CN114330834A - Charging pile power consumption prediction method based on self-updating cubic exponential smoothing method - Google Patents
Charging pile power consumption prediction method based on self-updating cubic exponential smoothing method Download PDFInfo
- Publication number
- CN114330834A CN114330834A CN202111465442.XA CN202111465442A CN114330834A CN 114330834 A CN114330834 A CN 114330834A CN 202111465442 A CN202111465442 A CN 202111465442A CN 114330834 A CN114330834 A CN 114330834A
- Authority
- CN
- China
- Prior art keywords
- updating
- power consumption
- charging pile
- value
- self
- 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
- 238000009499 grossing Methods 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 67
- 230000002159 abnormal effect Effects 0.000 claims abstract description 36
- 238000010219 correlation analysis Methods 0.000 claims abstract description 13
- 230000005611 electricity Effects 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000001932 seasonal effect Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 4
- 230000007774 longterm Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 230000003068 static effect Effects 0.000 description 8
- 230000007547 defect Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011221 initial treatment Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a charging pile power consumption prediction method based on a self-updating cubic exponential smoothing method, which comprises the following steps: acquiring historical time sequence power consumption data of the charging pile, and collecting abnormal conditions; obtaining the influence degree of the abnormal condition on the power consumption of the charging pile through correlation analysis, and obtaining a correlation coefficient; automatically updating the smoothing coefficient in the cubic exponential smoothing method by using the correlation coefficient, and determining the optimal smoothing coefficient by comparing the precision before and after updating; and (3) performing medium-long term prediction on the power consumption of the charging pile based on a cubic exponential smoothing method using an optimal smoothing coefficient to obtain a corresponding prediction result. Compared with the prior art, the method can effectively analyze the influence of internal correlation and external influence factors of the time sequence on the electricity utilization condition of the charging pile, generate corresponding correlation coefficients according to the influence, improve the prediction effect through continuous iteration, reduce the prediction error caused by abnormal fluctuation of the sequence, and further ensure the accuracy and reliability of the electricity consumption prediction of the charging pile.
Description
Technical Field
The invention relates to the technical field of power consumption prediction, in particular to a charging pile power consumption prediction method based on a self-updating cubic exponential smoothing method.
Background
With the rapid development of electric vehicles, the charging pile is widely applied as a matched capital construction device. The input and the alternating current power grid lug connection of filling electric pile, the output all is equipped with charging plug and is used for charging for electric automobile, fills the power consumption action of electric pile and receives the strong influence of factors such as user's action, geography, environment, season, and the prediction to filling electric pile power consumption and the relevant research to the electric wire netting influence is less at present.
The power consumption condition of charging pile belongs to time series, and common time series models are roughly divided into three types: a weighted moving average model, an exponential smoothing model, an ARIMA autoregressive differential model. Generally, there will be irregular components in any time series, periodicity is usually not considered in business and administrative data, only trend and seasonal components are considered. The time series not containing the trend and seasonal components, i.e., the stationary time series, contains only the random component, as long as the random fluctuation can be eliminated by smoothing. Therefore, such prediction methods are also called smooth prediction methods.
The exponential smoothing method is an algorithm in a time series prediction method, and data predicted by the algorithm refers to a group of predicted values of the same data variable which are arranged according to the occurrence sequence of events. The time series is characterized by two: one is to predict the development of future data according to the trend of historical data; the second is that the data has variability and four types of trend, periodicity, randomness and comprehensiveness. The number of iterations of the exponential smoothing method is one to three for the conventional exponential smoothing method. The first and second exponential smoothing is suitable for a linear model, and when the nonlinear trend of the power consumption changes, the second exponential smoothing has errors, so that a third exponential smoothing method is introduced. The cubic exponential smoothing method is to perform a smoothing calculation again on the basis of the secondary smoothing. The traditional cubic exponential smoothing method adopts static parameters and static coefficients to be brought into an operation model. The static cubic exponential smoothing model has the advantage that the static cubic exponential smoothing model retains seasonal information on the basis of quadratic exponential smoothing, so that the static cubic exponential smoothing model can predict time series with seasonality. Cubic exponential smoothing adds a new parameter to represent the trend after smoothing. Because the traditional cubic exponential smoothing method does not consider the deviation existing outside, the static cubic exponential smoothing method is widely applied to stable data models. Moreover, in the conventional static cubic exponential smoothing model, the smoothing coefficient is a fixed constant, and once the coefficient is determined, dynamic correction cannot be performed, which results in that prediction cannot be accurately and reliably performed in medium-long term prediction of complex data.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a method for predicting the power consumption of a charging pile based on a self-updating cubic exponential smoothing method.
The purpose of the invention can be realized by the following technical scheme: a charging pile electricity consumption prediction method based on a self-updating cubic exponential smoothing method comprises the following steps:
s1, acquiring historical time sequence electricity consumption data of the charging pile, and collecting abnormal conditions;
s2, obtaining the influence degree of the abnormal conditions on the power consumption of the charging pile through correlation analysis based on the historical time sequence power consumption data and the collected abnormal conditions, and displaying the influence degree by using a correlation coefficient to obtain a correlation coefficient;
s3, automatically updating the smoothing coefficient in the cubic exponential smoothing method by using the correlation coefficient, and determining the optimal smoothing coefficient through precision comparison before and after updating;
and S4, performing medium and long term prediction on the power consumption of the charging pile based on a cubic exponential smoothing method using the optimal smoothing coefficient to obtain a corresponding prediction result.
Further, the abnormal condition is policy information data, equipment data and seasonal variation interference data when data fluctuates sharply.
Further, the step S2 is specifically to perform correlation analysis by using pearson correlation coefficients.
Further, the specific process of step S2 is as follows: and performing independent hot coding based on the collected abnormal conditions, and performing correlation analysis on the power consumption data of the occurrence time point corresponding to the abnormal conditions by combining the historical time series power consumption data to obtain a correlation coefficient.
Further, the correlation coefficient is specifically:
wherein x is the electricity consumption at the time point of occurrence of the abnormal condition, y is the one-hot coded value of the abnormal condition, xμ、yμThe mean values of x and y, respectively, and the standard deviations of x and y, respectively.
Further, the step S3 specifically includes the following steps:
s31, conducting preliminary prediction on the historical time series electricity consumption data, and when the data of the Nth period is predicted, using the data of the first three marked time points of the predicted time point as the update training starting point [ X ]N-3,XN-2,XN-1]；
S32, setting training iteration times and an initial value of a self-updating smoothing coefficient alpha;
s33, creating a coefficient value array F [ m, n ], and storing an initial value of alpha, wherein m is a predicted charging pile, and n is a predicted time point;
s34, storing the correlation coefficient into Fm, n to be used as a condition of self-updating data, and carrying out self-updating processing on the data predicted at the later stage;
s35, setting the predicted time length T, and determining the time point of the electric quantity used by the charging pile through self-updating processing;
s36, calculating to obtain a predicted value, judging whether self-updating is needed or not, if so, updating alpha, and then executing a step S37;
otherwise, calculating a correlation coefficient, and then executing step S37;
s37, calculating an error value between the predicted value and the actual value, if the error value is smaller than a set error threshold value, indicating that the current alpha is the optimal smooth coefficient, otherwise, executing the step S38;
and S38, resetting the distance of the traversal search alpha to 0.01, recalculating the error value between the predicted value and the actual value, and selecting the alpha corresponding to the error value below the set optimization threshold value to serve as the optimal smoothing coefficient.
Further, the smoothing coefficient is specifically a weight coefficient, and includes a weight coefficient corresponding to time series stability and a weight coefficient corresponding to time series abnormality.
Further, the formula for calculating the predicted value in step S36 is:
xt+T＝At+BtT+CtT2
At＝3St (1)-3St (2)+St (3)
wherein, XN-3,XN-2,XN-1For predicting data at the first three time points of the data, the three predicted parameters are updated every time the predicted value is calculated, and the constant At、Bt、CtThe coefficients are exponentially smoothed for the t-th cycle, and are different for each prediction due to the self-updating of the coefficients and parameters,
respectively, the first, second and third exponential smoothing values, alpha, of the t-th periodm,nIs a weight coefficient under m and n, namely a correlation coefficient P, xtIs the actual value of the current t period, xt+TIs the predicted value of the future T + T period.
Further, the error value in step S37 includes a mean square error and a relative error, and the error value in step S38 is specifically a relative error.
Further, the calculation formula of the mean square error is as follows:
St＝xt+T
the calculation formula of the relative error is as follows:
wherein s istTo predict value, xtThe smaller the mean square error is, the higher the accuracy is; and delta is an actual relative error, usually output by percentage, represents an absolute error between a true value and a predicted value of the power consumption, and L is true power consumption data, wherein the smaller the relative error value is, the higher the prediction accuracy is.
Compared with the prior art, the method improves the traditional cubic exponential smoothing method, provides the cubic exponential smoothing method with the self-updating smoothing coefficient, updates the parameters in an iterative mode, can focus data sections with abnormal fluctuation in the time sequence charging pile power consumption data, performs correlation analysis on the condition possibly causing fluctuation, and generates the correlation coefficient according to the correlation coefficient, so that the internal correlation and the external influence possibly suffered by a power consumption curve are comprehensively considered, the influence of the internal correlation and the external influence factors in the time sequence on the charging pile power consumption condition can be effectively analyzed, the corresponding self-updating coefficient is generated according to the correlation coefficient, the prediction effect is improved through continuous iteration, the prediction error caused by sequence fluctuation is reduced, and accurate and reliable prediction on the power consumption is guaranteed.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
fig. 2 is a schematic diagram of an application process of the embodiment.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
Examples
As shown in fig. 1, a method for predicting charging pile power consumption based on a self-updating cubic exponential smoothing method includes the following steps:
s1, obtaining historical time sequence power consumption data of the charging pile, and collecting abnormal conditions, wherein the abnormal conditions are policy information data, equipment data and seasonal variation interference data when the data fluctuate severely;
s2, based on the historical time-series power consumption data and the collected abnormal conditions, obtaining the degree of influence of the abnormal conditions on the power consumption of the charging pile through correlation analysis, and presenting the degree of influence by a correlation coefficient to obtain a correlation coefficient, in this embodiment, a pearson correlation coefficient is used to perform correlation analysis, based on the collected abnormal conditions, first performing unique heat coding, and then performing correlation analysis on the power consumption data at the occurrence time point corresponding to the abnormal conditions in combination with the historical time-series power consumption data to obtain the correlation coefficient:
wherein x is the electricity consumption at the time point of occurrence of the abnormal condition, y is the one-hot coded value of the abnormal condition, xμ、yμThe mean values of x and y are respectively, and the delta x and the delta y are respectively the standard deviation of x and y;
s3, automatically updating the smoothing coefficient in the cubic exponential smoothing method by using the correlation coefficient, and determining the optimal smoothing coefficient by comparing the precision before and after updating, specifically:
s31, conducting preliminary prediction on the historical time series electricity consumption data, and when the data of the Nth period is predicted, using the data of the first three marked time points of the predicted time point as the update training starting point [ X ]N-3,XN-2,XN-1]；
S32, setting training iteration times and an initial value of a self-updated smoothing coefficient alpha, wherein the smoothing coefficient is specifically a weight coefficient and comprises a weight coefficient corresponding to time sequence stability and a weight coefficient corresponding to time sequence abnormity;
s33, creating a coefficient value array F [ m, n ], and storing an initial value of alpha, wherein m is a predicted charging pile, and n is a predicted time point;
s34, storing the correlation coefficient into Fm, n to be used as a condition of self-updating data, and carrying out self-updating processing on the data predicted at the later stage;
s35, setting the predicted time length T, and determining the time point of the electric quantity used by the charging pile through self-updating processing;
s36, calculating to obtain a predicted value, judging whether self-updating is needed or not, if so, updating alpha, and then executing a step S37;
otherwise, calculating a correlation coefficient, and then executing step S37;
the specific formula for calculating the predicted value is as follows:
xt+T＝At+BtT+CtT2
At＝3St (1)-3St (2)+St (3)
wherein, XN-3,XN-2,XN-1Updating the three predicted parameters for predicting the data at the first three time points of the data every time the predicted value is calculated; constant At、Bt、CtThe coefficient is an exponential smoothing coefficient of the t period, and is different in each prediction due to self-updating of the coefficient and the parameter;
respectively, the first, second and third exponential smoothing values, alpha, of the t-th periodm,nIs a weight coefficient under m and n, namely a correlation coefficient P, xtIs the actual value of the current t period, xt+TThe predicted value of the future T + T period is obtained;
s37, calculating an error value (including a mean square error and a relative error) between the predicted value and the actual value, if the error value is smaller than a set error threshold, indicating that the current alpha is the optimal smoothing coefficient, otherwise, executing a step S38, wherein the mean square error has a calculation formula:
St＝xt+T
the relative error is calculated as:
in the formula, stTo predict value, xtThe smaller the mean square error is, the higher the accuracy is; delta is the actual relative error, usually output by percentage, delta represents the absolute error of the true value and the predicted value of the power consumption, L is the true power consumption data, and the smaller the relative error value is, the higher the prediction accuracy is;
and S38, resetting the distance of the traversal search alpha to 0.01, recalculating the error value between the predicted value and the actual value (only calculating the relative error), and selecting the alpha corresponding to the error value below the set optimization threshold value as the optimal smooth coefficient.
And S4, performing medium and long term prediction on the power consumption of the charging pile based on a cubic exponential smoothing method using the optimal smoothing coefficient to obtain a corresponding prediction result.
The technical scheme provides a charging pile power consumption prediction method based on a self-updating cubic exponential smoothing method, and the method comprises the steps of firstly, utilizing a Pearson correlation coefficient to carry out correlation analysis to obtain the influence degree of abnormal conditions on the charging pile power consumption, and presenting the influence degree by the correlation coefficient; then, automatically updating the smoothing index in the exponential smoothing method by using the correlation coefficient, and comparing the precision to determine an optimal value; and finally, performing medium-long term prediction by using a cubic exponential smoothing method after coefficient updating.
According to the technical scheme, the model is subjected to iterative training by constructing the prediction model so as to complete the prediction of the power consumption of the charging pile, as shown in fig. 2:
inputting data acquired by a charging pile into a processing module, screening to obtain historical time series power consumption data, and transmitting the historical time series power consumption data to an input part of a model;
step two, performing initial treatment on the dataStep analysis, when predicting the data of the Nth period, using the data of the first three marked time points of the predicted time point as the model training starting point [ X ]N-3,XN-2,XN-1]；
Setting a self-updating weight coefficient alpha, and preferentially selecting empirical golden ratio coefficients 0.618 and 0.382 as values of abnormal and stable time series at an initial value;
step four, establishing a coefficient value array F [ m, n ], and storing an initial value of alpha, wherein m is a predicted charging pile, and n is a predicted time point;
step five, collecting the abnormal conditions of the policy, the equipment condition and the seasonal variation interference when the data are fluctuated severely, carrying out single-hot coding, and carrying out correlation analysis on the power consumption at the occurrence time point to obtain a correlation coefficient P, wherein the formula is as follows:
wherein x is the electricity consumption at the time point of occurrence of the abnormal condition, y is the one-hot coded value of the abnormal condition, and xμ、yμThe mean values of x and y are respectively, and the delta x and the delta y are respectively the standard deviation of x and y;
step six, storing the obtained correlation coefficient P into F [ m, n ] as a condition for self-updating data, and performing self-updating processing on the data predicted at the later stage;
step seven, setting a predicted time length T and determining a self-updating processed charging pile power consumption time point;
step eight, calculating to obtain a predicted value, predicting and judging whether to perform self-updating, specifically judging whether abnormal data which enable the sequence to be unstable exist in the predicted time length T, if so, judging that the self-updating is needed, otherwise, judging that the self-updating is not needed, wherein the calculation formula of the predicted value is as follows:
xt+T＝At+BtT+CtT2
At＝3St (1)-3St (2)+St (3)
wherein, XN-3,XN-2,XN-1Updating the three predicted parameters for predicting the data at the first three time points of the data every time the predicted value is calculated; constant At、Bt、CtThe coefficient is an exponential smoothing coefficient of the t period, and is different in each prediction due to self-updating of the coefficient and the parameter;
respectively, the first, second and third exponential smoothing values, alpha, of the t-th periodm,nIs a weight coefficient under m and n, namely a correlation coefficient P, xtIs the actual value of the current t period, xt+TThe predicted value of the future T + T period is obtained;
calculating the mean square error and the relative error between the predicted value and the actual value, verifying the exponential smoothing coefficient alpha, and then performing iteration, wherein the formula is as follows:
St＝xt+T
wherein s istTo predict value, xtThe smaller the mean square error is, the higher the accuracy is; and delta is an actual relative error, usually output by percentage, represents an absolute error between a true value and a predicted value of the power consumption, and L is true power consumption data, wherein the smaller the relative error value is, the higher the prediction accuracy is.
Step ten, if the mean square error and the relative error are not satisfied, resetting the traversal searched interval to 0.01, calculating the relative error, selecting a weight coefficient alpha satisfying the relative error below 0.1, wherein the traversal searched object is alpha, if the error value corresponding to the updated alpha obtained by the correlation coefficient does not satisfy the set error threshold, trying to calculate +0.01 every time to traverse the new alpha until the corresponding relative error satisfies the set optimization threshold condition;
step eleven, predicting the power consumption based on the weight coefficient alpha obtained finally from the updating;
step twelve, calculating the prediction result again through the evaluation index, and outputting the prediction result and the evaluation index result for analysis.
In conclusion, the technical scheme can effectively analyze the influence of the internal correlation and the external influence factors of the time sequence on the power utilization condition of the charging pile, generate the corresponding correlation coefficient according to the influence, improve the prediction effect through continuous iteration, reduce the prediction error caused by abnormal fluctuation of the sequence, and provide data support for construction planning of electric power companies.
The self-updating cubic exponential smoothing method provided by the application is improved one by one aiming at the defects that the smoothing coefficient of the traditional three-order model is static and fixed, the fluctuation inflection point caused by time change cannot be found, the external factors existing along with the time change are not considered, and the influence of historical data at different times on future predicted data cannot be predicted, so that the capability of processing abnormal data is increased to improve the reliability of prediction. The self-updating cubic exponential smoothing overcomes the defect that the traditional method cannot accurately predict in a medium-long term, so that the error and mean square deviation are reduced in the medium-long term prediction of complex data.
Claims (10)
1. A charging pile electricity consumption prediction method based on a self-updating cubic exponential smoothing method is characterized by comprising the following steps:
s1, acquiring historical time sequence electricity consumption data of the charging pile, and collecting abnormal conditions;
s2, obtaining the influence degree of the abnormal conditions on the power consumption of the charging pile through correlation analysis based on the historical time sequence power consumption data and the collected abnormal conditions, and displaying the influence degree by using a correlation coefficient to obtain a correlation coefficient;
s3, automatically updating the smoothing coefficient in the cubic exponential smoothing method by using the correlation coefficient, and determining the optimal smoothing coefficient through precision comparison before and after updating;
and S4, performing medium and long term prediction on the power consumption of the charging pile based on a cubic exponential smoothing method using the optimal smoothing coefficient to obtain a corresponding prediction result.
2. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 1, wherein the abnormal conditions are policy information data, equipment data and seasonal variation interference data when data fluctuates sharply.
3. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 1, wherein the step S2 is specifically to perform correlation analysis by using a pearson correlation coefficient.
4. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 3, wherein the specific process of the step S2 is as follows: and performing independent hot coding based on the collected abnormal conditions, and performing correlation analysis on the power consumption data of the occurrence time point corresponding to the abnormal conditions by combining the historical time series power consumption data to obtain a correlation coefficient.
5. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 4, wherein the correlation coefficient is specifically as follows:
wherein x is the electricity consumption at the time point of occurrence of the abnormal condition, y is the one-hot coded value of the abnormal condition, xμ、yμThe mean values of x and y, respectively, and the standard deviations of x and y, respectively.
6. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 5, wherein the step S3 specifically comprises the following steps:
s31, conducting preliminary prediction on the historical time series electricity consumption data, and when the data of the Nth period is predicted, using the data of the first three marked time points of the predicted time point as the update training starting point [ X ]N-3，XN-2，XN-1]；
S32, setting training iteration times and an initial value of a self-updating smoothing coefficient alpha;
s33, creating a coefficient value array F [ m, n ], and storing an initial value of alpha, wherein m is a predicted charging pile, and n is a predicted time point;
s34, storing the correlation coefficient into Fm, n to be used as a condition of self-updating data, and carrying out self-updating processing on the data predicted at the later stage;
s35, setting the predicted time length T, and determining the time point of the electric quantity used by the charging pile through self-updating processing;
s36, calculating to obtain a predicted value, judging whether self-updating is needed or not, if so, updating alpha, and then executing a step S37;
otherwise, calculating a correlation coefficient, and then executing step S37;
s37, calculating an error value between the predicted value and the actual value, if the error value is smaller than a set error threshold value, indicating that the current alpha is the optimal smooth coefficient, otherwise, executing the step S38;
and S38, resetting the distance of the traversal search alpha to 0.01, recalculating the error value between the predicted value and the actual value, and selecting the alpha corresponding to the error value below the set optimization threshold value to serve as the optimal smoothing coefficient.
7. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 6, wherein the smoothing coefficients are weight coefficients specifically, and include weight coefficients corresponding to time series stability and weight coefficients corresponding to time series abnormity.
8. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 6, wherein the formula for calculating the predicted value in the step S36 is as follows:
xt+T＝At+BtT+CtT2
At＝3St (1)-3St (2)+St (3)
wherein, XN-3，XN-2，XN-1For predicting data at the first three time points of the data, the three predicted parameters are updated every time the predicted value is calculated, and the constant At、Bt、CtThe coefficients are exponentially smoothed for the t-th cycle, and are different for each prediction due to the self-updating of the coefficients and parameters,
respectively, the first, second and third exponential smoothing values, alpha, of the t-th periodm，nIs a weight coefficient under m and n, namely a correlation coefficient P, xtIs the actual value of the current t period, xt+TIs the predicted value of the future T + T period.
9. The method as claimed in claim 6, wherein the error value in step S37 includes a mean square error and a relative error, and the error value in step S38 is specifically a relative error.
10. The method for predicting the power consumption of the charging pile based on the self-updating cubic exponential smoothing method as claimed in claim 9, wherein the calculation formula of the mean square error is as follows:
st＝xt+T
the calculation formula of the relative error is as follows:
wherein s istTo predict value, xiThe smaller the mean square error is, the higher the accuracy is; and delta is an actual relative error, usually output by percentage, represents an absolute error between a true value and a predicted value of the power consumption, and L is true power consumption data, wherein the smaller the relative error value is, the higher the prediction accuracy is.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111465442.XA CN114330834A (en) | 2021-12-03 | 2021-12-03 | Charging pile power consumption prediction method based on self-updating cubic exponential smoothing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111465442.XA CN114330834A (en) | 2021-12-03 | 2021-12-03 | Charging pile power consumption prediction method based on self-updating cubic exponential smoothing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114330834A true CN114330834A (en) | 2022-04-12 |
Family
ID=81049569
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111465442.XA Pending CN114330834A (en) | 2021-12-03 | 2021-12-03 | Charging pile power consumption prediction method based on self-updating cubic exponential smoothing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114330834A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117171503A (en) * | 2023-11-03 | 2023-12-05 | 深圳市深创高科电子有限公司 | Intelligent prediction method for battery charging time |
| CN117312803A (en) * | 2023-11-29 | 2023-12-29 | 广州市悦到信息科技有限公司 | Digital processing method and system for cloud cooperative passenger transport service |
| CN117849652A (en) * | 2024-03-07 | 2024-04-09 | 华星动力(江苏)有限公司 | Charging and discharging detection method and detection system for charging pile |
| CN117848985A (en) * | 2024-03-06 | 2024-04-09 | 杭州泽天春来科技股份有限公司 | Gas concentration analysis method and device |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105761489A (en) * | 2016-04-13 | 2016-07-13 | 南京邮电大学 | Cubic exponential smoothing optimal method of traffic flow prediction |
-
2021
- 2021-12-03 CN CN202111465442.XA patent/CN114330834A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105761489A (en) * | 2016-04-13 | 2016-07-13 | 南京邮电大学 | Cubic exponential smoothing optimal method of traffic flow prediction |
Non-Patent Citations (2)
| Title |
|---|
| 蔡松桓: "充电桩数据挖掘与充电量预测算法研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》, vol. 2021, no. 02, 15 February 2021 (2021-02-15), pages 034 - 701 * |
| 黄伟建: "动态三次指数平滑法的火电厂发电量预测研究", 《现代电子技术》, vol. 43, no. 17, 1 September 2020 (2020-09-01), pages 147 - 150 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117171503A (en) * | 2023-11-03 | 2023-12-05 | 深圳市深创高科电子有限公司 | Intelligent prediction method for battery charging time |
| CN117171503B (en) * | 2023-11-03 | 2024-03-19 | 深圳市深创高科电子有限公司 | Intelligent prediction method for battery charging time |
| CN117312803A (en) * | 2023-11-29 | 2023-12-29 | 广州市悦到信息科技有限公司 | Digital processing method and system for cloud cooperative passenger transport service |
| CN117312803B (en) * | 2023-11-29 | 2024-03-15 | 广州市悦到信息科技有限公司 | Digital processing method and system for cloud cooperative passenger transport service |
| CN117848985A (en) * | 2024-03-06 | 2024-04-09 | 杭州泽天春来科技股份有限公司 | Gas concentration analysis method and device |
| CN117849652A (en) * | 2024-03-07 | 2024-04-09 | 华星动力(江苏)有限公司 | Charging and discharging detection method and detection system for charging pile |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114330834A (en) | Charging pile power consumption prediction method based on self-updating cubic exponential smoothing method | |
| CN110674604B (en) | Transformer DGA data prediction method based on multi-dimensional time sequence frame convolution LSTM | |
| US10331810B2 (en) | Method for determining a model of an output quantity of a technical system | |
| CN112098848A (en) | Battery discharge capacity prediction method and system and readable storage medium | |
| CN115169818B (en) | Technical engineering measuring and calculating method and system based on digital modeling | |
| CN112418921A (en) | Power demand prediction method, device, system and computer storage medium | |
| CN112381279B (en) | Wind power prediction method based on VMD and BLS combined model | |
| CN115495991A (en) | Rainfall interval prediction method based on time convolution network | |
| CN113449919B (en) | Power consumption prediction method and system based on feature and trend perception | |
| CN115270239A (en) | Bridge reliability prediction method based on dynamic characteristics and intelligent algorithm response surface method | |
| CN114066699A (en) | Carbon emission measuring and calculating method and device and terminal equipment | |
| CN116728783B (en) | Simulation method and system based on 3D printer | |
| CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
| CN114548494B (en) | Visual cost data prediction intelligent analysis system | |
| CN113297744B (en) | Charging pile data cleaning method suitable for error monitoring calculation and charging station | |
| CN116316600A (en) | Wind power day-ahead prediction self-adaptive filtering method based on seasonal typing | |
| CN116561569A (en) | Industrial power load identification method based on EO feature selection and AdaBoost algorithm | |
| CN109886495A (en) | Electric grid investment prediction technique based on gray model | |
| CN113722308B (en) | Acceleration response data complement method and device based on EEMD-MultiCNN-LSTM | |
| CN115545490A (en) | Power station engineering cost analysis method and device based on small sample machine learning | |
| CN115222114A (en) | Automobile part assembly index value prediction method, terminal device and storage medium | |
| CN114254801A (en) | Industrial user power consumption long-term prediction method and device based on similarity measurement | |
| CN111985721A (en) | Enterprise invoicing amount prediction method based on seasonal time sequence | |
| CN117977584B (en) | Power load probability prediction method, system, medium, device and program product | |
| CN115828736B (en) | EEMD-PE-LSTM-based short-time ship traffic flow prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |