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000005259 measurement Methods 0.000 abstract description 3
- 230000002195 synergetic effect Effects 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000013277 forecasting method Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation 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)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a medium and long term load prediction method, which is based on the Glan's causal test and the LSTM, and utilizes the synergetic and integral test in the measurement economics to find out the economic factors having a long term equilibrium relationship with the electric quantity; then determining economic factors which are beneficial to electric quantity prediction by using a Glange causal test method so as to reduce the number of input variables of the prediction model; and finally, inputting the economic factor data into an LSTM model for load prediction. According to the medium and long term load prediction method, a method of combining the Grave causal relationship test and the LSTM time sequence prediction model is introduced into a multivariable system, a medium and long term load prediction model which is not easy to overfit and high in expandability is constructed, the medium and long term load is predicted by using the model, and the predicted medium and long term load has high accuracy.
Description
Technical Field
The application relates to the technical field of intelligent power prediction, in particular to a medium-long term load prediction method.
Background
The medium-long term power load prediction is to determine the load value at a certain future time under the condition of meeting certain precision by using a reliable method and means according to historical load values and fully considering system characteristics, natural and social conditions. At present, the medium and long term load prediction methods include regression analysis, time series analysis, gray prediction and the like based on mathematical statistics, and in addition, some intelligent prediction methods such as artificial neural network, support vector machine, combined prediction and the like.
Since the accuracy of medium and long term load prediction is influenced by various random factors such as politics, economy, population, climate and the like, the single prediction method is difficult to improve the prediction accuracy. For example, when the time-series exponential smoothing method is applied to the medium-and long-term load prediction, it is assumed that the past load change law extends into the future, and the influence of the change of random variables such as economy and climate on the load change cannot be considered, and therefore, the prediction result has certain fluctuation. For another example, when a neural network model and a medium-and-long-term load prediction model are applied, the neural network model has the disadvantages of ambiguous meaning, sensitivity to initial values, poor adaptability to emergency events, and often needs a large amount of data when input variables are excessive.
Therefore, a medium-and-long-term load prediction method with higher prediction accuracy is urgently needed.
Disclosure of Invention
The application provides a medium-and-long-term load prediction method, which aims to solve the problem that the prediction accuracy of the existing load prediction method needs to be improved.
The application provides a medium and long term load prediction method, which comprises the following steps:
determining economic factors having a long-term balance relation with the power load by utilizing the co-integration test;
determining a causal relationship between the power load and economic relevant factors by adopting a Glange causal test method, and screening out economic factors which are strongly relevant to the power load, namely relevant economic factors;
and (4) carrying out normalization processing on the related economic factor data and the historical load data, and inputting the normalized data into an LSTM model to predict the medium and long term load value.
In some embodiments, the economic factors include regional production total, fixed investment total, social consumer retail total, residential consumption price index, industry growth, value growth rate, labor, and population.
In some embodiments, the economic factors that have a long-term balance relationship with the electrical load are determined using a co-integration test, including:
and carrying out stability test and coordination relation test on the time sequence of the economic factors to determine the economic factors which have long-term equilibrium relation with the power load.
In some embodiments, a granger causal test method is used to determine a causal relationship between the power load and the economic-related factors, and screen out economic factors that are strongly correlated to the power load, i.e., the correlated economic factors, including,
performing Glangel causal test on the time sequence Xi of the economic factors and the historical load data Yi, and establishing the following two vector autoregressive models:
wherein alpha isj、ajAnd bjIs the coefficient of the model, m is the order of the model, t is the time, j is a constant, εYAnd εY|XIs the residual error of the model;
and (3) judging whether the X → Y has a Glanberg causal relationship or not by utilizing the prediction result of the vector autoregressive model and comparing the variance of the residual error of the vector autoregressive model, and screening out economic factors which have strong correlation with the power load, namely related economic factors.
In some embodiments, the prediction result of the vector autoregressive model is used, and the variance of the residuals of the vector autoregressive model is compared to determine whether X → Y has a Glanberg causal relationship, including,
the calculation formula of the glange causal index GCI is:
if var (ε) is satisfiedY|X)<var(εY) I.e. GCIX→Y>0, then the statistical gram positive causal relationship of Y → X is determined.
In some embodiments, the LSTM model includes:
the hidden layer of LSTM contains input gates itAnd an output gate otForgetting door ftAnd cell status ciWherein, in the step (A),
ft=σ(Wfxt+Ufht-1+bf)
it=σ(Wixt+Uiht-1+bi)
ot=σ(Woxt+Uoht-1+bo)
in the formula: w is an input weight matrix of the hiding unit; u is an output weight matrix; b is a bias vector; subscripts f, i and o represent a forgetting gate, an input gate and an output gate;representing a point-by-point product operation; σ (-) is the activation function;
there are 2 activation functions as follows:
an output layer: for obtaining the final output result.
In some embodiments, the method for predicting the medium-long term load further comprises evaluating the prediction result of the medium-long term load value, including,
taking the root mean square error RMSE and the average absolute percentage error MAPE as evaluation indexes, and the calculation formula is specifically as follows:
wherein, yiWhich represents the actual value of the measured value,representing the predicted value, n representing the number of predicted samples,
the closer the RMSE and MAPE values are to zero, the closer the predicted result of the medium and long term load value is judged to be the real load value.
The application provides a medium-long term load prediction method, which is based on the Glange causal test and the LSTM, and utilizes the synergetic and integral test in the measurement economics to find out the economic factors having a long-term equilibrium relationship with the electric quantity; then determining economic factors which are beneficial to electric quantity prediction by using a Glange causal test method so as to reduce the number of input variables of the prediction model; and finally, inputting the economic factor data into an LSTM model for load prediction. In the medium and long term load prediction method, a method of combining the Grave causal relationship test and the LSTM time sequence prediction model is introduced into a multivariable system, a medium and long term load prediction model which is not easy to overfit and high in expandability is constructed, the medium and long term load is predicted by utilizing the model, and the predicted medium and long term load has high accuracy.
Drawings
In order to more clearly illustrate the technical solutions in the present disclosure, the drawings needed to be used in some embodiments of the present disclosure will be briefly described below, and it is apparent that the drawings in the following description are only drawings of some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art according to the drawings. Furthermore, the drawings in the following description may be regarded as schematic diagrams, and do not limit the actual size of products, the actual flow of methods, the actual timing of signals, and the like, involved in the embodiments of the present disclosure.
FIG. 1 is a flow chart of a long term load forecasting method of the present application;
FIG. 2 is a diagram of the basic structure of the LSTM of the present application over a network.
Detailed Description
In order to improve the accuracy of medium and long term load prediction, the application provides a medium and long term load prediction method. Fig. 1 is a flowchart of a medium-and-long-term load prediction method in the present application, and with reference to fig. 1, the medium-and-long-term load prediction method provided in the present application specifically includes the following steps.
And S100, determining economic factors having a long-term balance relation with the power load by utilizing a co-integration test.
In this application, economic factor (x)1、x2……xn) The method comprises a plurality of economic parameters, wherein the economic parameters comprise regional production total value, fixed investment total amount, social consumer product retail total amount, resident consumption price index, industry added value growth speed, labor force and population. Of course, those skilled in the art can adjust the parameters specifically included in the economic factors according to actual needs, such as increasing the economic development status, economic structure, residential income, consumer structure, etc., which all fall within the protection scope of the present application.
In this application, utilize the cooperation to examine and confirm that there is the economic factor of long-term equilibrium relation with power load, specifically include: and carrying out stability test and coordination relation test on the time sequence of the economic factors to determine the economic factors which have long-term equilibrium relation with the power load. The method mainly comprises the following steps that a coordinated relation test mainly means that some non-stationary time series variables can guarantee a stationary relation under linear combination in a certain form, and correlation analysis is mainly performed on the relation between each economic factor of a region and the power consumption of the whole society of the region in the example. It should be noted that both the stationarity check and the coordination relationship check belong to common checking methods in the metrology economics, and detailed description of specific implementation processes will not be provided herein.
And S200, determining the causal relationship between the power load and the economic relevant factors by adopting a Glange causal test method, and screening out the economic factors which are strongly relevant to the power load, namely relevant economic factors.
The granger causal test is that the change of a variable value can generate numerical influence on another variable value, and in the application, the granger causal relationship test among the variables is needed to verify and analyze the authenticity and long-term dynamic effect of the economic factors on the medium-long term load prediction.
The method comprises the following steps of determining a causal relationship between the power load and economic relevant factors by adopting a Glange causal test method, screening out economic factors strongly relevant to the power load, namely determining economic factors which are helpful for electric quantity prediction by using the relevant economic factors, and reducing the number of input variables of a prediction model.
Performing Glangel causal test on the time sequence Xi of the economic factors and the historical load data Yi, and establishing the following two vector autoregressive models:
wherein alpha isj、ajAnd bjIs the coefficient of the model, m is the order of the model, t is the time, j is a constant, εYAnd εY|XIs the residual error of the model;
and (3) judging whether the X → Y has a Glanberg causal relationship or not by utilizing the prediction result of the vector autoregressive model and comparing the variance of the residual error of the vector autoregressive model, and screening out economic factors which have strong correlation with the power load, namely related economic factors.
In this example, the causal relationship is obtained by determining whether there is a Glanberg causal relationship between X → YEconomic factor (x)1、x2……xp) Then, an economic factor having a strong correlation with the power load, i.e., a correlated economic factor (x), is screened out1、x2……xq) Here, n, p, and q are positive integers, and n.gtoreq.p.gtoreq. In the present application, the basic idea of the granger causal analysis is that if the prediction of the future medium and long term load Y using the time series Xi of the economic factors and the historical load information Yi is superior to the prediction of the future medium and long term load Y using only the historical load information Yi, then the time series X is the granger cause of the time series Y.
In the present application, the method for determining whether X → Y has a Glanberg causal relationship by using the prediction result of the vector autoregressive model and comparing the variance of the residuals of the vector autoregressive model specifically includes,
the calculation formula of the glange causal index GCI is:
if var (ε) is satisfiedY|X)<var(εY) I.e. GCIX→Y>0, then the statistical gram positive causal relationship of Y → X is determined.
Step S300, the relevant economic factor data and the historical load data are normalized and then input into an LSTM long-short term memory model (LSTM) to predict the medium-long term load value.
Fig. 2 is a basic structure diagram of the LSTM of the present application via a network, and as shown in fig. 2, the LSTM model of the present application includes:
the hidden layer of LSTM contains input gates itAnd an output gate otForgetting door ftAnd cell status ciWherein, in the step (A),
ft=σ(Wfxt+Ufht-1+bf)
it=σ(Wixt+Uiht-1+bi)
ot=σ(Woxt+Uoht-1+bo)
in the formula: w is an input weight matrix of the hiding unit; u is an output weight matrix; b is a bias vector; subscripts f, i and o represent a forgetting gate, an input gate and an output gate;representing a point-by-point product operation; σ (-) is the activation function;
in order to enhance the nonlinear function of the network, the following 2 activation functions are provided:
an output layer: for obtaining the final output result, i.e. the long-term load predicted value.
In this application, the method for predicting the medium-and-long term load further includes step S400 of evaluating the prediction result of the medium-and-long term load value, including taking the root mean square error RMSE and the average absolute percentage error MAPE as evaluation indexes, and the calculation formula specifically includes:
wherein, yiWhich represents the actual value of the measured value,representing the predicted value, and n is the number of predicted samples.
The closer the RMSE and MAPE values are to zero, the closer the prediction result of the medium and long-term load value is to the real load value, and the higher the accuracy of the model is.
The medium and long-term load forecasting method utilizes the synergistic test in the measurement economics to find out the economic factors which have long-term balance relation with the medium and long-term load, utilizes the Glange causal test method to determine the economic factors which are beneficial to load forecasting, reduces the number of input variables of a forecasting model, reduces the complexity of the model, utilizes the LSTM model to forecast the medium and long-term load, can process the long-term sequence data forecasting problem, and has the advantages of high accuracy, difficulty in overfitting, high expandability and the like. The medium-and-long-term load prediction method improves the prediction accuracy of medium-and-long-term loads, and enhances the robustness and generalization capability of the model, so that the method has a wider application prospect.
The foregoing is illustrative of the best mode of the invention and details not described herein are within the common general knowledge of a person of ordinary skill in the art. The scope of the present invention is defined by the appended claims, and any equivalent modifications based on the technical teaching of the present invention are also within the scope of the present invention.
Claims (7)
1. A method for predicting medium and long term load is characterized by comprising the following steps:
determining economic factors having a long-term balance relation with the power load by utilizing the co-integration test;
determining a causal relationship between the power load and economic relevant factors by adopting a Glange causal test method, and screening out economic factors which are strongly relevant to the power load, namely relevant economic factors;
and (4) carrying out normalization processing on the related economic factor data and the historical load data, and inputting the normalized data into an LSTM model to predict the medium and long term load value.
2. The method of claim 1, wherein the economic factors include regional production total, capital investment total, social consumer retail total, residential price index, industry growth, value growth rate, labor force, and population.
3. The method for predicting the medium and long term load according to claim 1, wherein the step of determining the economic factors having the long term balance relationship with the power load by using the coordination test comprises the following steps:
and carrying out stability test and coordination relation test on the time sequence of the economic factors to determine the economic factors which have long-term equilibrium relation with the power load.
4. The method of claim 1, wherein a gram cause and effect test method is used to determine the cause and effect relationship between the power load and the economic factors, and to screen out the economic factors strongly correlated to the power load, i.e. the correlated economic factors, including,
performing Glangel causal test on the time sequence Xi of the economic factors and the historical load data Yi, and establishing the following two vector autoregressive models:
wherein alpha isj、ajAnd bjAre the coefficients of the model and are,m is the order of the model, t is the time, j is a constant, εYAnd εY|XIs the residual error of the model;
and (3) judging whether the X → Y has a Glanberg causal relationship or not by utilizing the prediction result of the vector autoregressive model and comparing the variance of the residual error of the vector autoregressive model, and screening out economic factors which have strong correlation with the power load, namely related economic factors.
5. The method for predicting the medium and long term load according to claim 4, wherein the method for determining whether the Glanberg causal relationship exists between X → Y by comparing the variance of the residuals of the vector autoregressive model and the prediction results of the vector autoregressive model comprises,
the calculation formula of the glange causal index GCI is:
if var (ε) is satisfiedY|X)<var(εY) I.e. GCIX→YIf the ratio is more than 0, the gram-positive causal relationship of Y → X in the statistical sense is judged.
6. The method of predicting long and medium term load according to claim 1, wherein the LSTM model comprises:
the hidden layer of LSTM contains input gates itAnd an output gate otForgetting door ftAnd cell status ciWherein, in the step (A),
ft=σ(Wfxt+Ufht-1+bf)
it=σ(Wixt+Uiht-1+bi)
ot=σ(Woxt+Uoht-1+bo)
in the formula: w is an input weight matrix of the hiding unit; u is an output weight matrix; b is a bias vector; subscripts f, i and o represent a forgetting gate, an input gate and an output gate;representing a point-by-point product operation; σ (-) is the activation function;
there are 2 activation functions as follows:
an output layer: for obtaining the final output result.
7. The method of predicting a medium-and-long term load according to claim 1, further comprising evaluating a prediction result of the medium-and-long term load value, including,
taking the root mean square error RMSE and the average absolute percentage error MAPE as evaluation indexes, and the calculation formula is specifically as follows:
wherein, yiWhich represents the actual value of the measured value,representing the predicted value, n representing the number of predicted samples,
the closer the RMSE and MAPE values are to zero, the closer the predicted result of the medium and long term load value is judged to be the real load value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111419900.6A CN113919599A (en) | 2021-11-26 | 2021-11-26 | Medium-and-long-term load prediction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111419900.6A CN113919599A (en) | 2021-11-26 | 2021-11-26 | Medium-and-long-term load prediction method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113919599A true CN113919599A (en) | 2022-01-11 |
Family
ID=79248413
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111419900.6A Pending CN113919599A (en) | 2021-11-26 | 2021-11-26 | Medium-and-long-term load prediction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113919599A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114579407A (en) * | 2022-05-05 | 2022-06-03 | 北京航空航天大学 | Causal relationship inspection and micro-service index prediction alarm method |
-
2021
- 2021-11-26 CN CN202111419900.6A patent/CN113919599A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114579407A (en) * | 2022-05-05 | 2022-06-03 | 北京航空航天大学 | Causal relationship inspection and micro-service index prediction alarm method |
CN114579407B (en) * | 2022-05-05 | 2022-08-23 | 北京航空航天大学 | Causal relationship inspection and micro-service index prediction alarm method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110223196B (en) | Anti-electricity-stealing analysis method based on typical industry feature library and anti-electricity-stealing sample library | |
CN112735097A (en) | Regional landslide early warning method and system | |
CN113408869A (en) | Power distribution network construction target risk assessment method | |
CN113627735A (en) | Early warning method and system for safety risk of engineering construction project | |
CN113516845A (en) | Road monthly traffic volume prediction method based on SARIMA-NAR combined model | |
CN113506121A (en) | Analysis method and device for price influence factors | |
CN114595861A (en) | MSTL (modeling, transformation, simulation and maintenance) and LSTM (least Square TM) model-based medium-and-long-term power load prediction method | |
CN113919599A (en) | Medium-and-long-term load prediction method | |
CN114819395A (en) | Industry medium and long term load prediction method based on long and short term memory neural network and support vector regression combination model | |
CN115438897A (en) | Industrial process product quality prediction method based on BLSTM neural network | |
Rodrigues et al. | A system for analysis and prediction of electricity-load streams | |
CN113779879A (en) | Medium-and-long-term electricity utilization abnormity detection method based on LSTM-seq2seq-attention model | |
CN112101673A (en) | Power grid development trend prediction method and system based on hidden Markov model | |
CN114970745B (en) | Intelligent security and environment big data system of Internet of things | |
CN115689331A (en) | Power transmission and transformation project quantity rationality analysis method based on MLP | |
CN115510763A (en) | Air pollutant concentration prediction method and system based on data-driven exploration | |
CN113190535B (en) | Gas concentration prediction method based on improved grey-long and short-term memory neural network combined model | |
Khamis et al. | Time Series Forecasting of Malaysia Producer Price Index using ARIMA and Grey Models | |
Chattopadhyay et al. | Poverty index with time-varying consumption and income distributions | |
CN114048925A (en) | Power grid comprehensive operation early warning method and device and terminal equipment | |
Chatigny et al. | A variable-order regime switching model to identify significant patterns in financial markets | |
Fillah et al. | Prediction of Water Quality Index using Deep Learning in Mining Company | |
CN111353523A (en) | Method for classifying railway customers | |
Alamsyah et al. | Comparison of Predictions Foreign Currency Exchange Rates (USD, JPY, EUR, GBP, CHF, CAD) to Rupiah (IDR) using Artificial Neural Network and Long Short-Term Memory | |
Vika et al. | Forecasting the Albanian Time Series with Linear and Nonlinear Univariate Models |
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 |