CN111259537B - Road surface performance prediction method based on VAR (variable-value response) multivariate time sequence - Google Patents
Road surface performance prediction method based on VAR (variable-value response) multivariate time sequence Download PDFInfo
- Publication number
- CN111259537B CN111259537B CN202010037298.9A CN202010037298A CN111259537B CN 111259537 B CN111259537 B CN 111259537B CN 202010037298 A CN202010037298 A CN 202010037298A CN 111259537 B CN111259537 B CN 111259537B
- Authority
- CN
- China
- Prior art keywords
- road surface
- var
- surface performance
- pavement
- performance indexes
- 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.)
- Active
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a road surface performance prediction method based on a VAR multivariate time sequence, which comprises the following steps: step 1), dividing road sections; step 2), selecting a plurality of pavement performance indexes; step 3), analyzing the influence of correlation of the road surface index based on the VAR; and 4) establishing a road surface performance index prediction model. The step 3) comprises the following steps: step 3.1), a VAR model is established for the road surface performance indexes with long-term correlation, and the influence among the road surface performance indexes is analyzed through variance decomposition; and 3.2) selecting the pavement performance indexes which have influence on the pavement performance indexes exceeding the preset percentage value as predicted pavement performance indexes, and screening out other pavement performance indexes. The pavement performance prediction method based on the VAR multivariate time sequence screens the factors with the predicted performance by applying the numerical analysis aiming at the correlation among the factors in the VAR multivariate time sequence, thereby greatly improving the accuracy of pavement performance prediction and the reliability when being applied to pavement preventive maintenance decision.
Description
Technical Field
The invention relates to the technical field of road monitoring, in particular to a road surface performance prediction method based on a VAR (variable-value automatic response) multivariate time sequence.
Background
The prediction of the road performance has crucial significance for the implementation of preventive maintenance. The traditional single variable time sequence pavement performance prediction method can only predict a future value according to the historical change of a certain pavement performance index, and neglects other influence factors. In the future, a certain road surface performance index is usually related to the current value of the performance index and other road surface performance indexes. Therefore, if various road surface performance index values are taken into consideration at present, the mathematical model is used for prediction, and the model precision can be improved.
Therefore, the invention provides a road surface performance prediction method adopting a multivariate time sequence. The VAR model is most commonly used in the multivariate time sequence, and the current-stage value is predicted by carrying out regression on the same-order lag variables of all variables, so that the dynamic relation among the endogenous variables can be estimated, and the influence of the exogenous variables on the endogenous variables can be reflected. In order to further ensure the prediction accuracy of the model, the VAR model is firstly adopted to analyze the influence effect among variables, and a brand-new multivariate time sequence prediction method is constructed based on the analysis result of the influence effect, so that the disturbance effect of various factors on the road performance is comprehensively analyzed, and the performance of the road performance prediction model is improved.
Disclosure of Invention
The invention aims to provide a road surface performance prediction method based on a VAR multi-element time sequence, and the accuracy and the reliability of road surface performance prediction are improved.
In order to achieve the purpose, the road surface performance prediction method based on the VAR multivariate time sequence comprises the following steps: step 1), dividing road sections;
step 2), selecting a plurality of pavement performance indexes;
step 3), analyzing the influence of correlation of the road surface index based on the VAR;
step 4), establishing a road surface performance index prediction model;
the step 3) further comprises the following steps:
step 3.1), building a VAR model for the pavement performance indexes with long-term correlation, and analyzing the influence among the pavement performance indexes through variance decomposition;
and 3.2) selecting the pavement performance indexes which have influence on the pavement performance indexes exceeding the preset percentage value as predicted pavement performance indexes, and screening out other pavement performance indexes.
Preferably, in step 3.2), AIC calculation is performed on the road surface performance indexes with long-term correlation by using different p values of autoregressive coefficients, a p value corresponding to the smallest AIC value is selected, a VAR (p) model is constructed, and the influence between the road surface performance indexes is obtained through variance decomposition.
Preferably, in the step 3.2), the preset percentage value is 10%.
As a preferable scheme, in the step 4), different autoregressive coefficient p values are adopted to perform AIC calculation, the autoregressive coefficient p value corresponding to the minimum AIC value is taken to construct a VAR (p) model, and data of the original road surface performance index is substituted into data for calculating the future road surface performance index.
Preferably, the step 1) further comprises the following steps:
step 1.1), extracting original data of a plurality of pavement performance indexes of the current pavement from a database;
and 1.2) dividing the original pavement performance index into equidistant road sections according to the same factors in a datamation mode.
Preferably, in step 1.1), the original data of the road surface performance indexes extracted from the database are arranged according to the year, when multiple groups of data appear in the same year, the multiple groups of data are averaged, and if data are missing, the data are supplemented by an interpolation method.
Preferably, in the step 1.2), the factors comprise the same structure, the same curing measures and time, and the equal distance is 100 meters.
Preferably, the step 2) further comprises the following steps:
step 2.1), carrying out stability inspection on the original data of a plurality of road surface performance indexes in the road section, if not, carrying out differential calculation until the stability inspection is passed;
and 2.2) performing collaborative analysis on the data of the plurality of road surface performance indexes which pass the stability test, and selecting the road surface performance indexes with long-term correlation as predicted road surface performance indexes.
Preferably, in the step 2.1), performing stability ADF inspection on the road surface performance index data in the road section, if not, performing differential calculation until a stability inspection result is less than 0.05; in the step 2.2), when the road surface performance index passing stability test is subjected to the coordination test result of less than 0.05, the road surface performance index data is considered to have long-term correlation and is determined as the road surface performance index for prediction.
Preferably, the predicted road surface performance indexes are PCI, IRI and RD;
the variation Δ IRI of the IRI prediction time is calculated according to the following formula:
ΔIRI(t)＝-0.0426IRI(t-1)+0.1202ΔRD(t-1)；
the change value Δ RD at the RD predicted time is calculated according to the following equation:
ΔRD(t)＝4.9184IRI(t-1)+0.1939ΔRD(t-1)；
the change Δ RD at the PCI predicted time is calculated according to the following equation:
ΔPCI(t)＝-2.3529ΔIRI(t-1)-0.5323ΔRD(t-1)+0.4575PCI(t-1)；
where t denotes the predicted time and t-1 denotes the previous time at an interval of t.
The beneficial effects of the invention are: according to the pavement performance prediction method based on the VAR multivariate time sequence, the factors with the predicted performance are screened by using numerical analysis aiming at the correlation among the factors in the VAR multivariate time sequence, so that the accuracy of pavement performance prediction and the reliability when the method is applied to pavement preventive maintenance decision are greatly improved.
Drawings
Fig. 1 is a flowchart of a road surface performance prediction method based on VAR multivariate time series according to a preferred embodiment of the present invention.
Fig. 2 is a comparison graph before and after an IRI predicted value is performed by using the VAR multivariate time series-based road surface performance prediction method according to the preferred embodiment of the present invention.
Fig. 3 is a comparison graph before and after the RD prediction value is performed by using the road surface performance prediction method based on VAR multivariate time series according to the preferred embodiment of the present invention.
FIG. 4 is a diagram of PCI unit and multivariate time series prediction value comparison.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments.
In order to solve the problems, the invention provides a road surface performance prediction method based on a VAR multivariate time sequence. VAR (Vector Auto Regression) is commonly used to predict interconnected time series systems and to analyze the dynamic effects of random perturbations on variable systems. The VAR circumvents the need to structure the model by constructing the model with each endogenous variable in the system as a function of the hysteresis values of all endogenous variables in the system.
The road surface performance prediction method based on the VAR multivariate time sequence firstly adopts the VAR model to analyze the influence effect among variables, and a brand new multivariate time sequence prediction method is constructed on the basis, so that the time sequence prediction precision can be improved.
Please refer to fig. 1, which is a schematic flow chart of a road surface performance prediction method based on VAR multivariate time series according to a preferred embodiment of the present invention, including the following steps: step 1), road section division; step 2), selecting a plurality of pavement performance indexes; step 3), analyzing the influence of correlation of the road surface indexes based on the VAR; and 4) establishing a road surface performance index prediction model.
Wherein, in the step 1), the method specifically comprises the following steps:
step 1.1), extracting the original data of a plurality of pavement performance indexes of the current pavement from the database.
Extracting original data of a plurality of pavement performance indexes of the current pavement from a database, arranging various performance data according to the year, and averaging the plurality of groups of data when the plurality of groups of data appear in the same year.
Step 1.2), dividing the road sections into sections of 100 meters per section according to the factors of the same structure, the same maintenance measures, time and the like.
The road sections are divided into sections of 100 meters each according to the factors of the same structure, the same maintenance measures, time and the like.
And 1.3) if data are missing, supplementing the data through an interpolation method, such as an average value method.
If data loss occurs in a certain year, the data is supplemented by an interpolation method of averaging the previous value and the next value.
Wherein, in the step 2), the method specifically comprises the following steps:
and 2.1) carrying out stability inspection on the original data of a plurality of road surface performance indexes in the road section, if not, carrying out differential calculation until the stability inspection is passed.
And (3) carrying out stability ADF (automatic document feeder) inspection (unit root inspection) on various performance data in the road section, if the performance data are not stable, carrying out differential calculation until the stability inspection is passed, namely, the inspection result p-value is less than 0.05.
And 2.2) performing collaborative analysis on the multiple road surface performance index data passing through the stability test, and selecting the road surface performance index having a long-term correlation relationship.
And performing co-integration inspection on the various types of performance data passing the stability inspection, and when the inspection result p-value is less than 0.05, determining that the performance data have long-term correlation and determining the performance data as the road surface performance index for modeling.
Wherein, in the step 3), the method specifically comprises the following steps:
and 3.1) establishing a VAR model for the pavement performance indexes with long-term correlation, and analyzing the influence effect among all factors through variance decomposition.
And for the road surface performance indexes with long-term correlation, different autoregressive coefficient p values are adopted to carry out AIC calculation, the p value corresponding to the minimum AIC value is taken to construct a VAR (p) model, and the influence effect of each road surface performance index on each other is obtained through variance decomposition.
And 3.2) selecting other road surface performance indexes which have influence on the road surface performance index to be predicted over a certain percentage value, considering that the road surface performance index over the percentage value is effective on the predicted road surface performance index, and screening out the other road surface performance indexes as bad disturbance.
And selecting other road surface performance indexes which have more than 10% influence on the road surface performance index to be predicted, considering that the road surface performance index which exceeds the value is beneficial to improving the prediction result on the predicted road surface performance index, and screening out the other road surface performance indexes as bad disturbance.
In the step 4), a VAR model based on the screened road surface performance index data is constructed, and future road surface performance data is predicted.
And (2) adopting an AIC information criterion (Akaike information criterion) and adopting different autoregressive coefficient p values to calculate AIC, and taking the autoregressive coefficient p value corresponding to the minimum AIC value to construct a VAR (P) model, and substituting original data to realize the prediction of future pavement performance index data.
The following is an application of the road surface performance prediction method based on the VAR multivariate time sequence in a specific example.
And step 1), assuming basic data, carrying out time sequence arrangement and road section division.
And 2) selecting IRI (international flatness index), RD (track depth) and PCI (road surface condition index) data in the road section performance.
ADF inspection is carried out on the three types of data, and the data p-value >0.05 in the original time sequence is found, and the data is not stable, so that the data is subjected to first-order difference, inspected again and passed. And carrying out pairwise integration test on IRI, RD and PCI data, wherein p-value is less than 0.05, and long-term correlation exists between the IRI, RD and PCI data.
The VAR (1) model was constructed by performing AIC calculation on a VAR model with an autoregressive coefficient p =0,1 and finding that the AIC is the minimum value when the autoregressive coefficient p = 1.
And 3) calculating to obtain the relevant parameters of the VAR model, and obtaining the influence effect of each road surface performance index on each other by adopting variance decomposition.
ΔIRI(t)＝-0.2085ΔIRI(t-1)+0.1055ΔRD(t-1)-0.0243PCI(t-1)；
ΔRD(t)＝2.1674ΔIRI(t-1)-0.0784ΔRD(t-1)-0.4440PCI(t-1)；
ΔPCI(t)＝-2.3529ΔIRI(t-1)-0.5323ΔRD(t-1)+0.4575PCI(t-1)。
Wherein t represents the time corresponding to the solved road surface performance index, and t-1 represents the last time spaced from the time t. Typically, t is the year and t-1 is the last year.
For this model, the effect of the various road surface performance indicators on each other can be obtained by variance decomposition, as shown in table 1.
TABLE 1 Effect of various pavement Performance indicators on each other
Therefore, it is considered that the IRI and RD can be predicted from each other, and the predicted result of PCI is still determined by IRI, RD and PCI.
The VAR model (1) was constructed by performing AIC calculation on the IRI and RD VAR models with an autoregressive coefficient p =0,1 and finding that the minimum AIC is obtained when the autoregressive coefficient p = 1.
And 4), calculating to obtain VAR model related parameters to obtain a VAR model related formula.
ΔIRI(t)＝-0.0426IRI(t-1)+0.1202ΔRD(t-1)；
ΔRD(t)＝4.9184IRI(t-1)+0.1939ΔRD(t-1)。
For this model, the change values Δ IRI, Δ RD of the future year can be predicted, and compared with the predicted result in the previous step, see fig. 2 and fig. 3.
Because PCI and PCI, IRI, RD all have comparatively close relation, PCI prediction formula after screening still is:
ΔPCI(t)＝-2.3529ΔIRI(t-1)-0.5323ΔRD(t-1)+0.4575PCI(t-1)。
therefore, a unit time sequence prediction is used to compare with a multi-element time sequence prediction result of the PCI, and an ARIMA (p, l, q) model (differential Integrated Moving Average Autoregressive model, also called Integrated Moving Average Autoregressive model) of the unit time sequence is established. Where p denotes an autoregressive coefficient, l denotes a difference order, and q denotes a moving average coefficient. Firstly, obtaining a stable PCI time sequence through difference, secondly, selecting different (p, q) values to carry out AIC calculation, and constructing an ARIMA model by taking an autoregressive coefficient p and a moving average coefficient q corresponding to the minimum AIC value. Through the difference sequence stationary discrimination and the AIC criterion, the formula for obtaining the unit time sequence ARIMA (3, 1) is as follows:
ΔPCI(t)-0.3858ΔPCI(t-1)-0.6502ΔPCI(t-2)-0.4981ΔPCI(t-3)-1ε(t-1)-1.7214。
for this model, the PCI values in the time series of units for the future year can be predicted, see fig. 4.
Compared with the traditional unit or multivariate time sequence, the pavement performance prediction method based on the VAR multivariate time sequence screens the factors with prediction performance by applying numerical analysis on the correlation among the factors in the VAR multivariate time sequence, thereby greatly improving the accuracy of pavement performance prediction and the reliability when being applied to pavement preventive maintenance decision.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that various changes and modifications can be made by those skilled in the art without departing from the spirit of the invention, and these changes and modifications are all within the scope of the invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (8)
1. A road surface performance prediction method based on VAR multi-element time sequence comprises the following steps:
step 1), dividing road sections;
step 2), selecting a plurality of pavement performance indexes;
step 3), analyzing the influence of correlation of the road surface index based on the VAR;
step 4), establishing a road surface performance index prediction model;
the step 3) further comprises the following steps:
step 3.1), a VAR model is established for the road surface performance indexes with long-term correlation, and the influence among the road surface performance indexes is analyzed through variance decomposition;
step 3.2), selecting the pavement performance indexes which have influence on the pavement performance indexes exceeding the preset percentage value as predicted pavement performance indexes, and screening out other pavement performance indexes;
in the step 4), different autoregressive coefficient p values are adopted to perform AIC calculation, the autoregressive coefficient p value corresponding to the minimum AIC value is taken to construct a VAR (p) model, and the data of the original pavement performance index is brought into the calculation of the future pavement performance index data;
wherein the predicted road surface performance indexes are PCI, IRI and RD;
the variation Δ IRI at the IRI prediction time is calculated according to the following formula:
ΔIRI(t)＝-0.0426IRI(t-1)+0.1202ΔRD(t-1)；
the change value Δ RD at the RD prediction time is calculated according to the following equation:
ΔRD(t)＝4.9184IRI(t-1)+0.1939ΔRD(t-1)；
the change Δ RD at the PCI predicted time is calculated according to the following equation:
ΔPCI(t)＝-2.3529ΔIRI(t-1)-0.5323ΔRD(t-1)+0.4575PCI(t-1)；
where t denotes the predicted time and t-1 denotes the last time at an interval of one from t.
2. The VAR multivariate time series-based road surface performance prediction method as defined in claim 1, wherein: in the step 3.2), different autoregressive coefficient p values are adopted to perform AIC calculation on the road surface performance indexes with long-term correlation, the p value corresponding to the minimum AIC value is taken to construct a VAR (p) model, and the influence among the road surface performance indexes is obtained through variance decomposition.
3. The VAR multivariate time series-based road surface performance prediction method as defined in claim 1, wherein: in the step 3.2), the preset percentage value is 10%.
4. The VAR multivariate time series-based road surface performance prediction method as defined in claim 1, wherein: the step 1) further comprises the following steps:
step 1.1), extracting original data of a plurality of pavement performance indexes of the current pavement from a database;
and 1.2) dividing the original pavement performance index into equidistant road sections according to the same factors in a datamation mode.
5. The VAR multivariate time series-based road surface performance prediction method as set forth in claim 4, wherein: in the step 1.1), the original data of the pavement performance indexes are extracted from the database, the various performance data are arranged according to the year, when multiple groups of data appear in the same year, the multiple groups of data are averaged, and if data are missing, the data are supplemented through an interpolation method.
6. The VAR multivariate time series-based road surface performance prediction method as defined in claim 4, wherein: in step 1.2), the factors include the same structure, the same maintenance measures and time, and the equal distance is 100 meters.
7. The VAR multivariate time series-based road surface performance prediction method as defined in claim 1, wherein: the step 2) further comprises the following steps: step 2.1), carrying out stability inspection on the original data of a plurality of road surface performance indexes in the road section, if not, carrying out differential calculation until the stability inspection is passed;
and 2.2) performing collaborative analysis on the data of the plurality of road surface performance indexes passing the stability test, and selecting the road surface performance indexes with long-term correlation as predicted road surface performance indexes.
8. The VAR multivariate time series-based road surface performance prediction method according to claim 7, characterized in that: in the step 2.1), stability ADF inspection is carried out on the road surface performance index data in the road section, if the road surface performance index data are not stable, differential calculation is carried out until a stability inspection result is less than 0.05; in the step 2.2), when the road surface performance index passing stability test is subjected to the coordination test result of less than 0.05, the road surface performance index data is considered to have long-term correlation and is determined as the road surface performance index for prediction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010037298.9A CN111259537B (en) | 2020-01-14 | 2020-01-14 | Road surface performance prediction method based on VAR (variable-value response) multivariate time sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010037298.9A CN111259537B (en) | 2020-01-14 | 2020-01-14 | Road surface performance prediction method based on VAR (variable-value response) multivariate time sequence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111259537A CN111259537A (en) | 2020-06-09 |
| CN111259537B true CN111259537B (en) | 2023-03-31 |
Family
ID=70946952
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010037298.9A Active CN111259537B (en) | 2020-01-14 | 2020-01-14 | Road surface performance prediction method based on VAR (variable-value response) multivariate time sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111259537B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154292A (en) * | 2017-12-20 | 2018-06-12 | 中公高科养护科技股份有限公司 | A kind of method for building up and system of the prediction model of Pavement Performance |
| CN108898838A (en) * | 2018-08-03 | 2018-11-27 | 首都经济贸易大学 | A kind of aerodrome traffic congestion prediction technique and device based on LSTM model |
| CN108921688A (en) * | 2018-07-02 | 2018-11-30 | 阿里巴巴集团控股有限公司 | Construct the method and device of prediction model |
| CN109993223A (en) * | 2019-03-26 | 2019-07-09 | 南京道润交通科技有限公司 | Pavement Condition prediction technique, storage medium, electronic equipment |
-
2020
- 2020-01-14 CN CN202010037298.9A patent/CN111259537B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154292A (en) * | 2017-12-20 | 2018-06-12 | 中公高科养护科技股份有限公司 | A kind of method for building up and system of the prediction model of Pavement Performance |
| CN108921688A (en) * | 2018-07-02 | 2018-11-30 | 阿里巴巴集团控股有限公司 | Construct the method and device of prediction model |
| CN108898838A (en) * | 2018-08-03 | 2018-11-27 | 首都经济贸易大学 | A kind of aerodrome traffic congestion prediction technique and device based on LSTM model |
| CN109993223A (en) * | 2019-03-26 | 2019-07-09 | 南京道润交通科技有限公司 | Pavement Condition prediction technique, storage medium, electronic equipment |
Non-Patent Citations (5)
| Title |
|---|
| 单岗 ; 阙家奇 ; 王飞 ; 朱震海 ; 董侨 ; 王茜 ; .基于多元回归模型的高速公路养护路段长期性能研究.公路.2018,(03),全文. * |
| 敬超 ; 张金喜 ; .沥青路面性能预测研究综述.中外公路.2017,(05),全文. * |
| 梅沁 ; 刘宴先 ; 景小楠 ; .基于多因素VAR分析的上海GDP预测.湖北工业大学学报.2011,(03),全文. * |
| 汤晴 ; 吴大鸿 ; .沥青路面使用性能及其预测研究.公路工程.2015,(05),全文. * |
| 马士宾 ; 陈奕 ; 孙敬福 ; 虞秋富 ; .路面检测周期对使用性能预测影响研究.中外公路.2014,(06),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111259537A (en) | 2020-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3289420B1 (en) | Computer system and method for causality analysis using hybrid first-principles and inferential model | |
| Arguez et al. | The definition of the standard WMO climate normal: The key to deriving alternative climate normals | |
| JP6141235B2 (en) | How to detect anomalies in time series data | |
| CN108241901B (en) | Transformer early warning evaluation method and device based on prediction data | |
| CN111310981B (en) | Reservoir water level trend prediction method based on time series | |
| CN108830417B (en) | ARMA (autoregressive moving average) and regression analysis based life energy consumption prediction method and system | |
| CN113239132B (en) | Online out-of-tolerance identification method for voltage transformer | |
| CN117196353A (en) | Environmental pollution assessment and monitoring method and system based on big data | |
| CN111310990A (en) | Improved gray combination model-based track quality prediction method and system | |
| CN114066262A (en) | Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium | |
| CN110298765B (en) | Power distribution network power consumption abnormality detection method based on objective correlation factors | |
| Elshenawy et al. | Automatic imputation of missing highway traffic volume data | |
| CN113539382B (en) | Early warning positioning method and system for key technological parameters of dimethyl phosphite | |
| CN111259537B (en) | Road surface performance prediction method based on VAR (variable-value response) multivariate time sequence | |
| CN113269327A (en) | Flow anomaly prediction method based on machine learning | |
| CN116384843A (en) | Energy efficiency evaluation model training method and monitoring method for digital energy nitrogen station | |
| CN108123436B (en) | Voltage out-of-limit prediction model based on principal component analysis and multiple regression algorithm | |
| CN114564487B (en) | Meteorological raster data updating method combining forecast prediction | |
| Zhang et al. | Determining statistical process control baseline periods in long historical data streams | |
| CN114970311A (en) | Method for establishing remote module life prediction model and life prediction method | |
| CN114580151A (en) | Water demand prediction method based on gray linear regression-Markov chain model | |
| CN110532698B (en) | Industrial equipment vibration characteristic value trend prediction method based on data model | |
| CN113283157A (en) | System, method, terminal and medium for predicting life cycle of intelligent stamping press part | |
| CN114048592A (en) | Finish rolling whole-flow distributed operation performance evaluation and non-optimal reason tracing method | |
| Soller et al. | Adaptive error prediction for production lines with unknown dependencies |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |