CN117852691A - Cloud edge cooperative bus load prediction method and system - Google Patents
Cloud edge cooperative bus load prediction method and system Download PDFInfo
- Publication number
- CN117852691A CN117852691A CN202311750021.0A CN202311750021A CN117852691A CN 117852691 A CN117852691 A CN 117852691A CN 202311750021 A CN202311750021 A CN 202311750021A CN 117852691 A CN117852691 A CN 117852691A
- Authority
- CN
- China
- Prior art keywords
- bus load
- load prediction
- model
- data
- cloud
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000011156 evaluation Methods 0.000 claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 7
- 238000013210 evaluation model Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000007637 random forest analysis Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000005096 rolling process Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000007726 management method Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 238000005192 partition Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008140 language development Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of data processing, and particularly relates to a cloud edge cooperative bus load prediction method and a cloud edge cooperative bus load prediction system, wherein the method comprises the steps of constructing a dynamic modeling evaluation feature set and screening out an influence factor set of bus load; calculating the bus load prediction accuracy; an evaluation mechanism of dynamic modeling evaluation is constructed, and a data acquisition and data uploading function is realized at the side end; the collected bus load data and weather information data are sent to a cloud server, and data processing is achieved at the cloud; and receiving the processed data at the edge, and operating a bus load prediction model to perform reasoning so as to obtain a bus load prediction result. According to the cloud edge cooperative bus load prediction method based on the dynamic modeling evaluation mechanism, the calculated amount caused by repeated training of the bus load prediction model can be effectively reduced, the timely and efficient capability of collecting the edge data is fully exerted through cloud edge cooperation, and the timeliness of prediction is improved.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a cloud edge cooperative bus load prediction method and system.
Background
The bus load prediction can be used for power grid dynamic state evaluation, power grid safety check, reactive power optimization, energy-saving power generation scheduling and other aspects, and is an important supporting means for improving the refinement and intelligent level of power grid scheduling. The number of buses is large, a traditional bus load prediction model is trained and constructed in a daily rolling mode, the model training process is long in time consumption, and a large amount of calculation force is needed. The bus load prediction reasoning process requires little calculation power and has higher requirements on reasoning performance.
Disclosure of Invention
In order to solve or improve the problems, the invention provides a cloud edge cooperative bus load prediction method and a cloud edge cooperative bus load prediction system, and the specific technical scheme is as follows:
the invention provides a cloud edge cooperative bus load prediction method, which comprises the following steps:
s1, constructing a dynamic modeling evaluation feature set, screening out an influence factor set of bus load, and marking the influence factor set as { f } i -respectively calculate f i The median, maximum and minimum values of the last 24 hours are expressed asCalculating f i The median, maximum and minimum of the last 3 days are expressed as +.>Calculating f i The median, maximum and minimum of the last 7 days are expressed as +.>
S2, calculating the bus load prediction accuracy of the date t, using the obtained actual influence factor value of the date t at the date t+1, respectively using models obtained through t-7, t-3 and t-1 training to infer and output bus load prediction results of the date t, respectively calculating the accuracy of 3 models, and marking as p t-7 、p t-3 、p t-1 ；
S3, constructing an evaluation mechanism for dynamic modeling evaluation, wherein Splicing the two groups into a matrix M, constructing a label set, and respectively calculating p t-7 And p t-3 Whether or not to be goodAt p t-1 If the model is superior to the model, the model is marked as 1, otherwise, the model is marked as 0, the label set is marked as L, and the matrix M and the label set L are trained by adopting a random forest algorithm to obtain an evaluation model E for dynamic modeling evaluation;
step S4, realizing data acquisition and data uploading functions at the side end, uploading the acquired bus load data and weather information data to a cloud server, and realizing data processing at the cloud;
and S5, receiving the processed data at the edge, and operating a bus load prediction model to perform reasoning so as to obtain a bus load prediction result.
Preferably, the step S1 includes: historical data of the last 3 years is selected, a dynamic modeling evaluation feature set is constructed, and the temperature (f 1 ) Humidity (f) 2 ) Wind speed (f) 3 ) Rainfall (f) 4 ) Irradiation value (f) 5 ) Cloud (f) 6 ) User production plan (f) 7 ) Power grid outage planning (f) 8 ) As an influencing factor of the bus load; and calculates the corresponding f i Median, maximum, minimum, f for the last 24 hours i Median, maximum, minimum, f for the last 3 days i The median, maximum, minimum of the last 7 days, and save the data to the database table tb_feature by day.
Preferably, the step S2 includes: selecting historical data of the last 3 years, calculating the accuracy of daily bus load prediction, respectively calculating the accuracy of bus load prediction results obtained by reasoning through t-7, t-3 and t-1 training models, calculating the accuracy of 3 models, and recording as p t-7 、p t-3 、p t-1 And storing the accuracy data to a database table tb_accuracy according to days.
Preferably, the step S3 includes: constructing an evaluation mechanism of dynamic modeling evaluation, and respectively calculating p based on a database table tb_accuracy obtained by S2 t-7 And p t-3 Whether or not to be better than p t-1 A list of saved results is newly added in tb_accuracy, wherein the calculation rule is as follows: if the value is better, the value is marked as 1, otherwise, the value is marked as 0;
and executing connection query on the tb_feature and the tb_accuracy based on the date to form a data set S, dividing the S into a training set and a testing set, and training the data set by adopting a random forest algorithm to obtain a dynamic modeling evaluation model E.
Preferably, the step S4 includes: the Java language is adopted to develop the functions of data acquisition and data uploading, the Java language is deployed at the side end, and the acquired bus load data and weather information data are uploaded to the cloud server.
Preferably, the step S4 includes: and (3) adopting the evaluation model E to evaluate whether rolling training is needed, if the model is needed to be retrained, executing operations such as model training, model downloading and the like, completing construction of the bus load prediction model based on cloud computing resources, and downloading the constructed model to the side.
Preferably, said and underfilling the built model to the marginal end comprises: and (5) mounting the built model to the edge in a mode of a Docker mirror image.
Preferably, based on a side Docker container management platform, the receiving of a Docker load prediction model is realized, and the bus load prediction model is operated to perform reasoning so as to obtain a bus load prediction result.
The invention provides a cloud edge cooperative bus load prediction system, which comprises:
a first module for executing step S1, constructing dynamic modeling evaluation feature set, screening out influence factor set of bus load, and recording as { f } i -respectively calculate f i The median, maximum and minimum values of the last 24 hours are expressed asCalculating f i The median, maximum and minimum of the last 3 days are expressed as +.>Calculating f i The median, maximum and minimum of the last 7 days are expressed as +.>
A second module for executing step S2, calculating the accuracy of bus load prediction of date t, using the obtained actual influence factor value of date t at date t+1, respectively using models obtained by training t-7, t-3 and t-1 to make reasoning and output the bus load prediction result of date t, respectively calculating the accuracy of 3 models, and recording as p t-7 、p t-3 、p t-1 ；
A third module for executing step S3 to construct an evaluation mechanism for dynamic modeling evaluationSplicing the two groups into a matrix M, constructing a label set, and respectively calculating p t-7 And p t-3 Whether or not to be better than p t-1 If the model is superior to the model, the model is marked as 1, otherwise, the model is marked as 0, the label set is marked as L, and the matrix M and the label set L are trained by adopting a random forest algorithm to obtain an evaluation model E for dynamic modeling evaluation;
the fourth module is used for executing the step S4, realizing the functions of data acquisition and data uploading at the side end, uploading the acquired bus load data and weather information data to a cloud server, and realizing data processing at the cloud;
and a fifth module, configured to execute step S5, receive the processed data at the edge, and operate the bus load prediction model to perform reasoning, so as to obtain a bus load prediction result.
The beneficial effects of the invention are as follows: according to the cloud edge cooperative bus load prediction method based on the dynamic modeling evaluation mechanism, the calculated amount caused by repeated training of the bus load prediction model can be effectively reduced, the timely and efficient capability of collecting the edge data is fully exerted through cloud edge cooperation, and the timeliness of prediction is improved.
Drawings
FIG. 1 is a schematic diagram of a cloud-edge cooperative bus load prediction mechanism based on a dynamic modeling evaluation mechanism according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
In order to solve the technical problems mentioned in the background, the invention provides a cloud edge cooperative bus load prediction method, which comprises the following steps:
s1, constructing a dynamic modeling evaluation feature set, screening out an influence factor set of bus load, and marking the influence factor set as { f } i -respectively calculate f i The median, maximum and minimum values of the last 24 hours are expressed asCalculating f i The median, maximum and minimum of the last 3 days are expressed as +.>Calculating f i Recently 7The median, maximum and minimum of the day are expressed as +.>
S2, calculating the bus load prediction accuracy of the date t, using the obtained actual influence factor value of the date t at the date t+1, respectively using models obtained through t-7, t-3 and t-1 training to infer and output bus load prediction results of the date t, respectively calculating the accuracy of 3 models, and marking as p t-7 、p t-3 、p t-1 ；
S3, constructing an evaluation mechanism for dynamic modeling evaluation, wherein Splicing the two groups into a matrix M, constructing a label set, and respectively calculating p t-7 And p t-3 Whether or not to be better than p t-1 If the model is superior to the model, the model is marked as 1, otherwise, the model is marked as 0, the label set is marked as L, and the matrix M and the label set L are trained by adopting a random forest algorithm to obtain an evaluation model E for dynamic modeling evaluation;
step S4, realizing data acquisition and data uploading functions at the side end, uploading the acquired bus load data and weather information data to a cloud server, and realizing data processing at the cloud;
and S5, receiving the processed data at the edge, and operating a bus load prediction model to perform reasoning so as to obtain a bus load prediction result.
The step S1 includes: historical data of the last 3 years is selected, a dynamic modeling evaluation feature set is constructed, and the temperature (f 1 ) Humidity (f) 2 ) Wind speed (f) 3 ) Rainfall (f) 4 ) Irradiation value (f) 5 ) Cloud (f) 6 ) User production plan (f) 7 ) Power grid outage planning (f) 8 ) As an influencing factor of the bus load; and calculates the corresponding f i Median, maximum, minimum of the last 24 hours,f i median, maximum, minimum, f for the last 3 days i The median, maximum, minimum of the last 7 days, and save the data to the database table tb_feature by day.
The step S2 includes: selecting historical data of the last 3 years, calculating the accuracy of daily bus load prediction, respectively calculating the accuracy of bus load prediction results obtained by reasoning through t-7, t-3 and t-1 training models, calculating the accuracy of 3 models, and recording as p t-7 、p t-3 、p t-1 And storing the accuracy data to a database table tb_accuracy according to days.
The step S3 includes: constructing an evaluation mechanism of dynamic modeling evaluation, and respectively calculating p based on a database table tb_accuracy obtained by S2 t-7 And p t-3 Whether or not to be better than p t-1 A list of saved results is newly added in tb_accuracy, wherein the calculation rule is as follows: if the value is better, the value is marked as 1, otherwise, the value is marked as 0;
and executing connection query on the tb_feature and the tb_accuracy based on the date to form a data set S, dividing the S into a training set and a testing set, and training the data set by adopting a random forest algorithm to obtain a dynamic modeling evaluation model E.
The step S4 includes: the Java language is adopted to develop the functions of data acquisition and data uploading, the Java language is deployed at the side end, and the acquired bus load data and weather information data are uploaded to the cloud server.
The step S4 includes: and (3) adopting the evaluation model E to evaluate whether rolling training is needed, if the model is needed to be retrained, executing operations such as model training, model downloading and the like, completing construction of the bus load prediction model based on cloud computing resources, and downloading the constructed model to the side.
The method for underfilling the constructed model to the edge comprises the following steps: and (5) mounting the built model to the edge in a mode of a Docker mirror image.
Based on the side-end Docker container management platform, receiving a Docker load prediction model is realized, and the bus load prediction model is operated to perform reasoning so as to obtain a bus load prediction result.
The invention provides a cloud edge cooperative bus load prediction system, which comprises:
a first module for executing step S1, constructing dynamic modeling evaluation feature set, screening out influence factor set of bus load, and recording as { f } i -respectively calculate f i The median, maximum and minimum values of the last 24 hours are expressed asCalculating f i The median, maximum and minimum of the last 3 days are expressed as +.>Calculating f i The median, maximum and minimum of the last 7 days are expressed as +.>
A second module for executing step S2, calculating the accuracy of bus load prediction of date t, using the obtained actual influence factor value of date t at date t+1, respectively using models obtained by training t-7, t-3 and t-1 to make reasoning and output the bus load prediction result of date t, respectively calculating the accuracy of 3 models, and recording as p t-7 、p t-3 、p t-1 ；
A third module for executing step S3 to construct an evaluation mechanism for dynamic modeling evaluationSplicing the two groups into a matrix M, constructing a label set, and respectively calculating p t-7 And p t-3 Whether or not to be better than p t-1 If the model is superior to the model, the model is marked as 1, otherwise, the model is marked as 0, the label set is marked as L, and the matrix M and the label set L are trained by adopting a random forest algorithm to obtain an evaluation model E for dynamic modeling evaluation;
the fourth module is used for executing the step S4, realizing the functions of data acquisition and data uploading at the side end, uploading the acquired bus load data and weather information data to a cloud server, and realizing data processing at the cloud;
and a fifth module, configured to execute step S5, receive the processed data at the edge, and operate the bus load prediction model to perform reasoning, so as to obtain a bus load prediction result.
Examples
The invention provides a cloud edge cooperative bus load prediction mechanism based on a dynamic modeling evaluation mechanism as shown in fig. 1:
the system comprises two ends, a cloud end and an edge end, wherein the edge end performs data acquisition and data uploading; carrying out data processing, feature engineering, modeling evaluation and model training on the cloud; and (5) continuing model receiving at the edge, predicting the bus load, and outputting a prediction result.
The invention provides a cloud edge cooperative bus load prediction method based on a dynamic modeling evaluation mechanism, which comprises the following steps:
step S1, a dynamic modeling evaluation feature set is constructed, and the specific process is as follows:
firstly, acquiring bus load prediction related data of the last 3 years from a cloud big data platform. Comprises bus active power, temperature (f 1 ) Humidity (f) 2 ) Wind speed (f) 3 ) Rainfall (f) 4 ) Irradiation value (f) 5 ) Cloud (f) 6 ) User production plan (f) 7 ) Power grid outage planning (f) 8 ) Etc. Writing data processing SQL, and respectively calculating f 1 f 2 f 3 f 4 f 5 f 6 f 7 f 8 Median, maximum, minimum for the last 24 hours, the last 3 days, the last 7 days. The calculation is performed by taking the bus as an object and taking the natural day as a partition, and the calculation result is stored in the database table tb_feature by taking the bus and the date as a primary key.
Step S2: calculating the prediction accuracy of the daily bus load, wherein the specific process is as follows:
and taking the bus as an object and taking the natural day as a partition, respectively calculating the load prediction accuracy of each bus in each day, calculating the accuracy by using the reasoning results of the 3 models of t-7, t-3 and t-1, and storing the accuracy data into a database table tb_accuracy according to the day. The accuracy calculation formula is as follows:
wherein (1)>
Wherein p is i For the daily comprehensive prediction precision of the bus i, E i,k The deviation rate is predicted for the load of bus i in period k. L'. i,k For the load predictive value of the bus load i in the period k, L i,k The actual load value of the single busbar load i in the period k. L is a load reference value, and the value rule is as follows:
L i, k>20MW，L＝L i, k；L i, k<=20mw, l=main transformer capacity 50%;
step S3: an evaluation mechanism for dynamic modeling evaluation is constructed, and the specific process is as follows:
based on database table tb_accuracy obtained in S2, p is calculated respectively t-7 And p t-3 Whether or not to be better than p t-1 A new column of save results is added at tb_accuracy. The calculation rule is as follows: if better then 1 is noted, otherwise 0 is noted. The join query is performed on tb_feature, tb_accuracy based on date to construct the dataset S.
And dividing the S into a training set and a testing set, and training the data set by adopting a random forest algorithm to obtain a dynamic modeling evaluation model E. The random forest algorithm partial super parameters are set as follows:
| n_estimators | 300 |
| criterion | gini |
| max_depth | 15 |
| min_samples_split | 5 |
| min_samples_leaf | 5 |
step S4: based on the e file format, a Java language development data acquisition function and a cloud OSS service development data uploading function are adopted. The method comprises the steps of deploying at a side end, and sending collected busbar load data, meteorological information data and the like to a cloud OSS server;
step S5: based on the cloud big data platform, data processing is achieved, and whether rolling training is needed or not is evaluated by adopting an evaluation model E in the step S3. If the model needs to be retrained, performing model training, model downloading and other operations, completing the construction of a bus load prediction model based on cloud computing resources, and downloading the constructed model to an edge in a dock mirror mode;
step S6: based on the edge-end dock container management platform, receiving of a dock load prediction model is achieved, and reasoning is conducted on the operation load prediction model to obtain a bus load prediction result.
Those of ordinary skill in the art will appreciate that the elements of the examples described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or combinations of both, and that the elements of the examples have been described generally in terms of functionality in the foregoing description to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in this application, it should be understood that the division of units is merely a logic function division, and there may be other manners of division in practical implementation, for example, multiple units may be combined into one unit, one unit may be split into multiple units, or some features may be omitted.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention, and are intended to be included within the scope of the appended claims and description.
Claims (9)
1. The cloud edge cooperative bus load prediction method is characterized by comprising the following steps of:
s1, constructing a dynamic modeling evaluation feature set, screening out an influence factor set of bus load, and marking the influence factor set as { f } i -respectively calculate f i The median, maximum and minimum values of the last 24 hours are expressed asCalculating f i The median, maximum and minimum of the last 3 days are expressed as +.>Calculating f i The median, maximum and minimum of the last 7 days are expressed as +.>
S2, calculating the bus load prediction accuracy of the date t, and using the obtained data at the date t+1The actual value of the influence factor from the date t is used as input, load prediction models of t-7, t-3 and t-1 are used for predicting bus loads of the date t, and the accuracy of the 3 model prediction values is calculated and recorded as p t-7 、p t-3 、p t-1 ；
S3, constructing an evaluation mechanism for dynamic modeling evaluation, wherein Splicing the two groups into a matrix M, constructing a label set, and respectively calculating p t-7 And p t-3 Whether or not to be better than p t-1 If the model is superior to the model, the model is marked as 1, otherwise, the model is marked as 0, the label set is marked as L, and the matrix M and the label set L are trained by adopting a random forest algorithm to obtain an evaluation model E for dynamic modeling evaluation;
step S4, realizing data acquisition and data uploading functions at the side end, uploading the acquired bus load data and weather information data to a cloud server, and realizing data processing at the cloud;
and S5, receiving the processed data at the edge, and operating a bus load prediction model to perform reasoning so as to obtain a bus load prediction result.
2. The cloud edge cooperative bus load prediction method according to claim 1, wherein the step S1 includes:
selecting historical data of the last 3 years, constructing a dynamic modeling evaluation feature set, and obtaining the temperature f 1 Humidity f 2 Wind speed f 3 Rainfall f 4 Irradiation value f 5 Cloud quantity f 6 User production plan f 7 Power grid outage plan f 8 As an influencing factor of the bus load;
and calculates the corresponding f i Median, maximum, minimum, f for the last 24 hours i Median, maximum, minimum, f for the last 3 days i Median of last 7 daysMaximum, minimum, and save the data to database table tb_feature by day.
3. The cloud edge cooperative bus load prediction method according to claim 2, wherein the step S2 includes:
selecting historical data of the last 3 years, calculating the accuracy of daily bus load prediction, respectively calculating the accuracy of bus load prediction results obtained by reasoning through t-7, t-3 and t-1 training models, calculating the accuracy of 3 models, and recording as p t-7 、p t-3 、p t-1 And storing the accuracy data to a database table tb_accuracy according to days.
4. The cloud edge cooperative bus load prediction method according to claim 3, wherein the step S3 includes:
constructing an evaluation mechanism of dynamic modeling evaluation, and respectively calculating p based on a database table tb_accuracy obtained by S2 t-7 And p t-3 Whether or not to be better than p t-1 A list of saved results is newly added in tb_accuracy, wherein the calculation rule is as follows: if the value is better, the value is marked as 1, otherwise, the value is marked as 0;
and executing connection query on the tb_feature and the tb_accuracy based on the date to form a data set S, dividing the S into a training set and a testing set, and training the data set by adopting a random forest algorithm to obtain a dynamic modeling evaluation model E.
5. The cloud edge cooperative bus load prediction method according to claim 4, wherein the step S4 includes:
the Java language is adopted to develop the functions of data acquisition and data uploading, the Java language is deployed at the side end, and the acquired bus load data and weather information data are uploaded to the cloud server.
6. The cloud edge cooperative bus load prediction method according to claim 5, wherein the step S4 includes:
and (3) adopting the evaluation model E to evaluate whether rolling training is needed, if the model is needed to be retrained, executing operations such as model training, model downloading and the like, completing construction of the bus load prediction model based on cloud computing resources, and downloading the constructed model to the side.
7. The cloud-edge collaborative bus load prediction method according to claim 6, wherein the downloading the constructed model to an edge comprises:
and (5) mounting the built model to the edge in a mode of a Docker mirror image.
8. The cloud-edge collaborative bus load prediction method according to claim 7 is characterized in that based on an edge-side Docker container management platform, receiving of a Docker load prediction model is achieved, and the bus load prediction model is operated for reasoning to obtain a bus load prediction result.
9. Cloud edge cooperative bus load prediction system, which is characterized by comprising:
a first module for executing step S1, constructing dynamic modeling evaluation feature set, screening out influence factor set of bus load, and recording as { f } i -respectively calculate f i The median, maximum and minimum values of the last 24 hours are expressed asCalculating f i The median, maximum and minimum of the last 3 days are expressed as +.>Calculating f i The median, maximum and minimum of the last 7 days are expressed as +.>
A second module for executing step S2 and calculating bus load prediction accuracy of date t, and using the obtained date at date t+1Actual influence factor values of period t are respectively inferred by using models obtained through t-7, t-3 and t-1 training, bus load prediction results of date t are output, and accuracy of 3 models is respectively calculated and recorded as p t-7 、p t-3 、p t-1 ；
A third module for executing step S3 to construct an evaluation mechanism for dynamic modeling evaluationSplicing the two groups into a matrix M, constructing a label set, and respectively calculating p t-7 And p t-3 Whether or not to be better than p t-1 If the model is superior to the model, the model is marked as 1, otherwise, the model is marked as 0, the label set is marked as L, and the matrix M and the label set L are trained by adopting a random forest algorithm to obtain an evaluation model E for dynamic modeling evaluation;
the fourth module is used for executing the step S4, realizing the functions of data acquisition and data uploading at the side end, uploading the acquired bus load data and weather information data to a cloud server, and realizing data processing at the cloud;
and a fifth module, configured to execute step S5, receive the processed data at the edge, and operate the bus load prediction model to perform reasoning, so as to obtain a bus load prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311750021.0A CN117852691A (en) | 2023-12-19 | 2023-12-19 | Cloud edge cooperative bus load prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311750021.0A CN117852691A (en) | 2023-12-19 | 2023-12-19 | Cloud edge cooperative bus load prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117852691A true CN117852691A (en) | 2024-04-09 |
Family
ID=90545716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311750021.0A Pending CN117852691A (en) | 2023-12-19 | 2023-12-19 | Cloud edge cooperative bus load prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117852691A (en) |
-
2023
- 2023-12-19 CN CN202311750021.0A patent/CN117852691A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhu et al. | Using a self-adaptive grey fractional weighted model to forecast Jiangsu’s electricity consumption in China | |
| Wang et al. | Forecasting China's oil consumption: a comparison of novel nonlinear-dynamic grey model (GM), linear GM, nonlinear GM and metabolism GM | |
| Andresen et al. | Validation of Danish wind time series from a new global renewable energy atlas for energy system analysis | |
| WO2021213192A1 (en) | Load prediction method and load prediction system employing general distribution | |
| CN108921358B (en) | Prediction method, prediction system and related device of power load characteristics | |
| CN101551884A (en) | A fast CVR electric load forecast method for large samples | |
| CN111191811A (en) | Cluster load prediction method and device and storage medium | |
| Xiao et al. | Short-term power load interval forecasting based on nonparametric Bootstrap errors sampling | |
| CN114372360A (en) | Method, terminal and storage medium for power load prediction | |
| CN104199870A (en) | Method for building LS-SVM prediction model based on chaotic search | |
| Du et al. | Prediction of electricity consumption based on GM (1, Nr) model in Jiangsu province, China | |
| Zeng et al. | Forecasting the total energy consumption in China using a new-structure grey system model | |
| CN115481918A (en) | Active sensing and predictive analysis system for unit state based on source network load storage | |
| Cai et al. | Gray wolf optimization-based wind power load mid-long term forecasting algorithm | |
| CN110991768A (en) | Electric quantity prediction method and system | |
| CN114881528A (en) | BIM-based assembly type building quality safety control method, system, equipment and medium | |
| Kim et al. | Tutorial on time series prediction using 1D-CNN and BiLSTM: A case example of peak electricity demand and system marginal price prediction | |
| Shang et al. | A novel model based on multiple input factors and variance reciprocal: application on wind speed forecasting | |
| CN112348352B (en) | Big data analysis-based automatic generation method for electric power budget proposal scheme | |
| WO2024082836A1 (en) | Optimization method for comprehensive benefit evaluation scheme for water-wind-photovoltaic energy storage multi-energy complementary system | |
| CN117852691A (en) | Cloud edge cooperative bus load prediction method and system | |
| CN110991748A (en) | Short-term load prediction method for urban power grid | |
| CN113837486B (en) | RNN-RBM-based distribution network feeder long-term load prediction method | |
| Pang et al. | A photovoltaic power predicting model using the differential evolution algorithm and multi-task learning | |
| Sun et al. | A new paradigm for simulating and forecasting China’s economic growth in the medium and long term |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |