CN111897800A - Electric vehicle charging facility recommendation method and system based on federal learning - Google Patents
Electric vehicle charging facility recommendation method and system based on federal learning Download PDFInfo
- Publication number
- CN111897800A CN111897800A CN202010778276.8A CN202010778276A CN111897800A CN 111897800 A CN111897800 A CN 111897800A CN 202010778276 A CN202010778276 A CN 202010778276A CN 111897800 A CN111897800 A CN 111897800A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- pile
- data set
- characteristic data
- federal
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000012549 training Methods 0.000 claims abstract description 51
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 230000015654 memory Effects 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 230000004927 fusion Effects 0.000 abstract description 4
- 239000000284 extract Substances 0.000 description 6
- 230000002776 aggregation Effects 0.000 description 5
- 238000004220 aggregation Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000003442 weekly effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 101100217298 Mus musculus Aspm gene Proteins 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
Images
Classifications
-
- 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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
- G06F16/212—Schema design and management with details for data modelling support
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L53/00—Methods of charging batteries, specially adapted for electric vehicles; Charging stations or on-board charging equipment therefor; Exchange of energy storage elements in electric vehicles
- B60L53/60—Monitoring or controlling charging stations
- B60L53/66—Data transfer between charging stations and vehicles
-
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- 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
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F15/00—Coin-freed apparatus with meter-controlled dispensing of liquid, gas or electricity
- G07F15/003—Coin-freed apparatus with meter-controlled dispensing of liquid, gas or electricity for electricity
- G07F15/005—Coin-freed apparatus with meter-controlled dispensing of liquid, gas or electricity for electricity dispensed for the electrical charging of vehicles
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/7072—Electromobility specific charging systems or methods for batteries, ultracapacitors, supercapacitors or double-layer capacitors
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02T90/10—Technologies relating to charging of electric vehicles
- Y02T90/12—Electric charging stations
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02T90/10—Technologies relating to charging of electric vehicles
- Y02T90/16—Information or communication technologies improving the operation of electric vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Water Supply & Treatment (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Fuzzy Systems (AREA)
- Mechanical Engineering (AREA)
- Transportation (AREA)
- Power Engineering (AREA)
- Public Health (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses an electric vehicle charging facility recommendation method and system based on federal learning, wherein the method comprises the following steps: acquiring a pile characteristic data set and a vehicle characteristic data set, wherein the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to charging pile operation, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle enterprise data platform data and are related to vehicle charging; carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model; acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in a list of charging piles to be recommended; and inputting the current pile characteristic data set and the current vehicle characteristic data set into a federal recommendation model to obtain a recommendation result. By implementing the method and the system, the fusion use of the pile operation data platform data and the vehicle-enterprise data platform data is realized, the model characteristics are expanded, and the personalized recommendation of the charging facilities is realized.
Description
Technical Field
The invention relates to the technical field of Internet of things, in particular to a federal learning-based electric vehicle charging facility recommendation method and system.
Background
With the development of new energy technology, the market share of electric vehicles is higher and higher, and the number of charging facilities is continuously increased. However, it is statistical that the utilization rate of charging service facilities of each operator is currently less than 15%. Therefore, each charging facility operator starts to build an own operation data platform, accesses the pile side data and provides charging facility recommendation service for the user, but the current state analysis and recommendation of the charging facility mainly depend on the pile side data and do not consider the individual requirements of the user. In addition, the vehicle enterprises with the electric vehicle user side data are not related to the requirements of charging facility recommendation and the like for the user data use, so that the pile side operation data platform cannot know the user data of all vehicle users. Therefore, the existing recommendation method for providing the charging facility for the user is difficult to realize the personalized recommendation requirement of the user on the charging facility.
Disclosure of Invention
Therefore, the technical problem to be solved by the invention is to overcome the defect that the charging facility recommendation method in the prior art is difficult to realize the personalized recommendation requirement of the user on the charging facility, so that the electric vehicle charging facility recommendation method and system based on federal learning are provided.
In a first aspect, an embodiment of the present invention provides an electric vehicle charging facility recommendation method based on federal learning, including:
acquiring a pile characteristic data set and a vehicle characteristic data set, wherein the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to charging pile operation, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle-enterprise data platform data and are related to vehicle charging; carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model; acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in a list of charging piles to be recommended; and inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain a recommendation result.
In an embodiment, the pile characteristics include user identifiers, the car characteristics include car identifiers, and the federal recommendation algorithm is used to perform federal training on the pile characteristic data set and the car characteristic data set to obtain a federal recommendation model, including: carrying out entity alignment on a user identifier in the pile characteristics and a vehicle identifier in the vehicle characteristics corresponding to the same charging process to obtain aligned pile characteristics and vehicle characteristics; updating the pile feature data set by using the aligned pile features, and updating the vehicle feature data set by using the aligned vehicle features; and carrying out federal training on the updated pile characteristic data set and the updated vehicle characteristic data set to obtain the federal recommendation model.
In an embodiment, the federal training is performed on the updated pile characteristic data set and the vehicle characteristic data set to obtain the federal recommendation model, and the method includes: step S231: according to the characteristic association relation, the pile characteristic data set is decomposed into pile self-owned characteristics and cross characteristics, and the vehicle characteristic data set is decomposed into vehicle self-owned characteristics and cross characteristics; step S232: decomposing the cross feature into a pile-side feature and a car-side feature; step S233: respectively calculating a pre-estimation value, a loss function value and a vehicle characteristic gradient corresponding to the vehicle self-characteristic and the vehicle side characteristic by using a preset federal recommendation model, and respectively calculating a loss function value and a pile characteristic gradient corresponding to the pile self-characteristic and the pile side characteristic according to the pre-estimation value; step S234: and updating the parameters of the preset federal recommended model according to the vehicle characteristic gradient and the pile characteristic gradient, and returning to the step S233 until the preset training requirement is met.
In an embodiment, the inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain a recommendation result includes: inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain predicted values corresponding to all charging piles in a list of the charging piles to be recommended; sorting the predicted values from large to small; and obtaining the recommendation result according to the sorting result.
In a second aspect, an embodiment of the present invention provides an electric vehicle charging facility recommendation system based on federal learning, including: the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a pile characteristic data set and a vehicle characteristic data set, the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to the operation of a charging pile, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle-enterprise data platform data and are related to the charging of a vehicle; the model training module is used for carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model; the second acquisition module is used for acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in the list of the charging piles to be recommended; and the processing module is used for inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain a recommendation result.
In one embodiment, the pile features include a user identification, the car features include a car identification, and the model training module includes: the alignment submodule is used for carrying out entity alignment on the user identification in the pile characteristics and the vehicle identification in the vehicle characteristics corresponding to the same charging process to obtain the aligned pile characteristics and vehicle characteristics; the updating submodule is used for updating the pile characteristic data set by adopting the aligned pile characteristics and updating the vehicle characteristic data set by adopting the aligned vehicle characteristics; and the training submodule is used for carrying out federal training on the updated pile characteristic data set and the updated vehicle characteristic data set to obtain the federal recommendation model.
In one embodiment, the training submodule includes: the first decomposition unit is used for decomposing the pile feature data set into pile self-owned features and cross features according to the feature association relation, and decomposing the vehicle feature data set into vehicle self-owned features and cross features; a second decomposition unit for decomposing the cross feature into a pile-side feature and a vehicle-side feature; the calculation unit is used for calculating a pre-estimation value, a loss function value and a vehicle characteristic gradient corresponding to the vehicle owned characteristic and the vehicle side characteristic respectively by using a preset federal recommendation model, and calculating a loss function value and a pile characteristic gradient corresponding to the pile owned characteristic and the pile side characteristic respectively according to the pre-estimation value; and the updating unit is used for updating the parameters of the preset federal recommended model by the vehicle characteristic gradient and the pile characteristic gradient and returning the parameters to the calculating unit until the preset training requirements are met.
In one embodiment, the processing module includes: the input sub-module is used for inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain the corresponding predicted values of all the charging piles in the list of the charging piles to be recommended; the sorting submodule is used for sorting the predicted values from large to small; and the recommending submodule is used for obtaining the recommending result according to the sorting result.
In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, where computer instructions are stored, and the computer instructions are configured to cause the computer to execute the federal learning electric vehicle charging facility recommendation method according to the first aspect of the embodiment of the present invention.
In a fourth aspect, an embodiment of the present invention provides a computer device, including: the electric vehicle charging facility recommendation method based on federal learning is implemented by executing computer instructions stored in a memory and a processor, wherein the memory and the processor are connected with each other in a communication mode, and the processor executes the computer instructions so as to implement the electric vehicle charging facility recommendation method based on federal learning according to the first aspect of the embodiment of the invention.
The technical scheme of the invention has the following advantages:
according to the electric vehicle charging facility recommendation method based on federal learning, a pile characteristic data set and a vehicle characteristic data set are obtained, wherein the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to charging pile operation, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle enterprise data platform data and are related to vehicle charging; carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model; acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in a list of charging piles to be recommended; and inputting the current pile characteristic data set and the current vehicle characteristic data set into a federal recommendation model to obtain a recommendation result. The federal recommendation algorithm is adopted to conduct federal training on the pile characteristic data set and the vehicle characteristic data set to obtain a federal recommendation model, fusion use of pile operation data platform data and vehicle-enterprise data platform data is achieved, model characteristics are expanded, available relevant dimensions of the recommendation algorithm are enriched, and personalized recommendation of charging facilities is achieved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a specific example of a federal learning-based electric vehicle charging facility recommendation method in an embodiment of the present invention;
FIG. 2 is a related platform related to a federal learning-based electric vehicle charging facility recommendation method in an embodiment of the invention;
FIG. 3 is a flow chart of another specific example of a federal learning-based electric vehicle charging facility recommendation method in an embodiment of the present invention;
FIG. 4 is a flowchart illustrating another exemplary method for federally learned electric vehicle charging facility recommendation in accordance with an embodiment of the present invention;
FIG. 5 is a flowchart illustrating another exemplary method for federally learned electric vehicle charging facility recommendation in accordance with an embodiment of the present invention;
FIG. 6 is a functional block diagram of a specific example of a Federal learning-based electric vehicle charging facility recommendation system in an embodiment of the present invention;
FIG. 7 is a functional block diagram of a specific example of a model training module in an embodiment of the present invention;
FIG. 8 is a functional block diagram of a specific example of a training submodule in an embodiment of the present invention;
FIG. 9 is a functional block diagram of a specific example of a processing module in an embodiment of the present invention;
fig. 10 is a block diagram of a specific example of a computer device according to an embodiment of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it should be noted that the terms "first", "second", and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
The embodiment of the invention provides a federal learning-based electric vehicle charging facility recommendation method, and a specific charging facility is explained by taking a charging pile as an example, which is only taken as an example and is not limited to the example. As shown in fig. 1, the federal learning-based electric vehicle charging facility recommendation method includes the following steps:
step S1: and acquiring a pile characteristic data set and a vehicle characteristic data set, wherein the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to the operation of a charging pile, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle enterprise data platform data and are related to the charging of a vehicle.
In one embodiment, as shown in fig. 2, a platform recommended for an electric vehicle charging facility includes: a pile operation data platform, a vehicle-enterprise data platform and a third party platform. In practical application, the pile characteristic data set and the vehicle characteristic data set can be extracted by the pile operation data platform and the vehicle-enterprise data platform respectively according to platform data of the pile operation data platform and the vehicle-enterprise data platform, so that platform data are prevented from leaking, and data privacy and safety are protected. The pile data operation platform extracts a pile characteristic data set aiming at historical data of the pile data operation platform and is used for supporting state analysis and recommendation of piles, and the vehicle-enterprise data platform extracts a vehicle characteristic data set aiming at historical data of the vehicle-enterprise data platform and is used for supporting personalized recommendation of electric vehicle charging service. The electric vehicle government supervision platform or other related and credible third-party platforms are used as coordinators for managing key and feature aggregation in the subsequent federal learning process. In an embodiment of the present invention, a pile feature data set includes: stake data characteristic, user's characteristic and order the electricity characteristic, the car characteristic data set includes: a vehicle static characteristic, a vehicle charging characteristic, and a driving characteristic.
Specifically, the pile operation data platform mainly extracts user characteristics such as user identification, average weekly charging cost, average weekly charging capacity, latest weekly charging cost, latest weekly charging capacity, pile identification charged for the latest 3 times, user registration place, common charging payment mode, vehicle type, endurance mileage, battery capacity and the like; the charging method comprises the following steps of charging pile identification, average charging times per hour, average charging cost per hour, average charging electric quantity per hour, average charging cost per time, average charging duration per time, longitude and latitude, commissioning time, charging types (direct current and alternating current), parking cost, service cost, identification of affiliated charging stations, power of the charging stations, affiliated areas, catering quantity near the charging stations, physical education facilities quantity, living service quantity, shopping place quantity and other pile data characteristics. In addition, the pile operation data platform also extracts order features, including charging user identification, charging pile identification, charging starting and stopping time, charging time interval, charging electric quantity, charging electric charge, charging service charge and the like.
The vehicle-enterprise data platform mainly extracts vehicle static characteristics including vehicle identification, electric vehicle endurance mileage, vehicle battery capacity, charging interface type, vehicle type, usage (whether operation properties exist) and the like; driving characteristics such as driving habits (whether an air conditioner switch is normally opened), average speed and power; and extracting vehicle charging characteristics (including vehicle identification, fuzzy charging start time, fuzzy charging end time and charging position) based on the vehicle by using the position of the vehicle at the interval of 5s, the charging proportion (charging is normally to SOC), the charging state and the traveling state report information.
And the pile operation data platform also extracts a label from the order characteristics for displaying whether the charging pile is in a charging state. Specifically, the charge order is displayed with a label sample label of 1 as the positive sample of the data set. And generating a negative sample by adopting a heuristic strategy, bringing the pile which is in an idle state in the charging start time and within the range of 3KM around the pile in the charging order into a negative sample set, supplementing the same user identification, user characteristics and order characteristics as the corresponding positive sample, replacing the pile data characteristics with the characteristics of the negative sample pile, and then marking the sample label as 0.
Step S2: and carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model.
In a specific embodiment, as shown in fig. 3, the obtaining of the federal recommended model in step S2 specifically includes the following steps:
step S21: and carrying out entity alignment on the user identification in the pile characteristics and the vehicle identification in the vehicle characteristics corresponding to the same charging process to obtain the aligned pile characteristics and vehicle characteristics.
In the embodiment of the invention, in the federal training process, because the ID of the pile operation data platform is the user identifier and the ID of the vehicle-enterprise data platform is the vehicle identifier, a uniform sample ID needs to be generated in order to realize entity alignment. Specifically, one charge order can be uniquely determined by 4 characteristics of the charge position (longitude and latitude), the charge start time (expiration to minute), and the charge duration (minute), so that the sample id (hid) is generated by using hash algorithms (SM3, SHA1, SHA224, SHA256, SHA384, SHA512, and other hash algorithms, which are only used as an example and not limited thereto, and the peg operation data platform and the vehicle-enterprise data platform are agreed: h (charge start time, longitude, latitude, charge duration). In order to facilitate entity alignment during federal training, the charging start-stop time and the charging duration acquired by the pile operation data platform and the vehicle-enterprise data platform are reserved to the dimension of minutes.
In practical application, the process of entity alignment can be completed through interaction between the two platforms, so that data leakage is further avoided, data security is guaranteed, and efficiency is improved. The pile operation data platform and the vehicle-enterprise data platform respectively have own sample ID sets, and the sample ID sets are assumed to have sample numbers mcp、mvpAnd (4) respectively. The specific entity alignment process is as follows:
firstly, the pile operation data platform generates an RSA key pair, sends the public keys (e, n) to the vehicle-enterprise data platform, and reserves the private keys (d, n). And the pile operation data platform generates a random number list for the IDs in the own sample ID set. Then, carrying out hash operation of encrypting the random number and multiplying the random number by hid: ei=randi e*H(hidi) Wherein E isiAs a result of the operation, HidiFor the ith sample, randiRandom number generated for the ith sample, i e {1,2, … …, mcp}。
Pile operating data platform will EiBack to the mobile enterprise data platform, the mobile enterprise data platform pair EiDecrypting to obtain DiAnd after hashing the own sample ID set, using a private key to sign and then hashing to obtain Fj: fj=H((H(Hidj))d) Wherein HidjFor the jth sample, j ∈ {1,2, … …, mvp}. Then, the vehicle-enterprise data platform calculates the obtained result DiAnd FjAnd transmitting the data back to the pile operating data platform.
Finally, the pile operating data platform will { H (D)i/randi)|i=1,2,……,mcpAnd FjAnd (4) taking the intersection to obtain a sample ID set shared by two sides. The pile operation data platform sends the intersection back to the vehicle-enterprise data platform, and the vehicle-enterprise data platform can also deduce to obtain a common ID set. The two parties can deduce the corresponding relation of the user identification and the vehicle identification according to the common ID set, and then the aligned pile characteristics and the aligned vehicle characteristics can be obtained. By adopting the RSA algorithm and the Hash algorithm, the design realizes the alignment of the safety entities without common type identification, reduces the information interaction times, reduces the possibility of data leakage and ensures the privacy safety of users.
Step S22: and updating the pile characteristic data set by adopting the aligned pile characteristics, and updating the vehicle characteristic data set by adopting the aligned vehicle characteristics.
In the embodiment of the invention, the pile operation data platform reserves the mapping between the Hid list and the user identifier according to the aligned pile characteristics, and the vehicle-enterprise data platform reserves the mapping between the Hid and the vehicle identifier according to the aligned vehicle characteristics, so that the pile characteristic data set updating process and the vehicle characteristic data set updating process are completed.
Step S23: and carrying out federal training on the updated pile characteristic data set and the updated vehicle characteristic data set to obtain a federal recommendation model.
In the embodiment of the present invention, as shown in fig. 4, the federal training is performed on the updated pile characteristic data set and the vehicle characteristic data set to obtain a federal recommendation model, which includes the following steps:
step S231: according to the characteristic association relation, the pile characteristic data set is decomposed into pile self-owned characteristics and cross characteristics, and the vehicle characteristic data set is decomposed into vehicle self-owned characteristics and cross characteristics.
Specifically, the federal learning-based factorization machine decomposes the pile feature data set into pile-forming self-owned features and cross features, and decomposes the vehicle feature data set into vehicle self-owned features and cross features. The pile self-characteristic is the characteristic of the whole pile operation data platform part, for example, the user identification is the pile self-characteristic; the vehicle self-characteristic is a characteristic of the whole vehicle enterprise data platform part, for example, a vehicle identifier is the vehicle self-characteristic; the cross feature is a feature of crossing between two parts of the pile operation data platform and the vehicle-enterprise data platform, and the feature of both the two parts, such as the charging start-stop time, the charging period and the like, is the cross feature. The method has the advantages that the pile feature data set is decomposed into pile-forming self-owned features and cross features by adopting a factor decomposition machine algorithm, and the vehicle feature data set is decomposed into vehicle self-owned features and cross features, so that compared with a single classification algorithm, the learning of the cross features is increased, and the accuracy of a recommendation algorithm is improved.
Step S232: the cross feature is decomposed into a picket-side feature and a car-side feature.
Specifically, the features of the intersection between the two parties are decomposed into pile side features in the part of the pile operation data platform, and the features of the intersection between the two parties are decomposed into car side features in the part of the car-enterprise data platform.
Step S233: and respectively calculating a pre-estimation value, a loss function value and a vehicle characteristic gradient corresponding to the vehicle self-characteristic and the vehicle side characteristic by using a preset federal recommendation model, and respectively calculating a loss function value and a pile characteristic gradient corresponding to the pile self-characteristic and the pile side characteristic according to the pre-estimation value.
Specifically, the vehicle-enterprise data platform calculates the pre-estimation value, the loss function value and the vehicle characteristic gradient of the vehicle-enterprise data platform, and sends the pre-estimation value, the loss function value and the vehicle characteristic gradient to the pile operation data platform, and the pile operation data platform calculates the loss function value and the pile characteristic gradient corresponding to the pile self-characteristic and the pile side characteristic according to the pre-estimation value sent by the vehicle-enterprise data platform. The above-mentioned calculation methods for the prediction value, the loss function and the feature gradient are all the prior art, and are not described herein again.
Step S234: and updating the parameters of the preset federal recommended model according to the vehicle characteristic gradient and the pile characteristic gradient, and returning to the step S233 until the preset training requirement is met.
Specifically, the vehicle-enterprise data platform sends the encrypted vehicle characteristic gradient added with the security aggregation mask to a third-party platform, and the pile operation data platform sends the encrypted pile characteristic gradient added with the security aggregation mask and the loss function value to the third-party platform. And after the third-party platform completes decryption, calculating the vehicle characteristic gradient and the pile characteristic gradient in a gathering manner. And then, adding the updated vehicle characteristic gradient and pile characteristic gradient with corresponding security aggregation masks, and respectively sending the updated vehicle characteristic gradient and pile characteristic gradient to the vehicle enterprise data platform and the pile operation data platform. And (5) updating the model after removing the mask code by the vehicle-enterprise data platform and the pile operation data platform, and returning to the step (S233) until the preset training requirement is met. In the embodiment of the invention, a third-party platform is used for carrying out feature calculation summary and model parameter transmission in a homomorphic encryption and safety aggregation based mode, so that the federal training and the federal reasoning of a recommendation algorithm under the condition that data does not leave the local are realized, and the user privacy is protected.
Wherein, predetermine the training requirement and include: the pile operation data platform calculates that the difference value of the loss functions corresponding to the self-characteristics of the piles of the front and rear wheels and the vehicle side characteristics is smaller than a first preset threshold value; or the number of rounds is greater than the first preset number of rounds for control; or the third-party platform calculates that the difference value of the overall losses of the front wheel and the rear wheel is smaller than a second preset threshold value; or the number of rounds is greater than a second preset number of rounds for control, wherein the first preset threshold, the second preset threshold, the first number of rounds and the second number of rounds may be set according to actual needs, which is not limited in the present invention. And when the model meets the preset training requirement, finishing the training to obtain the final federal recommended model.
Step S3: and acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in the list of the charging piles to be recommended.
In one embodiment, the request for recommendation of the charging pile is typically initiated by a user through the pile operating data platform. Based on the user identification and the current position of the initiating user, the pile operation data platform firstly pulls all the charging piles within a preset range (such as 5KM) to form a list of charging piles to be recommended, and then acquires current pile characteristic data sets and current vehicle characteristic data sets corresponding to all the charging piles in the list of charging piles to be recommended.
Step S4: and inputting the current pile characteristic data set and the current vehicle characteristic data set into a federal recommendation model to obtain a recommendation result.
In a specific embodiment, as shown in fig. 5, inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain a recommendation result includes the following steps:
step S41: and inputting the current pile characteristic data set and the current vehicle characteristic data set into a federal recommendation model to obtain the predicted values corresponding to all the charging piles in the list of the charging piles to be recommended.
In the embodiment of the invention, when the current pile characteristic data set and the current vehicle characteristic data set are input into the federal recommendation model, the pile operation data platform acquires the user identification and the charging pile identification in the charging pile list from the current pile characteristic data set and the current vehicle characteristic data set, calculates the self-owned characteristics and the cross characteristics of the pile by using the federal recommendation model, and encrypts the calculation result to the third-party platform. And simultaneously, the pile operation data platform sends the Hid list corresponding to the user identification to the vehicle-enterprise data platform. And the vehicle-enterprise data platform side calculates the own characteristics and the cross characteristics of the vehicles by using the vehicle characteristics according to the Hid list sent by the pile operation data platform and corresponding to the vehicle identification, and encrypts the calculation result and sends the calculation result to a third-party platform. And the third-party platform collects all characteristic calculations to obtain a predicted value, and encrypts and sends the predicted value to the pile operation data platform.
Step S42: and sorting the predicted values from large to small.
In the embodiment of the invention, the pile operation data platform decrypts the predicted value which is collected and sent by the third-party platform to obtain the recommended predicted value, and if the numerical range [0,1] of the predicted value is assumed, the charging piles in the charging pile list are sorted in the reverse order of the recommended predicted value, and the sorting result is output.
Step S43: and obtaining a recommendation result according to the sorting result.
In the embodiment of the invention, the personalized recommendation result for the user is obtained according to the sorting result, the first ranking result is the best recommendation result, and the user can freely select the charging facilities in the recommendation result list. Meanwhile, N (for example, N is 10) charging piles before sorting may be taken from the sorting result according to the preset number for recommendation, and the like, which is not limited by the present invention.
According to the electric vehicle charging facility recommendation method based on federal learning, a pile characteristic data set and a vehicle characteristic data set are obtained, wherein the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to charging pile operation, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle enterprise data platform data and are related to vehicle charging; carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model; acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in a list of charging piles to be recommended; and inputting the current pile characteristic data set and the current vehicle characteristic data set into a federal recommendation model to obtain a recommendation result. The federal recommendation algorithm is adopted to conduct federal training on the pile characteristic data set and the vehicle characteristic data set to obtain a federal recommendation model, fusion use of pile operation data platform data and vehicle-enterprise data platform data is achieved, model characteristics are expanded, available relevant dimensionalities of the recommendation algorithm are enriched, and personalized recommendation of charging facilities is achieved.
An embodiment of the present invention further provides an electric vehicle charging facility recommendation system based on federal learning, as shown in fig. 6, including:
the first acquisition module 1 is used for acquiring a pile characteristic data set and a vehicle characteristic data set, wherein the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to the operation of a charging pile, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle-enterprise data platform data and are related to the charging of a vehicle. For details, refer to the related description of step S1 in the above method embodiment, and are not described herein again.
And the model training module 2 is used for carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model. For details, refer to the related description of step S2 in the above method embodiment, and are not described herein again.
And the second acquisition module 3 is used for acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in the list of the charging piles to be recommended. For details, refer to the related description of step S3 in the above method embodiment, and are not described herein again.
And the processing module 4 is used for inputting the current pile characteristic data set and the current vehicle characteristic data set into a federal recommendation model to obtain a recommendation result. For details, refer to the related description of step S4 in the above method embodiment, and are not described herein again.
The electric vehicle charging facility recommendation system based on the federal learning obtains a pile characteristic data set and a vehicle characteristic data set by applying an electric vehicle charging facility recommendation method based on the federal learning, wherein the pile characteristic data set is a set formed by pile characteristics related to charging pile operation and extracted from pile operation data platform data, and the vehicle characteristic data set is a set formed by vehicle characteristics related to vehicle charging and extracted from vehicle enterprise data platform data; carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model; acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in a list of charging piles to be recommended; and inputting the current pile characteristic data set and the current vehicle characteristic data set into a federal recommendation model to obtain a recommendation result. The federal recommendation algorithm is adopted to conduct federal training on the pile characteristic data set and the vehicle characteristic data set to obtain a federal recommendation model, fusion use of pile operation data platform data and vehicle-enterprise data platform data is achieved, model characteristics are expanded, available relevant dimensionalities of the recommendation algorithm are enriched, and personalized recommendation of charging facilities is achieved.
In a specific embodiment, the pile characteristics include a user identifier, the car characteristics include a car identifier, and the model training module 2, as shown in fig. 7, includes:
and the alignment submodule 21 is configured to perform entity alignment on the user identifier in the pile feature and the vehicle identifier in the vehicle feature corresponding to the same charging process, so as to obtain an aligned pile feature and vehicle feature. For details, refer to the related description of step S21 in the above method embodiment, and are not described herein again.
And the updating submodule 22 is used for updating the pile characteristic data set by using the aligned pile characteristics and updating the vehicle characteristic data set by using the aligned vehicle characteristics. For details, refer to the related description of step S22 in the above method embodiment, and are not described herein again.
And the training submodule 23 is configured to perform federal training on the updated pile characteristic data set and the updated vehicle characteristic data set to obtain a federal recommendation model. For details, refer to the related description of step S23 in the above method embodiment, and are not described herein again.
In one embodiment, as shown in fig. 8, the training submodule 23 includes:
the first decomposition unit 231 is configured to decompose the pile feature data set into pile own features and cross features according to the feature association relationship, and decompose the vehicle feature data set into vehicle own features and cross features. For details, refer to the related description of step S231 in the above method embodiment, and are not repeated herein.
And a second decomposition unit 232 for decomposing the cross feature into a pile-side feature and a car-side feature. For details, refer to the related description of step S232 in the above method embodiment, and are not repeated herein.
The calculating unit 233 is configured to calculate a pre-estimation value, a loss function value, and a vehicle feature gradient corresponding to the vehicle-owned feature and the vehicle-side feature respectively by using a preset federal recommended model, and calculate a loss function value and a pile feature gradient corresponding to the pile-owned feature and the pile-side feature respectively according to the pre-estimation value. For details, refer to the related description of step S233 in the above method embodiment, and are not repeated herein.
And the updating unit 234 is used for updating the parameters of the preset federal recommended model by the vehicle characteristic gradient and the pile characteristic gradient and returning the parameters to the calculating unit until the preset training requirements are met. For details, refer to the related description of step S234 in the above method embodiment, and are not repeated herein.
In one embodiment, as shown in fig. 9, the processing module 4 includes:
and the input submodule 41 is used for inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain the corresponding predicted values of all the charging piles in the list of the charging piles to be recommended. For details, refer to the related description of step S41 in the above method embodiment, and are not described herein again.
And the sorting submodule 42 is used for sorting the predicted values from large to small. For details, refer to the related description of step S42 in the above method embodiment, and are not described herein again.
And the recommending submodule 43 is used for obtaining a recommending result according to the sorting result. For details, refer to the related description of step S43 in the above method embodiment, and are not described herein again.
An embodiment of the present invention provides a computer device, as shown in fig. 10, the device may include a processor 61 and a memory 62, where the processor 61 and the memory 62 may be connected by a bus or in another manner, and fig. 10 takes the connection by the bus as an example.
The processor 61 may be a Central Processing Unit (CPU). The Processor 61 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or combinations thereof.
The memory 62, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as the corresponding program instructions/modules in embodiments of the present invention. The processor 61 executes various functional applications and data processing of the processor by executing the non-transitory software programs, instructions and modules stored in the memory 62, namely, the method for recommending an electric vehicle charging facility based on federal learning in the above method embodiment.
The memory 62 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor 61, and the like. Further, the memory 62 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 62 may optionally include memory located remotely from the processor 61, and these remote memories may be connected to the processor 61 via a network. Examples of such networks include, but are not limited to, the internet, intranets, mobile communication networks, and combinations thereof.
One or more modules are stored in the memory 62 and, when executed by the processor 61, perform the federal learning based electric vehicle charging facility recommendation method in the embodiment shown in fig. 1-5.
The details of the computer device can be understood by referring to the corresponding descriptions and effects in the embodiments shown in fig. 1-5, and are not described herein again.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program that can be stored in a computer-readable storage medium and that when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications of the invention may be made without departing from the spirit or scope of the invention.
Claims (10)
1. An electric vehicle charging facility recommendation method based on federal learning is characterized by comprising the following steps:
acquiring a pile characteristic data set and a vehicle characteristic data set, wherein the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to charging pile operation, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle-enterprise data platform data and are related to vehicle charging;
carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model;
acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in a list of charging piles to be recommended;
and inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain a recommendation result.
2. The federal learning-based electric vehicle charging facility recommendation method as claimed in claim 1, wherein the post features include user identifiers, the vehicle features include vehicle identifiers, and the federal recommendation algorithm is used for federal training of the post feature data set and the vehicle feature data set to obtain a federal recommendation model, including:
carrying out entity alignment on a user identifier in the pile characteristics and a vehicle identifier in the vehicle characteristics corresponding to the same charging process to obtain aligned pile characteristics and vehicle characteristics;
updating the pile feature data set by using the aligned pile features, and updating the vehicle feature data set by using the aligned vehicle features;
and carrying out federal training on the updated pile characteristic data set and the updated vehicle characteristic data set to obtain the federal recommendation model.
3. The federal learning-based electric vehicle charging facility recommendation method as claimed in claim 2, wherein the federal recommendation model is obtained by federally training the updated pile characteristic data set and vehicle characteristic data set, and comprises:
step S231: according to the characteristic association relation, the pile characteristic data set is decomposed into pile self-owned characteristics and cross characteristics, and the vehicle characteristic data set is decomposed into vehicle self-owned characteristics and cross characteristics;
step S232: decomposing the cross feature into a pile-side feature and a car-side feature;
step S233: respectively calculating a pre-estimation value, a loss function value and a vehicle characteristic gradient corresponding to the vehicle self-characteristic and the vehicle side characteristic by using a preset federal recommendation model, and respectively calculating a loss function value and a pile characteristic gradient corresponding to the pile self-characteristic and the pile side characteristic according to the pre-estimation value;
step S234: and updating the parameters of the preset federal recommended model according to the vehicle characteristic gradient and the pile characteristic gradient, and returning to the step S233 until the preset training requirement is met.
4. The federal learning-based electric vehicle charging facility recommendation method as claimed in claim 1, wherein the step of inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain a recommendation result comprises the steps of:
inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain predicted values corresponding to all charging piles in a list of the charging piles to be recommended;
sorting the predicted values from large to small;
and obtaining the recommendation result according to the sorting result.
5. An electric vehicle charging facility recommendation system based on federal learning, comprising:
the system comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a pile characteristic data set and a vehicle characteristic data set, the pile characteristic data set is a set formed by pile characteristics which are extracted from pile operation data platform data and are related to the operation of a charging pile, and the vehicle characteristic data set is a set formed by vehicle characteristics which are extracted from vehicle-enterprise data platform data and are related to the charging of a vehicle;
the model training module is used for carrying out federal training on the pile characteristic data set and the vehicle characteristic data set by adopting a federal recommendation algorithm to obtain a federal recommendation model;
the second acquisition module is used for acquiring current pile characteristic data sets and current vehicle characteristic data sets corresponding to all charging piles in the list of the charging piles to be recommended;
and the processing module is used for inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain a recommendation result.
6. The federal learning based electric vehicle charging facility recommendation system of claim 5, wherein the post characteristics include a user identification, the car characteristics include a car identification, the model training module comprises:
the alignment submodule is used for carrying out entity alignment on the user identification in the pile characteristics and the vehicle identification in the vehicle characteristics corresponding to the same charging process to obtain the aligned pile characteristics and vehicle characteristics;
the updating submodule is used for updating the pile characteristic data set by adopting the aligned pile characteristics and updating the vehicle characteristic data set by adopting the aligned vehicle characteristics;
and the training submodule is used for carrying out federal training on the updated pile characteristic data set and the updated vehicle characteristic data set to obtain the federal recommendation model.
7. The federal learning based electric vehicle charging facility recommendation system of claim 6, wherein the training sub-module comprises:
the first decomposition unit is used for decomposing the pile feature data set into pile self-owned features and cross features according to the feature association relation, and decomposing the vehicle feature data set into vehicle self-owned features and cross features;
a second decomposition unit for decomposing the cross feature into a pile-side feature and a vehicle-side feature;
the calculation unit is used for calculating a pre-estimation value, a loss function value and a vehicle characteristic gradient corresponding to the vehicle owned characteristic and the vehicle side characteristic respectively by using a preset federal recommendation model, and calculating a loss function value and a pile characteristic gradient corresponding to the pile owned characteristic and the pile side characteristic respectively according to the pre-estimation value;
and the updating unit is used for updating the parameters of the preset federal recommended model by the vehicle characteristic gradient and the pile characteristic gradient and returning the parameters to the calculating unit until the preset training requirements are met.
8. The federal learning based electric vehicle charging facility recommendation system of claim 6, wherein the processing module comprises:
the input sub-module is used for inputting the current pile characteristic data set and the current vehicle characteristic data set into the federal recommendation model to obtain the corresponding predicted values of all the charging piles in the list of the charging piles to be recommended;
the sorting submodule is used for sorting the predicted values from large to small;
and the recommending submodule is used for obtaining the recommending result according to the sorting result.
9. A computer readable storage medium having stored thereon computer instructions for causing a computer to execute the federal learning electric vehicle charging facility recommendation method as claimed in any of claims 1-4.
10. A computer device, comprising: a memory and a processor, the memory and the processor being communicatively coupled to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the federal learning based electric vehicle charging facility recommendation method as claimed in any of claims 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010778276.8A CN111897800B (en) | 2020-08-05 | 2020-08-05 | Electric vehicle charging facility recommendation method and system based on federal learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010778276.8A CN111897800B (en) | 2020-08-05 | 2020-08-05 | Electric vehicle charging facility recommendation method and system based on federal learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111897800A true CN111897800A (en) | 2020-11-06 |
CN111897800B CN111897800B (en) | 2021-06-22 |
Family
ID=73246700
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010778276.8A Active CN111897800B (en) | 2020-08-05 | 2020-08-05 | Electric vehicle charging facility recommendation method and system based on federal learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111897800B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112560088A (en) * | 2020-12-11 | 2021-03-26 | 同盾控股有限公司 | Knowledge federation-based data security exchange method and device and storage medium |
CN112632380A (en) * | 2020-12-24 | 2021-04-09 | 北京百度网讯科技有限公司 | Training method of interest point recommendation model and interest point recommendation method |
CN113808289A (en) * | 2021-09-15 | 2021-12-17 | 上海氢巅能源信息科技有限公司 | Charging and parking method based on distributed energy service diagram and storage medium |
CN113815464A (en) * | 2021-10-26 | 2021-12-21 | 广州小鹏汽车科技有限公司 | Charging pile recommendation method and device, vehicle and storage medium |
CN113887811A (en) * | 2021-10-13 | 2022-01-04 | 江苏明茂新能源科技有限公司 | Charging pile data management method and system |
CN116051235A (en) * | 2022-12-30 | 2023-05-02 | 中山大学·深圳 | Information processing method, device and medium for network about vehicle aggregation platform based on federal learning |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8533224B2 (en) * | 2011-05-04 | 2013-09-10 | Google Inc. | Assessing accuracy of trained predictive models |
CN110014907A (en) * | 2017-08-21 | 2019-07-16 | 现代自动车株式会社 | Charge control method, electric vehicle and the charging equipment using this method |
CN111079022A (en) * | 2019-12-20 | 2020-04-28 | 深圳前海微众银行股份有限公司 | Personalized recommendation method, device, equipment and medium based on federal learning |
CN111428137A (en) * | 2020-03-25 | 2020-07-17 | 全球能源互联网研究院有限公司 | Recommendation method and recommendation device for electric vehicle charging facilities |
-
2020
- 2020-08-05 CN CN202010778276.8A patent/CN111897800B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8533224B2 (en) * | 2011-05-04 | 2013-09-10 | Google Inc. | Assessing accuracy of trained predictive models |
CN110014907A (en) * | 2017-08-21 | 2019-07-16 | 现代自动车株式会社 | Charge control method, electric vehicle and the charging equipment using this method |
CN111079022A (en) * | 2019-12-20 | 2020-04-28 | 深圳前海微众银行股份有限公司 | Personalized recommendation method, device, equipment and medium based on federal learning |
CN111428137A (en) * | 2020-03-25 | 2020-07-17 | 全球能源互联网研究院有限公司 | Recommendation method and recommendation device for electric vehicle charging facilities |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112560088A (en) * | 2020-12-11 | 2021-03-26 | 同盾控股有限公司 | Knowledge federation-based data security exchange method and device and storage medium |
CN112560088B (en) * | 2020-12-11 | 2024-05-28 | 同盾控股有限公司 | Knowledge federation-based data security exchange method, device and storage medium |
CN112632380A (en) * | 2020-12-24 | 2021-04-09 | 北京百度网讯科技有限公司 | Training method of interest point recommendation model and interest point recommendation method |
CN113808289A (en) * | 2021-09-15 | 2021-12-17 | 上海氢巅能源信息科技有限公司 | Charging and parking method based on distributed energy service diagram and storage medium |
CN113887811A (en) * | 2021-10-13 | 2022-01-04 | 江苏明茂新能源科技有限公司 | Charging pile data management method and system |
CN113815464A (en) * | 2021-10-26 | 2021-12-21 | 广州小鹏汽车科技有限公司 | Charging pile recommendation method and device, vehicle and storage medium |
CN116051235A (en) * | 2022-12-30 | 2023-05-02 | 中山大学·深圳 | Information processing method, device and medium for network about vehicle aggregation platform based on federal learning |
Also Published As
Publication number | Publication date |
---|---|
CN111897800B (en) | 2021-06-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111897800B (en) | Electric vehicle charging facility recommendation method and system based on federal learning | |
Yang et al. | Toward trustworthy vehicular social networks | |
Sun et al. | Security and privacy preservation in fog-based crowd sensing on the internet of vehicles | |
CN105799530B (en) | Information processor, motor-driven movable body and discharge control method | |
Zeitouni | Random walks in random environment | |
CN107786639A (en) | A kind of electric automobile networked system and its method of work based on block chain technology | |
CN112200046B (en) | Driving behavior recognition method and device, electronic equipment and storage medium | |
US9286485B2 (en) | Using trust points to provide services | |
CN114205091B (en) | Network authentication and key negotiation method for automatic driving vehicle based on chaotic mapping | |
CN116192419B (en) | Application program data safety protection method and device based on Internet of things card | |
US20130317666A1 (en) | Vehicle information providing apparatus | |
CN102404312B (en) | Points of trust is used to provide service | |
CN112906139A (en) | Vehicle fault risk assessment method and device, electronic equipment and storage medium | |
CN115484025A (en) | Vehicle encrypted communication method and device | |
Wang et al. | Charging station recommendation for electric vehicle based on federated learning | |
WO2019227597A1 (en) | Chatting robot-based navigation method and system | |
US11626977B2 (en) | Out-of-band key splitting and key derivation | |
US20220281340A1 (en) | Battery preservation amid transport disuse | |
CN113783685A (en) | System for protecting Internet of vehicles data by using block chain | |
CN116994423A (en) | Lightweight, privacy-protected lead vehicle selection system and method in vehicle formation | |
CN116245247A (en) | Safety prediction method, platform, equipment and medium based on federal learning | |
CN113343956B (en) | Road condition information prediction method and device, storage medium and electronic equipment | |
EP4101679A1 (en) | Management device, management system, management method, and program | |
EP3960530A1 (en) | Apparatus, method and computer program for determining a plurality of information of a location of a charging station | |
CN111429724B (en) | Traffic data deduplication obtaining 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |