CN117035285B - Method for guiding user to change power based on real-time recommendation mode on big data line - Google Patents
Method for guiding user to change power based on real-time recommendation mode on big data line Download PDFInfo
- Publication number
- CN117035285B CN117035285B CN202310916500.9A CN202310916500A CN117035285B CN 117035285 B CN117035285 B CN 117035285B CN 202310916500 A CN202310916500 A CN 202310916500A CN 117035285 B CN117035285 B CN 117035285B
- Authority
- CN
- China
- Prior art keywords
- user
- battery
- site
- big data
- real
- 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
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000006243 chemical reaction Methods 0.000 claims abstract description 23
- 230000005611 electricity Effects 0.000 claims abstract description 11
- 238000007405 data analysis Methods 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 8
- 230000001360 synchronised effect Effects 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000004806 packaging method and process Methods 0.000 claims description 2
- 238000007619 statistical method Methods 0.000 abstract description 2
- 238000013500 data storage Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012217 deletion Methods 0.000 description 3
- 230000037430 deletion Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
- G06F16/275—Synchronous replication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0208—Trade or exchange of goods or services in exchange for incentives or rewards
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Computing Systems (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for guiding a user to change electricity based on a real-time recommendation mode on a big data line, which comprises the steps of collecting the electricity consumption of the user, a driving path, batteries and information of a battery changing cabinet, and recording the battery inventory condition of each battery changing cabinet station; according to the big data analysis, finding out user groups with low power consumption and battery stock-rich power conversion sites in each hour; establishing a model for the users according to the collected data, wherein the model of each user comprises expected rewards and confidence intervals of each site; for each site of each user, calculating a confidence upper bound as a selection basis; selecting a site with the highest confidence upper bound as a recommended site for a user who cannot exchange full electricity in a period of time; short-chain contacts are generated to guide the user to change power. The battery replacement cabinet station with low electric quantity and rich battery inventory can be taken out per hour according to big data statistical analysis, and the battery replacement cabinet station is recommended to the user optimal station in real time to replace the battery near the frequent battery replacement cabinet or near the driving path.
Description
Technical Field
The invention belongs to the technical field of power conversion of a power conversion cabinet, and particularly relates to a method for guiding a user to perform power conversion based on a real-time recommendation mode on a large data line.
Background
In the two-wheel vehicle battery replacement industry, the battery idle rate is higher, and traditional manual selection battery replacement station's mode wastes time and energy, can't accomplish the higher battery replacement cabinet of intelligent guide user's nearby battery electric quantity, has increased the battery idle rate and can't satisfy the user and trade the electric quantity demand. In order to solve the problems, the invention develops a method for guiding a user to change electricity based on a real-time recommendation mode on a big data line based on big data and artificial intelligence technology.
Disclosure of Invention
The invention aims to: in order to overcome the defects in the prior art, the invention provides a method for guiding a user to change electricity based on a real-time recommendation mode on a big data line, which can be used for taking out user groups with low electric quantity and battery stock-rich electricity changing cabinet stations every hour according to big data statistical analysis, and recommending the battery to the optimal stations of the user to change the battery in real time near the frequently-removed electricity changing cabinet or near a driving path.
The technical scheme is as follows: in order to achieve the above purpose, the method for guiding the user to change the power based on the real-time recommendation mode on the big data line comprises the following specific steps:
step S1: collecting the power consumption, the driving path, the batteries and the battery changing cabinet information of a user, and recording the battery inventory condition of each battery changing cabinet station;
step S2: storing the acquired information into a large database in a real-time access or off-line synchronous mode;
step S3: according to the big data analysis, finding out user groups with low power consumption and battery stock-rich power conversion sites in each hour;
step S4: based on the collected data, a model is built for the users, each user's model containing the expected rewards and confidence intervals for each site:
for each user u and site i, the expected reward is denoted as E, the expected reward being an estimate of the probability that user i gets full power on site u;
for each user u and site i, the confidence interval is denoted CI, the confidence interval being an uncertainty range estimate for the desired reward;
step S5: initializing an initial expected reward E and a confidence interval CI for each user for each site:
for user u, if going to station i, setting the initial expected reward to be a proper fixed value;
for user u, if site i is not visited, the initial confidence interval is set to be positive infinity, indicating uncertainty of initial reward estimation;
step S6: for each site i of each user u, calculating a confidence upper bound as a selection basis;
step S7: after the user changes the electricity, calculating actual rewards according to whether the user successfully changes to full electricity, and updating rewards estimated values and confidence intervals of the recommended sites by the user by using a Bayesian theorem;
step S8: for users who can not change full power in a period of time, selecting a site with the highest confidence upper bound as a recommended site;
step S9: selecting a site with the highest confidence upper bound as a potential high-availability site pushed to a user, namely a recommended site, packaging a link of site information through a short-chain service, generating a short link, and enabling the short-chain contact to reach the user in a push and short message combined mode;
step S10: and the user opens a short link at the handheld terminal point, reaches a designated site according to the guidance of the short link page information, and guides the user to perform power conversion operation.
Further, in step S1, the information content of the collected battery includes a battery number, longitude, latitude, electric quantity percentage and current capacity, and the information content of the collected battery exchange cabinet includes a cabinet number, a cabinet grid number, operation time, operation type, battery number, operation user ID and battery electric quantity.
Further, in the big data analysis of step S3, the data needs to be preprocessed, where the preprocessing includes removing duplicate data, processing missing values, processing outliers, data format conversion, feature selection, feature conversion, and data set division.
Further, the data set division includes division of training sets and test sets, specifically: the cleaned dataset is divided into a training set and a testing set for model training and evaluation.
Further, after data preprocessing, data analysis is performed, and the analysis process is as follows:
a) Calculating the average value of the power taken out per hour: according to the user ID and the time stamp, calculating the average value of the power taken out per hour so as to find out the user group with low power taking out;
b) Calculating a statistical index of battery inventory: for each battery-changing cabinet station, calculating the average value, the median and the total statistical index of the battery inventory to find out the station with rich inventory.
Further, in step S6, the calculation formula of the confidence upper bound is:
UCB(t) = E(t) + c * sqrt(ln(t) / N(t))
where UCB (t) is the confidence upper bound for selecting the station at time step t, E (t) is the expected reward estimate for the station at time step t, N (t) is the number of times the station was selected at time step t, and c is the exploration factor used to control the extent of exploration.
Further, in step S7, uncertainty of the desired prize, denoted Beta, is represented by Beta distribution, where a and b are parameters of the distribution;
observing the number of times of successfully changing to full power from the data is k, and the number of times of failure is n-k, wherein n is the total number of times;
according to the bayesian theorem, a posterior distribution is calculated to update the estimate of the expected rewards: parameters of the posterior distribution are a '=a+k and b' =b+ (n-k), the updated posterior distribution is used to calculate an estimate of the desired prize, the mean or median of the distribution is used;
the estimated value of the expected prize is e=a '/(a ' +b ');
confidence interval ci= [ q_low, q_high ], where q_low and q_high are the lower and upper quantiles, respectively, of the posterior distribution.
Further, in step S9, the push notification is a main notification channel, the short message notification is a supplementary notification channel, and the short message notification can improve the visibility and the arrival rate of the notification.
The beneficial effects are that: according to the invention, the battery replacement cabinet sites with low electric quantity and rich battery inventory can be taken out per hour according to big data statistics analysis, and the battery replacement cabinet sites are recommended to the user optimal sites in real time near the frequently-removed battery replacement cabinet or near the driving path, so that the scheduling efficiency is improved, the electric quantity removed by the user is increased, and the battery idling rate is reduced.
Drawings
FIG. 1 is a block diagram of the method steps of the invention;
FIG. 2 is a block diagram of a real-time access scheme in data storage;
FIG. 3 is a block diagram of an offline synchronization scheme in data storage;
FIG. 4 is a schematic diagram of user expectations and confidence intervals at different sites.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
As shown in fig. 1, the method for guiding the user to change power based on the real-time recommendation mode on the big data line comprises the following specific steps:
step S1: and collecting the power consumption, the driving path, the batteries and the battery changing cabinet information of the user, and recording the battery inventory condition of each battery changing cabinet station.
In step S1, the information content of the collected battery includes a battery number, a longitude, a latitude, an electric quantity percentage, and a current capacity, and the information content of the collected battery exchange cabinet includes a cabinet number, a cabinet grid number, an operation time, an operation type, a battery number, an operation user ID, and a battery electric quantity.
Step S2: and storing the acquired information into a large database in a real-time access or off-line synchronous mode.
Data storage mode one: the real-time access, as shown in figure 2,
data format:
| attributes of | Paraphrasing meaning | Remarks |
| imei | IEMI of device | |
| protocol_type | Reporting protocol type | |
| message | Device reporting content | Unprocessed messages |
| topic | Device heartbeat source |
(1) The data center confirms the reporting protocol and the message channel through developing and configuring the Flink task;
(2) The data center receives the push data of the corresponding kafka;
(3) The data center stores the original data;
(4) And the data center analyzes the received message according to a preset protocol to finish the storage process.
And a second data storage mode: offline synchronization, as shown in figure 3,
configuring a data source and a target source: the synchronous data source and the target source are configured in the configuration center, and the synchronous data source and the target source comprise data source types, connection information, table names and the like.
Defining a synchronization task: according to the service requirement, defining the data source and target source to be synchronized, data column, synchronous start-stop position, etc.
Generating a data synchronization script: and generating a corresponding data synchronization script according to the task definition.
Performing data synchronization: and executing data synchronization operation according to the generated script, and synchronizing the appointed source data into the target data source.
Step S3: and according to the big data analysis, finding out the user group with low power consumption and the battery stock-rich battery replacement site.
When analyzing the big data in the step S3, the data needs to be preprocessed, wherein the preprocessing comprises the steps of removing repeated data, processing missing values, processing abnormal values, data format conversion, feature selection, feature conversion and data set division.
1. Duplicate data is removed: checking whether repeated records exist in the data, if so, performing a deduplication operation, and reserving a unique record.
2. Processing the missing values: checking whether there are missing values in the data, the missing values may be processed by the following strategy:
2.1, deletion of missing values: if the proportion of missing values is small, the record containing the missing values may be selected for deletion.
2.2, filling in missing values: and filling the missing values by adopting an interpolation method according to the types and data distribution of the features and adopting a mean value, a median value and a mode value.
3. Processing outliers: checking whether there are outliers in the data, the outliers can be processed using the following method:
3.1, deleting abnormal values: if the number of outliers is small, the record containing outliers may be selected for deletion;
3.2, replacing outliers: and replacing the abnormal value with a reasonable value according to the distribution characteristics of the data.
4. Data format conversion: the data is converted into a format suitable for analysis, and the date and time field is converted into a time stamp or a specific time format.
5. Feature selection:
5.1, selecting characteristics related to the target: selecting features related to the amount of power taken out per hour and battery inventory, such as user ID, time stamp, etc., according to the analysis target;
5.2, eliminating irrelevant features: features that are present independent or redundant of the analysis targets are culled to reduce data dimensionality and complexity.
6. Feature conversion:
6.1, timestamp conversion: if the time stamps are provided in a different format or granularity, they are converted to time stamps of the required hours level;
6.2, feature coding: for classification variables, tag coding is used to convert them into a numerical representation for use in subsequent analysis.
7. Data set partitioning:
the data set division comprises the steps of dividing a training set and a testing set, and specifically comprises the following steps: the cleaned dataset is divided into a training set and a testing set for model training and evaluation.
After data preprocessing, data analysis is carried out, and the analysis process is as follows:
a) Calculating the average value of the power taken out per hour: according to the user ID and the time stamp, calculating the average value of the power taken out per hour so as to find out the user group with low power taking out;
b) Calculating a statistical index of battery inventory: for each battery-changing cabinet station, calculating the average value, the median and the total statistical index of the battery inventory to find out the station with rich inventory.
Step S4: as shown in FIG. 4, based on the collected data, a model is built for users, each user's model containing the expected rewards and confidence intervals for each site:
for each user u and site i, the expected reward is denoted as E (u, i), which is an estimate of the probability that user i gets full power on site u;
for each user u and site i, the confidence interval is denoted CI (u, i), the confidence interval being an uncertainty range estimate for the desired reward;
step S5: initializing an initial expected reward E (u, i) and a confidence interval CI (u, i) for each user for each site:
for user u, if going to station i, setting the initial expected reward to be a proper fixed value;
for user u, if site i is not visited, the initial confidence interval is set to be positive infinity, indicating uncertainty of initial reward estimation;
step S6: for each site i for each user u, a confidence upper bound is calculated as a basis for selection.
In step S6, the calculation formula of the confidence upper bound is:
UCB(t) = E(t) + c * sqrt(ln(t) / N(t))
where UCB (t) is the confidence upper bound for selecting the station at time step t, E (t) is the expected reward estimate for the station at time step t, N (t) is the number of times the station was selected at time step t, and c is the exploration factor used to control the extent of exploration.
Step S7: and after the user changes the power, calculating actual rewards according to whether the user successfully changes to full power, and updating the rewards estimated value and the confidence interval of the user on the recommended site by using the Bayesian theorem.
In step S7, the uncertainty of the desired prize is represented by a Beta distribution, denoted Beta (a, b), where a and b are parameters of the distribution;
observing the number of times of successfully changing to full power from the data is k, and the number of times of failure is n-k, wherein n is the total number of times;
according to the bayesian theorem, a posterior distribution is calculated to update the estimate of the expected rewards: parameters of the posterior distribution are a '=a+k and b' =b+ (n-k), the updated posterior distribution is used to calculate an estimate of the desired prize, the mean or median of the distribution is used;
the estimated value of the expected prize is E (u, i) =a '/(a ' +b ');
confidence interval CI (u, i) = [ q_low, q_high ], where q_low and q_high are the lower and upper quantiles, respectively, of the posterior distribution.
Step S8: for users who can not change full power in a period of time, selecting a site with the highest confidence upper bound as a recommended site;
step S9: the site with the highest confidence upper bound is selected as a potential high-availability site pushed to a user, namely a recommended site, the link of the site information is packaged through short-chain service, a short link is generated, and the short link is contacted with the user in a push and short message combined mode.
In step S9, the push notification is a main notification channel, the short message notification is a supplementary notification channel, and the short message notification can improve the visibility and the arrival rate of the notification.
push notification: and the push notification function in the application is used for sending the push content and the short chains to the user, wherein the push content comprises recommended sites, estimated acquired electric quantity and guidance for checking details by using the short chains.
And (3) short message notification: the short message can be used as a supplementary channel to send a notification of the power change to the user, including the abstract and short chain of the push content.
Step S10: and the user opens a short link at the handheld terminal point, reaches a designated site according to the guidance of the short link page information, and guides the user to perform power conversion operation.
The user opens the application and performs the power change operation, and the details are as follows:
navigating to a power conversion page: when a user clicks a short chain in a push notification or a short message, the application can directly navigate the user to a power conversion page in the application so as to reduce the operation steps of the user and improve the user experience;
(ii) providing a clear interface and guidance: on the power conversion page, the application interface displays nearby power conversion sites, the residual capacity of the sites, the distance, navigation options and the like, so that a user can conveniently select a proper power conversion site and quickly arrive through a navigation function;
(III) guiding a user to perform power change operation: the application interface displays the idle grid of the site, indicates the correct use mode of the user, provides operation feedback and ensures that the user successfully performs power change operation.
The advantages of the invention are summarized as follows:
1) The battery replacement cabinet sites with low electric quantity and rich battery inventory can be taken out per hour according to big data statistics analysis, and the battery replacement can be recommended to the user optimal sites in real time near the frequently-removed battery replacement cabinet or near the driving path, so that the dispatching efficiency is improved;
2) The average power taken out by the user is improved by about 10%, and more optimal power exchange stations are recommended in real time, so that the user can be effectively guided to complete more battery replacement, and the use experience and effect of the user are improved;
3) The idle battery of the station of the battery changing cabinet is reduced by about 5%, the battery changing process is optimized, the idle rate of the battery is reduced, and the efficiency and the service level of the two-wheel vehicle battery changing industry are improved.
The foregoing is only a preferred embodiment of the invention, it being noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the invention.
Claims (8)
1. The method for guiding the user to change the power based on the real-time recommendation mode on the big data line is characterized by comprising the following steps of: the method comprises the following specific steps:
step S1: collecting the power consumption of a user, a driving path, batteries and battery changing cabinet information, and recording the battery inventory condition of each battery changing cabinet station;
step S2: storing the acquired information into a large database in a real-time access or off-line synchronous mode;
step S3: according to the big data analysis, finding out user groups with low power consumption and battery stock-rich power conversion sites in each hour;
step S4: based on the collected data, a model is built for the users, each user's model containing the expected rewards and confidence intervals for each site:
for each user u and site i, the expected reward is denoted as E (u, i), which is an estimate of the probability that user u gets full power on site i;
for each user u and site i, the confidence interval is denoted CI (u, i), the confidence interval being an uncertainty range estimate for the desired reward;
step S5: initializing the expected rewards E (u, i) and confidence intervals CI (u, i) of each user for each site:
for user u, if the user goes to station i, setting the initial expected reward as a fixed value;
for user u, if site i is not visited, setting the initial confidence interval to be positive infinity, and indicating uncertainty of initial expected reward estimation;
step S6: for each site i of each user u, calculating a confidence upper bound as a selection basis;
step S7: after the user changes the electricity, calculating actual rewards according to whether the user successfully changes to full electricity, and updating expected rewards estimated values and confidence intervals of the user on the recommended sites by using a Bayesian theorem;
step S8: for users who can not change full power in a period of time, selecting a site with the highest confidence upper bound as a recommended site;
step S9: selecting a site with the highest confidence upper bound as a potential high-availability site pushed to a user, namely a recommended site, packaging a link of site information through a short-chain service, generating a short link, and enabling the short-chain contact to reach the user in a push and short message combined mode;
step S10: and the user opens a short link at the handheld terminal point, reaches a designated site according to the guidance of the short link page information, and guides the user to perform power conversion operation.
2. The method for guiding the user to change power based on the real-time recommendation mode on the big data line according to claim 1, wherein the method is characterized in that: in step S1, the information content of the collected battery includes a battery number, a longitude, a latitude, an electric quantity percentage, and a current capacity, and the information content of the collected battery exchange cabinet includes a cabinet number, a cabinet grid number, an operation time, an operation type, a battery number, an operation user ID, and a battery electric quantity.
3. The method for guiding the user to change power based on the real-time recommendation mode on the big data line according to claim 1, wherein the method is characterized in that: when analyzing the big data in the step S3, the data needs to be preprocessed, wherein the preprocessing comprises the steps of removing repeated data, processing missing values, processing abnormal values, data format conversion, feature selection, feature conversion and data set division.
4. The method for guiding a user to change power based on real-time recommendation mode on a big data line according to claim 3, wherein the method comprises the following steps: the data set division comprises the steps of dividing a training set and a testing set, and specifically comprises the following steps: the cleaned dataset is divided into a training set and a testing set for model training and evaluation.
5. The method for guiding a user to change power based on real-time recommendation mode on a big data line according to claim 3, wherein the method comprises the following steps: after data preprocessing, data analysis is carried out, and the analysis process is as follows:
a) Calculating the average value of the power taken out per hour: according to the operation user ID and the time stamp, calculating the average value of the power taken out per hour so as to find out the user group with low power;
b) Calculating a statistical index of battery inventory: for each battery-changing cabinet station, calculating the average value, the median and the total statistical index of the battery inventory to find out the station with rich inventory.
6. The method for guiding the user to change power based on the real-time recommendation mode on the big data line according to claim 1, wherein the method is characterized in that: in step S6, the calculation formula of the confidence upper bound is:
UCB(t) = E(t) + c * sqrt(ln(t) / N(t))
where UCB (t) is the confidence upper bound for selecting the station at time step t, E (t) is the expected reward estimate for the station at time step t, N (t) is the number of times the station was selected at time step t, and c is the exploration factor used to control the extent of exploration.
7. The method for guiding the user to change power based on the real-time recommendation mode on the big data line according to claim 1, wherein the method is characterized in that: in step S7, the uncertainty of the desired prize is represented by a Beta distribution, denoted Beta (a, b), where a and b are parameters of the distribution;
observing the number of times of successfully changing to full power from the data is k, and the number of times of failure is n-k, wherein n is the total number of times;
according to the bayesian theorem, a posterior distribution is calculated to update the estimate of the expected rewards: parameters of the posterior distribution are a '=a+k and b' =b+ (n-k), the updated posterior distribution is used to calculate an estimate of the desired prize, the mean or median of the distribution is used;
the estimated value of the expected prize is E (u, i) =a '/(a ' +b ');
confidence interval CI (u, i) = [ q_low, q_high ], where q_low and q_high are the lower and upper quantiles, respectively, of the posterior distribution.
8. The method for guiding the user to change power based on the real-time recommendation mode on the big data line according to claim 1, wherein the method is characterized in that: in step S9, the push notification is a main notification channel, and the short message notification is a supplementary notification channel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310916500.9A CN117035285B (en) | 2023-07-25 | 2023-07-25 | Method for guiding user to change power based on real-time recommendation mode on big data line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310916500.9A CN117035285B (en) | 2023-07-25 | 2023-07-25 | Method for guiding user to change power based on real-time recommendation mode on big data line |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117035285A CN117035285A (en) | 2023-11-10 |
| CN117035285B true CN117035285B (en) | 2024-04-12 |
Family
ID=88601429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310916500.9A Active CN117035285B (en) | 2023-07-25 | 2023-07-25 | Method for guiding user to change power based on real-time recommendation mode on big data line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117035285B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619091A (en) * | 2019-08-14 | 2019-12-27 | 深圳易马达科技有限公司 | Method for recommending power exchange cabinet and terminal equipment |
| CN111602159A (en) * | 2018-03-20 | 2020-08-28 | 本田技研工业株式会社 | Server and management system |
| CN112477635A (en) * | 2020-11-30 | 2021-03-12 | 浙江吉利控股集团有限公司 | Method, device and equipment for supplementing electric quantity of battery and storage medium |
| CN114692927A (en) * | 2020-12-31 | 2022-07-01 | 奥动新能源汽车科技有限公司 | Battery replacement station recommendation method and system, electronic device and storage medium |
| CN115359646A (en) * | 2022-08-18 | 2022-11-18 | 网电楚创智慧能源湖北有限公司 | Battery replacement vehicle scheduling method and system based on vehicle-mounted terminal |
| CN115431830A (en) * | 2022-08-04 | 2022-12-06 | 广东易积网络股份有限公司 | Express enterprise-oriented battery replacement system and method and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10288439B2 (en) * | 2017-02-22 | 2019-05-14 | Robert D. Pedersen | Systems and methods using artificial intelligence for routing electric vehicles |
-
2023
- 2023-07-25 CN CN202310916500.9A patent/CN117035285B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111602159A (en) * | 2018-03-20 | 2020-08-28 | 本田技研工业株式会社 | Server and management system |
| CN110619091A (en) * | 2019-08-14 | 2019-12-27 | 深圳易马达科技有限公司 | Method for recommending power exchange cabinet and terminal equipment |
| CN112477635A (en) * | 2020-11-30 | 2021-03-12 | 浙江吉利控股集团有限公司 | Method, device and equipment for supplementing electric quantity of battery and storage medium |
| CN114692927A (en) * | 2020-12-31 | 2022-07-01 | 奥动新能源汽车科技有限公司 | Battery replacement station recommendation method and system, electronic device and storage medium |
| CN115431830A (en) * | 2022-08-04 | 2022-12-06 | 广东易积网络股份有限公司 | Express enterprise-oriented battery replacement system and method and storage medium |
| CN115359646A (en) * | 2022-08-18 | 2022-11-18 | 网电楚创智慧能源湖北有限公司 | Battery replacement vehicle scheduling method and system based on vehicle-mounted terminal |
Non-Patent Citations (1)
| Title |
|---|
| 基于强化学习的电动汽车换电站实时调度策略优化;张文昕 等;《电力自动化设备》;20221031;第42卷(第10期);第134-141页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117035285A (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8983671B2 (en) | Data collecting apparatus and data collecting method | |
| CN103747523A (en) | User position predicating system and method based on wireless network | |
| CN115640914A (en) | Intelligent gas storage optimization method, internet of things system, device and medium | |
| CN103731916A (en) | Wireless-network-based user position predicting system and method | |
| Shipman et al. | We got the power: Predicting available capacity for vehicle-to-grid services using a deep recurrent neural network | |
| Unterluggauer et al. | Short‐term load forecasting at electric vehicle charging sites using a multivariate multi‐step long short‐term memory: A case study from Finland | |
| CN109146129A (en) | Charging pile group recommended method can use stake prediction, result acquisition methods and controller | |
| CN117035285B (en) | Method for guiding user to change power based on real-time recommendation mode on big data line | |
| CN116599160B (en) | Active sensing method and system for new energy station cluster and new energy station | |
| CN112766599A (en) | Intelligent operation and maintenance method based on deep reinforcement learning | |
| CN116090702B (en) | ERP data intelligent supervision system and method based on Internet of things | |
| CN115860278A (en) | Motor assembly production management method and system based on data analysis | |
| CN114143822B (en) | Flow management method, operation management platform, charging pile and storage medium | |
| CN101441454A (en) | Low-frequency unloading and fault circuit tagmeme on-line monitoring implementing method | |
| CN112560325B (en) | Prediction method, system, equipment and storage medium for electricity conversion service | |
| CN113642248A (en) | Method and device for evaluating remaining service time of positioning equipment | |
| CN112800102A (en) | Alarm correlation calculation method and device and calculation equipment | |
| CN112925831A (en) | Big data mining method and big data mining service system based on cloud computing service | |
| CN116993085B (en) | Method for improving battery utilization efficiency based on charge replacement consumption algorithm | |
| CN112895967B (en) | Method, system, medium, and device for predicting remaining service time of battery replacement mileage | |
| CN117853274B (en) | New energy automobile fills electric pile operation data analysis management system based on big data | |
| CN115963408B (en) | Energy storage power station single battery fault early warning system and method | |
| CN117913996B (en) | Intelligent monitoring management method and system for operation of power distribution cabinet based on data analysis | |
| CN107465531A (en) | Power generation dispatching method and device | |
| CN116451828A (en) | Method for adaptively diagnosing zero low-power charging pile |
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 |