Unmanned intelligent wind and light power storage and supply container type seawater and brine desalination system and method

Abstract

The application is a kind of unmanned intelligent wind and light power storage and supply container type seawater and brine desalination system and method, the system comprises an energy supply subsystem, a desalination treatment subsystem, a water tank storage subsystem, an intelligent inspection and security subsystem, a three-layer heterogeneous robot collaborative operation system, a standardized quick-mounting interface and an RFID integrated management system. Through the robot collaboration of air global perception, ground armed patrol and cabinet fine operation, combined with adaptive energy scheduling algorithm and remote digital twin platform, the application realizes unmanned autonomous operation of the whole system, module quick replacement and flexible expansion, can adapt to seawater / high-salt brine dual water source and output industrial, agricultural, living or direct drinking water, and has the advantages of strong scene adaptability, high unmanned degree, high energy saving and efficiency, etc.

Description

Technical Field

[0001] This invention relates to the fields of water treatment and new energy technology, and in particular to an unmanned intelligent wind, solar, energy storage and power supply containerized seawater and brine desalination system and method. Background Technology

[0002] With the increasing scarcity of freshwater resources globally, seawater desalination and high-salinity brine (brine) desalination have become important ways to solve water shortage problems. Especially in areas with scarce freshwater resources such as remote islands, inland deserts, and coastal farms, the demand for miniaturized, rapidly deployable, and unattended desalination plants is becoming increasingly urgent.

[0003] Currently, several containerized seawater desalination technology solutions have been proposed. For example, a smart microgrid system and a seawater desalination system can be integrated into a transportable container, using wind-solar-storage hybrid power to provide a stable power supply for the desalination equipment; another solution uses solar photovoltaic panels in conjunction with battery banks to power the reverse osmosis membrane modules; yet another solution combines a wind-solar hybrid power generation system with a seawater desalination system and configures remote monitoring capabilities.

[0004] However, when the above solution is applied to remote, uninhabited, or sparsely populated areas, the following technical drawbacks still exist:

[0005] Poor adaptability to different scenarios: Most existing solutions are designed for a single water source (seawater) and a single scenario (island), which cannot meet the desalination needs of inland high-salinity brine (brine), and it is also difficult to switch flexibly between agricultural irrigation water, industrial water and drinking water. A single system is difficult to meet the differentiated water needs of different regions and different users.

[0006] Low level of automation: While existing solutions offer remote monitoring capabilities, they are limited to fixed sensor data acquisition and alarm functions, lacking mobile autonomous inspection capabilities. This prevents detailed inspection of internal instrument readings, valve status, and pipeline leaks, and also hinders security protection of the surrounding environment. When equipment is deployed in remote, unattended areas, blind spots and security vulnerabilities exist, making it difficult to detect and address equipment malfunctions or external intrusions in a timely manner.

[0007] Inconvenient module replacement and maintenance: Existing containerized units mostly adopt an integrated design with fixed connections between units, lacking standardized quick-disassembly and assembly interfaces. When a module (such as reverse osmosis membrane module, pretreatment unit, energy storage battery, etc.) fails or needs to be upgraded, it is difficult to replace it quickly on-site. It usually requires professional personnel and special tools, resulting in long equipment downtime, high maintenance costs, and difficulty in adapting to the needs of unmanned operation and maintenance.

[0008] High cost of cross-scenario deployment: Existing solutions require customized design for different deployment locations (islands, deserts, coastal farms, overseas), and cannot achieve rapid on-site configuration and flexible capacity expansion after standardized production, resulting in high equipment manufacturing costs, long delivery cycles, and difficulty in large-scale promotion.

[0009] Insufficient energy dispatch and robot collaboration: Existing solutions lack adaptive energy dispatch algorithms for wind, solar and energy storage fluctuations and robot charging needs, and also lack collaborative operation and maintenance mechanisms among drones, humanoid robots and robot dogs, making it difficult to achieve truly unmanned autonomous operation.

[0010] Therefore, there is an urgent need to develop an intelligent desalination device system that can adapt to various water source types and geographical environments, has unattended autonomous inspection and security functions, supports rapid on-site module replacement, and can achieve standardized mass production and flexible expansion. Summary of the Invention

[0011] The present invention aims to address the shortcomings of the prior art by providing an unmanned intelligent wind, solar, energy storage and power supply containerized seawater and brine desalination system and method.

[0012] To achieve the above objectives, this invention employs the following technical solution: an unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system, comprising:

[0013] The energy supply subsystem includes wind turbine generators, photovoltaic arrays, energy storage units, and smart microgrid controllers. The wind turbine generators, photovoltaic arrays, and energy storage units are connected through the smart microgrid controller to form a wind-solar-storage complementary off-grid or grid-connected dual-mode power supply system. The energy storage units and smart microgrid controllers are integrated into the power supply container.

[0014] The desalination subsystem includes a pretreatment module, two sets of interchangeable ultrafiltration modules, two sets of interchangeable reverse osmosis modules, a posttreatment module, and an edge cloud brain. The desalination subsystem is integrated into the desalination container. Its inlet can be selectively connected to seawater or high-salinity brine, and its outlet has interfaces for industrial freshwater, agricultural freshwater, domestic freshwater, or direct drinking water.

[0015] The water tank storage subsystem includes a foldable TPU raw water tank and a TPU fresh water tank, which are integrated into the water tank container.

[0016] The intelligent inspection and security subsystem includes a drone inspection unit, a humanoid robot inspection unit, and a robot dog patrol unit. Each unit is housed in an independent charging compartment within the power supply container, and all three are equipped with automatic start / stop and automatic charging functions. The drone inspection unit is used to conduct aerial inspections of the photovoltaic array, wind turbine generators, and the surrounding environment of the system, identifying equipment anomalies and intrusion behaviors. The humanoid robot inspection unit is used to perform plug-in / plug-out connections, module replacements, pipeline tightening, and instrument reading operations for equipment within the de-icing container and / or the power supply container. The robot dog patrol unit is used to proactively warn, drive away, or prevent foreign objects from entering the restricted area.

[0017] The drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit are all connected to the intelligent microgrid controller, edge cloud brain, and remote monitoring and operation and maintenance platform via wireless communication, forming a three-layer heterogeneous robot collaborative operation and maintenance system of aerial global perception, ground armed patrol, and cabinet precision operation.

[0018] All functional units within and between the energy supply subsystem, desalination subsystem, and intelligent inspection and security subsystem are connected using standardized quick-connect interfaces. These interfaces include pipe interfaces, cable quick connectors, and mechanical quick-locking mechanisms, and are designed to be recognizable and operable by humanoid robot inspection units.

[0019] All functional devices and modules within the system are equipped with RFID electronic tags. The drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit interact with the edge cloud brain through RFID technology to achieve integrated management of equipment identification, maintenance record traceability, and self-positioning and navigation of each unit.

[0020] Specifically, the energy storage unit uses a lithium iron phosphate battery pack, encapsulated in an aluminum alloy casing, and integrates a battery management system, energy storage converter, thermal management system, and fire protection system. The casing protection level is no less than IP56, and the corrosion resistance level meets the ISO12944 C5-M standard.

[0021] Specifically, the smart microgrid controller employs an adaptive energy dispatch algorithm based on wind and solar resource prediction and energy storage state of charge, including:

[0022] In the daytime dispatch phase: after obtaining meteorological forecast data, wind and solar power are predicted through a neural network model. Combined with the load demand prediction of the desalination subsystem, a multi-objective optimization algorithm is used to solve the 24-hour energy storage charging and discharging plan.

[0023] Intraday rolling correction phase: Based on real-time collected photovoltaic output, wind power output, load power and energy storage state of charge, the prediction error is calculated, and the charging and discharging commands are adaptively corrected through model prediction control;

[0024] Real-time power balance control phase: Calculate net power in seconds and dynamically switch between energy storage charging, energy storage discharging, limiting new energy output, or cutting off some loads according to the zone control rules of energy storage charge status.

[0025] The charging priority management module dynamically allocates charging windows based on the charge status and task urgency of the drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit, and supports the humanoid robot inspection unit and robot dog patrol unit to charge or swap batteries for the drone inspection unit outdoors.

[0026] Specifically, the reverse osmosis module of the desalination subsystem uses two sets of vertical 8040 type reverse osmosis membrane modules. The two sets of membrane modules can be switched or operated in parallel. Seawater desalination membranes or brackish water desalination membranes are selectively configured according to the water source type. The outlet of the reverse osmosis module is connected to the post-treatment module, which includes a mineralization device, an ultraviolet sterilizer, and an activated carbon sterilizer. Each device is equipped with a bypass valve, which can be bypassed or activated according to the target water quality.

[0027] The desalination subsystem also includes an energy recovery device, which is an isobaric pressure exchanger. Its inlet is connected to the concentrate discharge end of the reverse osmosis module, and its outlet is connected to the inlet end of the reverse osmosis module. The energy recovery efficiency is not less than 95%.

[0028] The desalination subsystem also includes a concentrated brine treatment unit, comprising a foldable TPU drying tank and a scraper-type salt collector. The TPU drying tank is folded and stored inside a water tank container during transport, with unfolded dimensions of 20m × 10m × 0.3m. The scraper-type salt collector is installed at one end of the foldable TPU drying tank and is used to scrape the crystallized salt to the collection tank.

[0029] Specifically, the container structures of power supply containers, desalination containers, and water tank containers are as follows:

[0030] The power supply container is divided into three areas along its length. The first area is divided into two layers: the upper layer houses the emergency backup diesel generator set, and the lower layer houses the energy storage unit. The second area houses the intelligent microgrid controller, photovoltaic inverter control cabinet, and wind power control cabinet. The third area is divided into two sections: the upper section houses the independent charging compartment for the drone inspection unit, and the lower section is divided into two areas, which respectively house the independent charging compartments for the humanoid robot inspection unit and the robot dog patrol unit. Automatic doors connect the areas for the humanoid robot inspection unit to pass through.

[0031] The desalination container is divided into three zones along its length. The first zone is divided into upper and lower levels. The upper level houses the pretreatment transition water tank, and the lower level is the raw water treatment zone. This zone is further divided into two sections along the center line, with a passageway in the middle. One side is equipped with an ultrafiltration module, a pre-filter titanium alloy impeller high-pressure pump, a titanium alloy self-cleaning screen corrosion-resistant self-priming pump, and a security filter. The other side is equipped with a self-cleaning disc filter, a multi-media filter, and a pre-filter activated carbon filter. The second zone has a primary reverse osmosis transition water tank at the top and a reverse osmosis zone at the bottom. It is also divided into two sections along the center line, with a passageway in the middle. One side is equipped with a primary reverse osmosis module and a secondary pressurized water pump. The first section includes a secondary reverse osmosis module, a freshwater output pump, an online water quality monitor, an ultraviolet disinfection device, and a post-activated carbon filter. On the other side are a post-titanium alloy impeller high-pressure pump, a ceramic rotor ERD pressure exchanger, a scale inhibitor and disinfectant pump, a membrane cleaning pump, a scale inhibitor and disinfectant liquid tank, a membrane cleaning liquid tank, a mixing and injection pump, an intelligent mixing and blending system, and a mixing and mineralization device. The third section has a desalination transition water tank at the top and is divided into two areas along the centerline, with a passageway in the middle. One side has an edge-controlled cloud platform, and the other side has an intelligent spare parts warehouse and a maintenance workbench. Automatic doors allow humanoid robot inspection units to pass between the areas.

[0032] The water tank container is divided into 50m sections by partitions. 3 Raw water tank area and 25m 3 In the freshwater tank area, both tanks are TPU water bag structures that are folded and stored during transportation. After being unfolded on site, they are connected to the desalination container via quick-connect couplings. When multiple containers are stacked, a retractable and deployable elevator is provided for humanoid robot inspection units and robot dog patrol units to allow them to move between different container levels.

[0033] Specifically, the drone inspection unit includes a drone, a first camera, a first infrared thermal imager, a first RFID reader mounted on the drone, and an automatic charging nest located in the drone's corresponding independent charging compartment. The automatic charging nest has an automatic charging interface that matches the drone. The first RFID reader is used to read RFID electronic tags on devices or modules to assist in identification and positioning. The drone is also equipped with a positioning module and a communication module for communicating with the edge cloud brain and transmitting the images and thermal data collected by the first camera and the first infrared thermal imager to the edge cloud brain. The edge cloud brain has a built-in AI visual recognition algorithm for identifying photovoltaic array hot spots, wind turbine blade cracks, pipeline leaks, and intrusion targets in the surrounding environment.

[0034] Specifically, the humanoid robot inspection unit includes a humanoid robot, a second camera and a second infrared thermal imager mounted on the humanoid robot's head, a robotic hand mounted on the end of its arm, and a second RFID reader mounted on its head or arm; the humanoid robot's corresponding independent charging compartment is equipped with a robot charging pile, which has a power supply interface, and the humanoid robot has a power receiving interface; the humanoid robot is used to enter and exit the automatic doors with dust-proof compartments on the power supply container or desalination container, and moves along the central aisle and modular interval aisles inside the container; the humanoid robot has the capability to inspect, repair, maintain, and replace parts for the drone inspection unit and the robot dog patrol unit;

[0035] The humanoid robot uses a built-in AI visual recognition algorithm to identify equipment malfunctions, pipeline leaks, and instrument parameters. It uses a second RFID reader to read and calibrate RFID tags to correct its navigation position, reads RFID electronic tags on the equipment to obtain its identity and maintenance information, and writes new data to the RFID electronic tags after maintenance is completed.

[0036] The humanoid robot has an internal library of embodied operation models and connects wirelessly to the smart microgrid controller, edge cloud brain, and remote monitoring and maintenance platform. When maintenance or crisis management is required, it sequentially obtains operation instructions from the embodied model library, the smart microgrid controller or edge cloud brain, and the remote monitoring and maintenance platform, or receives AR annotation guidance from remote personnel through digital twin models. The operation process and results are automatically generated into reports, which are reported to the remote monitoring and maintenance platform and backed up simultaneously in the smart microgrid controller or edge cloud brain.

[0037] Specifically, the robot dog patrol unit includes a robot dog, a non-lethal weapon system mounted on the robot dog, a third camera, a third infrared thermal imager, a third RFID reader, and a scanning radar linkage interface that wirelessly connects to a scanning radar deployed at a high point on the system perimeter; the non-lethal weapon system includes at least one of a laser dazzler, a sonic deflector, and an electromagnetic jammer; the robot dog's independent charging compartment is equipped with a charging station, which has a power supply interface, and the robot dog has a power receiving interface; the third RFID reader is used to read RFID electronic tags on the device or path to assist in navigation and identification;

[0038] The activation of the non-lethal weapon system is subject to dual authorization from the edge cloud brain or the remote monitoring and operation and maintenance platform: when the robot dog autonomously patrols and detects an intrusion target or receives a radar alarm, it will automatically issue a voice warning. If the intrusion target continues to approach, it will send a request to the edge cloud brain, and the non-lethal weapon system will be activated after authorization is obtained.

[0039] Specifically, the three-layer heterogeneous robot collaborative operation and maintenance system is configured as follows: the drone inspection unit performs a global scan of the photovoltaic array, wind turbine generator set, and system perimeter according to a preset route, generates a hot spot distribution map and a 3D environmental map, and distributes the optimal inspection path to the humanoid robot inspection unit and the robot dog patrol unit through the edge cloud brain; when the drone inspection unit detects equipment abnormality, the edge cloud brain dispatches the humanoid robot inspection unit to enter the corresponding container to perform detailed inspection and maintenance; when the robot dog patrol unit detects intrusion or receives a radar alarm, the robot dog patrol unit initiates a drive-away procedure and simultaneously notifies the drone inspection unit to take off for tracking and filming, and the drone inspection unit transmits the real-time images back to the remote monitoring and operation and maintenance platform.

[0040] A containerized method for desalination of seawater and brine using unmanned intelligent wind, solar, and energy storage power supply includes the following steps:

[0041] S1: Select the corresponding reverse osmosis membrane module type according to the water source type of the deployment location, and assemble the desalination treatment subsystem through standardized quick-installation interfaces;

[0042] S2: The intelligent microgrid controller supplies power to the desalination subsystem based on an adaptive energy dispatch algorithm;

[0043] S3: Raw water is sequentially treated through a pretreatment module, an ultrafiltration module, a reverse osmosis module, and a post-treatment module to produce fresh water of the target quality. The activation or bypass status of the post-treatment module is selected according to water demand: when drinking water is needed, the mineralization device and ultraviolet sterilizer are activated; when agricultural irrigation water is needed, the post-treatment module is partially bypassed to retain an appropriate amount of minerals; when industrial water is needed, the post-treatment module is completely bypassed.

[0044] S4: The drone inspection unit takes off to conduct a global inspection of the photovoltaic array, wind turbine generator and system perimeter. It uses AI visual recognition algorithms to identify equipment abnormalities and intrusion behaviors, and sends the inspection path to the humanoid robot inspection unit and / or robot dog patrol unit through the edge cloud brain.

[0045] S5: The robot dog patrol unit patrols the ground along the planned path. When it detects an intrusion, it activates the non-lethal weapon system to issue a warning and drive away the intruder, and notifies the drone inspection unit to take off and track it.

[0046] S6: When an equipment malfunction is detected, the edge cloud brain dispatches a humanoid robot inspection unit to enter the corresponding container. The humanoid robot identifies the RFID tag on the faulty module through the second RFID reader, uses a robotic arm to unlock the mechanical quick-locking mechanism, pulls out the faulty module or disconnects the quick-connect connector, inserts a new module or reconnects the pipeline, and performs locking.

[0047] S7: During operation, the humanoid robot receives AR annotation guidance from remote experts through AR glasses and a digital twin platform. The operation process and results are automatically generated into reports and uploaded to the remote monitoring and maintenance platform, and written into the RFID tags of fault modules and new modules.

[0048] The beneficial effects of this invention are:

[0049] Multi-scenario adaptability and multi-purpose functionality: Through modular design, optional reverse osmosis membrane components (seawater membrane / brine membrane), and configurable bypass for post-treatment modules, a standardized system can quickly adapt to various geographical environments such as islands, deserts, coastal farms, and overseas locations, as well as various water source types such as seawater and high-salt brine. It can produce industrial water, agricultural irrigation water, domestic water, or direct drinking water according to user needs, achieving a single system that is universally applicable to multiple scenarios, significantly reducing equipment customization costs and delivery cycles.

[0050] Unmanned intelligent operation and maintenance, three-layer heterogeneous robot collaboration: A three-layer heterogeneous robot collaborative operation and maintenance system, comprising aerial global perception (drones), ground armed patrols (robot dogs), and in-container precision operations (humanoid robots), combined with a remote monitoring and operation and maintenance platform, achieves unmanned operation of the entire system. Drones are responsible for large-scale rapid inspections and global path planning; robot dogs are responsible for security and ground patrols; and humanoid robots are responsible for in-depth inspections and module replacements inside the containers. With clear division of labor, information sharing, and task collaboration, unmanned autonomous operation and maintenance are achieved.

[0051] Standardized quick-installation interfaces support rapid module replacement and expansion: The entire system features standardized quick-installation interfaces (pipe interfaces, cable quick connectors, and mechanical quick-locking mechanisms). These interfaces are specifically designed to be recognizable and operable by humanoid robots, allowing any module to be quickly disassembled and replaced on-site by a humanoid robot. Faulty modules are returned to the factory for repair, significantly reducing equipment downtime. System capacity can be flexibly expanded by adding parallel modules to meet water demand of varying scales.

[0052] Energy self-sustainability and intelligent dispatch: Utilizing an off-grid power supply scheme that integrates wind, solar, and energy storage, the system does not rely on an external power grid and fully leverages the abundant wind and solar energy resources of the deployment site. The intelligent microgrid controller employs a three-level adaptive energy dispatch algorithm consisting of day-ahead scheduling, intraday correction, and real-time control. A dedicated robot charging priority management module is also designed to prioritize the energy needs of security and inspection robots while ensuring de-emphasis on production, achieving zero-carbon emissions and low-cost autonomous power supply.

[0053] Energy-efficient and with a high water recovery rate: Utilizing an isobaric pressure exchanger with an energy recovery efficiency of no less than 95%, combined with low-energy reverse osmosis membrane technology, the energy consumption per ton of water can be as low as 3.8-4.0 kWh, saving 40%-60% energy compared to traditional desalination processes. The backwash water recovery design of the pretreatment and ultrafiltration modules ensures an overall water recovery rate of no less than 70%. Concentrated brine is further utilized through a drying pond to produce industrial salt products, achieving near-zero emissions.

[0054] Integrated RFID management with full lifecycle traceability: All key equipment, modules, instruments, quick-connect connectors, and even inspection path points within the system are equipped with RFID electronic tags. Robots use RFID readers to automatically identify equipment, write maintenance records in real time, and dynamically correct location navigation. Combined with edge cloud brains and remote operation and maintenance platforms, this forms a digital management system for the entire lifecycle of equipment, providing a data foundation for predictive maintenance and remote scheduling.

[0055] Remote digital twin and AR collaboration enables high efficiency in human-machine collaboration: It allows remote experts to view the real-time status of on-site equipment in a three-dimensional virtual space through a digital twin platform and AR glasses, and remotely guide humanoid robots to perform complex maintenance operations through AR annotations, which greatly reduces the reliance on on-site professionals and improves the efficiency and accuracy of fault handling. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the overall system structure of the quick-installation parallel connection of the present invention;

[0057] Figure 2 This is a front view of the internal partitions of the power supply container of the present invention;

[0058] Figure 3 This is a top view of the internal partitions of the power supply container of the present invention;

[0059] Figure 4 This is a front view of the internal partitions of the container for the desalination process according to the present invention;

[0060] Figure 5 This is a schematic diagram of the raw water treatment area and reverse osmosis area inside the desalination container of the present invention;

[0061] Figure 6 This is a process flow diagram of the seawater and brine desalination system of the present invention;

[0062] Figure 7 This is a flowchart illustrating the inspection, maintenance, and repair process of the intelligent inspection and security subsystem of this invention.

[0063] Figure 8 This is a diagram illustrating the remote digital twin and AR collaboration of the present invention.

[0064] In the diagram: 1-Wind turbine generator set; 2-Photovoltaic array; 3-Power supply container; 30-Energy storage unit; 31-Intelligent microgrid controller; 32-Emergency backup diesel generator set; 33-Photovoltaic inverter control cabinet; 34-Wind power control cabinet; 35-Independent charging compartment; 4-Desalination container; 40-Pre-treatment transition water tank; 41-Raw water treatment area; 411-Ultrafiltration module; 412-Pre-treatment titanium alloy impeller high-pressure pump; 413-Titanium alloy corrosion-resistant self-priming pump with self-cleaning screen; 414-Security filter; 415-Self-cleaning disc filter; 416-Multi-media filter; 417-Pre-treatment activated carbon filter; 42-Transition water tank after primary reverse osmosis; 43-Reverse osmosis zone; 4301-Primary reverse osmosis module; 4302-Secondary pressurized water high-pressure pump; 4303 - Secondary reverse osmosis module; 4304- Freshwater output pump; 4305- Online water quality analyzer; 4306- Ultraviolet disinfection device; 4307- Post-activated carbon filter; 4308- Post-titanium alloy impeller high-pressure pump; 4309- Ceramic rotor ERD pressure exchanger; 4310- Antiscalant and disinfectant pump; 4311- Membrane cleaning pump; 4312- Antiscalant and disinfectant liquid tank; 4313- Membrane cleaning liquid tank; 4314- Mixing and injection pump; 4315- Intelligent mixing and blending system; 4316- Mixing and mineralization device; 44- Desalination transition water tank; 45- Edge cloud brain control cabinet; 46- Intelligent spare parts warehouse; 47- Maintenance workbench; 5- Water tank container; 51- Raw water tank area; 52- Freshwater tank area; 6- Foldable TPU drying pool;

[0065] The following will describe in detail, with reference to the accompanying drawings, embodiments of the present invention. Detailed Implementation

[0066] The present invention will be further described below with reference to embodiments:

[0067] like Figures 1-8 As shown, an unmanned intelligent wind-solar-storage-powered containerized seawater and brine desalination system includes an energy supply subsystem, a desalination treatment subsystem, a water tank storage subsystem, and an intelligent inspection and security subsystem. Specifically, it consists of three 40-foot high-cube containers (power supply container 3, desalination treatment container 4, and water tank container 5), a 100kWp photovoltaic array 2 that can be quickly installed on-site, a 100kW wind turbine generator set 1 (optional, depending on the wind resources of the deployment site), and a deployable concentrated brine drying device with crystallization salt collection. The container's anti-corrosion outer wall is coated with a polyurethane heavy-duty anti-corrosion coating or marine heavy-duty anti-corrosion paint (such as epoxy paint), with a salt spray test duration of ≥2000 hours. It also features a 2mm aerogel insulation layer, maintains positive pressure for dust prevention, and is equipped with a dehumidification system to adapt to harsh environments such as islands and deserts.

[0068] The energy supply subsystem includes wind turbine generator 1 (a 100kW permanent magnet direct-drive variable pitch wind turbine generator, with a tower height of 30-36 meters and a rotor diameter of 21-25 meters, which can be deployed on islands, coastal farms, or the plains of southern Xinjiang where photovoltaics cannot be used, to complement photovoltaics), photovoltaic array 2 (100kWp, 550Wp monocrystalline silicon × 182 blocks, occupying an area of ​​approximately 500-800 square meters, with optimal ground tilt angle for installation), energy storage unit 30 (a 200kWh / 100kW PCS lithium iron phosphate battery energy storage system, integrated in a 20-foot or 40-foot container, with charge and discharge management functions coordinated with wind turbine generator 1 and photovoltaic array 2), and intelligent microgrid controller 31. Wind turbine generator 1, photovoltaic array 2, and energy storage unit 30 are connected through intelligent microgrid controller 31 to form a wind-solar-storage complementary off-grid or grid-connected dual-mode power supply system, and energy storage unit 30 and intelligent microgrid controller 31 are integrated in power supply container 3.

[0069] The energy storage unit 30 uses a lithium iron phosphate battery pack, which is encapsulated in an aluminum alloy shell and integrates a battery management system (BMS), an energy storage converter (PCS), a thermal management system, and a fire protection system. The shell protection level is not lower than IP56, and the corrosion resistance level meets the ISO 12944 C5-M standard.

[0070] The smart microgrid controller 31 employs an adaptive energy dispatch algorithm based on wind and solar resource prediction and energy storage state of charge (SOC), including:

[0071] The daytime scheduling phase (executed once every 24 hours): After obtaining meteorological forecast data, wind and solar power are predicted through a neural network model. Combined with the load demand prediction of the desalination subsystem, a multi-objective optimization algorithm is used to solve the 24-hour energy storage charging and discharging plan.

[0072] Intraday rolling correction phase (executed every 5-15 minutes): Based on real-time collected photovoltaic output, wind power output, load power and energy storage state of charge, the prediction error is calculated, and the charging and discharging commands are adaptively corrected through model prediction control.

[0073] Real-time power balance control phase: Calculate net power in seconds and dynamically switch between energy storage charging, energy storage discharging, limiting new energy output, or cutting off some loads according to the zone control rules of energy storage charge status.

[0074] The charging priority management module dynamically allocates charging windows based on the charge status and task urgency of the drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit, and supports the humanoid robot inspection unit and robot dog patrol unit to charge or swap batteries for the drone inspection unit outdoors.

[0075] The desalination subsystem adopts a modular container structure and is integrated into the desalination container 4. Its inlet can be selectively connected to seawater or high-salinity brine, and its outlet has interfaces for industrial freshwater, agricultural freshwater, domestic freshwater, or direct drinking water, which are output through standardized quick-connect interfaces.

[0076] Specifically, the desalination subsystem includes a pretreatment module (multi-media filter 416, pre-activated carbon filter 417, and security filter 414, integrated at the front end of the desalination container 4), two sets of interchangeable ultrafiltration modules (8-inch ultrafiltration membrane modules, serving as a pretreatment barrier for reverse osmosis; the two sets can be switched for online cleaning), two sets of interchangeable reverse osmosis modules (vertical 8040 type reverse osmosis membrane modules; the two sets can be switched or operated in parallel; SW series seawater desalination membranes or BW series brackish water desalination membranes can be selectively configured according to the water source type), a posttreatment module (mineralization device, ultraviolet sterilizer, and activated carbon sterilizer; each device is equipped with a bypass valve, which can be bypassed or activated according to the target water quality), and an edge brain (deployed in the edge brain control cabinet 45).

[0077] The outlet of the reverse osmosis module is connected to the post-treatment module. When drinking water needs to be produced, the mineralization unit and ultraviolet sterilizer are activated to replenish beneficial minerals and kill bacteria. When agricultural irrigation water needs to be produced, the post-treatment module is partially bypassed to retain an appropriate amount of minerals. When industrial water (such as boiler feedwater) needs to be produced, the post-treatment module is completely bypassed, retaining only the necessary desalination treatment.

[0078] The desalination subsystem also includes an energy recovery device, which is an isobaric pressure exchanger (such as the ERIPX series). Its inlet end is connected to the concentrate discharge end of the reverse osmosis module, and its outlet end is connected to the inlet end of the reverse osmosis module. The energy recovery efficiency is not less than 95%.

[0079] The desalination subsystem also includes a concentrated brine treatment unit, comprising a foldable TPU drying tank 6 and a scraper-type salt collector. The TPU drying tank is folded and stored in a water tank container 5 during transport, with folded dimensions of 2.4m × 2.4m × 1m and unfolded dimensions of 20m × 10m × 0.3m (200 square meters). The scraper-type salt collector, installed at one end of the foldable TPU drying tank 6 and powered by solar energy, scrapes crystalline salt to a collection tank. The produced industrial-grade crude salt has a purity of no less than 85%, achieving resource utilization and near-zero emissions of the concentrated brine.

[0080] The water tank storage subsystem includes a collapsible TPU raw water tank and a TPU fresh water tank, integrated within water tank container 5. The interior of water tank container 5 is divided into 50m³ sections by partitions. 3 Raw water tank areas 51 and 25m 3Freshwater tank area 52 contains two tanks, both of which are TPU water bag structures (polyether-type TPU reinforced soft water bags, resistant to hydrolysis, salt spray, and UV radiation). These tanks are folded for storage during transport, saving space. During transport, the container also carries photovoltaic modules, support structures, drying tanks, and other goods. Upon arrival at the site, the TPU water bags are unfolded and connected to the desalination treatment container 4 via quick-connect fittings. This design combines the transport container and the water storage container into one unit, significantly reducing logistics costs.

[0081] When a larger water storage capacity is required, two water tank containers 5 can also be configured, one as a raw water tank and the other as a fresh water tank, connected by a parallel quick-connect interface.

[0082] When multiple containers are stacked, a retractable and deployable elevator is provided for humanoid robot inspection units and robot dog patrol units to allow them to move between different container levels.

[0083] like Figure 2 , Figure 3 As shown, the interior of the power supply container 3 is divided into three areas along its length. The first area (6 meters in length) is divided into two layers: the upper layer (starting from 1.3 meters above the ground) is equipped with a steel load-bearing platform, which houses the emergency backup diesel generator set 32 ​​(100kW automatic diesel generator set, optional), and is equipped with slide rails, hangers, flexible exhaust ducts, and intelligent spare parts and tool cabinets; the lower layer's bottom platform houses the energy storage unit 30 (lithium iron phosphate battery pack and BMS, PCS, thermal management, and fire protection system) and its intelligent spare parts and tool cabinet.

[0084] The second area (3 meters in length) houses the intelligent microgrid controller 31, the photovoltaic inverter control cabinet 33 (integrating inverter, combiner box, data acquisition unit, air conditioner, and monitoring system), and the wind power control cabinet 34, along with their intelligent spare parts boxes.

[0085] The third area (3 meters in length) has an independent charging compartment 35 for drone inspection units suspended at the top (starting 2 meters from the ground). This compartment is equipped with rails and electric push rods, which can push the drone airport out of the box for drone take-off and landing. The drone airport has an automatic charging function. The lower part is divided into two sections: the left section contains an independent charging compartment 35 for humanoid robot inspection units (including an automatic charging pile and spare parts box), and the right section contains an independent charging compartment 35 for robot dog patrol units (including an automatic charging pile and spare parts box).

[0086] The areas are separated by metal and insulating materials, and each partition wall has an automatic door in the middle for robot passage, which facilitates the humanoid robot to move between the areas.

[0087] like Figure 4 , Figure 5As shown, the interior of the desalination container 4 is divided into three areas along its length. The first area (3 meters long) is divided into upper and lower layers: the upper layer (starting from 1.9 meters above the floor) houses the pretreatment transition water tank 40 (a welded partition with an internal TPU water bag type tank); the lower layer is the raw water treatment area 41, which is divided into left and right sections along the center line, with a passage in the middle. The left section is equipped with: 2×6 AF-8040 ultrafiltration modules 411, an automatic inlet / outlet water switching valve, a pre-mounted titanium alloy impeller high-pressure pump 412, a titanium alloy corrosion-resistant self-priming pump with self-cleaning mesh 413 (3kW), and a security filter 414; the right section is equipped with: a 316L stainless steel self-cleaning disc filter 415 (100μm), a multi-media filter 416, and a pre-mounted activated carbon filter 417.

[0088] The second zone (7 meters in length): The top is equipped with a transition water tank 42 after the first reverse osmosis stage (a welded partition with an internal TPU water bag type tank); the lower part is the reverse osmosis zone 43, divided into left and right sections along the center line, with a passageway in the middle. The right section contains: 2×12 vertical RO-8040 primary reverse osmosis modules 4301 and their connectors and automatic switching valves, 2 secondary high-pressure water pumps 4302 (one in use and one on standby), 2×6 RO-8040 secondary reverse osmosis modules 4303, automatic inlet and outlet water switching valves, 2 freshwater output pumps 4304 (one in use and one on standby), an online water quality analyzer 4305, an ultraviolet disinfection device 4306, and a post-activated carbon filter 4307; the left section... It is equipped with: two rear-mounted titanium alloy impeller high-pressure pumps 4308 (one for use and one for standby), a ceramic rotor ERD pressure exchanger 4309, a scale inhibitor and disinfectant pump 4310, a membrane module cleaning pump 4311, a scale inhibitor and disinfectant liquid tank 4312, a membrane module cleaning liquid tank 4313, two mixing and injection pumps 4314 (one for use and one for standby), an intelligent mixing and mixing system 4315 (for the mixing of trace elements in agricultural water), and a mixing and mineralization device 4316 (for the mineralization of drinking water).

[0089] The third area (2 meters in length): The top is equipped with a desalination transition water tank 44 (TPU water bag type water tank); the lower part is divided into left and right sections along the center line, with a passage in the middle. The left section is equipped with an edge cloud brain control cabinet 45 (built-in edge cloud brain and PLC touch screen), and the right section is equipped with an intelligent spare parts warehouse 46 and a maintenance workbench 47.

[0090] Automatic doors connecting different areas allow humanoid robot inspection units to pass through, and all partitions are made of metal anti-corrosion, fireproof, and heat-insulating materials.

[0091] The intelligent inspection and security subsystem includes a drone inspection unit, a humanoid robot inspection unit (equipped with multimodal sensors and AI visual recognition algorithms), and a robot dog patrol unit (equipped with multimodal sensors and AI visual recognition algorithms). Each unit is housed in an independent charging compartment 35 within the power supply container 3, and all three units have automatic start-stop and automatic charging functions. The drone inspection unit is used to conduct aerial inspections of the photovoltaic array 2, wind turbine generator 1, and the surrounding environment of the system, identifying equipment anomalies and intrusion behaviors. The humanoid robot inspection unit is used to perform plug-in / plug-out connections, module replacements, pipeline tightening, and instrument reading operations for the equipment in the de-emphasis treatment container 4 and / or the power supply container 3. The robot dog patrol unit is used to proactively warn, drive away, or refuse entry to foreign objects that intrude into the warning area. The intelligent inspection and security subsystem also includes fixed cameras, intrusion sensors, audible and visual alarms, and a full-area long-range scanning radar system deployed at key parts of the device, as well as a remote monitoring and maintenance platform (based on 4G / 5G or satellite communication, enabling real-time monitoring of equipment status, predictive maintenance alarms, and the issuance of module replacement scheduling instructions).

[0092] The UAV inspection unit, humanoid robot inspection unit, and robot dog patrol unit are all wirelessly connected to the intelligent microgrid controller 31, the edge cloud brain, and the remote monitoring and maintenance platform, forming a three-layer heterogeneous robot collaborative maintenance system with aerial global perception, ground armed patrol, and in-cabinet precision operation. The specific configuration of the three-layer heterogeneous robot collaborative maintenance system is as follows: the UAV inspection unit performs a global scan of the photovoltaic array 2, wind turbine generator 1, and system perimeter along a preset route, generating a heat spot distribution map and a 3D environmental map, and sends the optimal inspection path to the humanoid robot inspection unit and robot dog patrol unit via the edge cloud brain; when the UAV inspection unit detects equipment abnormalities, the edge cloud brain dispatches the humanoid robot inspection unit to enter the corresponding container to perform precision inspection and repair; when the robot dog patrol unit detects intrusion or receives a radar alarm, the robot dog patrol unit initiates a drive-away procedure and simultaneously notifies the UAV inspection unit to take off for tracking and filming, and the UAV unit transmits real-time images back to the remote monitoring and maintenance platform.

[0093] All functional units within and between the energy supply subsystem, desalination subsystem, and intelligent inspection and security subsystem utilize standardized quick-connect interfaces. These interfaces include pipe connectors, cable quick-connect fittings, and mechanical quick-locking mechanisms, and are designed to be recognizable and operable by the humanoid robot inspection unit. Any module can be independently disassembled and replaced, supporting rapid on-site replacement of faulty modules and system expansion. Specifically, the standardized quick-connect interfaces include:

[0094] Hybrid power and communication connectors between containers: adopts military-grade / marine-grade circular bayonet connectors (such as XC / YMG series), five-key anti-misfit bayonet quick-lock structure, protection level IP67 / IP68, salt spray resistance not less than 1000 hours;

[0095] Water quick connectors between containers: adopt a large-diameter, leak-free quick-connect structure, made of 316L stainless steel or duplex stainless steel, with a pressure resistance of not less than 1.6MPa, pipe diameter DN25-DN100, and the three water lines of raw water (blue), fresh water (green), and concentrated brine (red) are distinguished by color.

[0096] Sanitary clamps / ferrules inside the container: used for high-pressure pipeline connections (pressure resistance of the high-pressure section before the RO membrane is not less than 10MPa), made of 316L stainless steel, and duplex stainless steel is used for the high-pressure section before the RO membrane.

[0097] Mechanical quick-locking mechanism: including rotary latch or eccentric cam structure, with robot vision-recognizable QR code or color block markings, and a handle or knob that can be plugged into by a robotic arm;

[0098] Photovoltaic bracket quick-connection: adopts composite material quick-connect brackets with a lifespan of no less than 30 years, C5 anti-corrosion rating, and no exposed metal parts;

[0099] Photovoltaic electrical connection: Uses MC4 photovoltaic quick-connect connector, IP68 protection rating, resistant to salt spray and ultraviolet rays;

[0100] Energy storage / inverter connection: Adopts Anderson high-power connectors with stacked busbars, with interlocking to prevent reverse connection and supports hot-swapping.

[0101] All functional devices and modules within the system are equipped with RFID electronic tags. The drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit interact with the edge cloud brain via RFID technology, achieving integrated management of equipment identification, maintenance record traceability, and self-positioning and navigation for each unit. Specifically, the tiered deployment of RFID electronic tags includes:

[0102] Anti-metal RFID tags are fixed to the casings or brackets of the following equipment: the frame of each photovoltaic module, wind power generation nacelle, TPU raw water tank, TPU fresh water tank, TPU concentrated brine drying pool, scraper salt collector, diesel generator set 32, all filter housings, ultrafiltration membrane module pressure vessel, RO membrane module pressure vessel, all water pumps, energy recovery device, cleaning and disinfection device tank, intelligent mixing and blending system 4315, and blending and mineralization device 4316; used to store the unique identifier of the equipment, production date, installation date, and maintenance records;

[0103] Small, anti-metal RFID tags are affixed to the surface of the following instruments: flow meters, pressure sensors, conductivity meters, laser water quality analyzers, and TDS / pH online monitors; used to store instrument model, calibration date, and next calibration reminder;

[0104] Flexible or weather-resistant RFID tags are wrapped or affixed to the housings of all quick-connect and hybrid connectors to record connection counts, the most recent insertion / removal time, and connection status confirmation.

[0105] RFID tags, either buried or wall-mounted, are installed or affixed to fixed points along the inspection path inside the power supply container 3 and the desalination container 4, for robot positioning, navigation, and path correction.

[0106] Small RFID tags are attached to each electric valve and actuator to record valve status, number of operations, and maintenance records.

[0107] The drone inspection unit includes a drone, a first camera, a first infrared thermal imager, a first RFID reader mounted on the drone, and an automatic charging nest located in the independent charging compartment 35 corresponding to the drone. The automatic charging nest has an automatic charging interface that matches the drone. The first RFID reader is used to read RFID electronic tags on devices or modules to assist in identification and positioning. The drone is also equipped with a positioning module and a communication module for communicating with the edge cloud brain and transmitting the images and thermal data collected by the first camera and the first infrared thermal imager to the edge cloud brain. The edge cloud brain has a built-in AI visual recognition algorithm for identifying photovoltaic array hot spots, wind turbine blade cracks, pipeline leaks, and intrusion targets in the surrounding environment.

[0108] The humanoid robot inspection unit includes a humanoid robot, a second camera and a second infrared thermal imager mounted on the head of the humanoid robot, a robotic hand mounted on the end of the arm, and a second RFID reader mounted on the head or arm; the humanoid robot's independent charging compartment 35 is equipped with a robot charging pile, which has a power supply interface, and the humanoid robot has a power receiving interface; the humanoid robot is used to enter and exit the automatic door with a dust-proof compartment on the power supply container 3 or the desalination container 4, and moves along the central aisle and modular interval aisles inside the container; the humanoid robot has the ability to inspect, repair, maintain, and replace parts of the drone inspection unit and the robot dog patrol unit;

[0109] The humanoid robot uses a built-in AI visual recognition algorithm to identify equipment malfunctions, pipeline leaks, and instrument parameters. It uses a second RFID reader to read and calibrate RFID tags to correct its navigation position, reads RFID electronic tags on the equipment to obtain its identity and maintenance information, and writes new data to the RFID electronic tags after maintenance is completed.

[0110] The humanoid robot has an internal library of embodied operation models and connects to the smart microgrid controller 31, the edge cloud brain, and the remote monitoring and maintenance platform via wireless communication. When maintenance or crisis management is required, it sequentially obtains operation instructions from the embodied model library, the smart microgrid controller 31 or the edge cloud brain, and the remote monitoring and maintenance platform, or receives AR annotation guidance from remote personnel through the digital twin model. The operation process and results are automatically generated into a report, which is reported to the remote monitoring and maintenance platform and backed up at the same time in the smart microgrid controller 31 or the edge cloud brain.

[0111] The robot dog patrol unit includes a robot dog, a non-lethal weapon system mounted on the robot dog, a third camera, a third infrared thermal imager, a third RFID reader, and a scanning radar linkage interface that wirelessly connects to a scanning radar deployed at a high point on the system perimeter; the non-lethal weapon system includes at least one of a laser dazzler, a sonic deflector, and an electromagnetic jammer; the robot dog's independent charging compartment 35 contains a robot dog charging station with a power supply interface, and the robot dog has a power receiving interface; the third RFID reader is used to read RFID electronic tags on the equipment or path to assist in navigation and identification;

[0112] The activation of the non-lethal weapon system is subject to dual authorization from the edge cloud brain or the remote monitoring and operation and maintenance platform: when the robot dog autonomously patrols and detects an intrusion target or receives a radar alarm, it will automatically issue a voice warning. If the intrusion target continues to approach, it will send a request to the edge cloud brain, and the non-lethal weapon system will be activated after authorization is obtained.

[0113] The inspection, maintenance, and repair process is as follows: The intelligent management center of the island or desert distributed sea / brine desalination system (cloud task allocation layer can be accessed via 5G / satellite) is responsible for global task decomposition, priority allocation, and multi-robot task scheduling. The intelligent microgrid controller and the sea / brine desalination edge cloud brain (edge ​​planning layer) are responsible for trajectory planning, dynamic obstacle avoidance, and multi-robot collaborative path generation. The robot intelligent controller (local control) is responsible for robot underlying motion control, sensor data acquisition, and command execution. A Thing Model is used to abstractly model the capabilities of all robots, uniformly defining attributes (battery power, attitude), services (takeoff, photography, movement), and events (low battery alarm, equipment abnormality). Multi-target inspection path planning is performed based on an improved maximum-minimum ant colony algorithm to minimize the total inspection distance and achieve balance among sub-paths. The D-CAPT algorithm is applied to achieve "local communication + local optimization," where robots only cooperate locally when they meet, reducing dependence on the central node and improving the system's fault resistance.

[0114] Drones are responsible for aerial inspections (photovoltaic modules, wind power equipment, park perimeter) and cross-regional material transportation. They can also provide global environmental perception information to humanoid robots and robot dogs from the air. Their mission types include high-priority aerial inspections of photovoltaic / wind power systems, autonomous drone takeoff, and execution along designated routes. Humanoid robots only intervene in joint air-ground inspections in case of anomalies. Drones take off first to perform global scanning and path planning, then distribute the optimal route to the ground robot. Drones are responsible for large-scale environmental monitoring (photovoltaic panel hotspot detection, overall equipment layout), long-distance delivery of parts and tools, and aerial hazard warnings. Drones conduct 15-20 minute inspections daily: thermal imaging scanning of photovoltaic modules, visible light photography, foreign object detection on module surfaces, surface inspection of wind power equipment blades, tower appearance inspection, and nacelle temperature monitoring. In case of suspicious behavior or escalating threats, drones monitor and issue warnings from the air, while maintaining contact with remote control personnel through a two-way communication system.

[0115] Humanoid robots are responsible for delicate operations within the container (instrument readings, valve operation, equipment insertion / removal, precision maintenance and repair), undertaking the most intricate operational tasks. Task type: Priority is delicate maintenance within the container; the system only dispatches a humanoid robot after a drone or robot dog detects an anomaly. The humanoid robot is responsible for close-range detailed inspections within the container (instrument readings, valve positions, equipment temperatures). The humanoid robot can perform complex operations within the container (such as replacing RO membranes, tightening quick-connect fittings). It assists in transporting robot dogs or drones when encountering complex terrain, malfunctions, or power outages; it monitors, maintains, and repairs drones and robot dogs. A container internal inspection is conducted every 4 hours: 30-40 minutes checking instrument readings, valve status, RO membrane pressure, and electrical cabinet temperature; it performs tasks such as inserting / removing connectors, tightening fittings, replacing filter / membrane modules, and maintaining electrical control modules within the container. In the event of a sudden equipment failure or leak, a humanoid robot is dispatched from the cloud, a drone continuously monitors the leak area, and a robot dog is responsible for delineating a safe isolation zone.

[0116] The robot dog is responsible for large-scale ground patrols, foreign object removal (equipped with non-lethal weapons), routine inspections, and providing charging / resupply for drones. Task types: Highest priority emergency security incidents (with drone aerial support if necessary), the robot dog responds quickly and performs close-range detailed inspections outside the container (instrument readings, valve positions, equipment temperatures); a fusion of genetic and ant colony algorithms optimizes the path, enabling the robot dog to recharge drones at designated locations. Every 30 minutes, the robot dog patrols the outside of the container: 5 minutes at a time, including perimeter surveillance, intrusion detection, non-lethal drone aerial monitoring and early warning; lower priority routine ground patrols, automatically returning to charging when battery is low. In the event of human intrusion / animal intrusion, the robot dog autonomously detects the intrusion and activates non-lethal weapons to issue voice warnings, laser dazzle, electromagnetic interference, or sonic repellent.

[0117] A containerized method for desalination of seawater and brine using unmanned intelligent wind, solar, and energy storage power supply includes the following steps:

[0118] S1: Select the corresponding reverse osmosis membrane module type (SW series or BW series) according to the water source type of the deployment location (seawater or high-salinity brine), and assemble the desalination treatment subsystem through standardized quick-installation interfaces;

[0119] S2: The intelligent microgrid controller 31, based on the adaptive energy dispatch algorithm, dynamically dispatches the wind turbine generator 1, photovoltaic array 2 and energy storage unit 30 according to wind and solar resource prediction and energy storage SOC to provide stable power supply for the desalination subsystem.

[0120] S3: Raw water is sequentially treated through a pretreatment module, an ultrafiltration module, a reverse osmosis module, and a post-treatment module to produce fresh water of the target quality. The activation or bypass status of the post-treatment module is selected according to water demand: when drinking water is needed, the mineralization device and ultraviolet sterilizer are activated; when agricultural irrigation water is needed, the post-treatment module is partially bypassed to retain an appropriate amount of minerals; when industrial water is needed, the post-treatment module is completely bypassed.

[0121] S4: The drone inspection unit takes off to conduct a global inspection of the photovoltaic array 2, wind turbine generator 1 and system perimeter. It uses AI visual recognition algorithm to identify equipment abnormalities and intrusion behaviors, and sends the inspection path to the humanoid robot inspection unit and / or robot dog patrol unit through the edge cloud brain.

[0122] S5: The robot dog patrol unit patrols the ground along the planned path. When it detects an intrusion, it activates the non-lethal weapon system to issue a warning and drive away the intruder, and notifies the drone inspection unit to take off to track and photograph the intruder. The real-time images are then transmitted back to the remote monitoring and maintenance platform.

[0123] S6: When an equipment malfunction is detected, the edge cloud brain dispatches a humanoid robot inspection unit to enter the corresponding container. The humanoid robot identifies the RFID tag on the faulty module through the second RFID reader, obtains the equipment identity and maintenance information, uses a robotic arm to unlock the mechanical quick-locking mechanism, pulls out the faulty module or disconnects the quick-connect connector, inserts a new module or reconnects the pipeline, and performs locking.

[0124] S7: For complex maintenance tasks, the humanoid robot receives AR annotation guidance from remote experts through AR glasses and a digital twin platform during the operation (the expert circles the operation point on the remote screen, and the AR annotation is superimposed on the robot's field of vision in real time). The operation process and results are automatically generated into a report and uploaded to the remote monitoring and maintenance platform, and written into the RFID tags of the fault module and the new module to form a closed-loop traceability.

[0125] This system employs a multi-robot coverage algorithm guided by the Energy Activity (EA) function for dynamic online path planning. Specifically, it includes: gridded environment modeling (using occupied grids to represent the coverage area, assigning an energy activity value to each grid to guide robots to prioritize coverage of highly active areas such as fault points and densely populated equipment areas); dynamic adaptation to terrain factors (fully considering terrain factors such as sand, gravel, steep slopes, and internal steps of containers, dynamically adjusting the path according to terrain complexity and equipment importance); online dynamic collaboration and collision avoidance (using the Artificial Potential Field Method (APF-CPP) to construct personalized coverage strategies through artificial potential fields, achieving adaptive task allocation); a conflict prevention coordination mechanism (using the MAC-Planner unified framework to integrate task allocation and path planning, updating task completion status in real time, effectively reducing conflicts); and collaborative obstacle avoidance and "dead zone" escape (robots collaborate through communication, collaboratively planning escape paths when trapped in "dead zones," and using a backtracking mechanism to optimize the path). Experimental results show that this algorithm can reduce the average path length by 5.2% and the number of turns by 30.5%.

[0126] The intelligent management platform of this system features remote digital twin and AR collaboration capabilities. A virtual model (rendered using glTF / Unity 3D modeling) is established, fully synchronized with the actual containerized seawater desalination equipment, including key components such as RO membrane modules, piping systems, electrical cabinets, and energy storage modules. Data collected by the robot, including equipment temperature, pressure, vibration, and images, is mapped to the digital twin model in real time. Maintenance personnel can view the equipment status from any location using a 3D visualization. When the humanoid robot discovers a problem during inspection, remote experts can view the equipment's 3D structure through the digital twin platform, use AR annotations to circle the problem area in the robot's field of view, and guide the robot to perform maintenance actions. The AR smart glasses are explosion-proof, waterproof, dustproof, anti-static, and corrosion-resistant, support 5G real-time collaboration (WebRTC / RTSP low-latency transmission), and support the display of synchronized 3D models of the real equipment and the on-site robot in virtual space. Experts can indirectly control the robot's actions by manipulating the virtual model.

[0127] This system employs multimodal fusion to ensure safety and reliability: For communication reliability, it utilizes multiple backups including 4G / 5G, satellite communication, local industrial Ethernet, fiber optic communication, and LoRa / NB-IoT, with critical commands undergoing dual-channel verification; for battery life assurance, the robot can automatically return to a wireless charging station, and drones can dock and charge on top of the robot (ground mobile resupply station mechanism); for data security, equipment operation and inspection data are encrypted and stored locally, and end-to-end encrypted transmission is used when uploading to the cloud; for redundant deployment, fixed monitoring cameras are deployed on each container as a supplement to robot inspections; for obstacle avoidance and collision prevention, it integrates laser radar... The system incorporates a multi-sensor fusion obstacle avoidance system using radar, depth cameras, and ultrasonic sensors. Regarding non-lethal weapon safety mechanisms, laser dazzlers (green and eye-safe), acoustic deflectors, and electromagnetic interference devices can only be activated after authorization and confirmation from remote operators. For fault self-recovery, the robot automatically reports a fault to the cloud, and the system reassigns tasks or commands it to return to the maintenance position. In terms of heat dissipation and cooling, the energy storage unit 30 and the backup diesel generator set 32 ​​all use a circulating water cooling system. All container exteriors are coated with aerogel insulation, the power distribution control cabinet uses circulating air cooling, and air conditioning is installed in the second area of ​​the power supply container 3 and the third area of ​​the desalination container 4.

[0128] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0129] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0130] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0131] The present invention has been described above by way of example. Obviously, the specific implementation of the present invention is not limited to the above-described manner. Any improvements made by adopting the inventive concept and technical solution of the present invention, or direct application to other occasions without modification, are all within the protection scope of the present invention.

Claims

1. An unmanned intelligent wind, solar, and energy storage power supply containerized seawater and brine desalination system, characterized in that: include: The energy supply subsystem includes a wind turbine generator (1), a photovoltaic array (2), an energy storage unit (30), and a smart microgrid controller (31). The wind turbine generator (1), the photovoltaic array (2), and the energy storage unit (30) are connected through the smart microgrid controller (31) to form a wind-solar-storage complementary off-grid or grid-connected dual-mode power supply system. The energy storage unit (30) and the smart microgrid controller (31) are integrated into the power supply container (3). The desalination subsystem includes a pretreatment module, two sets of interchangeable ultrafiltration modules, two sets of interchangeable reverse osmosis modules, a posttreatment module, and an edge cloud brain. The desalination subsystem is integrated into the desalination container (4). Its inlet can be selectively connected to seawater or high-salt brine, and its outlet has industrial freshwater, agricultural freshwater, domestic freshwater or direct drinking water interfaces. The water tank storage subsystem includes a foldable TPU raw water tank and a TPU fresh water tank, which are integrated into the water tank container (5); The intelligent inspection and security subsystem includes a drone inspection unit, a humanoid robot inspection unit, and a robot dog patrol unit. Each unit is installed in an independent charging compartment (35) in the power supply container (3), and all three have automatic start-stop and automatic charging functions. The drone inspection unit is used to conduct aerial inspections of the photovoltaic array (2), wind turbine generator set (1), and the surrounding environment of the system to identify equipment abnormalities and intrusion behaviors. The humanoid robot inspection unit is used to perform plug-in and plug-out connections, module replacements, pipeline tightening, and instrument reading operations for the equipment in the desalination container (4) and / or the power supply container (3). The robot dog patrol unit is used to actively warn, drive away, or refuse foreign objects that intrude into the warning area. The drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit are all connected to the intelligent microgrid controller (31), edge cloud brain, and remote monitoring and maintenance platform via wireless communication, forming a three-layer heterogeneous robot collaborative maintenance system of aerial global perception, ground armed patrol, and cabinet fine operation. All functional units within and between the energy supply subsystem, desalination subsystem, and intelligent inspection and security subsystem are connected using standardized quick-connect interfaces. These interfaces include pipe interfaces, cable quick connectors, and mechanical quick-locking mechanisms, and are designed to be recognizable and operable by humanoid robot inspection units. All functional devices and modules within the system are equipped with RFID electronic tags. The drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit interact with the edge cloud brain through RFID technology to achieve integrated management of equipment identification, maintenance record traceability, and self-positioning and navigation of each unit.

2. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The energy storage unit (30) uses a lithium iron phosphate battery pack, which is encapsulated in an aluminum alloy shell and integrates a battery management system, an energy storage converter, a thermal management system and a fire protection system. The shell protection level is not lower than IP56 and the corrosion resistance level meets the ISO 12944 C5-M standard.

3. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The intelligent microgrid controller (31) adopts an adaptive energy dispatch algorithm based on wind and solar resource prediction and energy storage state of charge, including: In the daytime dispatch phase: after obtaining meteorological forecast data, wind and solar power are predicted through a neural network model. Combined with the load demand prediction of the desalination subsystem, a multi-objective optimization algorithm is used to solve the 24-hour energy storage charging and discharging plan. Intraday rolling correction phase: Based on real-time collected photovoltaic output, wind power output, load power and energy storage state of charge, the prediction error is calculated, and the charging and discharging commands are adaptively corrected through model prediction control; Real-time power balance control phase: Calculate net power in seconds and dynamically switch between energy storage charging, energy storage discharging, limiting new energy output, or cutting off some loads according to the zone control rules of energy storage charge status. The charging priority management module dynamically allocates charging windows based on the charge status and task urgency of the drone inspection unit, humanoid robot inspection unit, and robot dog patrol unit, and supports the humanoid robot inspection unit and robot dog patrol unit to charge or swap batteries for the drone inspection unit outdoors.

4. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The desalination subsystem's reverse osmosis module uses two sets of vertical 8040 type reverse osmosis membrane modules. The two sets of membrane modules can be switched or operated in parallel. Seawater desalination membranes or brackish water desalination membranes are selectively configured according to the water source type. The outlet of the reverse osmosis module is connected to the post-treatment module, which includes a mineralization device, an ultraviolet sterilizer, and an activated carbon sterilizer. Each device is equipped with a bypass valve, which can be bypassed or activated according to the target water quality. The desalination subsystem also includes an energy recovery device, which is an isobaric pressure exchanger. Its inlet is connected to the concentrate discharge end of the reverse osmosis module, and its outlet is connected to the inlet end of the reverse osmosis module. The energy recovery efficiency is not less than 95%. The desalination subsystem also includes a concentrated brine treatment unit, which includes a foldable TPU drying tank (6) and a scraper salt collector. The TPU drying tank is folded and stored in a water tank container (5) during transportation. The scraper salt collector is installed at one end of the foldable TPU drying tank (6) to scrape the crystallized salt to the collection tank.

5. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The container structures of the power supply container (3), the desalination container (4), and the water tank container (5) are as follows: The power supply container (3) is divided into three areas along its length. The first area is divided into two layers: the upper layer houses the emergency backup diesel generator set (32), and the lower layer houses the energy storage unit (30). The second area houses the intelligent microgrid controller (31), the photovoltaic inverter control cabinet (33), and the wind power control cabinet (34). The third area is divided into two sections: the upper section houses the independent charging compartment (35) for the drone inspection unit, and the lower section is divided into two areas, which house the independent charging compartments (35) for the humanoid robot inspection unit and the robot dog patrol unit, respectively. Automatic doors for the humanoid robot inspection unit to pass through are provided between the areas. The desalination container (4) is divided into three areas along its length. The first area is divided into two layers. The upper layer is equipped with a pretreatment transition water tank (40), and the lower layer is the raw water treatment area (41). This area is divided into two sections along the center line, with a passage in the middle. One side is equipped with an ultrafiltration module (411), a pre-mounted titanium alloy impeller high-pressure pump (412), a titanium alloy self-cleaning mesh corrosion-resistant self-priming pump (413), and a security filter (414). The other side is equipped with a self-cleaning disc filter (415), a multi-media filter (416), and a pre-mounted activated carbon filter (417). The second area is equipped with a first-stage reverse osmosis transition water tank (42) at the top and a reverse osmosis area (43) at the bottom. It is divided into two sections along the center line, with a passage in the middle. One side is equipped with a first-stage reverse osmosis module (4301), a second-stage pressurized water high-pressure pump (4302), a second-stage reverse osmosis module (4303), and a freshwater conveyor. The system includes a pump (4304), an online water quality analyzer (4305), an ultraviolet disinfection device (4306), and a post-activated carbon filter (4307). On the other side, there is a post-titanium alloy impeller high-pressure pump (4308), a ceramic rotor ERD pressure exchanger (4309), a scale inhibitor and disinfectant pump (4310), a membrane module cleaning pump (4311), a scale inhibitor and disinfectant liquid tank (4312), a membrane module cleaning liquid tank (4313), a mixing and injection pump (4314), an intelligent mixing and mixing system (4315), and a mixing and mineralization device (4316). The top of the third area is equipped with a desalination transition water tank (44), and the lower part is divided into two areas along the center line. The middle is a passage, and one side is equipped with an edge cloud brain control cabinet (45). The other side is equipped with an intelligent spare parts warehouse (46) and a maintenance workbench (47). Automatic doors for humanoid robot inspection units to pass through are provided between the areas. The water tank container (5) is divided into 50m sections by partitions. 3 Raw water tank area (51) and 25m 3 Freshwater tank area (52), both tanks are TPU water bag type structures, which are folded and stored during transportation, and connected to the desalination container through quick-connect couplings after being unfolded on site; when multiple containers are stacked, a retractable and unfoldable elevator is provided for humanoid robot inspection units and robot dog patrol units to pass between different container levels.

6. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The drone inspection unit includes a drone, a first camera mounted on the drone, a first infrared thermal imager, a first RFID reader, and an automatic charging nest located in the drone's corresponding independent charging compartment (35). The automatic charging nest has an automatic charging interface that matches the drone. The first RFID reader is used to read RFID electronic tags on the device or module to assist in identification and positioning. The drone is also equipped with a positioning module and a communication module, which are used to communicate with the edge cloud brain and transmit the images and thermal images collected by the first camera and the first infrared thermal imager to the edge cloud brain. The edge cloud brain has a built-in AI visual recognition algorithm to identify photovoltaic array hot spots, wind turbine blade cracks, pipeline leaks, and intrusion targets in the surrounding environment.

7. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The humanoid robot inspection unit includes a humanoid robot, a second camera and a second infrared thermal imager installed on the head of the humanoid robot, a robotic hand installed at the end of the arm, and a second RFID reader installed on the head or arm; the humanoid robot is equipped with a robot charging pile in its independent charging compartment (35), the robot charging pile has a power supply interface, and the humanoid robot has a power receiving interface; the humanoid robot is used to enter and exit the automatic door with a dustproof room on the power supply container (3) or the desalination container (4), and moves along the middle channel and the module interval channel inside the container; the humanoid robot has the ability to inspect, repair, maintain and replace parts of the drone inspection unit and the robot dog patrol unit; The humanoid robot uses a built-in AI visual recognition algorithm to identify equipment malfunctions, pipeline leaks, and instrument parameters. It uses a second RFID reader to read and calibrate RFID tags to correct its navigation position, reads RFID electronic tags on the equipment to obtain its identity and maintenance information, and writes new data to the RFID electronic tags after maintenance is completed. The humanoid robot has an internal library of body-based operation models and connects to the smart microgrid controller (31), the edge cloud brain, and the remote monitoring and maintenance platform via wireless communication. When maintenance or crisis management is required, it can obtain operation instructions from the body model library, the smart microgrid controller (31) or the edge cloud brain, and the remote monitoring and maintenance platform in sequence, or receive AR annotation guidance from remote human through the digital twin model. The operation process and results are automatically generated into a report, which is reported to the remote monitoring and maintenance platform and backed up at the same time in the smart microgrid controller (31) or the edge cloud brain.

8. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The robot dog patrol unit includes a robot dog, a non-lethal weapon system mounted on the robot dog, a third camera, a third infrared thermal imager, a third RFID reader, and a scanning radar linkage interface that is wirelessly connected to a scanning radar deployed at a high point on the perimeter of the system; the non-lethal weapon system includes at least one of a laser dazzler, a sonic deflector, and an electromagnetic jammer; the robot dog's independent charging compartment (35) is equipped with a robot dog charging pile, the robot dog charging pile has a power supply interface, and the robot dog has a power receiving interface; the third RFID reader is used to read RFID electronic tags on the equipment or path to assist navigation and identification; The activation of the non-lethal weapon system is subject to dual authorization from the edge cloud brain or the remote monitoring and operation and maintenance platform: when the robot dog autonomously patrols and detects an intrusion target or receives a radar alarm, it will automatically issue a voice warning. If the intrusion target continues to approach, it will send a request to the edge cloud brain, and the non-lethal weapon system will be activated after authorization is obtained.

9. The unmanned intelligent wind-solar-storage power supply containerized seawater and brine desalination system according to claim 1, characterized in that, The three-layer heterogeneous robot collaborative operation and maintenance system is specifically configured as follows: the UAV inspection unit performs a global scan of the photovoltaic array (2), wind turbine generator (1) and system perimeter according to the preset route, generates a hot spot distribution map and a three-dimensional environment map, and sends the optimal inspection path to the humanoid robot inspection unit and the robot dog patrol unit through the edge cloud brain; when the UAV inspection unit finds equipment abnormality, the edge cloud brain schedules the humanoid robot inspection unit to enter the corresponding container to perform fine inspection and maintenance; when the robot dog patrol unit finds intrusion behavior or receives a radar alarm, the robot dog patrol unit starts the expulsion program and simultaneously notifies the UAV inspection unit to take off for tracking and shooting, and the UAV inspection unit transmits the real-time image back to the remote monitoring and operation and maintenance platform.

10. A method for unmanned intelligent wind-solar-storage-powered containerized seawater and brine desalination, comprising using the unmanned intelligent wind-solar-storage-powered containerized seawater and brine desalination system as described in any one of claims 1-9, characterized in that, Includes the following steps: S1: Select the corresponding reverse osmosis membrane module type according to the water source type of the deployment location, and assemble the desalination treatment subsystem through standardized quick-installation interfaces; S2: The intelligent microgrid controller (31) supplies power to the desalination subsystem based on an adaptive energy scheduling algorithm; S3: Raw water is sequentially treated through a pretreatment module, an ultrafiltration module, a reverse osmosis module, and a post-treatment module to produce fresh water of the target quality. The activation or bypass status of the post-treatment module is selected according to water demand: when drinking water is needed, the mineralization device and ultraviolet sterilizer are activated; when agricultural irrigation water is needed, the post-treatment module is partially bypassed to retain an appropriate amount of minerals; when industrial water is needed, the post-treatment module is completely bypassed. S4: The drone inspection unit takes off to conduct a global inspection of the photovoltaic array (2), wind turbine generator (1) and system perimeter. It identifies equipment abnormalities and intrusion behaviors through AI visual recognition algorithms and sends the inspection path to the humanoid robot inspection unit and / or robot dog patrol unit through the edge cloud brain. S5: The robot dog patrol unit patrols the ground along the planned path. When it detects an intrusion, it activates the non-lethal weapon system to issue a warning and drive away the intruder, and notifies the drone inspection unit to take off and track it. S6: When an equipment malfunction is detected, the edge cloud brain dispatches a humanoid robot inspection unit to enter the corresponding container. The humanoid robot identifies the RFID tag on the faulty module through the second RFID reader, uses a robotic arm to unlock the mechanical quick-locking mechanism, pulls out the faulty module or disconnects the quick-connect connector, inserts a new module or reconnects the pipeline, and performs locking. S7: During operation, the humanoid robot receives AR annotation guidance from remote experts through AR glasses and a digital twin platform. The operation process and results are automatically generated into reports and uploaded to the remote monitoring and maintenance platform, and written into the RFID tags of fault modules and new modules.

Images