An internet of things self-generating container AI selling system based on cloud data analysis

Abstract

The application is particularly a thing networking self-generating container AI selling system based on cloud data analysis, relating to the technical field of unmanned retail equipment, comprising: a full-shell radiation refrigeration-ultra-hydrophobic self-cleaning composite module is covered on the full-shell surface of the cabinet body, a galvanized steel plate is used as a base material to construct a reverse energy flow composite structure of outer self-cleaning, middle power generation and inner refrigeration, and rainwater self-cleaning, scattered light power generation and zero-energy-consumption passive cooling are simultaneously realized, thereby reducing the cabinet body cooling load and operation and maintenance cost from the source.The multi-faceted cadmium telluride power generation system is integrated with the full-shell radiation refrigeration structure in the application, and the overall energy efficiency performance is synergistically optimized from the two ends of energy supply gain and energy consumption reduction; the top power generation glass is combined with the side wall flexible power generation film, which can capture and utilize multi-form light energy such as direct light and scattered light, thereby improving the comprehensive utilization efficiency of light energy and providing a continuous clean energy supply for the equipment.

Description

Technical Field

[0001] This invention relates to the field of unmanned retail equipment technology, and in particular to an IoT-based self-generating vending machine AI vending system based on cloud data analysis. Background Technology

[0002] With the continuous development of unmanned retail, outdoor vending machines have been widely used in various scenarios such as urban commercial districts, transportation hubs, and scenic areas. To ensure the storage quality of food and beverage products, existing outdoor vending machines are generally equipped with compressor refrigeration systems and insulated enclosures. Most devices rely on the municipal power grid for continuous power supply, while some special locations supplement power by installing photovoltaic modules to expand the deployment range and adaptability of the equipment to different scenarios.

[0003] Regarding cabinet temperature control and energy utilization, existing equipment typically uses methods such as thickening the insulation layer and applying heat insulation film to block external heat transfer, which can reduce the operating load of the refrigeration system to some extent. Photovoltaic power supply solutions mostly use crystalline silicon modules with brackets installed on the top of the cabinet to realize the conversion and utilization of direct sunlight. Under outdoor high temperature and sun exposure conditions, the temperature rise caused by heat absorption on the cabinet surface will still increase the cooling energy consumption. At the same time, the scattered sunlight energy in areas such as the side walls of the cabinet is not fully utilized, indicating room for optimization in the synergy between energy supply and temperature control.

[0004] In terms of equipment operation and maintenance, dust and stains in the outdoor environment are easy to adhere to the surface of the cabinet, which usually requires regular manual cleaning to maintain the equipment condition. There is potential for optimization in operation and maintenance investment and cleaning cycle. At present, most networked vending machines have achieved basic operation status monitoring and sales data statistics, but in the refined operation dimensions such as dynamic energy consumption optimization, power generation performance prediction, and intelligent inventory scheduling, the ability of in-depth data analysis and global collaborative management can still be further improved. Summary of the Invention

[0005] The purpose of this invention is to provide an AI-powered vending system for self-generating IoT vending machines based on cloud data analysis in order to solve the above-mentioned problems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: An IoT-based self-generating vending machine AI vending system based on cloud data analysis includes: The integrated cabinet body and basic enclosure module serve as the physical load-bearing foundation of the system. It adopts a frame load-bearing and layered enclosure structure, and is equipped with adjustable merchandise shelves inside. The whole system has outdoor waterproof, anti-corrosion and thermal insulation properties. The multi-faceted cadmium telluride self-generating and energy storage power supply module is set on the top surface, side walls and back panel of the cabinet. The cadmium telluride double glass power generation glass on the top surface is the main power generation unit, and the flexible cadmium telluride thin film power generation layer on the side wall is the supplementary power generation unit. It is equipped with a high-voltage adaptable MPPT controller, lithium iron phosphate energy storage battery and mains power backup interface to form a multi-source power supply architecture. The full-shell radiative cooling-superhydrophobic self-cleaning composite module is covered on the entire shell surface of the cabinet. It uses galvanized steel plate as the base material to build a reverse energy flow composite structure with external self-cleaning, central power generation and internal cooling. It simultaneously realizes rainwater self-cleaning, scattered light supplementary power generation and zero-energy passive cooling, reducing the cabinet's cooling load and operation and maintenance costs from the source. The bottom-mounted independent cooling and heat dissipation and local electronic control interaction module is located in an independent compartment at the bottom of the cabinet. It integrates the cooling and heat dissipation system, local main control unit, gravity sensor settlement unit and human-machine security interaction unit to realize local autonomous temperature control of the cabinet and automatic settlement of goods. The cloud-based data analysis and interconnected intelligent operation module connects all vending machine terminals via wireless communication, gathers all operational data, and incorporates multi-dimensional intelligent analysis algorithms. It is paired with a full-link operation management backend to achieve interconnected intelligent control and refined operation of vending machines across the entire network.

[0007] Preferably, the main frame of the integrated cabinet body and the basic enclosure module is welded from galvanized rectangular steel pipes. The enclosure structure adopts a three-layer structure consisting of an outer functional layer, a polyurethane foam insulation layer, and a food-grade stainless steel inner liner. The front door adopts a thermally broken aluminum alloy door frame with double-layer tempered Low-E hollow glass. The hollow layer is filled with argon gas. The door is surrounded by double-ring magnetic sealing strips and has a built-in electromagnetic anti-theft lock and human body sensing components.

[0008] Preferably, the cabinet top of the multi-faceted cadmium telluride self-generating and energy storage power supply module is provided with a 6° single-slope slope that is lower in the front and higher in the back. The slope is installed with cadmium telluride double-glass power generation glass with a cut size of 700×850mm. It is sealed and fixed with aluminum profile frame, EPDM gasket and aluminum alloy pressure plate. A continuous water guide channel and drainage hole are opened at the front end of the slope. The power generation line is connected to the top electrical circuit.

[0009] Preferably, a flexible cadmium telluride thin-film power generation layer is bonded to the sidewall and backplate of the multi-faceted cadmium telluride self-generating and energy storage power supply module by means of a highly thermally and electrically conductive adhesive. The flexible thin-film power generation layer is composed of multiple CdTe thin-film modules connected in series and parallel, which are used to capture scattered light and side-sunlight to supplement power generation. The power generation lines are connected to the top electrical compartment for unified management.

[0010] Preferably, the multi-faceted cadmium telluride self-generating and energy storage power supply module is equipped with 4 independent MPPT controllers, which independently perform maximum power point tracking for the power generation units on the top surface and three side walls respectively; it is equipped with a 24V / 50Ah lithium iron phosphate battery pack and a BMS battery management system, with a built-in automatic power switching module that supports seamless switching between photovoltaic priority and grid priority power supply modes.

[0011] Preferably, the full-shell radiative cooling-superhydrophobic self-cleaning composite module consists of, from the outside to the inside, a superhydrophobic self-cleaning surface layer, a flexible power generation layer, an epoxy primer layer, a metal substrate layer, and an inner radiative cooling coating; the superhydrophobic surface layer has a water contact angle ≥152°, and the inner radiative cooling coating has an emissivity ≥98% and a solar reflectance ≥93% in the 8~13μm atmospheric transparency window band.

[0012] Preferably, the heat dissipation chamber of the bottom-mounted independent cooling and heat dissipation and local electronic control interaction module adopts a left-right partition layout, and is equipped with a unidirectional airflow duct with front-low air intake and rear-up air outlet. The air inlet is equipped with a dust filter, and the air outlet is equipped with a stepless speed-regulating cooling fan. A water collection tray and porous water-absorbing cotton are set inside the heat dissipation chamber, and the condensate is self-evaporated by using the hot air discharged by the condenser.

[0013] Preferably, the bottom-mounted independent cooling and heat dissipation and local electronic control interaction module adopts an ARM architecture industrial-grade main control unit with built-in adaptive temperature control algorithm and fault self-diagnosis function; a high-precision gravity sensor is configured under each shelf to identify the quantity and type of goods to be picked up and put down through a dynamic weight detection algorithm, so as to realize opening the door to pick up goods and automatic settlement.

[0014] Preferably, the cloud-based data analysis and interconnected intelligent operation module adopts a microservice architecture and a time-series database, uniformly accesses all energy, environmental, operational, and maintenance data from all terminals, supports multi-dimensional query statistics and automatic report generation, and transmits terminal data encrypted via the MQTT protocol.

[0015] Preferably, the cloud-based data analysis and interconnected intelligent operation module incorporates three core algorithms: intelligent energy consumption optimization, power generation performance operation and maintenance prediction, and intelligent sales prediction; it is equipped with a web-based and mobile operation management backend, supporting remote equipment monitoring, full-process work order management, product marketing management, and multi-level permission configuration.

[0016] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. This invention integrates a multi-faceted cadmium telluride power generation system with a full-shell radiative cooling structure, thereby synergistically optimizing the overall energy efficiency from both the power supply gain and energy consumption reduction ends. The combination of the top power generation glass and the side wall flexible power generation film can capture and utilize various forms of light energy, such as direct light and scattered light, thereby improving the comprehensive utilization efficiency of light energy and providing a continuous supply of clean energy for the equipment.

[0017] 2. This invention combines a superhydrophobic self-cleaning coating with a cloud-based intelligent operation system, which can effectively reduce equipment operation and maintenance costs and improve the level of operational precision. The self-cleaning structure on the shell surface can be washed away by rainwater, extending the interval between manual cleaning and reducing the manpower input for routine cleaning and maintenance. Attached Figure Description

[0018] Further details, features, and advantages of this application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which: Figure 1 This is a system flowchart of the present invention; Figure 2 This is a diagram of the vending machine of the present invention. Detailed Implementation

[0019] Several embodiments of this application will now be described in more detail with reference to the accompanying drawings to enable those skilled in the art to implement this application. This application may be embodied in many different forms and for various purposes and should not be limited to the embodiments set forth herein. These embodiments are provided to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. The embodiments described do not limit this application.

[0020] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having a meaning consistent with their meaning in the relevant field and / or the context of this specification, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0021] Example 1

[0022] Its specific implementation method is combined with the appendix Figure 1 and attached Figure 2 Please provide a detailed explanation.

[0023] In this embodiment, it includes: Module 1: Integrated Cabinet Body and Basic Enclosure Module This module serves as the physical foundation of the entire sales system. It adopts a design approach of "frame load-bearing + layered enclosure + modular assembly," taking into account outdoor structural strength, thermal insulation performance, and ease of assembly. All structural parameters have been specifically optimized for outdoor open-air scenarios.

[0024] 1. Main load-bearing frame construction process The main frame of the cabinet is constructed from 40×20×1.5mm galvanized rectangular steel pipes, with columns, beams, and top and bottom frames fully welded together. The weld height is ≥2mm, and a galvanized repair agent is evenly applied after welding to ensure a continuous anti-corrosion layer. The overall frame has undergone static load testing, and the top can withstand a uniformly distributed load of 150kg. The front can withstand a deformation of ≤1 / 500 under a level 12 wind load, meeting the requirements for extreme outdoor conditions such as typhoons and snow accumulation. Standardized installation slots are pre-installed on the inner side of the frame for quick installation of side wall panels, inner liner, and insulation layer, enabling modular assembly and increasing production efficiency by 40%.

[0025] 2. Layered enclosure and insulation structure The cabinet's side walls, back panel, top surface, and bottom surface all adopt a three-layer enclosure structure consisting of an outer functional layer, an insulation layer, and an inner liner, effectively blocking heat conduction paths. Thermal insulation layer: Utilizing a high-pressure polyurethane integral foaming process, the foaming material is a two-component mixture of black and white components. It is injected into the mold cavity under high pressure at a constant temperature of 42℃ and a pressure of 15MPa, with a curing time of 35 minutes. After foaming, the density stabilizes at 42±2kg / L. The thermal conductivity is ≤0.022W / (m·K). Positioning clips are pre-embedded in the inner wall of the shell before foaming to prevent board displacement during the foaming process, eliminate splicing gaps, and prevent cold bridges. The insulation layer thickness is uniformly 40mm. An additional 10mm thick extruded polystyrene board is added to areas prone to cold bridges, such as door frames and columns, reducing heat loss due to cold bridges by 60%.

[0026] Inner Liner: Made of 0.8mm thick food-grade 304 stainless steel plate, bent into one piece, with full argon arc welding at the joints, and the weld seams are ground and polished to Ra≤1.6μm, eliminating any unsanitary corners. The inner liner features 3 adjustable shelves with 50mm spacing between shelf support slots, allowing for 50mm height adjustments to accommodate products of different packaging sizes. Each shelf is made of food-grade ABS material with reinforced frame design, enabling a single-point load capacity ≥20kg and a uniformly distributed load capacity ≥50kg.

[0027] Inner liner accessories: Two side-emitting LED cold light strips are embedded in the top of the inner liner, with a color temperature of 6000K, a total power of 5W, an illuminance uniformity of ≥0.7, and no direct glare; the light strips are controlled by a human body sensor, turning off when no one is present and automatically turning on when the door is opened, reducing standby power consumption. A pre-drilled area for the evaporator is provided on the back of the inner liner, with a flat surface treatment to ensure a tight fit with the evaporator, improving heat exchange efficiency by 15%.

[0028] 3. Precision structure of the front door The door frame is constructed from extruded aluminum alloy with thermal break. The thermal break strip, made of PA66 + 25% glass fiber and 14.8mm wide, effectively blocks solid-borne heat conduction between the inner and outer sides of the frame. The inner cavity of the frame is simultaneously filled with polyurethane foam to further reduce heat loss. The door glass features a double-layered tempered Low-E hollow structure: the outer layer is Low-E tempered glass with online sunlight control coating, located inside the hollow layer to prevent abrasion; the inner layer is ordinary float tempered glass. The hollow layer is 12mm thick and filled with 92% pure argon gas, with the edges sealed using a combination of butyl rubber and structural adhesive. The overall heat transfer coefficient K≤1.8W / (m²·K), shading coefficient SC≤0.25, blocks over 92% of solar infrared radiation, and has a visible light transmittance ≥70%, balancing heat insulation and display performance. The door is equipped with double-ring magnetic EPDM sealing strips around its perimeter. The strips have a porous honeycomb structure and a compression set of ≤10%. When the door is closed, the magnetic attraction creates a double airtight seal, with a cold air leakage rate of ≤0.05 m³ / (h·m). The left side of the door features two stainless steel damping hinges with an opening angle of 110° and a 90° position locking function. The right side has a built-in electromagnetic anti-theft lock with a 12mm bolt travel, achieving a civilian level 2 anti-pry rating. It supports three unlocking methods: QR code authorization, remote unlocking via cloud, and local emergency mechanical unlocking.

[0029] 4. Bottom support and overall machine protection The cabinet base is equipped with four M16 adjustable leveling feet, with an adjustment range of ±30mm, adaptable to uneven outdoor ground. A 10mm thick rubber shock-absorbing pad is installed between the feet and the cabinet base plate to buffer compressor vibration and isolate moisture from the ground. The outer edge of the base plate is turned up 15mm to form a water-retaining edge, preventing ground water from seeping into the cabinet. The overall casing has an IPX4 protection rating, and the electrical compartment has an IP65 protection rating. All external seams are sealed with neutral silicone weather-resistant sealant, with uniform sealant width and no breaks. All metal sheets undergo double anti-corrosion treatment with galvanizing and primer, showing no red rust after ≥1000h neutral salt spray testing, and an outdoor design service life of ≥8 years.

[0030] Module 2: Multi-faceted Collaborative Cadmium Telluride Self-generating and Energy Storage Power Supply Module This module is the energy core of the system. It adopts a multi-source power supply architecture of "top surface main power generation + side wall supplementary power generation + energy storage buffer + mains power backup". Based on the low light advantage of cadmium telluride material, it can capture light energy at all times and in multiple directions, ensuring long-term continuous operation in the absence of mains power.

[0031] 1. Top-mounted cadmium telluride power generation glass subsystem The cabinet base is equipped with four M16 adjustable leveling feet, with an adjustment range of ±30mm, adaptable to uneven outdoor ground. A 10mm thick rubber shock-absorbing pad is installed between the feet and the cabinet base plate to buffer compressor vibration and isolate moisture from the ground. The outer edge of the base plate is turned up 15mm to form a water-retaining edge, preventing ground water from seeping into the cabinet. The overall casing has an IPX4 protection rating, and the electrical compartment has an IP65 protection rating. All external seams are sealed with neutral silicone weather-resistant sealant, with uniform sealant width and no breaks. All metal sheets undergo double anti-corrosion treatment with galvanizing and primer, showing no red rust after ≥1000h neutral salt spray testing, and an outdoor design service life of ≥8 years.

[0032] Module 2: Multi-faceted Collaborative Cadmium Telluride Self-generating and Energy Storage Power Supply Module This module is the energy core of the system. It adopts a multi-source power supply architecture of "top surface main power generation + side wall supplementary power generation + energy storage buffer + mains power backup". Based on the low light advantage of cadmium telluride material, it can capture light energy at all times and in multiple directions, ensuring long-term continuous operation in the absence of mains power.

[0033] 1. Top-mounted cadmium telluride power generation glass subsystem The cabinet top adopts a single-slope design, lower in the front and higher in the back, with a precise slope of 6°. This angle was optimized through multiple objectives: compared to a 0° horizontal top surface, drainage speed is increased by 62%, dust accumulation is reduced by 41%, annual solar radiation reception is increased by 8%, and the wind resistance coefficient is increased by no more than 5%, balancing drainage, self-cleaning, power generation efficiency, and structural safety.

[0034] Sloping base and drainage structure: The sloping base is integrally stamped from a 1.5mm thick galvanized steel plate, with a slope flatness tolerance of ≤±2mm / m, ensuring a complete fit of the power generation glass without any gaps. The sloping base has a 20mm upward-curving edge around its perimeter to form a water-blocking rim. A continuous U-shaped water-guiding channel is installed at the lowest point of the front end, with the channel sloping at a 1% angle to both ends to guide rainwater to collect at both ends. The drainage hole has a diameter of 10mm, with a 304 stainless steel dust filter installed at the opening using a snap-fit ​​design. The filter has a 2mm mesh size to intercept fallen leaves and large particles of sand and dust; the filter can be removed and cleaned by hand. Below the drainage hole, an 8mm PVC water pipe is connected, concealed along the inside of the cabinet column to the bottom side outlet, allowing rainwater to flow down the side wall and preventing it from washing over the front door and display screen.

[0035] Power generation glass assembly and sealing: Custom-cut 700×850mm cadmium telluride double-glass power generation glass is used. Both upper and lower encapsulation layers are 3.2mm thick tempered ultra-clear patterned glass, with a CdTe light-absorbing layer in between. The total module thickness is 7.8mm. Rated parameters: Maximum output power Pmax=62W, optimal operating voltage Vmpp=146V, optimal operating current Impp=0.42A, open circuit voltage Voc=183V, short circuit current Isc=0.48A, module conversion efficiency ≥10.4%, fire resistance rating A, and hail impact resistance conforms to IEC61646 standard. The assembly adopts a three-level waterproof sealing structure: The first layer consists of a 3mm thick EPDM rubber gasket between the aluminum alloy profile edging and the beveled base, with the compression controlled at 30%, to achieve buffering and the first layer of waterproofing; The second layer consists of aluminum alloy pressure plates spaced 150mm apart along the four sides, fixed with M4 stainless steel self-tapping screws, and a 2mm thick silicone buffer pad between the pressure plate and the glass to distribute stress and prevent the glass from breaking; The third layer consists of neutral silicone weather-resistant sealant applied to all joints, with a sealant width ≥8mm and a sealant thickness ≥3mm. A primer is applied before applying the sealant to ensure bonding strength.

[0036] Electrical connection: The back of the power generation glass integrates a junction box with a built-in bypass diode to prevent hot spot effect caused by local shading; the positive and negative leads use 2.5mm² tinned copper core photovoltaic cables, which enter the top electrical compartment through the IP68 waterproof gland on the sloping side, and the cable entry point is filled with waterproof sealant; after the cable enters the compartment, it is first connected to a 10A DC fuse, and then connected to the MPPT controller.

[0037] 2. Flexible Cadmium Telluride Thin Film Power Generation System Flexible cadmium telluride thin-film power generation layers are attached to the left and right side walls and the outer side of the back panel of the cabinet to supplement the power generation on the top surface. This fully utilizes scattered light, ground reflected light and morning and evening side light to generate electricity and improve the overall light energy utilization rate.

[0038] Thin-film module and bonding process: Utilizing polymer-based flexible CdTe thin-film batteries, each cell measures 300×600mm with a thickness ≤2mm, allowing for complete bonding to flat and slightly curved surfaces. Bonding is achieved using a high thermal and electrical conductivity acrylic adhesive via a roll-pressing process at a pressure of 0.3MPa, achieving a bonding accuracy ≥98%, with no bubbles or edge lifting. After curing, the peel strength is ≥15N / cm. Six modules are arranged on a single side wall, connected in a 3-series, 2-parallel configuration, providing a peak power of 25W per side. Eight modules are arranged on the back panel, connected in a 4-series, 2-parallel configuration, providing a peak power of 30W. The combined peak power supplement from all three sides is 80W, with daily supplementary power generation reaching 1.3 times that of the top side.

[0039] Performance and Wiring: The thin-film module has a spectral response range of 300-900nm, with higher response efficiency in the 500-700nm visible light band, making it particularly suitable for overcast and shaded environments with diffused light. Its low-light power generation performance is more than 20% better than that of crystalline silicon modules. The module leads are concealed along the side of the steel plate, passing through flame-retardant cable trays and converging into the top electrical compartment. The wiring joints are double-insulated with heat-shrink tubing and waterproof adhesive. The output voltage is matched to the top-mounted power-generating glass and can be uniformly connected to the MPPT controller.

[0040] 3. Energy Management and Energy Storage Subsystem All power generation, energy storage, and power distribution components are centrally located in an independent electrical compartment at the top rear. The compartment is IP65 protected, with built-in waterproof and breathable valves to balance air pressure and prevent condensation, and equipped with miniature temperature-controlled fans to ensure heat dissipation.

[0041] High-voltage adapted MPPT controller: Employs an industrial-grade 4-channel independent MPPT controller with an input voltage range of 60~200V, perfectly adapting to the high-voltage output characteristics of cadmium telluride modules. The 4 channels correspond to the top, left side wall, right side wall, and back panel of the 4 power generation units, independently performing maximum power point tracking (MPPT). Utilizing a variable step-size perturbation observation algorithm, the tracking efficiency is ≥99.5%, and the conversion efficiency is ≥97.5%. Compared to a single-channel MPPT, multi-channel independent tracking avoids mutual interference between modules with different orientations and light conditions, increasing overall power generation by over 18%. The controller features comprehensive protection against overvoltage, overcurrent, reverse connection, overheating, and short circuit, and supports RS485 communication for real-time power generation data upload.

[0042] Lithium iron phosphate battery pack and BMS management: Equipped with a 24V / 50Ah lithium iron phosphate battery pack, using 32650 power cells, 8-cell structure, rated energy 1.2kWh; cycle life ≥2000 cycles (80% DOD), operating temperature range -20~60℃. It is equipped with an integrated BMS battery management system, which monitors the voltage of each cell in the string, total voltage, charge / discharge current, and cell temperature in real time; protection thresholds: overcharge 3.65V / cell, over-discharge 2.5V / cell, overcurrent 50A, overtemperature 65℃; built-in active balancing function, balancing current ≥50mA, ensuring cell consistency and extending battery life. The battery pack is installed in the lower part of the electrical compartment, with a 10mm thick heat insulation pad to isolate the cabinet temperature rise. Under full charge, it can support continuous operation for more than 3 days in high-temperature summer conditions and more than 5 days in spring and autumn.

[0043] Power switching and distribution management: Integrated with multiple DC-DC converters, it converts 24V mains power to 12V and 5V levels to supply low-voltage loads such as main control, monitoring, lighting, and door locks. A 220V AC power backup interface is reserved, and a built-in relay automatic switching module ensures a switching time of ≤100ms, achieving seamless power supply connection. The system defaults to "PV priority" mode: PV power generation prioritizes supplying loads, with excess power charging the battery; after the battery is fully charged, excess energy is dissipated through a load-relief resistor; when power generation is insufficient, the battery provides supplementary power; when the battery SOC is below 20%, it automatically switches to AC power for supplementary charging. Manual switching to "AC priority" mode is supported, adapting to common AC power supply locations.

[0044] Module 3: Full-shell radiative cooling - superhydrophobic self-cleaning composite module This module is the core of the cabinet's passive energy saving. It adopts a reverse energy flow composite structure of "external self-cleaning, internal power generation, and internal cooling". It uses galvanized steel plate as the base material and simultaneously builds functional coatings on both the inner and outer sides. It achieves three major functions: self-cleaning to prevent dust accumulation, power generation by scattering light, and zero-energy radiation cooling, thereby reducing the cabinet's cooling load and operation and maintenance costs from the source.

[0045] 1. Superhydrophobic self-cleaning surface layer The outermost layer of the entire shell is coated with a transparent superhydrophobic fluorocarbon nano-coating with a thickness of 10~20μm and a visible light transmittance of ≥92%, which does not affect the light absorption of the underlying power generation film.

[0046] Materials and Processes: The coating is composed of fluorocarbon resin, silica micro / nano particles, and a crosslinking agent. It is formed through high-pressure air spraying, and after curing, a micro / nano binary rough structure is formed on the surface, simulating the lotus leaf effect. Key performance parameters: water contact angle ≥152°, roll-off angle ≤5°, oil-water separation coefficient >0.9. Rainwater rolls off the surface in a spherical shape, simultaneously carrying away dust and oil particles, achieving self-cleaning.

[0047] Weather resistance and lifespan: After 1000 hours of QUV-B ultraviolet aging test, the contact angle retention rate of the coating is ≥90%; it can withstand ≥10,000 manual washing cycles, and its outdoor service life is ≥5 years. Actual tests show that this coating can reduce dust accumulation on the sidewall surface by 72%, extending the manual cleaning cycle from once a month to once a quarter, and reducing annual maintenance costs by more than 60%; it also avoids efficiency degradation caused by dust accumulation obstructing the power generation film, improving power generation stability by 35% over long-term operation.

[0048] 2. Radiative Cooling Functional Layer The radiation cooling layer consists of an outer surface solar reflective layer and an inner surface infrared emitting layer, which together achieve the dual effect of "blocking external heat and radiating heat outward".

[0049] External reflective heat insulation: A composite reflective layer consisting of epoxy primer and radiative cooling intermediate coating is installed on the outer surface of the galvanized steel plate, below the flexible power generation layer, with a total thickness of 180-280μm. The epoxy primer is 60-100μm thick, ensuring corrosion resistance and adhesion; the radiative cooling intermediate coating is 120-180μm thick, composed of rutile titanium dioxide, barium sulfate micro powder, and acrylic resin, with a reflectivity of ≥93% for 0.3-2.5μm solar shortwave radiation, which can reflect the vast majority of solar heat and reduce heat absorption of the cabinet at the source.

[0050] Internal Radiative Heat Dissipation: A 120-180μm thick radiative cooling coating is sprayed onto the inner surface of the galvanized steel sheet. The coating is composed of silica microspheres, silicon carbide nanoparticles, and high weather-resistant resin. Core Performance: Normal emissivity ≥98% in the 813μm atmospheric transparency window band. Infrared radiation in this band can penetrate the atmosphere and directly reach outer space, dissipating heat conducted by the cabinet into outer space in the form of infrared radiation without a medium or energy consumption, achieving passive cooling.

[0051] Actual cooling effect: Under a summer midday solar irradiance of 1000W / ㎡, this composite coating can reduce the temperature of the inner wall of the cabinet by 35℃ compared to the ambient temperature, and the temperature of the inner surface of the outer shell of the cabinet by 810℃; correspondingly, the compressor operating load is reduced by 32%, resulting in an average daily energy saving of over 0.2kWh. At night, after the ambient temperature decreases, the radiative cooling effect is further enhanced, enabling compressor-free cooling.

[0052] 3. Collaborative Working Mechanism of Composite Structures The five-layer structure of the sidewalls and back panel—"superhydrophobic surface layer - flexible power generation layer - reflective primer layer - metal substrate - inner radiative cooling layer"—creates a unique "bidirectional outward energy flow," completely opposite to the "heat intrusion" characteristic of traditional cabinets. The three layers work together seamlessly. Cleaning Synergy: The outer superhydrophobic coating maintains surface cleanliness through rainwater self-cleaning, which not only avoids dust accumulation and heat absorption that aggravates temperature rise, but also ensures the light-receiving area of ​​the middle flexible power generation layer and maintains stable power generation efficiency.

[0053] Energy Synergy: The middle power generation layer absorbs visible light and converts it into electrical energy for output; the unabsorbed infrared heat and the heat conducted out of the cabinet are evenly diffused through the high thermal conductivity metal substrate and then emitted into outer space by the inner radiative cooling layer in the form of infrared radiation, forming a reverse energy flow of "electrical energy outward and heat outward", which changes the drawback of traditional cabinets where heat accumulates inward.

[0054] Synergistic thermal insulation: The outer reflective layer blocks the intrusion of short-wave solar radiation, the inner radiative layer continuously dissipates heat outward, and the intermediate metal substrate evens out local hot spots. Combined with the internal insulation layer, multiple thermal barriers are formed, significantly reducing the cooling load inside the cabinet. Non-planar structures such as cabinet columns, door frames, and bottom are simultaneously coated with epoxy primer, radiative cooling coating, and superhydrophobic topcoat to achieve full shell functional coverage and eliminate local thermal bridges.

[0055] Module 4: Bottom-mounted independent cooling and heat dissipation module with local electronic control interaction This module is the control core for local operation of the cabinet. It adopts a bottom-mounted heat dissipation design with "cooling and heating separation" and is equipped with an intelligent electronic control and gravity sensing settlement system to ensure efficient cooling, precise control, and convenient interaction. At the same time, it can achieve local autonomous operation and cloud linkage.

[0056] 1. Cooling system and independent heat dissipation duct All core components of the refrigeration system are located in an independent heat dissipation compartment at the bottom of the cabinet, which is completely separated from the upper storage compartment by a 20mm thick insulation partition to prevent the heat from back-heating the goods inside the cabinet and achieve cold and heat separation.

[0057] Refrigeration System Configuration: Utilizes a fully enclosed rotary inverter compressor with a rated power of 220W, a COP ≥ 1.8, and is charged with R134a environmentally friendly refrigerant at a charge of 180g. Equipped with high and low pressure protection switches, it automatically shuts down when high pressure > 1.6MPa or low pressure < 0.1MPa. The condenser is a wire-tube condenser with a heat exchange area ≥ 2.5㎡ and a fin spacing of 2mm, balancing heat exchange efficiency and dust prevention. The evaporator is a blown aluminum plate evaporator, integrally attached to the back of the inner liner, with a contact area ≥ 0.8㎡, ensuring uniform heat exchange without localized temperature differences. The system employs a φ1.8×1.5m capillary tube for precise throttling, matched to the compressor displacement, achieving a temperature control accuracy of ±1℃. The cabinet temperature can be freely set within the range of 2~12℃, suitable for various products such as beverages, snacks, and light meals.

[0058] The cooling duct is designed for a unidirectional flow: the cooling chamber is 200mm high and features a left-right partitioned layout, with the compressor compartment on the left and the condenser compartment on the right, separated by an arc-shaped baffle in the middle, creating a "front-to-bottom air intake, rear-to-top air exhaust" unidirectional flow. The front left air intake is equipped with a removable G3-grade dust filter with an effective air intake area ≥0.02㎡ and a filtration accuracy ≥10μm. A 24V DC axial fan with a rated airflow ≥120m³ / h and a pressure ≥80Pa is installed on the rear right side. Cool air enters through the air intake, first flowing over the compressor surface to remove heat, then being guided by the baffle to the condenser fins for heat exchange, and finally discharged horizontally by the fan. The duct has no vortex dead zones, preventing hot air backflow. The cooling fan uses PWM stepless speed regulation, automatically adjusting its speed according to the condensing temperature: it stops when the condensing temperature is <35℃, operates at low speed between 35~45℃, and operates at full speed above 45℃, balancing cooling efficiency and energy consumption.

[0059] 2. Condensate self-evaporation treatment mechanism The condensate produced by the evaporator is collected in a water collection tank at the bottom of the inner tank and then introduced into a drip tray inside the heat dissipation chamber via an 8mm food-grade silicone guide tube. The drip tray, made of galvanized steel, has an effective volume of 1.5L and is located directly in front of the condenser's air outlet, utilizing the 35-50℃ hot air discharged from the condenser to accelerate evaporation. The drip tray is lined with porous absorbent cotton to increase the evaporation area; the measured evaporation efficiency is ≥200ml / day, which can completely absorb the condensate during peak summer periods, eliminating the need for regular manual emptying. The drip tray is equipped with an overflow outlet and overflow pipe, automatically overflowing to the outside of the cabinet in case of excessive water during extreme rainstorms, preventing flooding and damage to electrical components.

[0060] 3. Local main control and intelligent temperature control It adopts an industrial-grade main control unit based on the ARM Cortex-A7 architecture, with a main frequency of 1GHz, 1GB of built-in memory + 8GB of eMMC storage, runs an embedded Linux system, and has an operating temperature of -30~70℃, making it suitable for extreme outdoor environments.

[0061] Data Acquisition and Control: It can connect to 12 analog input channels and 16 digital I / O interfaces to collect real-time data such as cabinet temperature and humidity, ambient temperature, cabinet surface temperature, photovoltaic power, battery SOC, compressor operating status, door status, gravity sensor, etc.; and directly control actuators such as compressor, cooling fan, lighting, and electromagnetic lock.

[0062] Adaptive temperature control algorithm: The built-in fuzzy PID temperature control algorithm dynamically adjusts the compressor start / stop threshold and operating frequency based on ambient temperature, cabinet surface temperature, and radiant cooling effect. When the ambient temperature is <20℃, the start / stop temperature difference is increased to 3℃ to reduce the number of start / stop cycles; when the ambient temperature is >35℃, the start / stop temperature difference is reduced to 1℃ to ensure stable temperature inside the cabinet. It can also receive optimization strategies from the cloud and execute commands such as cold storage and off-peak operation.

[0063] Fault self-diagnosis: It can identify 16 types of faults, such as compressor faults, sensor abnormalities, power generation abnormalities, door lock abnormalities, and battery faults, generate corresponding fault codes, display them on the local screen, and upload them to the cloud in real time; common minor faults can be automatically reset and retried, and serious faults will trigger alarms to notify maintenance personnel.

[0064] 4. Gravity-based calculation and human-computer security interaction Gravity-sensing checkout system: Each shelf is equipped with a high-precision cantilever beam gravity sensor, 24-bit AD sampling, recognition accuracy ±5g, and a single shelf weight of 50kg. It employs a dynamic weight detection algorithm, continuously sampling at a frequency of 20Hz during door opening and calculating the weight change after door closing using a filtering algorithm. This data is then matched against a product weight database to accurately identify the quantity and type of goods taken out and placed. It supports simultaneous taking and placing of multiple items with an accuracy rate ≥99.5%. In case of checkout anomalies, a verification mechanism is triggered, automatically retrieving camera footage for auxiliary verification, reducing theft and damage rates.

[0065] Human-computer interaction: A 7-inch capacitive touch screen is embedded in the middle of the door, which supports QR code payment, product information display, temperature display, and fault prompts; a built-in voice broadcast module provides voice guidance throughout the operation.

[0066] Security Monitoring: A 360° panoramic high-definition camera is installed on the sloping side of the top of the cabinet. It features 1080P resolution, a 2.8mm focal length, a 360° viewing angle, and supports infrared night vision with a night vision distance of 10m. Built-in human recognition and motion detection algorithms automatically capture and record video when events such as personnel lingering for more than 30 seconds, severe cabinet vibration, or abnormal door opening are detected. The video is uploaded to the cloud, simultaneously triggering on-site audio and visual alarms, and alarm information is pushed to maintenance personnel. Local storage is provided on a 32GB SD card with loop storage, ensuring video recordings are retained for at least 7 days. Alarm recordings are backed up in the cloud and can be viewed remotely in real time.

[0067] Module 5: Cloud Data Analysis and Interconnected Intelligent Operations This module is the "brain" and operational hub of the system. Based on a cloud-edge collaborative architecture, it enables interconnection, data aggregation, intelligent analysis, and refined operation of vending machines across the entire network, upgrading the independent operation of a single vending machine into a smart retail system driven by data across the entire network.

[0068] 1. Cloud-based data platform architecture The cloud platform employs a microservice distributed architecture, supporting concurrent access from millions of devices. Terminals connect to the cloud via a 4GCat.1 wireless communication module, using the MQTT protocol for data transmission with QoS=1 and TLS1.2 encryption to ensure reliable and secure data transmission. Regular operational data is uploaded every 5 minutes, and anomaly alarm data is reported in real time. Data is stored in a tiered manner according to category. Time series database: Stores high-frequency time series data such as power generation, temperature, electricity consumption, and operating time, and supports second-level query and aggregation calculation; Relational databases store structured data such as device files, product information, order data, user information, and maintenance records. Object storage: Stores unstructured files such as surveillance videos, captured images, and firmware upgrade packages. Data dimensions cover four main categories: Energy-related metrics: Real-time power generation, cumulative power generation, battery SOC, charge / discharge cycles, mains power consumption, compressor operating time, energy consumption per unit of product, etc. Environmental factors: ambient temperature and humidity, cabinet interior temperature and humidity, solar irradiance, cabinet surface temperature, rainfall, wind speed and direction, etc. Operational data: Sales details, inventory quantity, number of times the store is opened, user profiles, payment data, product turnover rate, regional foot traffic heat map, etc. Operations and maintenance (O&M) data includes: fault alarms, monitoring videos, power generation efficiency degradation, coating cleanliness estimation, equipment lifespan, and maintenance records. It supports multi-dimensional queries and statistics by equipment, region, time, and product category, and automatically generates daily / weekly / monthly / yearly reports.

[0069] 2. Core Intelligent Analysis Algorithm Three core algorithm models are built into the cloud to achieve data-driven intelligent optimization: Intelligent energy consumption optimization algorithm: An energy consumption prediction model is built based on an LSTM time series neural network. Input features include ambient temperature, solar irradiance, wind speed, date, cabinet set temperature, and historical energy consumption, with a prediction accuracy of ≥90%. The system dynamically issues temperature control strategies based on the prediction results: When a high temperature is predicted for the next day, the cabinet temperature is lowered to the set lower limit in advance during the surplus period of photovoltaic power generation, utilizing the cabinet's cold storage to reduce the peak load of the next day; combined with real-time power generation, the compressor start-stop threshold is dynamically adjusted to maximize the photovoltaic self-consumption ratio, which can reach over 95%.

[0070] Power generation performance and operation and maintenance prediction algorithms: Continuously comparing the power generation efficiency of similar equipment in the same region, and through meteorological data normalization processing to eliminate weather influences, accurately identifying equipment with abnormally declining efficiency; combining local dust and rainfall data, the algorithm estimates surface dust thickness and automatically generates on-demand cleaning work orders, replacing fixed-cycle cleaning and reducing operation and maintenance costs by more than 30%. Simultaneously, by long-term monitoring of battery charge and discharge curves and compressor operating parameters, a trend extrapolation algorithm predicts equipment aging and potential failures, proactively dispatching maintenance work orders to achieve preventative maintenance, increasing the average mean time between failures (MTBF) of equipment by 40%.

[0071] Intelligent sales forecasting and inventory management algorithm: Integrating 12 dimensions of features including historical sales, weather, holidays, weekdays, business district activities, and pedestrian traffic heatmaps, the XGBoost sales forecasting model is trained, achieving a daily sales forecast accuracy of ≥85% for a single SKU. The system automatically generates a replenishment list for each device, combining the location of maintenance personnel, vehicle load, and road congestion to plan the optimal replenishment route, reducing transportation costs; simultaneously, it tracks product expiration dates, providing early warnings for near-expiration products 30 days in advance, guiding promotional activities, and reducing product spoilage rates.

[0072] 3. End-to-end operation management backend Provides operators with a web-based management backend and a WeChat mini-program mobile application, covering the entire operational process: Equipment Overview and Monitoring: An electronic map displays the distribution of all equipment, with different colors indicating operating status (normal / alarm / offline). Click to view real-time parameters of a single unit; supports remote temperature adjustment, remote unlocking, remote restart, and remote firmware upgrade.

[0073] Data Reporting Center: Automatically generates four categories of reports: energy consumption, sales, power generation revenue, and operation and maintenance. Supports multi-dimensional filtering and Excel export to assist in operational decision-making.

[0074] Work order full-process management: Automatically generates cleaning, replenishment, and maintenance work orders, supports automatic and manual dispatching, and tracks work order progress, completion acceptance, and personnel performance statistics.

[0075] Product and Marketing Management: Product listing and delisting, price adjustments, membership system, and coupon activity configuration to support refined operations.

[0076] Multi-level access control: Supports multi-level account permissions such as super administrator, regional operation, and maintenance personnel to ensure data security.

[0077] 4. Multi-scenario cloud-edge collaborative operation mechanism The system supports four typical operating modes, with cloud and local collaborative control, adapting to different environments and working conditions: Sunny Daytime Mode: Photovoltaic power generation is at full power output, prioritizing the supply to the load, and surplus electricity is stored; cloud-based real-time cold storage strategies are issued to maximize energy utilization; radiative cooling continuously dissipates heat, and the compressor operates intermittently.

[0078] Cloudy low light mode: Cadmium telluride low light power generation supplements the load and makes up for the battery power gap; the cloud combines weather forecasts to dynamically adjust the energy consumption strategy, and appropriately relax the temperature control threshold when the battery is low to extend the battery life.

[0079] Nighttime energy-saving mode: Photovoltaic power generation stops, and battery power is supplied; the radiant cooling effect is enhanced at night, the compressor runs at a low frequency, and energy consumption is reduced by 40% compared to the daytime; the cloud summarizes the day's data and generates the next day's operation plan.

[0080] Mains-powered complementary mode: When the battery is low due to continuous rain, it automatically switches to mains power to replenish the battery; when the weather improves, it automatically switches back to photovoltaic priority, with no human intervention throughout the process.

[0081] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0082] It should be noted that, in this document, the use of relational terms such as "first" and "second" is merely for distinguishing one entity or operation from another, and does not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0083] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0084] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0085] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An IoT-based self-generating vending machine AI vending system based on cloud data analysis, characterized in that: include: The integrated cabinet body and basic enclosure module serve as the physical load-bearing foundation of the system. It adopts a frame load-bearing and layered enclosure structure, and is equipped with adjustable merchandise shelves inside. The whole system has outdoor waterproof, anti-corrosion and thermal insulation properties. The multi-faceted cadmium telluride self-generating and energy storage power supply module is set on the top surface, side walls and back panel of the cabinet. The cadmium telluride double glass power generation glass on the top surface is the main power generation unit, and the flexible cadmium telluride thin film power generation layer on the side wall is the supplementary power generation unit. It is equipped with a high-voltage adaptable MPPT controller, lithium iron phosphate energy storage battery and mains power backup interface to form a multi-source power supply architecture. The full-shell radiative cooling-superhydrophobic self-cleaning composite module is covered on the entire shell surface of the cabinet. It uses galvanized steel plate as the base material to build a reverse energy flow composite structure with external self-cleaning, central power generation and internal cooling. It simultaneously realizes rainwater self-cleaning, scattered light supplementary power generation and zero-energy passive cooling, reducing the cabinet's cooling load and operation and maintenance costs from the source. The bottom-mounted independent cooling and heat dissipation and local electronic control interaction module is located in an independent compartment at the bottom of the cabinet. It integrates the cooling and heat dissipation system, local main control unit, gravity sensor settlement unit and human-machine security interaction unit to realize local autonomous temperature control of the cabinet and automatic settlement of goods. The cloud-based data analysis and interconnected intelligent operation module connects all vending machine terminals via wireless communication, gathers all operational data, and incorporates multi-dimensional intelligent analysis algorithms. It is paired with a full-link operation management backend to achieve interconnected intelligent control and refined operation of vending machines across the entire network.

2. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 1, characterized in that, The main frame of the integrated cabinet body and the basic enclosure module is welded from galvanized rectangular steel pipes. The enclosure structure adopts a three-layer structure: an outer functional layer, a polyurethane foam insulation layer, and a food-grade stainless steel inner liner. The front door is made of thermally broken aluminum alloy frame with double-layer tempered Low-E hollow glass. The hollow layer is filled with argon gas. The door is surrounded by double-ring magnetic sealing strips and has a built-in electromagnetic anti-theft lock and human body sensing components.

3. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 1, characterized in that, The cabinet top of the multi-faceted cadmium telluride self-generating and energy storage power supply module is set with a 6° single slope that is lower in the front and higher in the back. The slope is installed with cadmium telluride double-glass power generation glass with a size of 700×850mm after cutting. It is sealed and fixed with aluminum profile frame, EPDM gasket and aluminum alloy pressure plate. A continuous water guide channel and drainage hole are opened at the front end of the slope. The power generation line is connected to the top electrical system.

4. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 3, characterized in that, The sidewalls and backplates of the multi-faceted cadmium telluride self-generating and energy storage power supply module are bonded with a flexible cadmium telluride thin-film power generation layer by a high thermal conductivity and electrical conductivity adhesive. The flexible thin-film power generation layer is composed of multiple CdTe thin-film modules connected in series and parallel, which are used to capture scattered light and side-sunlight to supplement power generation. The power generation lines are connected to the top electrical compartment for unified management.

5. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 1, characterized in that, The multi-faceted cadmium telluride self-generating and energy storage power supply module is equipped with 4 independent MPPT controllers, which independently perform maximum power point tracking for the power generation units on the top surface and three side walls; it is equipped with a 24V / 50Ah lithium iron phosphate battery pack and BMS battery management system, and has a built-in automatic power switching module to support seamless switching between photovoltaic priority and grid priority power supply modes.

6. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 1, characterized in that, The full-shell radiative cooling-superhydrophobic self-cleaning composite module consists of, from the outside to the inside, a superhydrophobic self-cleaning surface layer, a flexible power generation layer, an epoxy primer layer, a metal substrate layer, and an inner radiative cooling coating. The superhydrophobic surface layer has a water contact angle ≥152°, and the inner radiative cooling coating has an emissivity ≥98% and a solar reflectance ≥93% in the 8~13μm atmospheric transparency window band.

7. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 1, characterized in that, The bottom-mounted independent cooling and heat dissipation module with local electronic control interaction module adopts a left-right partition layout, with a unidirectional airflow duct that allows air to enter from the front and exit from the rear. The air inlet is equipped with a dust filter, and the air outlet is equipped with a stepless speed-regulating cooling fan. The heat dissipation chamber is equipped with a water collection tray and porous water-absorbing cotton, and the condensate is self-evaporated by using the hot air discharged by the condenser.

8. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 1, characterized in that, The bottom-mounted independent cooling and heat dissipation and local electronic control interaction module adopts an ARM architecture industrial-grade main control unit with built-in adaptive temperature control algorithm and fault self-diagnosis function; each shelf is equipped with a high-precision gravity sensor, which identifies the quantity and type of goods to be picked up and put down through a dynamic weight detection algorithm, enabling door opening for picking up goods and automatic checkout.

9. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 1, characterized in that, The cloud-based data analysis and interconnected intelligent operation module adopts a microservice architecture and time-series database, and uniformly accesses all energy, environmental, operational, and maintenance data from all terminals. It supports multi-dimensional query statistics and automatic report generation, and terminal data is transmitted in encrypted form via the MQTT protocol.

10. The IoT self-generating vending machine AI vending system based on cloud data analysis according to claim 9, characterized in that, The cloud-based data analysis and interconnected intelligent operation module incorporates three core algorithms: intelligent energy consumption optimization, power generation performance operation and maintenance prediction, and intelligent sales prediction. It is equipped with a web-based and mobile operation management backend, supporting remote equipment monitoring, full-process work order management, product marketing management, and multi-level permission configuration.

Images