Plug-and-play electric vehicle charging pile control method and control system

Abstract

The invention relates to the technical field of automobile charging, and discloses a plug-and-play electric automobile charging pile control method and system, and the method mainly comprises the following two steps: S1, inputting information for the first time, and carrying out the credible binding; and S2, plug-and-play non-inductive authentication is carried out. According to the method, payment verification is used as a triggering mechanism for credible binding of biological characteristics, and an edge-cloud cooperative computing architecture is fused, so that the problems of tedious first registration and low subsequent authentication efficiency in non-inductive charging of the electric vehicle are effectively solved. And during subsequent charging of the user, feature authentication of the user is directly realized during the period when the user pulls out the charging gun and inserts the charging gun into the vehicle, cloud matching is directly completed, charging is immediately carried out, and the user does not need to have redundant actions in the whole process. The plug-and-charge extremely convenient experience of registration after payment and charging after gun drawing of the user is achieved, and meanwhile millisecond response is ensured through multi-level authentication and localization processing.

Description

Technical Field

[0001] This invention relates to the field of vehicle charging technology, specifically to a control method and control system for electric vehicle charging piles that support plug-and-charge functionality. Background Technology

[0002] Currently, the mainstream authentication methods for electric vehicle charging are generally mobile phone scanning or mini-program verification. Mobile phone scanning payment is cumbersome and provides a poor experience when the network is weak. RFID cards are prone to loss, theft, and management costs. Operations such as scanning, swiping cards, and clicking on the app typically result in a single charging interaction time exceeding 30 seconds, and there are financial security risks such as easy loss of charging cards and easy copying of QR codes. In addition, new users need to download the app, enter their license plate number, and complete real-name authentication in advance, which is cumbersome, and the first-time user conversion rate is less than 40%.

[0003] Therefore, existing technologies cannot achieve seamless authentication and charging initiation within the extremely short time from "pulling out the gun" to "inserting," and cannot meet the demand for plug-and-charge. Summary of the Invention

[0004] To address the aforementioned problems in the existing technology, this invention provides a control method for electric vehicle charging piles that supports plug-and-charge functionality, possessing the advantage of plug-and-charge charging guns.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a control method for electric vehicle charging piles that support plug-and-charge functionality, comprising the following steps: Step S1: Initial information entry and trusted binding; ① When a user uses any charging gun for the first time and pulls it out, the sensor detects that the grip pressure is greater than the threshold, triggering a fingerprint collection request; at this time, although the user's fingerprint is not registered, the system enters "temporary authorization mode"; ② After the user inserts the charging gun into the vehicle, the charging station detects the CP signal and prompts the user via voice / light: "New users please complete payment verification first"; ③ The user selects a payment method and completes the payment on the charging pile screen. The cloud platform binds the payment order number with the charging gun device ID and timestamp to form a "registration token". ④ During the charging process, the fingerprint sensor continuously collects at least 3 fingerprint images from different angles. After the security chip extracts the feature values, it is encrypted with AES-256 using the "registration token" as the encryption key to generate a temporary template. ⑤ When charging is finished, the user returns the charging gun, and the edge gateway packages the encrypted template and payment order information and uploads it to the cloud; after the cloud verifies the payment is successful, it marks the template as "registered and trusted" and sends it to the "whitelist cache" of all charging piles in the network; ⑥ The user's mobile phone received a confirmation message: "Fingerprint has been bound, no need to scan the code for the next charge."

[0006] Step S2: Plug-and-Charge Seamless Authentication ① Triggering mechanism: When a registered user unplugs the charging gun at any station, the pressure sensor and capacitive sensor simultaneously wake up the fingerprint module; the fingerprint features are collected in real time while the user is holding the gun naturally. ② Multi-level matching: Local fast identification and comparison. If it passes, the gun end sends the feature value to the edge gateway for secondary confirmation through the gateway.

[0007] ③ Charging Start: After successful authentication, the charging pile automatically closes the relay and notifies the vehicle to start charging with a CC signal. The user's mobile phone will receive the charging notification silently.

[0008] Preferably, the multi-level matching includes: L1 Local Fast Denial: The gun-end security chip performs a lightweight comparison between the collected features and the built-in "negative example model". If the matching degree is greater than a low threshold (such as 30%), it is determined to be an unauthorized user and immediately goes into sleep mode to prevent accidental touch. L2 edge cache hit: If L1 passes, the charging gun sends the feature value to the edge gateway and compares it with the "high-frequency user fingerprint index" cached locally (such as users who have charged at this station in the past 30 days); if a hit occurs, charging is started directly with a latency of <500ms. L3 cloud-based secondary confirmation: If L2 fails to match, the gateway sends the feature value to the cloud via a TLS encrypted channel; the cloud searches the global database using "fuzzy matching + device UID constraint"; if a match is successful, the user information and charging instructions are returned, and the user's fingerprint is synchronized to the local cache, achieving "one-time cloud verification, subsequent edge access".

[0009] Preferably, the L2-level edge cache hit adopts a homomorphic encryption approximation strategy, which calculates and compares the weighted similarity of fingerprints in the encrypted domain, enabling fast comparison without the need for a cloud-based decryption key.

[0010] Preferably, the registration token is generated by hashing the payment order number, the charging gun device identifier, and the current timestamp, and is used as an encryption key to encrypt the biometric template collected for the first time.

[0011] Preferably, the local rapid rejection step includes: comparing the real-time collected biometric features with the pre-stored negative example model; if the abnormal score exceeds a first threshold, it is determined to be an illegal attempt and the process is terminated.

[0012] Preferably, the edge cache matching step employs a similarity measurement algorithm that combines the cosine similarity of the vector angle with the normalized Manhattan distance.

[0013] Preferably, the cloud-based secondary confirmation step adds a constraint on the correlation between the geographical location of the user's historically used devices and the location of the currently requested device, based on the results of the similarity search.

[0014] Preferably, after a user successfully completes global matching authentication through the cloud, the cloud synchronously sends the user's biometric summary information to the edge gateway where the current request is located to enrich its local cache.

[0015] The present invention also provides a control system used in a plug-and-charge electric vehicle charging pile control method, including an intelligent charging gun module, a locally deployed edge gateway with caching and computing capabilities, and a payment platform and cloud platform responsible for payment processing and global management.

[0016] Preferably, the intelligent charging gun module integrates a pressure and capacitance dual-mode wake-up sensor, a fingerprint acquisition sensor, a security chip, and a gun-end MCU for performing L1 level denial and feature encryption. The edge gateway is equipped with a high-frequency user cache and a local matching engine for performing L2 authentication and resuming transmission after network outage; The cloud platform includes a global template database, a fuzzy matching engine, and a whitelist synchronization service, which are used to perform L3 level authentication and network-wide template synchronization. The payment platform includes a payment interface for generating a payment order upon first use and binding the order information as a trusted anchor to a fingerprint template.

[0017] Compared with the prior art, the present invention provides a control method for electric vehicle charging piles that supports plug-and-charge charging, which has the following beneficial effects: By using payment verification as the trigger mechanism for trusted biometric binding and integrating an edge-cloud collaborative computing architecture, the problems of "cumbersome initial registration" and "low efficiency of subsequent authentication" in seamless electric vehicle charging are effectively solved. Furthermore, during subsequent charging, user authentication is achieved directly from the moment the user unplugs the charging gun into the vehicle, cloud matching is completed instantly, and charging begins immediately, requiring no additional user action. This achieves an extremely convenient "payment equals registration, unplug and charge" experience, while multi-level authentication and localized processing ensure millisecond-level response times. Attached Figure Description

[0018] Figure 1 This is an overall architecture diagram of the control system in an embodiment of the present invention; Figure 2 This is a flowchart of the plug-and-charge control method in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example 1

[0020] refer to Figure 1 This system consists of five parts: S100 Smart Charging Gun Module S101. Fingerprint Sensor: Employs an FPC1025 capacitive sensor with a resolution of 508 DPI and an array size of 192×192 pixels, supporting wet fingers and oily environments. The sensor is integrated into the grip area of ​​the gun handle, with a natural angle of approximately 15° to the finger, and a fingertip contact area of ​​≥80% when gripping.

[0021] S102. Pressure + Capacitive Dual-Mode Wake-up Module: The threshold of the pressure sensor (FSR400) is set to 0.5N, and the capacitive sensor detects a human body approaching at a distance of <5mm. The fingerprint module is only woken up when both are triggered simultaneously to prevent accidental touches.

[0022] S103. Security Chip SE: Integrated hardware encryption engine (supports AES-256, SHA-256, TRNG). SE memory is divided into: Boot area (immutable), feature extraction area, template encryption area, and negative example model area. All biometric data is processed within the SE; the main control MCU cannot read plaintext.

[0023] S104. Gun-end MCU: Responsible for CC / CP signal detection, relay control, and MQTT communication with the edge gateway. When the CP signal changes from 12V to 9V (vehicle connection indication), the MCU sends a CHARGE_START_IND command to the SE, triggering the authentication process.

[0024] S105. Cache area: 2KB FRAM is allocated within the SE to store encrypted whitelist templates (up to 20) for L2 level offline comparison.

[0025] S200 Edge Gateway S201. High-Frequency User Cache: Redis in-memory database, Key is Site ID: Fingerprint Hash, Value is User ID | Template Ciphertext | Access Timestamp, TTL=30 days. The caching strategy uses LRU, with a capacity limit of 1000 entries.

[0026] S202. Local Matching Engine: Based on NVIDIA Jetson Nano, it runs a lightweight feature matching algorithm (accelerated by cosine similarity) and supports 1000 matches per second.

[0027] S203. Resuming connection after network interruption: The gateway has a built-in 64GB eMMC. When the network is interrupted, the authentication requests are queued and stored. After the network is restored, the requests are automatically retried. The retry strategy is exponential backoff (1s, 2s, 4s... maximum interval of 5 minutes).

[0028] S300 Cloud Platform S301. Global Template Database: Uses a sharded MongoDB cluster.

[0029] S302. Fuzzy Matching Engine: Based on the Milvus vector database, using the IVF_PQ index, nlist=4096, during the query, first take nprobe=32, and then accurately compare the Top-100, to achieve second-level retrieval of tens of millions of templates.

[0030] S303. Whitelist Synchronization Service: Employs gRPC bidirectional streaming. When a user successfully completes L3 authentication at a certain site, the cloud proactively pushes a SYNC_TEMPLATE message to the gateway of that site and updates lastSyncStations.

[0031] S400 Payment Platform The system uses the Alipay / WeChat payment interface to generate an order. Users then scan the code to pay on the charging station screen (or the charging gun's OLED display). After successful payment, the payment platform pushes {orderNo, amount, payerOpenId} to the cloud via a webhook, and the cloud generates a registration token based on this information.

[0032] S500 Vehicle End Compliant with GB / T 18487.1-2015 standard, the system determines the gun connection status through the CC resistor value and obtains the maximum charging current through the CP duty cycle. This system reuses the CP signal as the authentication trigger, requiring no vehicle modification. Example 2

[0033] S1. Initial Information Entry and Trust Binding This process has 6 states, and the specific implementation steps are as follows: S1-1: Pressure-based wake-up and temporary authorization mode When the user first removes the charging gun, the pressure sensor detects a grip pressure P > 0.5N and the capacitive sensor detects the proximity of a human body, activating the fingerprint module (SE). Since the isRegistered flag is false at this time, the system enters TEMP_AUTH_MODE. In this mode, the SE generates a one-time ECC key pair (secp256r1 curve), with the public key denoted as ECC_PUB_TEMP and the private key denoted as ECC_PRIV_TEMP, which is never exported.

[0034] S1-2: Vehicle Connection and Payment Prompts When the user inserts the charging gun into the vehicle, the MCU detects that the CP signal has switched from 12V to 9V, confirming a successful physical connection. At this time, the MCU controls the voice module (SYN6658) to announce: "New users, please complete the payment verification first," while simultaneously displaying the payment QR code on the charging pile screen (or the OLED at the charging gun end).

[0035] S1-3: Payment Verification and Registration Token Generation Users select their payment method (Alipay / WeChat) on the charging station screen and complete the payment by calling the payment SDK (the amount can be set to 0.01 yuan for verification). After successful payment, the payment platform asynchronously notifies the cloud interface, and the cloud generates a registration token.

[0036] S1-4: Fingerprint Collection and Template Encryption During charging, the fingerprint sensor continuously collects data at 1-second intervals. The data collection process for each instance is as follows: Image preprocessing: The DSP within the SE performs baseline correction, denoising, and orientation field estimation on the original 192×192 image. A Gabor filter is used to extract texture, using the following formula: ; x'=xcosθ+ysinθ y'=-xsinθ+ycosθ Parameter settings: λ=8, θ∈[0,π), ψ=0, σ=4, γ=0.5.

[0037] Feature extraction: A minutiae extraction algorithm is used to detect endpoints and bifurcations. Each minutiae records its coordinates (x, y), direction θ, and type t. A minimum of 12 valid minutiae are required per acquisition.

[0038] Multi-angle fusion: Collect at least 3 fingerprints from different angles (the user naturally rotates their finger). SE registers and fuses the sets of detail points from multiple images to generate a unified feature vector F. VEC The dimension is 256 bytes.

[0039] Key derivation and encryption: The first 16 bytes of REG_TOKEN are used as the AES-256 key, and the last 16 bytes are used as the IV (initialization vector).

[0040] S1-5: Data Packaging and Uploading When charging is complete and the user returns the charging gun, the MCU detects the CP signal being disconnected and triggers the CHARGE_END event. The SE then packages the above data into a REG_PACKET: The packet was uploaded to the cloud via MQTT and uses TLS 1.3 mutual authentication.

[0041] S1-6: Cloud verification and whitelist synchronization After receiving REG_PACKET in the cloud, execute the following: Payment verification: Query the corresponding payment status in Redis to confirm that the payment has been made and has not expired.

[0042] Trusted token: After successful authentication, insert the user document into MongoDB, which includes the following fields: trustLevel: The string "HIGH" indicates a high trust level. registeredDevice: Device unique identifier deviceId lastAuthTime: Current timestamp, recording the time of the most recent authentication. After successful verification, insert a document into MongoDB, set trustLevel to HIGH, registeredDevice to deviceId, and record lastAuthTime.

[0043] Network-wide synchronization: The cloud pushes {deviceId, userId, templateCipher} to all edge gateways via gRPC streaming (network-wide synchronization), and the gateways store it in Redis with a TTL of 30 days. Initial synchronization uses "lazy loading": data is only cached at a particular site when a user performs a charging action, avoiding the waste of cold data.

[0044] User notification: The cloud will call the SMS / push service to send the message: "Fingerprint has been bound. No need to scan the code next time you charge." Simultaneously, users will see a "Charging fingerprint registration successful" note in their WeChat Pay / Alipay bill.

[0045] Example 3: S2. Plug-and-play seamless authentication refer to Figure 2 This process adopts a three-level hierarchical architecture, and the specific steps are as follows: S2-1: Dual-mode wake-up and feature acquisition When a registered user unplugs the charging gun at any station, both the pressure and capacitive sensors are triggered simultaneously, waking up the fingerprint module. At this time, the isRegistered flag is true, and the system enters QUICK_AUTH_MODE. The fingerprint sensor immediately acquires the fingerprint, requiring image acquisition and preprocessing to be completed within 500ms to generate the feature vector F. VEC_NEW .

[0046] S2-2: L1 Local Fast Rejection SE incorporates a negative example model, which is not a whitelist template but a low-dimensional feature synthesized from historical rejected fingerprints and attack samples (such as fake fingerprints). L1 alignment uses lightweight distance calculation. ; If similarity > L1 THRESHOLD (Default 0.30) If the error is determined to be a high-probability attack or accidental touch, SE will immediately go into sleep mode for 30 seconds and log the eventType: L1. REJECT This step involves very little computation (<10ms) and can filter out more than 95% of invalid wake-ups.

[0047] Negative example model update mechanism: L1 rejection logs are collected daily from all sites in the cloud, new negative examples centroid are trained using One-Class SVM, and distributed to the SE on the gun end via OTA.

[0048] S2-3: L2 edge cache hit If L1 passes, SE will F VEC_NEW Send to the edge gateway.

[0049] Upon successful L2 hit: The gateway sends an AUTH_SUCCESS message to the charging station, containing the userId and sessionToken. The charging station's MCU closes the relay to initiate charging. Simultaneously, the gateway updates the user's lastAccessTime in Redis and asynchronously notifies the cloud that the cache is valid.

[0050] The improved similarity measurement algorithm employed is a hybrid model that integrates the cosine similarity of the vector angle and the normalized Manhattan distance. This algorithm first measures the consistency in directional trends between two fingerprint feature vectors using cosine similarity to eliminate the overall numerical scaling effect caused by factors such as pressure intensity. Simultaneously, it evaluates the absolute differences in the specific values ​​of each dimension of the vectors by calculating the normalized Manhattan distance. The two are linearly combined through an adjustable weighting parameter to form a comprehensive similarity score.

[0051] The corresponding calculation formula for this algorithm is as follows: For two n-dimensional fingerprint feature vectors A and B, the formula for calculating their comprehensive similarity score S is: S = α· Cosine(A, B) + (1 - α) · [ 1-D m (A, B) ]; in, ; Cosine similarity measures the consistency of vector directions. ; To normalize the Manhattan distance, measure the absolute difference in values; α (0 ≤ α ≤ 1) is an adjustable weighting parameter used to balance the contribution ratio of the two metrics. This formula combines the two types of metrics complementaryly through linear weighting, ultimately outputting a unified and robust similarity score.

[0052] S2-4: L3 Cloud Secondary Confirmation If the L2 cache misses (cached out of sync or the user is visiting the site for the first time), the gateway will... VEC_NEW Send to the cloud-based verify interface via a TLS encrypted channel. Cloud processing: Fuzzy matching: Performs an ANN search in the Milvus vector database with parameters topK=1 and search_params={"nprobe":32}. Returns the most similar template and its ID.

[0053] Device UID Constraint: To prevent cross-device attacks (such as template theft and replay), the cloud checks whether the request source deviceId is in the user's registeredDevice list or lastSyncStations. If it is a new device for the first time, a secondary confirmation is triggered: a push notification is sent to the user's mobile phone asking "Do you want to charge at the new site?", and the user authorizes after clicking to confirm.

[0054] Similarity calculation: Calculate exact similarity in the plaintext domain:

[0055] like Authentication passed.

[0056] After L3 is successful: The cloud returns the userId and triggers cache preheating: the user template and deviceId are added to lastSyncStations, and a SYNC_TEMPLATE message is pushed to the current site gateway. Simultaneously, the cloud marks the user as a "high-frequency user," increasing their synchronization priority.

[0057] S2-5: Charging Start-up and Silent Notification Regardless of whether it's an L2 or L3 path, the MCU at the gun end executes the following after successful authentication: Relay closure: Drive the optocoupler MOC3063 to control the closure of the 30A relay, and the CC signal informs the vehicle through 4V PWM.

[0058] User notifications: If the user's app is active, a charging notification will be pushed via Google FCM / APNs from the cloud (optional, users can turn it off). Silent mode is used by default, requiring no user interaction for truly seamless operation.

[0059] In summary, by transforming the necessary payment step into a trusted identity binding process, "payment equals registration" is achieved, allowing users to complete biometric registration with zero additional steps. Subsequent use then becomes truly "plug and play."

[0060] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0061] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A control method for electric vehicle charging piles that support plug-and-charge functionality, characterized in that: Includes the following steps: Step S1: Initial information entry and trusted binding; ① When a user uses any charging gun for the first time and pulls it out, the sensor detects that the grip pressure is greater than the threshold, triggering a fingerprint collection request; at this time, although the user's fingerprint is not registered, the system enters "temporary authorization mode"; ② After the user inserts the charging gun into the vehicle and the charging station detects the CP signal, it will prompt via voice / light: "New users please complete payment verification first"; ③ The user selects a payment method and completes the payment on the charging pile screen. The cloud platform binds the payment order number with the charging gun device ID and timestamp to form a "registration token". ④ During the charging process, the fingerprint sensor continuously collects at least 3 fingerprint images from different angles. After the security chip extracts the feature values, it is encrypted with AES-256 using the "registration token" as the encryption key to generate a temporary template. ⑤ When charging is complete, the user returns the charging gun, and the edge gateway packages the encrypted template and payment order information and uploads it to the cloud; After successful payment verification in the cloud, the template is marked as "registered and trusted" and distributed to the "whitelist cache" of all charging piles in the network; ⑥ The user's phone receives a confirmation message: "Fingerprint has been bound, no need to scan the code for the next charge"; Step S2: Plug-and-play seamless authentication ① Triggering mechanism: When a registered user unplugs the charging gun at any station, the pressure sensor and capacitive sensor simultaneously wake up the fingerprint module; the fingerprint features are collected in real time while the user is holding the gun naturally. ② Multi-level matching: Local fast identification and comparison. If it passes, the gun end sends the feature value to the edge gateway for secondary confirmation through the gateway. ③ Charging Start: After successful authentication, the charging pile automatically closes the relay and notifies the vehicle to start charging with a CC signal. The user's mobile phone will receive the charging notification silently.

2. The control method for electric vehicle charging piles supporting plug-and-charge as described in claim 1, characterized in that: The multi-level matching includes: L1 Local Fast Denial: The gun-end security chip performs a lightweight comparison between the collected features and the built-in "negative example model". If the matching degree is greater than the low threshold, it is determined to be an unauthorized user and immediately goes into sleep mode to prevent accidental touch. L2 edge cache hit: If L1 passes, the gun sends the feature value to the edge gateway and compares it with the "high-frequency user fingerprint index" in the local cache; if a hit occurs, charging is started directly with a latency of <500ms. L3 cloud-based secondary confirmation: If L2 fails to match, the gateway sends the feature value to the cloud via a TLS encrypted channel; the cloud searches the global database using "fuzzy matching + device UID constraint"; if a match is successful, the user information and charging instructions are returned, and the user's fingerprint is synchronized to the local site cache, achieving "one-time cloud verification, subsequent edge access".

3. The electric vehicle charging pile control method supporting plug-and-charge as described in claim 2, characterized in that: The L2-level edge cache hit adopts a homomorphic encryption approximation strategy, which calculates and compares the weighted similarity of fingerprints in the encrypted domain, enabling fast comparison without the need for cloud decryption keys.

4. The control method for electric vehicle charging piles supporting plug-and-charge as described in claim 1, characterized in that: The registration token is generated by hashing the payment order number, the charging gun device identifier, and the current timestamp, and is used as an encryption key to encrypt the biometric template collected for the first time.

5. A control method for an electric vehicle charging pile supporting plug-and-charge as described in claim 2, characterized in that: The local rapid rejection step includes: comparing the real-time collected biometric features with the pre-stored negative example model; if the abnormal score exceeds the first threshold, it is determined to be an illegal attempt and the process is terminated.

6. A control method for an electric vehicle charging pile supporting plug-and-charge as described in claim 2, characterized in that: The edge cache matching step employs a similarity measurement algorithm that combines the cosine similarity of the vector angle with the normalized Manhattan distance.

7. A control method for an electric vehicle charging pile supporting plug-and-charge as described in claim 2, characterized in that: The cloud-based secondary confirmation step adds a constraint on the correlation between the geographical location of the user's historically used devices and the location of the currently requested device, based on the results of the similarity search.

8. A control method for an electric vehicle charging pile supporting plug-and-charge as described in claim 2, characterized in that: Once a user successfully completes global authentication via the cloud, the cloud synchronously sends the user's biometric summary information to the edge gateway where the current request is located to enrich its local cache.

9. The control system used in the plug-and-charge electric vehicle charging pile control method according to any one of claims 1-8, characterized in that: This includes a smart charging gun module, a locally deployed edge gateway with caching and computing capabilities, and a payment platform and cloud platform responsible for payment processing and overall management.

10. The control system used in the plug-and-charge electric vehicle charging pile control method according to claim 9, characterized in that: The intelligent charging gun module integrates a pressure and capacitance dual-mode wake-up sensor, a fingerprint sensor, a security chip, and a gun-end MCU, which are used to perform L1 level denial and feature encryption. The edge gateway is equipped with a high-frequency user cache and a local matching engine for performing L2 authentication and resuming transmission after network outage; The cloud platform includes a global template database, a fuzzy matching engine, and a whitelist synchronization service, which are used to perform L3 level authentication and network-wide template synchronization. The payment platform includes a payment interface for generating a payment order upon first use and binding the order information as a trusted anchor to a fingerprint template.

Images