CN108536652A - A kind of short-term vehicle usage amount prediction technique based on arma modeling - Google Patents
A kind of short-term vehicle usage amount prediction technique based on arma modeling Download PDFInfo
- Publication number
- CN108536652A CN108536652A CN201810213487.XA CN201810213487A CN108536652A CN 108536652 A CN108536652 A CN 108536652A CN 201810213487 A CN201810213487 A CN 201810213487A CN 108536652 A CN108536652 A CN 108536652A
- Authority
- CN
- China
- Prior art keywords
- data
- usage amount
- vehicle usage
- arma modeling
- vehicle
- 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
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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Operations Research (AREA)
- Economics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Traffic Control Systems (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Algebra (AREA)
Abstract
The short-term vehicle usage amount prediction technique based on arma modeling that the invention discloses a kind of, data cleansing is carried out to the vehicle usage amount data of history first, then attribute construction is carried out, then using week as cyclic indicator, it determines the exponent number of arma modeling, then the historical data of vehicle usage amount is used to be fitted arma modeling afterwards, obtain determining arma modeling, following several days vehicle usage amount prediction cases are obtained by the arma modeling, the dispensing for car operation quotient provides instruction.The present invention predicts that the arma modeling of vehicle usage amount is also that ensure that the validity of arma modeling in continuous adjusting and optimizing, has practical guiding value for the vehicle usage amount of short-term forecast;The present invention carrys out the reliability of evaluation model using the mean square error of model prediction result and history truthful data simultaneously, to the fitting quality of quantitative model, realizes that science adjusts model parameter.
Description
Technical field
The invention belongs to probability statistics and short-term forecast technical fields, and in particular to a kind of short-term vehicle based on arma modeling
Usage amount prediction technique.
Background technology
Vehicle usage amount is the important indicator evaluated user and use the vehicle frequency, can intuitively be reflected in region to vehicle
The degree of strength of demand, and for instructing Main Basiss source of the operator to vehicle injected volume in the region, therefore predict
Go out the vehicle usage amount of a certain region in a short time has actual guiding value to operator.
Autoregression slipping smoothness model (Autoregressive moving average model, ARMA), also referred to as
Box-Jenkins models, it is by autoregression model (Autoregressive model, AR) and moving average model(MA model)
(Moving average model, MA) is combined into, and is a kind of Time series analysis method of classics, has been widely used in
Economic analysis and market prediction field.
AR modelings are that current observation level can be depending on the observation level of its lag, it reflects economic variable
Current value and its past value relationship.AR models are past value only with it and random perturbation to the model set up.
Therefore, the visible formula of definition AR (p) (1) of p ranks AR models:
Wherein:IfThen it is known as centralization AR (p) models.
By formula (1) it is found that XtValue be preceding p phases Xt-1, Xt-2..., Xt-pMultiple linear regression, it is meant that it is current
Value is influenced by p past, sequence phase, error term εtWhat is represented is zero-mean white noise sequence.
MA models describe the deviation accumulation of autoregression part, it reflects economic variable current value and current and past
The relationship of error term.Therefore, the visible formula of definition MA (q) (2) of q ranks MA models:
Wherein:C is YtMean value, εtIt is current random disturbances error term i.e. zero-mean white noise sequence, θiIt is model
Parameter, εt-iIt is the disturbance term of preceding i phases, therefore current random disturbances are the random perturbation ε of preceding q phasest-1, εt-2..., εt-qIt is more
First linear function.
Arma modeling is exactly the combination of AR models and MA models, applies in general to the stationary selection that mean value is 0, then
Can derive ARMA (p, q) model containing p rank autoregression items and q rank rolling average items from formula (1) and formula (2) can
It is expressed as formula (3):
Arma modeling is established based on sequence based on certain level random fluctuation, for non-stationary time sequence
Row can become stationarity time series by way of one or many difference;Arma modeling shows a random time sequence
Row can be explained by the past of itself or lagged value and Disturbance.
Invention content
In view of above-mentioned, the short-term vehicle usage amount prediction technique based on arma modeling that the present invention provides a kind of is based on
The historical data of vehicle usage amount goes out the usage amount situation of the following several days vehicles by arma modeling short-term forecast, is operator
Vehicle injected volume provide advisory opinion.
A kind of short-term vehicle usage amount prediction technique based on arma modeling, includes the following steps:
(1) history of a large amount of vehicles in use of acquisition acquisition uses data, including vehicle GPS data, user's rental period, vehicle
The frequency and car operation data are rented, retains the history wherein within current time 1 year and be added to number using data
According in library;
(2) history in database is pre-processed using data, includes the processing to missing values and exceptional value, data
Cleaning, attribute construction and periodicity analysis;
(3) stationary test and white noise verification are carried out using data to the history for pre-processing rear vehicle;
(4) according to auto-correlation function image and partial autocorrelation function image determine arma modeling autoregression item exponent number p and
Rolling average item exponent number q, and then fitting is trained to arma modeling using data using by the vehicle history after examining;
(5) finally, the arma modeling obtained using training fitting completion carries out the following vehicle usage amount in a short time pre-
It surveys.
Further, the processing in the step (2) to missing values and exceptional value, including there are data for individual dates
The case where missing, needs that these dates corresponding vehicle usage amount is set to 0 to these dates progress data filling, while right
In there are data exceptions for individual dates the case where, these dates corresponding vehicle usage amount is equally set to 0.
Further, the data cleansing in the step (2) is filtered out history using data invalid in data,
Retain crucial car operation data;Attribute construction goes out vehicle rental amount and every monthly rent daily according to history using data statistics
Two new attributes of dosage;Periodicity analysis is i.e. using a week as the period, and analysis vehicle rental amount is with the presence or absence of periodically change
Change.
Further, the stationary test in the step (3) is i.e. single using ADF (Augment Dickey-Fuller)
Position root carries out stationary test, it is desirable that passes through when the corresponding probability value P > 0.05 of statistic and examines, otherwise the history of vehicle uses
Data are unstable, need to carry out ADF inspections again after first carrying out calculus of differences, until vehicle history using data be smoothly with
Until machine time series.
Further, the white noise verification in the step (3) is examined using Ljung-Box, it is desirable that statistic corresponds to
Probability value P < 0.05, it is determined that the history of vehicle using data be nonwhite noise data.
Further, the autoregression item exponent number p and rolling average item exponent number q that arma modeling is determined in the step (4), are adopted
It is that low order is gradually soundd out to high-order, i.e., arma modeling pair is calculated according to auto-correlation function image and partial autocorrelation function image
The AIC (Akaike information criterion, akaike information criterion) answered takes the minimum corresponding p and q conducts of AIC
The autoregression item exponent number p and rolling average item exponent number q of arma modeling, to establish the arma modeling for waiting for training fitting.
Further, the step (4) needs to calculate the prediction result of arma modeling during training arma modeling
It is sent out with vehicle usage amount with the mean square error between vehicle usage amount actual value if the mean square error is more than the threshold value of setting
The raw time is index, and the vehicle usage amount weight for keeping time of origin closer apart from current time is higher, and time of origin distance is current
Time it is remoter vehicle usage amount weight it is lower, by adjusting corresponding weight, input data is trained arma modeling again
Fitting;If the mean square error is less than the threshold value of setting, the training to arma modeling is completed.
The present invention can predict following several days vehicle usage amount situations using arma modeling, be car operation quotient's
It launches and instruction is provided.Predict that the arma modeling of vehicle usage amount is not cured model, over time, new
Historical data can be added in database, and historical data excessively remote can be removed, and every day is fitted the data of arma modeling
All it is therefore to predict that the arma modeling of vehicle usage amount is also that ensure that in continuous adjusting and optimizing by current newest
The validity of arma modeling has practical guiding value for the vehicle usage amount of short-term forecast.The present invention uses model prediction knot
The mean square error of fruit and history truthful data carrys out the reliability of evaluation model, to the fitting quality of quantitative model, realizes science
Adjust model parameter.Therefore, the present invention has following advantageous effects:
(1) the vehicle usage amount data that the present invention uses have apparent time series data feature, the data of vehicle usage amount
All it is newest data down to date, stale data, which will be removed, to be avoided affecting to prediction, therefore fitting ARMA out
Model is the vehicle prediction model of real-time update, can be instructed the injected volume of car operation quotient, and auxiliary resources rationally divides
With utilization.
(2) arma modeling of the present invention considers past or lagged value and the Disturbance of itself, to the stationarity time
Sequence has more good short-term forecast effect, and is easily achieved;Model order, mean square error are determined using AIC criterion simultaneously
Difference judges fitting effect, provides the foundation of judge for the accuracy and reliability of arma modeling.
Description of the drawings
Fig. 1 is the model treatment flow chart of present invention prediction vehicle usage amount.
Fig. 2 is modeling and the applicating flow chart of arma modeling of the present invention.
Specific implementation mode
In order to more specifically describe the present invention, below in conjunction with the accompanying drawings and specific implementation mode is to technical scheme of the present invention
It is described in detail.
Present invention is mainly applied to such as short-term forecasts at scenic spot, school, market vehicle usage amount in fixed area, pass through
Specific arma modeling is established to the data of nearest history vehicle usage amount, following several days vehicle usage amount situations are carried out
Short-term forecast.
The present invention predicts that the model treatment flow of vehicle usage amount is divided into 4 parts as shown in Figure 1, is that data are taken out respectively
It takes, Data Mining and pretreatment, modeling and application and result and feedback.
Step 1:Acquire the GPS data of vehicle in use, user rents the operation data of vehicle, user's rental period data,
Rent the essential informations such as the vehicle frequency.In data extraction process, using user rent vehicle operation data time as stroke
The same day newest data are added to set by point foundation, and the data of current time of adjusting the distance 1 year or more are rejected from set.
Step 2:The data acquired according to step 1 carry out missing values and outlier detection to it, this is because initial data
It is discrete data, there are the particular cases that individual number of days vehicle usage amounts are 0, and the corresponding date can be shown as in initial data
The phenomenon that shortage of data, therefore data filling is carried out firstly the need of the date lacked to these vehicle usage amounts, its vehicle is made
Dosage is set to 0;Meanwhile 0 also is set to by same day vehicle usage amount for the case where there are data exceptions;For vehicle usage amount number
It is filtered out according to middle invalid data, retains crucial car operation situation data.
Step 3:Next according to the rental period of initial data and rental amount respectively by daily and monthly carry out rental amount
Statistics constructs daily rental amount statistics and monthly rental amount counts the two new attributes, and discrete initial data is made to be converted into company
Continuous time series data.
Step 4:Periodicity analysis is carried out to time series data, according to attribute construction as a result, to judge that its overall trend becomes
Change, analysis vehicle rental amount whether there is cyclically-varying, be using week as cyclic indicator in the concrete realization.
Step 5:The flow chart of modeling with the application of data is as shown in Fig. 2, first carry out history vehicle usage amount data
Riding Quality Analysis, method are that the ADF unit root leveling styles used are examined, it is desirable that p>By examining when 0.05, if history vehicle
Usage amount data are unstable, need to carry out ADF inspections again after first carrying out calculus of differences, until vehicle usage amount data are steady
Random time sequence until.
Step 6:Judge whether vehicle usage amount data are white noise, are examined using Ljung-Box, it is desirable that p value is less than
0.05.Determine that vehicle usage amount time series data is nonwhite noise data.
Step 7:The auto-correlation function and nonautocorrelation function for calculating vehicle usage amount time series data judge its truncation of trailing
Feature is divided into three kinds of situations:(1) auto-correlation function trails, truncation when the nonautocorrelation function lag period is more than p, using AR (p) moulds
Type is fitted;(2) truncation truncation when the auto-correlation function lag period is more than q, nonautocorrelation function hangover, using MA (q) models fittings;
(3) auto-correlation function trails, and the hangover of nonautocorrelation function is fitted using ARMA (p, q) model.
Step 8:The image for analyzing auto-correlation function and partial autocorrelation function, is gradually soundd out from low order to high-order, is determined
Exponent number p, q of arma modeling are weighed using akaike information criterion (AIC criterion):
AIC=2k-2ln (L)
Wherein, k is model parameter number, and L is likelihood function.
When selecting best model from one group of alternative model, AIC (p, q) minimums, fitting effect are generally selected most
Excellent model is as determining model, it is determined that the arma modeling of exponent number, expression formula are as follows:
Wherein:It is the parameter of model, εt-iIt is the disturbance term of preceding i phases, Xt-iIt is preceding i phases lagged value, c is to lag the p phases
The mean value of value is added with the mean value of q phase disturbed values.
Step 9:The time series data of step 3 is imported to the arma modeling X for determining exponent numbertIn be fitted, obtain ARMA moulds
Type design parameter (i.e. determine parameter c, ε,θi), to obtain specific arma modeling, pass through specific ARMA moulds
Type predicts the following vehicle usage amount situation in a short time, and for car operation, quotient provides instruction.
Fit procedure includes error analysis and adjusting and optimizing, i.e., is made by the prediction data and vehicle that calculate arma modeling
With the mean square error of rate actual value, the threshold value of mean square error is given, if final mean square error is more than threshold value, is used with vehicle
Amount time of origin is index, higher, remoter apart from the current time vehicle of the vehicle usage amount weight closer apart from current time
Usage amount weight is lower, is fitted determines that the model of exponent number obtains new arma modeling again;If final mean square error is less than
Threshold value then exports the prediction result of the model.
The present invention predicts that the arma modeling of vehicle usage amount is not cured model, and over time, new goes through
History data can be added in database, and historical data excessively remote can be removed, and every day is fitted the data of arma modeling all
It is therefore to predict that the arma modeling of vehicle usage amount is also that ensure that ARMA in continuous adjusting and optimizing by current newest
The validity of model has practical guiding value for the vehicle usage amount of short-term forecast;It is true using model prediction result and history
The mean square error of real data carrys out the reliability of evaluation model, to the fitting quality of quantitative model, realizes science adjustment model ginseng
Number.
The above-mentioned description to embodiment can be understood and applied the invention for ease of those skilled in the art.
Person skilled in the art obviously easily can make various modifications to above-described embodiment, and described herein general
Principle is applied in other embodiment without having to go through creative labor.Therefore, the present invention is not limited to the above embodiments, ability
Field technique personnel announcement according to the present invention, the improvement made for the present invention and modification all should be in protection scope of the present invention
Within.
Claims (7)
1. a kind of short-term vehicle usage amount prediction technique based on arma modeling, includes the following steps:
(1) history of a large amount of vehicles in use of acquisition acquisition uses data, including vehicle GPS data, user's rental period, vehicle rent
With the frequency and car operation data, retains the history wherein within current time 1 year and be added to database using data
In;
(2) history in database is pre-processed using data, includes that the processing to missing values and exceptional value, data are clear
It washes, attribute construction and periodicity analysis;
(3) stationary test and white noise verification are carried out using data to the history for pre-processing rear vehicle;
(4) autoregression item exponent number p and the movement of arma modeling are determined according to auto-correlation function image and partial autocorrelation function image
Average item exponent number q, and then fitting is trained to arma modeling using data using by the vehicle history after examining;
(5) finally, obtained arma modeling is completed using training fitting to predict the following vehicle usage amount in a short time.
2. short-term vehicle usage amount prediction technique according to claim 1, it is characterised in that:To lacking in the step (2)
The processing of mistake value and exceptional value, including for individual dates there are shortage of data the case where, need to these dates carry out data
The case where these dates corresponding vehicle usage amount is set to 0 by filling, and there are data exceptions simultaneously for individual dates, equally
These dates corresponding vehicle usage amount is set to 0.
3. short-term vehicle usage amount prediction technique according to claim 1, it is characterised in that:Number in the step (2)
History is filtered out using data invalid in data according to cleaning, retains crucial car operation data;Attribute construction is
Go out vehicle rental amount and monthly two new attributes of rental amount daily using data statistics according to history;Periodicity analysis is i.e. with one
Week whether there is cyclically-varying as period, analysis vehicle rental amount.
4. short-term vehicle usage amount prediction technique according to claim 1, it is characterised in that:It is flat in the step (3)
Stability is examined carries out stationary test using ADF unit roots, it is desirable that passes through inspection when the corresponding probability value P > 0.05 of statistic
It tests, otherwise the history of vehicle is unstable using data, needs to carry out ADF inspections again after first carrying out calculus of differences, until vehicle
Until history is stable Random time sequence using data.
5. short-term vehicle usage amount prediction technique according to claim 1, it is characterised in that:It is white in the step (3)
Noise check is examined using Ljung-Box, it is desirable that the corresponding probability value P < of statistic 0.05, it is determined that the history of vehicle makes
It is nonwhite noise data with data.
6. short-term vehicle usage amount prediction technique according to claim 1, it is characterised in that:It is determined in the step (4)
The autoregression item exponent number p and rolling average item exponent number q of arma modeling, are gradually soundd out using low order to high-order, i.e., according to certainly
Correlation function image and partial autocorrelation function image calculate the corresponding AIC of arma modeling, take the minimum corresponding p and q conducts of AIC
The autoregression item exponent number p and rolling average item exponent number q of arma modeling, to establish the arma modeling for waiting for training fitting.
7. short-term vehicle usage amount prediction technique according to claim 1, it is characterised in that:The step (4) is in training
Need to calculate the mean square error between the prediction result and vehicle usage amount actual value of arma modeling during arma modeling, if
The threshold value that the mean square error is more than setting makes time of origin apart from current time then using vehicle usage amount time of origin as index
Closer vehicle usage amount weight is higher, and the time of origin vehicle usage amount weight remoter apart from current time is lower, passes through tune
Saving corresponding weight, input data is trained fitting to arma modeling again;If the mean square error is less than the threshold value of setting,
Complete the training to arma modeling.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810213487.XA CN108536652A (en) | 2018-03-15 | 2018-03-15 | A kind of short-term vehicle usage amount prediction technique based on arma modeling |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810213487.XA CN108536652A (en) | 2018-03-15 | 2018-03-15 | A kind of short-term vehicle usage amount prediction technique based on arma modeling |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108536652A true CN108536652A (en) | 2018-09-14 |
Family
ID=63483971
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810213487.XA Pending CN108536652A (en) | 2018-03-15 | 2018-03-15 | A kind of short-term vehicle usage amount prediction technique based on arma modeling |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108536652A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110182520A (en) * | 2019-05-29 | 2019-08-30 | 济南浪潮高新科技投资发展有限公司 | A kind of prediction technique of the coil of strip storage handling capacity based on fusion Time series forecasting model |
CN110491124A (en) * | 2019-08-19 | 2019-11-22 | 上海新共赢信息科技有限公司 | A kind of vehicle flow prediction technique, device, equipment and storage medium |
CN111191193A (en) * | 2020-01-17 | 2020-05-22 | 南京工业大学 | Long-term soil temperature and humidity high-precision prediction method based on autoregressive moving average model |
CN111768031A (en) * | 2020-06-24 | 2020-10-13 | 中电科华云信息技术有限公司 | Method for predicting crowd gathering tendency based on ARMA algorithm |
US20210109894A1 (en) * | 2019-10-11 | 2021-04-15 | Ikigai Labs Inc. | Automated customized modeling of datasets with intuitive user interfaces |
CN112785055A (en) * | 2021-01-18 | 2021-05-11 | 优必爱信息技术(北京)有限公司 | Vehicle refueling date prediction method and device |
CN113223291A (en) * | 2021-03-19 | 2021-08-06 | 青岛亿联信息科技股份有限公司 | System and method for predicting number of free parking spaces in parking lot |
CN113569491A (en) * | 2021-08-13 | 2021-10-29 | 江苏集萃智能光电系统研究所有限公司 | Analysis and correction method and device for wheel set size detection data |
CN114329347A (en) * | 2022-03-07 | 2022-04-12 | 广东电网有限责任公司 | Method and device for predicting metering error of electric energy meter and storage medium |
CN117149584A (en) * | 2023-10-31 | 2023-12-01 | 国合通用(青岛)测试评价有限公司 | Operation supervision system of multi-sample gauge length marking equipment based on big data |
CN118155423A (en) * | 2024-05-08 | 2024-06-07 | 四川智能交通系统管理有限责任公司 | Holiday long-term traffic situation prediction method and device based on time sequence model |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102519443A (en) * | 2011-11-26 | 2012-06-27 | 东南大学 | Method for recognizing and modifying abnormal measurement data of vehicle micro-mechanical gyroscope |
CN103996073A (en) * | 2014-05-06 | 2014-08-20 | 国家电网公司 | Photometric-network real-time-correction self-learning ARMA model photovoltaic power prediction method |
CN104657787A (en) * | 2015-02-03 | 2015-05-27 | 河海大学 | Wind power time series combined prediction method |
US20150170426A1 (en) * | 2013-12-16 | 2015-06-18 | Hyundai Motor Company | Method of calculating power consumption of air conditioner |
CN106920009A (en) * | 2017-03-03 | 2017-07-04 | 北京北青厚泽数据科技有限公司 | The Forecasting Methodology of hospital services amount |
CN107102969A (en) * | 2017-04-28 | 2017-08-29 | 湘潭大学 | The Forecasting Methodology and system of a kind of time series data |
CN107767659A (en) * | 2017-10-13 | 2018-03-06 | 东南大学 | Shared bicycle traffic attraction and prediction of emergence size method based on ARIMA models |
-
2018
- 2018-03-15 CN CN201810213487.XA patent/CN108536652A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102519443A (en) * | 2011-11-26 | 2012-06-27 | 东南大学 | Method for recognizing and modifying abnormal measurement data of vehicle micro-mechanical gyroscope |
US20150170426A1 (en) * | 2013-12-16 | 2015-06-18 | Hyundai Motor Company | Method of calculating power consumption of air conditioner |
CN103996073A (en) * | 2014-05-06 | 2014-08-20 | 国家电网公司 | Photometric-network real-time-correction self-learning ARMA model photovoltaic power prediction method |
CN104657787A (en) * | 2015-02-03 | 2015-05-27 | 河海大学 | Wind power time series combined prediction method |
CN106920009A (en) * | 2017-03-03 | 2017-07-04 | 北京北青厚泽数据科技有限公司 | The Forecasting Methodology of hospital services amount |
CN107102969A (en) * | 2017-04-28 | 2017-08-29 | 湘潭大学 | The Forecasting Methodology and system of a kind of time series data |
CN107767659A (en) * | 2017-10-13 | 2018-03-06 | 东南大学 | Shared bicycle traffic attraction and prediction of emergence size method based on ARIMA models |
Non-Patent Citations (3)
Title |
---|
WU JING JING等: "The New Vehicle Condition Monitoring Indicator VPR and its Short-Time Prediction ARMA Model", 《APPLIED MECHANICS AND MATERIALS》 * |
张致远: "时间序列分析在作业人员调控系统中的应用", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
苟斌娥: "时间序列分析法在教育研究中的应用", 《中国优秀硕士学位论文全文数据库基础科学辑》 * |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110182520B (en) * | 2019-05-29 | 2020-11-17 | 浪潮集团有限公司 | Steel coil storage throughput prediction method based on fusion time sequence prediction model |
CN110182520A (en) * | 2019-05-29 | 2019-08-30 | 济南浪潮高新科技投资发展有限公司 | A kind of prediction technique of the coil of strip storage handling capacity based on fusion Time series forecasting model |
CN110491124A (en) * | 2019-08-19 | 2019-11-22 | 上海新共赢信息科技有限公司 | A kind of vehicle flow prediction technique, device, equipment and storage medium |
CN110491124B (en) * | 2019-08-19 | 2020-09-08 | 上海新共赢信息科技有限公司 | Vehicle flow prediction method, device, equipment and storage medium |
US20210109894A1 (en) * | 2019-10-11 | 2021-04-15 | Ikigai Labs Inc. | Automated customized modeling of datasets with intuitive user interfaces |
US11995036B2 (en) * | 2019-10-11 | 2024-05-28 | Ikigai Labs Inc. | Automated customized modeling of datasets with intuitive user interfaces |
CN111191193A (en) * | 2020-01-17 | 2020-05-22 | 南京工业大学 | Long-term soil temperature and humidity high-precision prediction method based on autoregressive moving average model |
CN111768031B (en) * | 2020-06-24 | 2023-09-19 | 中电科华云信息技术有限公司 | Method for predicting crowd gathering trend based on ARMA algorithm |
CN111768031A (en) * | 2020-06-24 | 2020-10-13 | 中电科华云信息技术有限公司 | Method for predicting crowd gathering tendency based on ARMA algorithm |
CN112785055A (en) * | 2021-01-18 | 2021-05-11 | 优必爱信息技术(北京)有限公司 | Vehicle refueling date prediction method and device |
CN112785055B (en) * | 2021-01-18 | 2024-02-06 | 优必爱信息技术(北京)有限公司 | Method and equipment for predicting vehicle refueling date |
CN113223291A (en) * | 2021-03-19 | 2021-08-06 | 青岛亿联信息科技股份有限公司 | System and method for predicting number of free parking spaces in parking lot |
CN113223291B (en) * | 2021-03-19 | 2023-10-20 | 青岛亿联信息科技股份有限公司 | System and method for predicting number of idle parking spaces in parking lot |
CN113569491B (en) * | 2021-08-13 | 2023-08-18 | 江苏集萃智能光电系统研究所有限公司 | Wheel set size detection data analysis and correction method and device |
CN113569491A (en) * | 2021-08-13 | 2021-10-29 | 江苏集萃智能光电系统研究所有限公司 | Analysis and correction method and device for wheel set size detection data |
CN114329347B (en) * | 2022-03-07 | 2022-06-03 | 广东电网有限责任公司 | Method and device for predicting metering error of electric energy meter and storage medium |
CN114329347A (en) * | 2022-03-07 | 2022-04-12 | 广东电网有限责任公司 | Method and device for predicting metering error of electric energy meter and storage medium |
CN117149584A (en) * | 2023-10-31 | 2023-12-01 | 国合通用(青岛)测试评价有限公司 | Operation supervision system of multi-sample gauge length marking equipment based on big data |
CN118155423A (en) * | 2024-05-08 | 2024-06-07 | 四川智能交通系统管理有限责任公司 | Holiday long-term traffic situation prediction method and device based on time sequence model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108536652A (en) | A kind of short-term vehicle usage amount prediction technique based on arma modeling | |
CN109587713A (en) | A kind of network index prediction technique, device and storage medium based on ARIMA model | |
CN109615226A (en) | A kind of operation indicator method for monitoring abnormality | |
CN105305426B (en) | Mapreduce two-step short-period load prediction method based on deviation control mechanism | |
CN104574209B (en) | The modeling method of Medium Early Warning model is overloaded in a kind of city net distribution transforming again | |
EP2290603A1 (en) | Model optimization system using variable scoring | |
CN109598534A (en) | Information determines method and device, electronic equipment and storage medium | |
CN104899405A (en) | Data prediction method and system and alarming method and system | |
CN109376971A (en) | A kind of load curve forecasting method and system towards power consumer | |
CN108491682A (en) | Rainfall Runoff Forecasting | |
CN107563816A (en) | The Forecasting Methodology and system of the customer loss of e-commerce website | |
CN109697636A (en) | Merchant recommendation method, merchant recommendation device, electronic equipment and medium | |
CN105989535A (en) | Stock fluctuation ratio predicting method and system | |
CN109147324A (en) | A method of the traffic congestion probability forecast based on user feedback mechanisms | |
CN109886737A (en) | Needing forecasting method, device, electronic equipment and readable storage medium storing program for executing | |
CN113095597A (en) | Novel flood process design method suitable for small watershed without data | |
CN105653715A (en) | Pushing method and system for training courses | |
CN102279963A (en) | Two-level budget rationality inspection prompting and automatic optimization method, equipment thereof and system thereof | |
CN111415027A (en) | Method and device for constructing component prediction model | |
CN105740354A (en) | Adaptive potential Dirichlet model selection method and apparatus | |
CN111127074B (en) | Data recommendation method | |
CN111353625B (en) | Method, device, computer equipment and storage medium for predicting net point quantity | |
CN110069692A (en) | A kind of resource updates time forecasting methods and system towards open source information collection | |
CN105354644A (en) | Financial time series prediction method based on integrated empirical mode decomposition and 1-norm support vector machine quantile regression | |
CN102912760A (en) | Lowest navigable water level generating method of downstream river of reservoir |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180914 |