Port type power supply man-machine interaction smart electric meter and control system thereof

Abstract

The invention discloses a port type power supply man-machine interaction smart electric meter and a control system thereof, and belongs to the technical field of port power supply metering and energy management. Comprising an AC-DC dual-purpose power supply module, an intelligent metering module, an NFC interaction module, an AI edge calculation module, a national secret encryption module, a touch screen man-machine interaction module and a multi-mode communication module. The AC-DC dual-purpose power supply module is used for being connected with a 85-265 VAC power supply or a 100-1500 VDC power supply, uninterrupted power supply is achieved through an automatic switching circuit, and 12V / 5V / 3.3 V multi-gear voltage is output to be supplied to all functional modules; and the intelligent metering module adopts a high-precision metering chip and a sampling circuit to synchronously collect full electric power parameters including voltage, current, power, harmonic waves and power factors. The intelligent metering module adopts a high-precision metering chip (Saquan photoelectric HT7036) and a special sampling circuit to synchronously acquire full electric power parameters such as voltage, current, power, harmonic waves, power factors and the like, so that the monitoring limitation of a traditional single watt-hour meter is broken through.

Description

Technical Field

[0001] This invention relates to the field of port power supply metering and energy management technology, and in particular to a port-type power supply human-machine interactive smart meter and its control system. Background Technology

[0002] As global trade hubs, ports account for a significant proportion of energy consumption and carbon emissions, with diesel generator emissions from ships during berthing becoming a major source of pollution. Shore power systems, a core technology for green port construction, have been promoted in major ports nationwide, but their practical application faces numerous challenges: traditional metering relies on a single electricity meter, failing to collect key parameters such as voltage fluctuations and harmonic distortion rates, resulting in insufficient monitoring accuracy; ship power connection requires manual verification of identity and physical connection, taking 15-20 minutes and exhibiting low efficiency; power supply strategies depend on manual experience, easily leading to underutilization of resources and low resource efficiency; and the lack of accurate methods for calculating carbon emission reductions makes it difficult to meet green port certification requirements.

[0003] Existing smart meters are mostly suitable for routine residential or industrial scenarios, and have the following shortcomings: they only support single AC power supply, making them unsuitable for the hybrid AC / DC power supply environment of ports, which combines shore power and energy storage; they lack a rapid authentication mechanism for ships, resulting in poor interaction convenience; they do not integrate localized AI analysis capabilities, relying on cloud computing which leads to decision-making delays; data security protection is insufficient, making sensitive information easily leaked; and human-computer interaction functions are weak, making it inconvenient for port management personnel to operate and check status on-site. Therefore, there is an urgent need to develop a human-computer interactive smart meter and its control system that is suitable for port scenarios and combines accurate metering, convenient interaction, intelligent decision-making, and security protection. Summary of the Invention

[0004] Purpose of the invention: The purpose of this invention is to provide a port-type power supply human-machine interactive smart meter and its control system; which can solve the problems of inaccurate monitoring, cumbersome interaction, inefficient management, difficulty in carbon accounting, and data insecurity in port power supply systems.

[0005] Technical solution: To solve the above-mentioned technical problems, according to one aspect of the present invention, more specifically, a port-type power supply human-machine interaction smart meter and its control system, including: AC / DC dual-use power supply module, smart metering module, NFC interaction module, AI edge computing module, national cryptographic encryption module, touch screen human-machine interaction module, and multi-mode communication module; AC / DC dual-use power supply module: It is used to connect to 85-265VAC AC power or 100-1500VDC DC power, and achieves uninterrupted power supply through automatic switching circuit. It outputs 12V / 5V / 3.3V multi-level voltage to supply various functional modules. Intelligent metering module: It adopts a high-precision metering chip and sampling circuit to synchronously collect all electrical parameters, including voltage, current, power, harmonics, and power factor; NFC Interaction Module: Based on 13.56MHz near-field communication technology, it enables rapid identification and data interaction between ships and shore power piles; AI edge computing module: Built with domestically produced AI chip and lightweight algorithm model, it realizes ship power demand prediction, equipment anomaly detection, energy efficiency optimization scheduling and carbon emission reduction quantitative accounting; National Cryptographic Encryption Module: Employs SM2 / SM3 / SM4 national cryptographic algorithms to encrypt and store measurement data, identity information, and AI decision-making results during transmission; Touchscreen human-machine interaction module: Equipped with a 3.5-7 inch touchscreen, supporting real-time display of power parameters, historical data query, parameter setting and alarm viewing; Multi-mode communication module: integrates RS-485, Ethernet, and 4G / 5G wireless communication interfaces to realize local data interaction and remote cloud upload.

[0006] Furthermore, the AC / DC dual-use power supply module includes: an input unit, an automatic switching unit, a power conversion unit, and an overvoltage protection unit; Input unit: Supports a wide input range of 85-265V AC, 45-65Hz and 100-1500V DC, and is equipped with EMI filter circuit at the input terminal; The automatic switching unit uses a Hongfa HF46F / 012-HSTF relay to isolate the DC input when AC power is available. The power conversion unit uses a 650V / 20A SiC MOSFET from Jiejie Microelectronics to form a high-voltage BUCK circuit, outputting 12V voltage, which is regulated to 3.3V by a Sanbang Microelectronics SGM2039-3.3 LDO; The overvoltage protection unit is equipped with domestically produced varistors and TVS diodes to achieve input overvoltage protection.

[0007] Furthermore, the intelligent metering module includes: a voltage sampling unit, a current sampling unit, and a metering chip; Voltage sampling unit: It adopts a 1000:1 resistor voltage divider network and a Saint-Gobain SGM8110 isolated operational amplifier to output a 0-3V signal to the metering chip; Current sampling unit: 200A:100mA current transformer is used in AC scenarios, and ACS712 Hall sensor or 50mΩ shunt resistor + differential amplifier circuit is used in DC scenarios. Metering chip: Juquan Optoelectronics HT7036 is selected, which supports AC and DC waveform metering and communicates with the main controller through the SPI interface.

[0008] Furthermore, the AI ​​edge computing module uses domestic AI chips such as Cambricon MLU220 or Rockchip RK3399Pro, with a computing power of ≥0.5 TOPS, externally connected to GigaDevice GD25Q16 Flash storage for model weights, and has four built-in algorithm modules: load prediction module, anomaly detection module, energy efficiency optimization module, and carbon emission reduction accounting module. Load forecasting module: It adopts a lightweight LSTM model, inputs historical electricity consumption data, ship type, port berthing time and other parameters, and forecasts the electricity demand for the next 1-6 hours; Anomaly detection module: Combining isolated forest and one-dimensional convolutional neural network algorithms, it identifies anomalies such as voltage drop, harmonic exceedance, and equipment overload in real time; Energy efficiency optimization module: Based on the PPO reinforcement learning algorithm, dynamically adjust the synergistic strategy between shore power and energy storage; Carbon emission reduction accounting module: Based on GB / T 32151-2015 standard, it has a built-in emission factor database and automatically calculates CO2 and NOx emission reductions.

[0009] Furthermore, the NFC interaction module uses the domestically produced FM11NT0811 chip, supports the ISO14443A / B protocol, and connects to the PCB antenna through a π-type matching network composed of a 1µH inductor and 2×100pF capacitors, enabling rapid reading of ship identity information and power demand in 1-2 seconds.

[0010] Furthermore, the touchscreen human-computer interaction module uses Rockchip RK3399 or Allwinner A64 main control chip, supports LVGL graphical interface, and the touch method is compatible with finger and glove operation, with a resolution of 480×272 or 800×480.

[0011] According to another aspect of the present invention, a control system for a port-type power supply human-machine interactive smart meter is provided. The system is used to control the above-mentioned port-type power supply human-machine interactive smart meter, comprising: the smart meter and a side unit. The side unit serves as a port energy management platform, deploying complex algorithm models to perform long-term trend analysis, multi-device collaborative optimization, and iterative training of AI models. The smart meter and the edge device achieve encrypted data transmission through a multi-mode communication module, and the model trained on the edge device is sent to the edge device for updates via OTA.

[0012] Furthermore, the side unit includes a data processing engine, an equipment management module, a carbon emission reduction calculation center, and a model training platform, supporting four-level data visualization management of "port-dock-shore power pile-ship", and can generate multi-dimensional energy consumption and carbon emission analysis reports.

[0013] Furthermore, the control system also includes a security protection mechanism: the terminal side adopts local data desensitization and hierarchical access control, the edge side adopts two-way identity authentication, and data transmission and storage are encrypted through national cryptographic algorithms to prevent data tampering and unauthorized access.

[0014] Furthermore, the workflow of the control system includes: S1. Before a ship docks, it uses an NFC tag and a smart meter to complete identity verification and pre-registration of its electricity demand. S2. The smart meter collects the operating parameters and environmental data of the shore power system in real time, and after encryption, it is synchronized to the end-side AI module and the edge-side platform. S3, the edge AI module performs real-time load forecasting, anomaly detection and scheduling optimization, and displays the operating status and decision results through the touch screen; S4. The edge platform performs global data fusion analysis, generates optimization strategies and carbon emission reports, and distributes them to the edge for execution. S5. When equipment malfunctions, the system automatically triggers an alarm and pushes it to the administrator, supporting predictive maintenance.

[0015] Beneficial Effects: The intelligent metering module employs a high-precision metering chip (Juquan Optoelectronics HT7036) and a dedicated sampling circuit to simultaneously collect all electrical parameters, including voltage, current, power, harmonics, and power factor. This overcomes the monitoring limitations of traditional single-meter electricity meters and meets the high-precision metering requirements of mixed AC / DC shore power scenarios in ports. The voltage sampling unit utilizes a 1000:1 resistor voltage divider network combined with an isolated operational amplifier, while the current sampling unit features differentiated designs for AC / DC scenarios (AC transformer, DC Hall sensor / shunt resistor), ensuring accurate parameter acquisition under different power supply modes and effectively identifying hidden problems such as voltage fluctuations and harmonic distortion.

[0016] The NFC interaction module, based on 13.56MHz near-field communication technology and using the domestically produced FM11NT0811 chip, enables rapid reading of ship identity information and electricity demand within 1-2 seconds, replacing the cumbersome process of traditional manual verification and physical connection (originally taking 15-20 minutes). This improves operational efficiency by over 90% and reduces port operating labor costs. The touchscreen human-machine interface module supports 3.5-7 inch touchscreens, is compatible with finger and glove operation, and features an LVGL graphical interface that intuitively displays real-time parameters, historical data, and alarm information. It also supports on-site parameter settings, addressing the weaknesses of traditional electricity meters in terms of human-machine interaction and inconvenience for on-site operation. The load forecasting module accurately predicts electricity demand for the next 1-6 hours using an LSTM model, providing data support for the coordinated scheduling of shore power and energy storage, avoiding resource waste caused by over-engineering. The anomaly detection module combines isolated forests and one-dimensional convolutional neural networks to identify anomalies such as voltage drops, harmonic exceedances, and equipment overloads in real time. The carbon emission reduction accounting module strictly adheres to the GB / T32151-2015 standard, incorporates a complete emission factor database, and automatically calculates emission reductions for pollutants such as CO2 and NOx. This addresses the inaccuracies of traditional carbon accounting methods and their difficulty in meeting green port certification requirements, providing authoritative data support for port carbon footprint management. The national cryptographic encryption module employs SM2 / SM3 / SM4 national cryptographic algorithms to encrypt and store metering data, identity information, and AI decision results throughout the entire process. Data anonymization and hierarchical access control are implemented at the endpoint, while two-way authentication is used at the edge, effectively preventing data tampering, unauthorized access, and sensitive information leakage, meeting the security and compliance requirements of port energy management.

[0017] The AC / DC dual-use power supply module supports a wide input range of 85-265VAC AC and 100-1500VDC DC, achieving automatic switching via Hongfa relays. Combined with an overvoltage protection unit (domestic varistor + TVS diode), it adapts to the complex power supply environment of ports where shore power and energy storage are integrated, ensuring uninterrupted power supply. The multi-mode communication module integrates RS-485, Ethernet, and 4G / 5G interfaces, accommodating both local data exchange and remote cloud uploads, adapting to communication conditions in different areas of the port. The control system adopts an "end-side-edge" architecture. The edge-side port energy management platform supports four-level data visualization management of "port-terminal-dock-shore power pile-ship," generating multi-dimensional reports such as energy consumption trends, carbon emission analysis, and equipment status, achieving global collaborative optimization and refined management of port energy. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the system principle. Detailed Implementation

[0019] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] Example Module Name Specific configuration parameters AC / DC dual-use power supply module Input unit: 85-265VAC (45-65Hz), 100-1500VDC, EMI filter circuit; Automatic switching unit: Hongfa HF46F / 012-HSTF relay; Power conversion unit: Jiejie Microelectronics 650V / 20A SiC MOSFET, Shengbang Microelectronics SGM2039-3.3 LDO, output 12V / 5V / 3.3V; Overvoltage protection unit: Domestic 10D471K varistor + SMBJ60CA TVS diode. Intelligent metering module Voltage sampling: 1000:1 resistor divider network + SGM8110 isolated op-amp from Sanbang Microelectronics, output 0-3V; Current sampling: AC 200A:100mA current transformer, DC ACS712 Hall sensor + 50mΩ shunt resistor; Metering chip: Juquan Optoelectronics HT7036, SPI interface communication. NFC Interaction Module Chip: Domestic FM11NT0811, supports ISO14443A / B protocol; Matching network: 1µH inductor + 2×100pF capacitors (π type); PCB antenna: read / write distance 0-5cm, read time ≤1.5 seconds AI edge computing module Chip: Cambricon MLU220 (0.8 TOPS computing power); Storage: GigaDevice GD25Q16 Flash (16 Mbit); Algorithm modules: Lightweight LSTM, Isolation Forest + 1D-CNN, PPO reinforcement learning, GB / T 32151-2015 carbon accounting model National Cryptographic Encryption Module Algorithms: SM2 (Authentication), SM3 (Data Hash), SM4 (Symmetric Encryption); Encryption Rate: ≥10Mbps Touchscreen Human-Computer Interaction Module Main control chip: Rockchip RK3399; Touch screen: 7-inch, 800×480 resolution, supports finger / glove operation; Interface: LVGL graphical interface, including parameter display, history query, alarm list, and settings menu. Multi-mode communication module Interfaces: RS-485 (Modbus-RTU protocol), Ethernet (TCP / IP), 4G (Cat.1); Data upload rate: ≥1Mbps in 4G mode Control system deployment: End-side unit: 50 smart meters are deployed at 50 shore power piles at 10 wharves. Each meter corresponds to one ship connection point, enabling independent metering and local decision-making for each pile.

[0021] Side unit: The port energy management platform is deployed in the port data center, configured with 2 servers (master and backup mode), and integrates a data processing engine, equipment management module, carbon emission reduction calculation center and model training platform; the platform supports web and mobile access, and managers can monitor the operation status of shore power piles throughout the port in real time.

[0022] Communication link: The terminal side and the edge side achieve encrypted data transmission through Ethernet (dock area) and 4G (remote shore power pile), with data transmission latency ≤500ms; after the edge side model is iterated, it is sent to the terminal side via OTA, with an update time ≤10 minutes.

[0023] Security protection: The terminal side performs anonymization processing on the ship's identity information and electricity consumption data (only retaining the device ID and metering value, hiding the ship's privacy information), and sets three levels of access permissions for administrators, maintenance personnel, and ordinary users; the edge side uses digital certificates to realize two-way identity authentication between the terminal side and the edge side, and all data transmission and storage are encrypted with SM4.

[0024] Actual workflow: (1) Preparations before berthing (S1) After the vessel arrives at the port anchorage, the crew interacts with the smart meter of the target shore power station through the vessel's NFC tag (which is pre-loaded with information such as vessel ID, vessel type, rated power, and historical electricity consumption data). The meter completes identity recognition and pre-registration of electricity demand within 1.2 seconds and uploads the information to the side-side platform in encryption.

[0025] Based on the pre-registered information, combined with the shore power capacity of the terminal and the remaining energy storage capacity, the side platform generates a power supply plan in advance (such as allocating two shore power lines in parallel for large container ships and one shore power line for small general cargo ships).

[0026] (2) Data acquisition and synchronization (S2) After the ship berths, the crew connects to the physical power supply line, and the electricity meter automatically detects the power supply mode (AC / DC) and activates the smart metering module. AC scenario: 220VAC voltage and 150A current are collected via a current transformer, and the metering chip simultaneously calculates the active power as 33kW, power factor as 0.95, and third harmonic distortion rate as 2.1%. DC scenario: The ACS712 Hall sensor is used to collect 1000VDC voltage and 80A current, and the power is calculated to be 80kW, with voltage fluctuation ≤±2%.

[0027] The meter collects data once every second, and after being hashed by SM3 and encrypted by SM4, it is synchronized to the edge AI module and the edge platform.

[0028] (3) End-side local decision-making (S3) Load forecasting: The AI ​​edge computing module takes into account the ship's historical power consumption data (average power load of 60kW for the last 3 berthings), ship type (container ship), and berthing duration (24 hours), and uses the LSTM model to predict the power demand for the next 6 hours: 80kW for the first 1-2 hours (ship unloading equipment is operating at full capacity), and 40kW for the third-6th hours (only to maintain the ship's living power needs).

[0029] Anomaly Detection: During operation, the meter detected a sudden increase in current to 200A in a ship's power supply line (exceeding the rated current of 180A). The anomaly detection module identified it as "equipment overload" using the 1D-CNN algorithm, immediately triggering a local audible and visual alarm and displaying the alarm information ("Current overload: Current 200A, rated 180A") on the touch screen.

[0030] Energy efficiency optimization: Based on load forecast results, the PPO algorithm adjusts the energy storage coordination strategy: During the first 1-2 hours, the dock energy storage system is called to supplement power supply (energy storage output 20kW), and the shore power output is 60kW to avoid insufficient shore power capacity; During the third to sixth hours, only the shore power supply is 40kW, and the energy storage system is charged.

[0031] (4) Edge-side global optimization (S4) The side platform integrates real-time data from 50 electricity meters to generate a global energy consumption report: the total electricity consumption of shore power at 10 wharves on that day was 12,500 kWh, of which 8,200 kWh was AC power and 4,300 kWh was DC power. The Carbon Emission Reduction Calculation Center, based on the GB / T 32151-2015 standard and combined with the emission factors of ship diesel power generation (CO2: 2.67 kg / kWh, NOx: 0.03 kg / kWh), calculated the daily carbon emission reduction: CO2 reduction of 33,375 kg (12,500 kWh × 2.67 kg / kWh) and NOx reduction of 375 kg (12,500 kWh × 0.03 kg / kWh), generating a visualized carbon emission report for port green certification. The side platform discovered that the load rate of shore power pile No. 2 at Pier No. 3 had been below 30% for three consecutive days. An optimization strategy was issued: adjust the power supply priority of this shore power pile, giving priority to small vessels, and improve the utilization rate of shore power.

[0032] (5) Anomaly Handling and Maintenance (S5) In response to the aforementioned "equipment overload" alarm, the side platform simultaneously pushes alarm information to the mobile terminal of the maintenance personnel (including shore power pile number, anomaly type, and real-time parameters). The maintenance personnel arrive at the scene within 10 minutes and find that the current fluctuation is caused by a loose connection of the ship unloading equipment. After timely handling, the alarm is lifted. The system identifies the peak overload fault periods for a certain type of shore power pile (10:00-12:00 daily, when ships are unloading cargo in a concentrated manner) by statistically analyzing historical abnormal data, and formulates maintenance plans in advance (checking the wiring status before 9:30 daily).

[0033] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A port-type power supply human-machine interactive smart meter, characterized in that, include: AC / DC dual-use power supply module, smart metering module, NFC interaction module, AI edge computing module, national cryptographic encryption module, touch screen human-computer interaction module, multi-mode communication module; AC / DC dual-use power supply module: It is used to connect to 85-265VAC AC power or 100-1500VDC DC power, and achieves uninterrupted power supply through automatic switching circuit. It outputs 12V / 5V / 3.3V multi-level voltage to supply various functional modules. Intelligent metering module: It adopts a high-precision metering chip and sampling circuit to synchronously collect all electrical parameters, including voltage, current, power, harmonics, and power factor; NFC Interaction Module: Based on 13.56MHz near-field communication technology, it enables rapid identification and data interaction between ships and shore power piles; AI edge computing module: Built with domestically produced AI chip and lightweight algorithm model, it realizes ship power demand prediction, equipment anomaly detection, energy efficiency optimization scheduling and carbon emission reduction quantitative accounting; National Cryptographic Encryption Module: Employs SM2 / SM3 / SM4 national cryptographic algorithms to encrypt and store measurement data, identity information, and AI decision-making results during transmission; Touchscreen human-machine interaction module: Equipped with a 3.5-7 inch touchscreen, supporting real-time display of power parameters, historical data query, parameter setting and alarm viewing; Multi-mode communication module: integrates RS-485, Ethernet, and 4G / 5G wireless communication interfaces to realize local data interaction and remote cloud upload.

2. The port-type power supply human-machine interactive smart meter according to claim 1, characterized in that: The AC / DC dual-use power supply module includes: an input unit, an automatic switching unit, a power conversion unit, and an overvoltage protection unit; Input unit: Supports a wide input range of 85-265V AC, 45-65Hz and 100-1500V DC, and is equipped with EMI filter circuit at the input terminal; The automatic switching unit uses a Hongfa HF46F / 012-HSTF relay to isolate the DC input when AC power is available. The power conversion unit uses a 650V / 20A SiC MOSFET from Jiejie Microelectronics to form a high-voltage BUCK circuit, outputting 12V voltage, which is regulated to 3.3V by a Sanbang Microelectronics SGM2039-3.3 LDO; The overvoltage protection unit is equipped with domestically produced varistors and TVS diodes to achieve input overvoltage protection.

3. The port-type power supply human-machine interactive smart meter according to claim 1, characterized in that: The intelligent metering module includes: a voltage sampling unit, a current sampling unit, and a metering chip; Voltage sampling unit: It adopts a 1000:1 resistor voltage divider network and a Saint-Gobain SGM8110 isolated operational amplifier to output a 0-3V signal to the metering chip; Current sampling unit: 200A:100mA current transformer is used in AC scenarios, and ACS712 Hall sensor or 50mΩ shunt resistor + differential amplifier circuit is used in DC scenarios. Metering chip: Juquan Optoelectronics HT7036 is selected, which supports AC and DC waveform metering and communicates with the main controller through the SPI interface.

4. A port-type power supply human-machine interactive smart meter according to claim 1, characterized in that: The AI ​​edge computing module uses Cambricon MLU220 or Rockchip RK3399Pro domestic AI chips with a computing power of ≥0.5 TOPS. It is externally connected to GigaDevice GD25Q16 Flash to store model weights and has four built-in algorithm modules: load prediction module, anomaly detection module, energy efficiency optimization module, and carbon emission reduction accounting module. Load forecasting module: It adopts a lightweight LSTM model, inputs historical electricity consumption data, ship type, port berthing time and other parameters, and forecasts the electricity demand for the next 1-6 hours; Anomaly detection module: Combining isolated forest and one-dimensional convolutional neural network algorithms, it identifies anomalies such as voltage drop, harmonic exceedance, and equipment overload in real time; Energy efficiency optimization module: Based on the PPO reinforcement learning algorithm, dynamically adjust the synergistic strategy between shore power and energy storage; Carbon emission reduction accounting module: Based on GB / T 32151-2015 standard, it has a built-in emission factor database and automatically calculates CO2 and NOx emission reductions.

5. A port-type power supply human-machine interactive smart meter according to claim 1, characterized in that: The NFC interaction module uses the domestic FM11NT0811 chip, supports the ISO14443A / B protocol, and connects to the PCB antenna through a π-type matching network composed of a 1µH inductor and 2×100pF capacitors to achieve rapid reading of ship identity information and power demand in 1-2 seconds.

6. A port-type power supply human-machine interactive smart meter according to claim 1, characterized in that: The touchscreen human-computer interaction module uses Rockchip RK3399 or Allwinner A64 main control chip, supports LVGL graphical interface, and the touch method is compatible with finger and glove operation, with a resolution of 480×272 or 800×480.

7. A control system for a port-type power supply human-machine interactive smart meter, characterized in that: The system is used to control a port-type power supply human-machine interactive smart meter as described in claims 1-6, comprising: the smart meter and a side unit; The side unit serves as a port energy management platform, deploying complex algorithm models to perform long-term trend analysis, multi-device collaborative optimization, and iterative training of AI models. The smart meter and the edge device achieve encrypted data transmission through a multi-mode communication module, and the model trained on the edge device is sent to the edge device for updates via OTA.

8. The control system for a port-type power supply human-machine interactive smart meter according to claim 7, characterized in that: The side unit includes a data processing engine, an equipment management module, a carbon emission reduction calculation center, and a model training platform. It supports four-level data visualization management of "port-dock-shore power pile-ship" and can generate multi-dimensional energy consumption and carbon emission analysis reports.

9. The control system for a port-type power supply human-machine interactive smart meter according to claim 7, characterized in that: The control system also includes a security protection mechanism: the terminal side adopts local data desensitization and hierarchical access control, the edge side adopts two-way identity authentication, and data transmission and storage are encrypted with national cryptographic algorithms to prevent data tampering and unauthorized access.

10. The control system for a port-type power supply human-machine interactive smart meter according to claim 7, characterized in that: The workflow of the control system includes: S1. Before a ship docks, it completes identity verification and pre-registration of electricity demand through an NFC tag and a smart meter. S2. The smart meter collects the operating parameters and environmental data of the shore power system in real time, and after encryption, it is synchronized to the end-side AI module and the edge-side platform. S3, the edge AI module performs real-time load forecasting, anomaly detection and scheduling optimization, and displays the operating status and decision results through the touch screen; S4. The edge platform performs global data fusion analysis, generates optimization strategies and carbon emission reports, and distributes them to the edge for execution. S5. When equipment malfunctions, the system automatically triggers an alarm and pushes it to the administrator, supporting predictive maintenance.

Images