Kelp drying production process full-automatic management system based on artificial intelligence

Abstract

The invention relates to the technical field of kelp production intelligent management, in particular to a kelp drying production process full-automatic management system based on artificial intelligence. The system comprises the following modules: a kelp characteristic sensing module, a drying parameter planning module, an intelligent intervention adjustment module, a production parameter intelligent optimization module and a production visual management module, a double-line hanging multi-temperature-zone double-layer tunnel type drying structure is adopted, and a drying tunnel is divided into a plurality of temperature zones; different temperatures can be set for each temperature zone according to the kelp drying requirements, the temperature, humidity and kelp drying state of each temperature zone are monitored in real time, dynamic zoning and staged regulation and control are performed according to preset drying parameters, kelp automatic production operation parameters are intelligently learned, automatic production parameter feedback is performed, and the kelp drying efficiency is improved. The automatic control of kelp drying production and the management of the kelp drying production process are realized. According to the invention, full-automatic management of the kelp drying production process is realized through dynamic zoning and staged regulation and control.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent management of kelp production, and particularly to a fully automated management system for the kelp drying production process based on artificial intelligence. Background Art

[0002] Since the water content of fresh kelp is as high as about 90%, it is perishable and prone to spoilage. Traditionally, it is dried and processed into dried kelp for long-term storage and long-distance transportation. However, the traditional kelp drying and processing methods are difficult to meet the requirements of modern production and the market, mainly reflected in the following aspects: 1. The traditional kelp drying methods mainly rely on sun drying or simple drying equipment, which is greatly limited by weather conditions, with a long and unstable processing cycle. In rainy weather or seasons with high humidity, the kelp is prone to mildew and spoilage, causing serious economic losses. At the same time, open-air drying is easily contaminated by external factors such as dust and pests, making it difficult to ensure the hygienic quality and safety of the products. 2. The energy utilization rate of traditional drying equipment is low, resulting in a large amount of energy waste. Most drying equipment uses a single heating method and cannot accurately control the temperature according to the requirements at different stages of the kelp drying process, leading to uneven kelp quality, with some areas being over-dried while other areas have excessive moisture residue, which not only affects the product quality but also increases energy consumption. The heat energy is often not effectively recovered and utilized during the drying process, further reducing the energy efficiency. 3. The existing processing equipment has a low degree of automation and low production efficiency, and highly relies on manual operation. From the cleaning, hanging to drying, sorting, and packaging of kelp, there is a large amount of manual participation in each link, which not only increases the labor cost but also easily causes product quality fluctuations due to human factors. Therefore, there is an urgent need for a fully automated management system for the kelp drying production process based on artificial intelligence to achieve high efficiency, energy conservation, environmental protection, and stable improvement of product quality through intelligent control and precise energy supply. Summary of the Invention

[0003] Based on this, the present invention provides a fully automated management system for the kelp drying production process based on artificial intelligence to solve at least one of the above technical problems.

[0004] To achieve the above object, a fully automated management system for the kelp drying production process based on artificial intelligence, the fully automated management system for the kelp drying production process based on artificial intelligence adopts a double-line hanging multi-temperature zone double-layer tunnel drying structure, including the following modules:

[0005] The kelp characteristic perception module is used to analyze the kelp monitoring characteristics of single-piece / single-cluster fresh kelp units to obtain target kelp monitoring characteristic data; and generate the data of the kelp queue to be dried according to the target kelp monitoring characteristic data for hanging queue layout.

[0006] The drying parameter planning module is used to perform automated cleaning - air blowing pre - treatment production operations on single - piece / single - cluster fresh kelp units through the data of the kelp queue to be dried, and set the production operation parameters of multiple temperature zones according to the target kelp monitoring characteristic data to obtain the automated production operation parameters of kelp;

[0007] The intelligent intervention adjustment module is used to optimize the multi - temperature - zone kelp drying production based on the automated production operation parameters of kelp, and classify the grades of the dried kelp products to generate the classified data of the dried kelp products; among them, the optimization of the multi - temperature - zone kelp drying production includes the optimization of pre - heating and sweating production, the optimization of balanced dehumidification production, the optimization of temperature - rising drying production, and the optimization of temperature - decreasing shaping production;

[0008] The production parameter intelligent optimization module performs transfer learning based on the preset convolutional neural network model through the classified data of the dried kelp products and the automated production operation parameters of kelp, and gives automated production parameter feedback to obtain the intelligent production operation parameters;

[0009] The production visualization management module is used to monitor the kelp drying process in real time to achieve the comprehensive and transparent management of the kelp drying production process.

[0010] The fully automated management system for the kelp drying production process based on artificial intelligence has achieved a revolutionary improvement in kelp drying and processing through the coordinated action of multiple modules. The system's precise perception and analysis of kelp characteristics enable the equipment to perform personalized processing according to the actual characteristics of different batches of kelp, completely solving the problem that traditional drying methods cannot handle the differences in kelp materials, and significantly improving the uniformity and stability of product quality. The combination of the multi-temperature zone intelligent design and precise temperature control technology makes the drying process no longer dependent on weather conditions, shortening the traditional processing cycle of several days to within a few hours, and greatly improving production efficiency and production capacity. The multi-temperature zone precise control achieved by the system provides the most suitable temperature and humidity environment for different drying stages according to the moisture migration law during the kelp drying process, greatly improving the rehydration and taste of the product, shortening the drying cycle, and reducing energy consumption. In particular, the intelligent intervention and adjustment in the four key stages of preheating and sweating, balance dehumidification, heating and drying, and cooling and shaping ensure the balanced evaporation of moisture inside and outside the kelp, prevent the surface from forming a film, and form tiny channels conducive to moisture discharge, making the product texture more uniform. The implementation of the waste heat recycling strategy significantly reduces the system's energy consumption. Through intelligent heat energy allocation, precise compensation of waste heat from high-temperature areas to low-temperature areas is achieved, and the energy utilization efficiency is increased by about 30%, achieving the goal of energy conservation and environmental protection. The double-line hanging multi-temperature zone double-layer tunnel structure design improves the production capacity by more than 50% under the same floor area, realizing the efficient utilization of space resources. The production parameter optimization module based on the convolutional neural network has the ability of self-learning, and can continuously adjust and improve the process parameters according to historical production data. With the accumulation of production experience, the system performance continues to improve and the adaptability continues to increase. The system implements intelligent grading of the dried products, ensuring the standardization of the product quality supplied to the market, and improving the market competitiveness and added value of the products. The real-time visualization management platform makes the production process transparent. Managers can remotely monitor the status of each temperature zone, production progress, and energy consumption, discover and handle abnormal situations in a timely manner, and reduce operation risks. The entire system minimizes manual operations, and the full-process automation from kelp cleaning to packaging reduces the labor cost by 30%-50%, and at the same time eliminates the quality fluctuations caused by human factors. This system is not only applicable to large-scale kelp processing enterprises, but its modular design and mobility also enable it to meet the production needs of small and medium-sized scales, improving the applicability and promotion value of the equipment, and promoting the intelligent transformation of the kelp processing industry. Therefore, a fully automated management system for the kelp drying production process based on artificial intelligence of the present invention realizes precise identification and analysis of kelp characteristics, combines a convolutional neural network model to achieve parameter self-optimization, adopts a double-line hanging multi-temperature zone double-layer tunnel structure design, performs precise control in stages of preheating and sweating, balance dehumidification, heating and drying, and cooling and shaping of the kelp drying process, realizes waste heat recycling exchange and heat compensation between temperature zones, integrates real-time monitoring and a visualization management platform, and realizes full-process intelligent control from raw material processing to finished product grading. Brief Description of the Drawings

[0011] Figure 1 FIG. is a schematic diagram of the module process of the fully automated management system for the kelp drying production process based on artificial intelligence of the present invention;

[0012] Figure 2 is Figure 1 a detailed implementation process schematic diagram of the kelp characteristic perception module in the sea;

[0013] The realization, functional features and advantages of the object of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. Detailed Embodiments

[0014] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0015] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0016] It should be understood that although the terms "first", "second", etc. may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed related items.

[0017] To achieve the above object, please refer to Figures 1 to 2 , the present invention provides a fully automated management system for the kelp drying production process based on artificial intelligence. The fully automated management system for the kelp drying production process based on artificial intelligence adopts a double-line hanging multi-temperature zone double-layer tunnel drying structure and includes the following modules:

[0018] The kelp characteristic perception module is used to analyze the kelp monitoring characteristics of a single piece / cluster of fresh kelp units to obtain target kelp monitoring characteristic data; and perform a hanging queue layout according to the target kelp monitoring characteristic data to generate data of the kelp queue to be dried.

[0019] The drying parameter planning module is used to perform automated cleaning-blowing pre-treatment production operations on a single piece / cluster of fresh kelp units through the data of the kelp queue to be dried, and set multi-temperature zone production operation parameters according to the target kelp monitoring characteristic data to obtain kelp automated production operation parameters.

[0020] The intelligent intervention adjustment module is used to optimize the multi-temperature zone kelp drying production based on the kelp automated production operation parameters, and classify the grades of the dried kelp products to generate data on the classification of the dried kelp products; among them, the optimization of the multi-temperature zone kelp drying production includes preheating sweating production optimization, balanced dehumidification production optimization, heating and drying production optimization, and cooling and shaping production optimization.

[0021] The production parameter intelligent optimization module performs transfer learning based on the preset convolutional neural network model through the data on the classification of the dried kelp products and the kelp automated production operation parameters, and performs automated production parameter feedback to obtain intelligent production operation parameters.

[0022] The production visualization management module is used to monitor the kelp drying process in real time to achieve comprehensive and transparent management of the kelp drying production process.

[0023] In the embodiment of the present invention, the fully automated management system for the kelp drying production process based on artificial intelligence includes the following modules:

[0024] S1: The kelp characteristic perception module is used to analyze the kelp monitoring characteristics of a single piece / cluster of fresh kelp units to obtain target kelp monitoring characteristic data; and perform a hanging queue layout according to the target kelp monitoring characteristic data to generate data of the kelp queue to be dried.

[0025] In the embodiment of the present invention, before the kelp enters the drying production line, first, the image recognition unit takes high-resolution photos of each piece or cluster of fresh kelp. The image enhances the surface texture and color difference information through a multi-channel light source, and then is sent to the deep convolutional recognition model for feature extraction. The monitoring features extracted by the system include the length, width, thickness, color distribution, surface adhesion distribution, water-containing area distribution, etc. of the kelp. The thickness is scanned non-contact by a laser displacement sensor, and the surface humidity is analyzed regionally by a near-infrared reflection spectroscopy device. After all the feature data are fused and processed by the main control system, the target kelp monitoring feature data are generated. According to this feature data, the hanging layout scheduling system performs task planning, automatically controls the sorting unit on the hanging track, and automatically distributes the kelp to different hanging points according to the length and thickness. The spacing control range is 100 to 300 millimeters, generating structured data of the kelp queue to be dried.

[0026] S2: A drying parameter planning module, which is used to perform an automated cleaning-blowing pre-treatment production operation on a single piece / single cluster of fresh kelp unit through the data of the kelp queue to be dried, and set the production operation parameters of multiple temperature zones according to the target kelp monitoring feature data to obtain the automated production operation parameters of the kelp;

[0027] In the embodiment of the present invention, the drying parameter planning module first executes the automated cleaning-blowing pre-treatment process of the kelp. The cleaning system adopts a three-stage design: the first stage is high-pressure spraying (pressure 0.8 MPa, fresh water) to remove surface impurities; the second stage is ozone water spraying (concentration 0.5 ppm) for disinfection; the third stage is clean water flushing (pressure 0.5 MPa) to remove residues. The conveyor belt speed is controlled at 0.3 m / s to ensure that the cleaning time of each piece of kelp is not less than 45 seconds. The drying system consists of 12 centrifugal fans, with a wind pressure of 2000 Pa, a wind temperature of 35 °C, and a wind speed of 15 m / s. They are evenly arranged on both sides of the conveyor belt to form a 360-degree dead-angle-free blowing. Subsequently, the multi-temperature zone production operation parameter setting unit receives the target kelp monitoring feature data and starts the four-stage parameter planning. The parameter setting in the preheating stage is based on the average thickness and initial moisture content of the kelp, and determines the preheating temperature of 57 °C, humidity of 65%, the rotation speed of the circulating fan of 1300 r / min, and the preheating duration of 180 seconds. The parameters in the dehumidification stage are based on the surface moisture content distribution of the kelp, and set the dehumidification temperature of 50 °C, the initial air valve opening of the exhaust fan of 40%, the fresh air supply of 25 m³ / min, and the dehumidification duration of 300 seconds. The parameters in the drying stage are based on the wet weight and moisture content gradient of the kelp, and determine the target temperature of heating and drying of 85 °C, the main air duct wind speed of the strong exhaust fan of 9.5 m / s, and the drying duration of 480 seconds. The parameters in the shaping stage are based on the expected state after drying, and set the cooling temperature of 25 °C, the air supply wind speed of the cooling fan of 4.5 m / s, and the conveyor belt moving speed of 3.2 m / min. The system integrates the parameters of the four stages, generates a complete dataset of the automated production operation parameters of the kelp, and distributes them to each temperature zone control unit through the fieldbus network.

[0028] S3: The intelligent intervention and adjustment module is used to optimize the multi-temperature zone kelp drying production based on the kelp automated production operation parameters, classify the grades of the dried kelp products, and generate the classification data of the dried kelp products; perform heat compensation treatment for the waste heat exchange in the temperature zone according to the kelp automated production operation parameters, and construct a waste heat recycling strategy; among them, the optimization process of the multi-temperature zone kelp drying production includes preheating sweating production optimization, balanced dehumidification production optimization, heating and drying production optimization, and cooling and shaping production optimization.

[0029] In the embodiment of the present invention, the intelligent intervention and adjustment module realizes the whole-process dynamic optimization of the drying process, which is divided into four-stage coordinated control. The preheating sweating production optimization system constructs the first temperature zone environment field through 16 PT100 temperature sensors and 12 humidity sensors to monitor the surface temperature distribution and local humidity change of the kelp. When it is found that the surface temperature of the kelp is uneven and the temperature gradient exceeds 3 °C / cm, the system precisely adjusts the angles of 36 micro air vents and the power of 24 infrared heating units to achieve precise sweating control. The balanced dehumidification production optimization system deploys 8 high-precision weighing sensors and 16 near-infrared moisture content sensors in the second temperature zone to continuously monitor the water loss rate, and combines the microenvironment model constructed by 24 humidity sensors and 16 wind speed sensors to detect the abnormal dehumidification condition in real time. When detecting the abnormal condition of wet resistance accumulation type (local humidity is greater than 70%), the system immediately increases the air valve opening by 25% and reduces the temperature by 3 °C. The heating and drying production optimization system calculates the drying uniformity index based on the three-dimensional temperature field and wind speed field constructed by 56 temperature sensors and 32 wind speed sensors in the third temperature zone. When the index is lower than the preset threshold of 75, the system adjusts the hanging height of the kelp through 16 groups of chain lifting devices, moves the drying lag area up by 100 - 150 mm, and optimizes the heat energy utilization. The cooling and shaping production optimization system precisely executes the cooling curve control in the fourth temperature zone and creates a stable cooling environment through the water-cooled heat dissipation device and 16 centrifugal fans. The system deploys a high-precision sensor array at the discharge end to measure the final moisture content, appearance, shape, and texture characteristics, and classifies the kelp into four grades: special grade, first grade, second grade, and third grade according to the weighted scoring standard, generating the complete classification data of the dried kelp products.

[0030] S4: The production parameter intelligent optimization module is used to perform transfer learning based on the preset convolutional neural network model through the classification data of the dried kelp products and the kelp automated production operation parameters, and perform automated production parameter feedback to obtain the intelligent production operation parameters.

[0031] In the embodiment of the present invention, the core of the system is a preset improved ResNet-50 convolutional neural network model. The initial model is pre-trained with 5000 batches of historical production data, including the complete production parameters of kelp at each grade and the finished product quality data. The transfer learning unit receives the current grading data of the dried kelp finished products and the corresponding automated production operation parameters of the kelp, and constructs an input tensor containing 245 feature dimensions. The feature dimensions include production parameters such as the temperature curves of four temperature zones, humidity changes, wind speed adjustments, conveyor speeds, etc., and quality indicators such as the moisture content, hardness, rehydration ability, and color of the finished products. The learning process adopts the mini-batch learning method with a batch size of 32 and a learning rate of 0.001. The model is updated once every 50 batches of new data are added. The optimization goal is set to maximize the sum of the top-grade product rate and the first-grade product rate, while minimizing the energy consumption. Based on the updated model, the parameter optimization engine generates differentiated production parameter suggestions for kelp raw materials with different characteristics, and focuses on optimizing sensitive parameters such as key control points like the preheating temperature, dehumidifying wind speed, and drying duration. The system compares the production data of 30 consecutive batches. When it is found that the top-grade product rate increases by more than 2.5 percentage points or the energy consumption decreases by more than 5%, the optimized parameters are locked as the new standard process parameters. After the optimization results are verified by the quality inspection unit, an intelligent production operation parameter package containing the temperature adjustment values of the temperature zones, the wind speed correction coefficient, and the duration optimization ratio is formed and sent to each execution unit through the industrial Ethernet to achieve closed-loop continuous optimization.

[0032] S5: The production visualization management module is used to monitor in real time the operating status of each temperature zone, the processing progress of the kelp drying unit, and the system energy consumption in the double-line hanging multi-temperature zone double-layer tunnel drying structure, so as to achieve comprehensive and transparent management of the kelp drying production process.

[0033] In the embodiments of the present invention, the production visualization management module realizes comprehensive monitoring through a distributed data acquisition system. The system deploys 198 sensor nodes (96 for temperature, 48 for humidity, 32 for wind speed, and 22 for weight), with the sampling period ranging from 1 second to 60 seconds. The data is transmitted to the central server through the industrial Internet of Things protocol. The monitoring interface uses a 43-inch industrial touch screen with a resolution of 3840×2160 and a refresh rate of 60 Hz. The main interface displays a 3D model of a double-line hanging multi-temperature zone double-layer tunnel drying structure, and the temperature distribution is represented by color coding (blue-yellow-red corresponding to 20-65°C). The system updates 46 key process parameters of each temperature zone in real time, including temperature curves, humidity curves, energy consumption data, and anomaly alarms. The processing progress of the kelp drying unit is intuitively displayed through a Gantt chart, including information such as ID, moisture content change trend, estimated completion time, and quality prediction. The system's energy consumption monitoring uses multi-level statistics, from individual heating elements (accuracy of 0.1 kWh) to the overall system energy consumption, calculates the energy consumption index per unit product, and generates energy consumption reports per hour / day / week. The management module integrates 10 anomaly status recognition algorithms to early warn of equipment failures or parameter deviations and provides a historical data query function, supporting multi-dimensional filtering by time, batch, variety, etc.

[0034] Preferably, the double-line hanging multi-temperature zone double-layer tunnel drying structure includes:

[0035] Set two independently operating drying lines, each drying line is equipped with a hanging conveyor belt, and a plurality of hanging units are installed on the hanging conveyor belt to obtain a double-line hanging conveyor mechanism;

[0036] Each hanging unit is provided with a weighing sensor and a near-infrared moisture content sensor, and a 3D laser scanner is deployed at the entrance of the double-line hanging conveyor mechanism;

[0037] According to the double-line hanging conveyor mechanism, it is set as a tunnel structure, including upper and lower layers. Both the upper layer and the lower layer can be used for the kelp carried by the hanging units to pass through for drying, obtaining a double-layer reusable tunnel structure;

[0038] Along the length direction of the double-layer reusable tunnel structure, its internal space is divided into at least four physically isolated temperature zones with multiple independent temperature zones, obtaining multiple drying functional temperature zones; among them, the division of multiple independent temperature zones includes a first temperature zone for kelp preheating and sweating, a second temperature zone for balance dehumidification, a third temperature zone for heating and drying, and a fourth temperature zone for cooling and shaping.

[0039] Independently adjustable temperature sensors, humidity sensors, and wind speed sensors are distributed on the inner walls and tops of the multiple drying functional temperature zones to construct a drying temperature zone monitoring network.

[0040] In the embodiment of the present invention, the double-line hanging conveyor mechanism is made of 304 stainless steel, with two independent drying lines having a parallel spacing of 1.5 meters, and each drying line is 20 meters long. A chain-type hanging conveyor belt is installed on each drying line, and the conveyor belt is driven by a 22-kilowatt variable-frequency motor with a speed range of 0.5 - 3 meters per minute and an accuracy of ±0.05 meters per minute. A hanging unit connection point is set every 0.8 meters on the conveyor belt, with a total of 25 hanging points. The hanging unit adopts a quick-release stainless steel hook structure with a load-bearing capacity of 15 kilograms per single unit, and is fixedly connected to the conveyor belt through a special lock to ensure no detachment under high-temperature environments. The double-line conveyor system can operate independently or synchronously through a central PLC controller, and the operating status is displayed on the control panel in real time. A DS-1 type high-precision weighing sensor with a range of 0 - 20 kilograms and an accuracy of ±5 grams is integrated at the bottom of each hanging unit, and it adopts an IP67-level waterproof and dustproof design. An NIR-100 type near-infrared moisture content sensor is installed on the side of the unit, with a spectral range of 900 - 1700 nanometers, a sampling frequency of 10 times per second, and a moisture content detection accuracy of ±0.5%. An LT-300 three-dimensional laser scanner is fixedly installed at the entrance of the double-line hanging conveyor mechanism, with a scanning accuracy of 0.1 millimeter, a scanning frequency of 60 hertz, and a field of view angle of 120°×90°. The size and volume of kelp are calculated in real time through a three-dimensional point cloud image processing algorithm. All sensor data is transmitted to the data acquisition unit through the RS485 bus, and the sampling period is 1 second. The double-layer multiplexing tunnel structure shell is made of 100-millimeter-thick polyurethane sandwich insulation board, and the overall size is a cuboid with a length of 22 meters, a width of 4 meters, and a height of 3 meters. The inside of the tunnel is divided into upper and lower layers, each layer with a height of 1.4 meters and a layer spacing of 0.2 meters. Guide rail systems are equipped on both the upper and lower layers for the hanging units to carry kelp through. Automatic rolling doors are set at the entrance and exit of the tunnel, with a door size of 2 meters × 2 meters and a switch speed of 0.5 meters per second. The inner wall of the tunnel uses high-temperature-resistant reflective aluminum plates to enhance the thermal efficiency. Each layer is designed with a 5° micro-inclined structure to ensure the automatic drainage of condensed water. Maintenance doors are set on both sides of the tunnel, one every 5 meters, with a size of 1 meter × 2 meters. The 22-meter-long double-layer multiplexing tunnel structure is divided into four physically isolated temperature zones, and the physical isolation between each temperature zone is achieved by a silica gel curtain with a thickness of 30 millimeters. The adjustable height range of the opening of the curtain is 1.2 - 1.5 meters. The first temperature zone is 5 meters long, with the temperature controlled at 36 - 40°C and the relative humidity at 75 - 85%, for preheating and sweating of kelp; the second temperature zone is 5 meters long, with the temperature controlled at 42 - 46°C and the relative humidity controlled at 55 - 65%, for balancing and dehumidifying; the third temperature zone is 7 meters long, with the temperature controlled at 50 - 65°C and the relative humidity controlled at 35 - 45%, for heating and drying; the fourth temperature zone is 5 meters long, with the temperature gradually reduced to 28 - 32°C and the relative humidity controlled at 30 - 40%, for cooling and shaping. Each temperature zone is equipped with an independent hot air circulation system, and the heat source uses electric heating tubes with powers of 12 kilowatts, 15 kilowatts, 25 kilowatts, and 10 kilowatts respectively.The drying temperature zone monitoring network evenly arranges sensor nodes around the inner walls and at the top of each temperature zone. 8 PT100 temperature sensors are installed around the inner walls of each temperature zone, with a measurement range of 0 - 150 °C and an accuracy of ±0.1 °C; 4 SHT85 high-precision humidity sensors are installed at the top of each temperature zone, with a measurement range of 0 - 100%RH and an accuracy of ±1.5%RH; 2 FS450 thermal anemometers are installed at the inlet and outlet of each temperature zone, with a measurement range of 0 - 30 m / s and an accuracy of ±0.1 m / s. All sensor data is collected every 2 seconds through an industrial Internet of Things module. After being processed by the data preprocessing algorithm, it is transmitted to the central control system. On the inner walls of the tunnels in the first and third temperature zones, micro air vents with a diameter of 5 cm are installed at intervals of 30 cm on both sides of the coastal belt conveyor track. Each air vent is equipped with a wind deflector driven by a 15-watt servo motor, and the angle can be precisely adjusted within the range of 0 - 90 degrees. According to the moisture content data of the kelp detected by the infrared sensor in real time, the air vent array automatically adjusts the angle of the wind deflector through the PID control algorithm, so that the hot air blows directly towards the part of the kelp with a higher moisture content. On the top and both sides of the inner wall of the tunnel in the first temperature zone, 24 power-adjustable infrared radiation heating units are installed at equal intervals according to the width of the kelp conveyor belt. Each unit consists of a ceramic heating element and a focusing reflector. The power range of a single heating unit is 100 - 500 watts, and the power adjustment with an accuracy of 0.1 watt is achieved through a silicon-controlled rectifier. Main air valves with a DN200 diameter are installed in each temperature zone, driven by precision stepper motors, with a valve opening adjustment accuracy of 0.5% and a maximum air volume of 2000 cubic meters per hour. The second temperature zone is equipped with a centrifugal dehumidification fan with a rated power of 5 kW, and the rotational speed range is 800 - 3000 rpm, and the speed is precisely adjusted through a frequency converter. A 7.5-kW high-temperature fan is installed in the third temperature zone, with a maximum air temperature of 80 °C. A 3-kW low-temperature fan is used in the fourth temperature zone, and the air temperature is controlled at 30 - 35 °C. A plate heat exchanger is installed between the exhaust outlet of the third temperature zone and the intake inlet of the first temperature zone. The heat exchange area is 25 square meters, made of SUS316L stainless steel, with strong corrosion resistance. The internal design of the heat exchanger is a staggered flow channel structure, with the hot flow channel and the cold flow channel arranged alternately, and the width of a single flow channel is 8 mm. The exhaust hot air (temperature about 70 °C) and the fresh intake air (ambient temperature) exchange heat without mixing, and the heat recovery efficiency reaches 75%. The heat exchanger is equipped with temperature sensors to monitor the temperature difference between the inlet and outlet, and automatically adjusts the opening of the bypass valve according to the temperature difference to keep the inlet air temperature of the first temperature zone stable at 40 - 45 °C. The central control unit uses an industrial controller with a redundant architecture. The main controller is equipped with a quad-core processor, with a main frequency of 3.2 GHz and 32 GB of memory, and connects all actuators using EtherCAT fieldbus technology. Each actuator is equipped with an independent microcontroller, forming a three-level control network: central controller - regional controller - terminal actuator. The system response time is less than 20 milliseconds, and the control accuracy reaches ±0.5% of the set value.The system automatically generates an optimal drying curve based on factors such as product batch characteristics, environmental humidity, and the initial moisture content of kelp, and the actuator group works collaboratively to accurately track this curve.

[0041] Preferably, the kelp characteristic perception module is specifically:

[0042] S21: Use a three-dimensional laser scanner to perform non-contact contour scanning on a single piece / single cluster of fresh kelp units entering the automatic hanging station to obtain kelp three-dimensional morphology data;

[0043] S22: Perform three-dimensional surface reconstruction based on the kelp three-dimensional morphology data and estimate the macroscopic surface area of the kelp;

[0044] S23: Conduct kelp monitoring feature analysis based on the macroscopic surface area of the kelp to obtain target kelp monitoring feature data; among them, the target kelp monitoring feature data includes kelp wet weight data, target kelp average thickness data, and the moisture content at multiple points on the kelp surface;

[0045] S24: Evaluate the hanging spacing value between the front and rear hanging units based on the macroscopic surface area of the kelp to obtain the target kelp hanging spacing value;

[0046] S25: Control the hanging unit to adjust the spacing on the conveyor chain according to the target kelp hanging spacing value to obtain real-time hanging queue layout data;

[0047] S26: Match the spraying pressure and spraying action time according to the target kelp average thickness data and the macroscopic surface area of the kelp to obtain the spraying instruction for the cleaning area;

[0048] S27: Set the air knife drying parameters according to the target kelp monitoring feature data and the macroscopic surface area of the kelp to obtain the air knife drying control instruction for the kelp;

[0049] S28: Correlate the production instructions for the spraying instruction in the cleaning area and the air knife drying control instruction for the kelp through the real-time hanging queue layout data to obtain the data of the kelp queue to be dried.

[0050] In the embodiment of the present invention, before the kelp enters the automatic hanging station, a non-contact contour scan is performed using an LT-500 high-precision laser three-dimensional scanner. The scanner is arranged 1.2 meters above the front end of the hanging station, with a spatial resolution of 0.1 mm, a field of view of 120°×90°, and a sampling frequency of 120 Hz. The scan uses a 180° rotational scan method, and the motor drives the scan head to rotate at a speed of 15 revolutions per minute to obtain omnidirectional point cloud data. The scan process lasts for 3 seconds, and the amount of point cloud data obtained in a single scan reaches 3.6 million points. The point cloud data is transmitted to the processing unit in real time through industrial Ethernet. After being processed by the point cloud registration algorithm and removing environmental noise points, a complete three-dimensional topography data of the kelp is formed. The data is saved in the format of a three-dimensional matrix with a matrix size of 1024×1024×3, and the accuracy reaches the millimeter level. The obtained three-dimensional topography point cloud data of the kelp is processed by the Poisson surface reconstruction algorithm, and the reconstruction accuracy is set to 0.5 mm. First, noise reduction processing is performed on the point cloud data. The statistical analysis method within a spherical neighborhood with a radius of 2 mm is used to remove outliers, and connected points with a degree not less than 8 are retained. Then, a triangular mesh is constructed using a space division method with an octree depth of 10 to generate a complete surface model of the kelp. The surface area of the surface model is calculated by the discrete differential geometry method, and the specific calculation formula is , where S is the macroscopic surface area of the kelp, is the area of the i-th triangular patch, and the area unit is square centimeters. The accuracy of the calculation result is controlled within the range of ±1%, and at the same time, the data of the maximum length, maximum width, and surface undulation of the kelp are recorded. The wet weight data of the kelp is measured by a high-precision XS-500 electronic scale with an accuracy of ±0.5 g, and the measured value is transmitted to the database in real time. The average thickness of the kelp is obtained by measuring at 25 sampling points evenly divided on the surface grid of the kelp using an LDM-150 laser rangefinder. The measurement accuracy is 0.01 mm, and the arithmetic mean of the thicknesses of the sampling points is taken as the average thickness. The calculation formula is , where H is the average thickness of the kelp, is the thickness of the i-th measurement point, and n is the number of measurement points, which is 25. The moisture content on the surface of kelp is measured at 5 key points using a NIR-200 near-infrared spectrometer, with a wavelength range of 900 - 1700 nm, a sampling depth of 0.5 - 3 mm, a measurement accuracy of ±0.5%, and the average moisture content and standard deviation are calculated. The system sets the basic spacing value to 25 cm. The calculation formula for the target kelp hanging spacing value is: D_spacing = D_base + k × (S_measured - S_standard), where D_spacing is the target hanging spacing value (cm), D_base is the basic spacing value (25 cm), S_measured is the measured macroscopic surface area of kelp (cm²), S_standard is the standard kelp surface area (400 cm²), and k is the proportionality coefficient (0.05 cm / cm²). When S_measured < 200 cm², the spacing coefficient is adjusted to 0.025; when 200 ≤ S_measured < 600 cm², the spacing coefficient remains unchanged at 0.05; when S_measured ≥ 600 cm², the spacing coefficient increases to 0.075. Considering the air circulation requirements, the system sets the minimum hanging spacing to 15 cm and the maximum to 45 cm. When the calculation result exceeds the range, the system automatically limits it within the valid range. The spacing evaluation considers the surface area difference between the front and rear two kelp units, and takes the arithmetic mean of the two spacing values as the final adjustment value to ensure the balance of the overall hanging queue spacing. The system converts the target kelp hanging spacing value into the pulse number of the conveyor chain stepper motor. The calculation formula is: N_pulse = D_spacing × P, where N_pulse is the stepper motor pulse number, D_spacing is the target hanging spacing value (cm), and P is the transmission ratio (200 pulses / cm). The system sends a positioning command to the stepper motor controller through the RS485 bus to control the precise movement of the conveyor chain to the target position, and the positioning accuracy is controlled within ±1 mm. During the positioning process, the Hall sensor monitors the actual displacement of the chain in real time to form a closed-loop feedback and correct the cumulative error. After positioning is completed, the system records the absolute position coordinates of this hanging unit on the conveyor chain (the starting point of the chain is 0), and generates real-time hanging queue layout data including position, spacing, and timestamp. At the same time, the encoder continuously monitors the running speed of the conveyor chain. When the speed fluctuation exceeds ±2%, automatic compensation adjustment is performed to ensure the uniform running of the overall hanging queue. The calculation formula for the spray pressure is: P_spray = P_base × (d_average / d_standard) × (S_measured / S_standard)^0.5, where P_spray is the spray pressure (MPa), P_base is the basic spray pressure (0.25 MPa), d_average is the average thickness of kelp (mm), d_standard is the standard thickness (2 mm), S_measured is the measured surface area (cm²), and S_standard is the standard surface area (400 cm²). The calculation formula for the spray action time is: T_spray = T_base × (d_average / d_standard)^1.5 × (S_measured / S_standard)^0.3, where T_spray is the spray action time (s), and T_base is the basic spray time (3 s).The system sets the output pressure of the variable-frequency water pump (range: 0.1 - 0.5 MPa, adjustment step: 0.05 MPa) and the opening time of the solenoid valve (range: 1 - 6 s, accuracy: 0.1 s) according to the calculation results. The cleaning area is equipped with 6 groups of rotating nozzles, with a nozzle diameter of 1.2 mm and an installation spacing of 10 cm, forming a cross-type spraying network. The system accurately controls the start and stop time of the nozzles according to the kelp position information, reduces the loss of cleaning liquid medicine, and ensures that the cleaning uniformity reaches over 95%. The formula for setting the air knife wind speed parameter is: V wind speed = V base × (W average / W standard)^0.7 × (S measured / S standard)^0.4, where V wind speed is the air knife wind speed (m / s), V base is the base wind speed (25 m / s), W average is the average moisture content (%), W standard is the standard moisture content (85%), S measured is the measured surface area (cm²), and S standard is the standard surface area (400 cm²). The formula for setting the air knife action distance parameter is: L distance = L base × (d average / d standard)^0.5, where L distance is the distance between the air knife and the kelp surface (cm), L base is the base distance (8 cm), d average is the average thickness of the kelp (mm), and d standard is the standard thickness (2 mm). The system adjusts the fan speed (range: 1000 - 3000 rpm, adjustment step: 50 rpm) through the frequency converter and adjusts the air knife height (range: 5 - 15 cm, adjustment accuracy: 0.5 cm) through the servo mechanism according to the calculation results. The air knife system adopts an upper and lower double-layer design, with 5 groups of linear airflow air knives configured on each layer. The air knife length is 50 cm, and the air outlet width is 1.5 mm, forming a uniform airflow field covering the entire conveyor belt width. The generated air knife drying control instruction contains three core parameters: wind speed, distance, and action time, which accurately match the drying requirements of kelp with different characteristics. The system adopts a double-buffer queue structure. The main queue stores the kelp position information, and the sub-queue stores the corresponding processing instructions. The association algorithm binds all the processing instructions of the same kelp unit with the kelp ID as the index key to form an instruction chain. The system updates the queue status every 100 ms, calculates the estimated time for each kelp unit to reach each workstation according to the actual speed of the conveyor chain, and the system sends control instructions to the target workstation 5 s in advance.

[0051] Preferably, the kelp monitoring feature analysis based on the macroscopic surface area of the kelp includes:

[0052] Weigh a single piece / cluster of fresh kelp units using a weighing sensor to generate kelp wet weight data;

[0053] Evaluate the average thickness of the kelp based on the kelp wet weight data and the macroscopic surface area of the kelp to obtain the target kelp average thickness data;

[0054] Perform multi-point near-infrared spectral rapid scanning on a single piece / cluster of fresh kelp units using a near-infrared moisture sensor to generate kelp surface spectral data;

[0055] Perform baseline drift correction and filtering on the spectral data of the kelp surface to generate corrected spectral data of the kelp surface;

[0056] Identify characteristic bands for the corrected spectral data of the kelp surface through a preset kelp spectral calibration information library, and evaluate the average initial moisture content on the kelp surface to obtain the moisture content at multiple points on the kelp surface.

[0057] In the embodiment of the present invention, when the kelp is transported to the automatic hanging station, a HS-3000 high-precision weighing sensor is used to weigh a single-piece / single-cluster fresh kelp unit. The range of this sensor is 0-5 kg, the accuracy is ±0.5 g, the sensitivity is 2 mV / V, and the response time is less than 100 ms. The weighing sensor is installed at the bottom of the hanging unit, adopting a four-point support structure, and measuring the weight through the principle of the Stern-Whitney bridge. During the weighing process, the system collects 300 data points, the sampling frequency is 100 Hz, lasting for 3 s. After removing the highest 5% and the lowest 5% of the outliers, the arithmetic mean is taken to obtain the wet weight data W of the kelp (unit: g). The measured data is converted by a 24-bit high-precision analog-to-digital converter and then transmitted to the central processing unit through the RS485 bus, updated in real time to the kelp characteristic database, and associated with a unique kelp unit identification code. According to the obtained wet weight data W of the kelp (unit: g) and the macroscopic surface area S of the kelp (unit: cm²), analyze the average thickness H of the kelp (unit: mm). The system automatically analyzes and calculates the results. If the thickness value is less than 0.5 mm or greater than 5 mm, trigger the anomaly detection mechanism and re-collect the data; if there are three consecutive anomalies, a warning message will be displayed on the industrial control computer. Use a NIR-2000 near-infrared moisture sensor to perform multi-point scanning on a single-piece / single-cluster fresh kelp unit. The spectral range of this sensor is 850-2500 nm, the spectral resolution is 2 nm, the signal-to-noise ratio is 5000:1, and the sampling speed is 50 ms / point. The scanning device is driven by a stepping motor, and 9 fixed sampling points are set on the kelp surface along a "Z" -shaped path. The sampling point spacing is 1 / 3 of the maximum length and 1 / 3 of the maximum width of the kelp. Each sampling point is measured 3 times and averaged to ensure data stability. The sensor probe maintains a fixed distance of 5 mm from the kelp surface, which is controlled in real time by a precision distance sensor. Each scan obtains the reflection spectral data of 1024 wavelength points, forming a 9×1024 spectral matrix, which is transmitted to the spectral analysis module through a high-speed optical fiber, and the data transmission rate is 100 Mbit / s. First, perform baseline drift correction on the collected spectral data of the kelp surface, and use the polynomial fitting method to remove baseline noise. The specific steps are as follows: Determine that the baseline anchor points are at 1000, 1200, 1800, and 2300 nm, and use the cubic spline function for baseline fitting. The fitting formula is , where is the baseline value at wavelength λ, λ is the wavelength, , , , is the fitting coefficient. The fitting coefficient is solved by the least squares method, and the complete baseline curve is calculated. After correction, the spectrum , where is the original spectral reflectance, is the reflectance after baseline correction. In the second stage, filtering is performed. The Savitzky-Golay smoothing filter algorithm is used, the window width is set to 15 data points, and the polynomial order is 3. The filtering formula is: , where is the spectral value of the i-th point after filtering, the value of j ranges from -n to n, Rc(i + j) is the spectral value of the (i + j)-th point after correction, is the convolution coefficient, and n is the window half-width. Outliers are removed using the standard deviation determination method, and the threshold is set to 3 times the standard deviation. After completion of correction and filtering, the system normalizes the spectral data of 9 measurement points to generate calibrated spectral data of the kelp surface in a standard format, and the data accuracy reaches ±0.001 reflectance units. The system identifies characteristic bands for the calibrated spectral data of the kelp surface through a preset kelp spectral calibration information database. This calibration information database contains standard spectral data of 500 kelp samples with different moisture contents (35% - 95%). The moisture content true values of the samples are calibrated by the oven drying method at 105°C. A spectral - moisture content mathematical model is established. The continuous wavelet transform method is used for characteristic band identification, and it is determined that 1450 nm and 1940 nm are the main moisture absorption peaks, and 970 nm is the secondary peak. The moisture content is calculated using a multiple linear regression model: , where W is the moisture content (%), , , , , , are the calibrated spectral reflectances at the corresponding wavelengths respectively, , , , are the regression coefficients (92.3, -45.6, -32.8, -15.2 respectively). The system calculates the moisture content for each of the 9 measurement points and judges the uniformity: If the difference between the maximum value and the minimum value exceeds 10%, it is marked as "non-uniform sample", and the drying parameters need to be specially adjusted. The system calculates the average value of the 9 points as the average initial moisture content of the kelp, and at the same time records the moisture content distribution map to form a complete multi-point moisture content dataset of the kelp surface.

[0058] Preferably, the automated production operation parameters of kelp include the reference data for preheating operation in Zone 1, the reference data for dehumidification operation in Zone 2, the reference data for drying operation in Zone 3, and the reference data for shaping operation in Zone 4. The drying parameter planning module is specifically:

[0059] Automatically clean and blow-dry pre-treatment production operations on single-piece / single-cluster fresh kelp units using the data of the kelp team to be dried, generating kelp pre-cleaning status data;

[0060] Perform one-zone preheating target duration processing based on the kelp pre-cleaning status data and the target kelp average thickness data in the target kelp monitoring feature data, generating one-zone preheating target duration data;

[0061] Set the rotation speed of the one-zone preheating circulation fan according to the one-zone preheating target duration data, and integrate the operation control benchmarks according to the one-zone preheating target duration data to obtain one-zone preheating operation benchmark data;

[0062] Plan the dehumidification target temperature, the initial air valve target opening percentage of the exhaust fan, and the dehumidification treatment duration in the second temperature zone based on the moisture content at multiple points on the kelp surface in the target kelp monitoring feature data to obtain two-zone dehumidification operation benchmark data;

[0063] Set the heating and drying target temperature, the target air velocity value in the main air duct of the powerful exhaust fan, and the heating and drying treatment duration based on the wet weight data of the kelp in the target kelp monitoring feature data to obtain three-zone drying operation benchmark data;

[0064] Set the cooling temperature and the target air velocity of the cooling fan, and perform processing on the target moving speed of the conveyor belt based on the target kelp monitoring feature data to obtain four-zone shaping operation benchmark data.

[0065] In the embodiment of the present invention, based on the data of the seaweed queue to be dried, the system automatically starts the cleaning-blowing pre-treatment process. The seaweed queue to be dried is placed on a mesh stainless steel conveyor belt in a single-piece / single-cluster manner by an automatic sorting mechanism, and the conveyor belt speed is set at 0.5 meters per minute. During the cleaning process, a high-pressure fresh water spraying system is used, the spraying pressure is maintained at 0.6 MPa, the nozzle spacing is 10 cm, and the spraying time is 60 seconds to ensure that the salt and impurities attached to the surface of the seaweed are fully removed. Subsequently, a multi-stage centrifugal high-speed fan array dries the seaweed, the fan speed is 3000 revolutions per minute, the wind speed reaches 15 meters per second, and the drying time is 45 seconds. During the process, distributed optical sensors and infrared moisture sensors collect the appearance characteristics and surface moisture distribution data of the seaweed in real time, and generate a pre-cleaning state data matrix containing information such as seaweed morphological parameters, surface moisture distribution maps, and salt content. By sequentially inputting single-piece or single-cluster fresh seaweed units into the fully automatic transmission track system, a multi-head high-pressure water knife device is used to comprehensively clean the surface of the seaweed. After the cleaning is completed, a multi-nozzle air knife system is used for surface drying. During the cleaning process, an image recognition camera is used to obtain the residual stains on the surface of the seaweed, and a pre-treatment state recognition module generates the pre-cleaning state data of the seaweed. Subsequently, the thickness data of the target seaweed is collected. The thickness measurement is carried out by a laser displacement sensor, and the sensor accurately outputs the average thickness value of each piece of seaweed in millimeters. The preheating target duration takes the thickness parameter as the input variable and is calculated by the following empirical formula: preheating duration T_pre (seconds) = k_pre × H, where T_pre represents the preheating target duration in zone one, H is the average thickness of the seaweed (millimeters), and k_pre is a preset empirical constant with a value range of 8 - 12, and the average value obtained by fitting historical data is 10. T_pre is output to the preheating control system to set the speed of the circulating hot air fan, and the fan speed V_pre is controlled by a frequency converter, with the set range being 800 - 1600 revolutions per minute to meet the heat supply within the range of T_pre. The operation control reference integration module summarizes T_pre, V_pre, and the hot air temperature parameter θ_pre (initially set at 55 degrees Celsius) to form the preheating operation reference data in zone one. An infrared moisture meter installed between the outlet of the preheating zone and the inlet of the dehumidifying zone is used to perform non-contact moisture content detection on multiple areas of the seaweed surface to obtain the multi-point moisture content data on the seaweed surface. The system collects no less than 5 measurement point data for each piece of seaweed and calculates the average moisture content W_deh (in mass percentage). According to the value of W_deh, the dehumidifying target temperature θ_deh (degrees Celsius), the initial air valve opening P_deh (percentage) of the exhaust fan, and the dehumidifying treatment duration D_deh (seconds) are set through the dehumidifying strategy model. The calculation formula is as follows: θ_deh = θ_deh0 - ×(W_deh-W_targ), P_deh = P_base + μ×(W_deh-W_targ), D_deh = ν×(W_deh-W_targ); where: θ_deh0 is the default dehumidification temperature (50 °C), W_targ is the target moisture content (65%), P_base is the basic damper opening (30%), , μ, and ν are system empirical coefficients, which are set to 0.5, 1.2, and 10 respectively. All parameters are obtained through fitting analysis of historical production data. The dehumidification area uses a heat pump dehumidification device. The temperature is controlled by automatically adjusting the heater power through a PID regulator, and the damper opening is controlled by an electric actuator to adjust the damper angle. The dehumidification operation reference data is uniformly generated by the control system and sent to the execution unit in the dehumidification area. Before the kelp unit that has completed dehumidification treatment enters the three-zone drying area, the wet weight M_wet (in grams) of a single piece of kelp is measured in real time using a dynamic weighing module. An artificial intelligence prediction model is used to model the relationship between M_wet and the drying target, and the target temperature θ_dry for heating and drying, the main air duct wind speed V_dry, and the drying treatment duration D_dry are set. The model uses multiple regression to fit the following functional relationship: θ_dry = θ_ref + α×lg(M_wet / M_targ), V_dry = V_base + β×(M_wet-M_targ), D_dry = γ× , where: θ_ref is the preheating reference temperature in the first zone (reference value), M_targ is the target mass after drying (set to 15 grams), α, β, and γ are fitting coefficients, which are set to 5, 0.2, and 15 respectively, and V_base is the basic wind speed (3 m / s). The three-zone drying area uses a multi-segment electric heating module, and the temperature control accuracy is maintained within ±1 °C. The main air duct is closed-loop regulated through a wind speed sensor. The drying duration is accurately controlled in seconds, and the system control module integrates θ_dry, V_dry, and D_dry to form the three-zone drying operation reference data. After the kelp drying is completed, it enters the shaping area. First, the current kelp temperature θ_curr (in °C) is detected through a thermocouple temperature sensor. The system sets the target wind speed V_cool (in m / s) for the cooling fan according to the difference Δθ = θ_curr-θ_cool between θ_curr and the target cooling temperature θ_cool (set value of 30 °C). The specific calculation method is: V_cool = V_min + σ×Δθ, where: V_min is the minimum wind speed (2 m / s), and σ is the adjustment coefficient, which is set to 0.3. The cooling fan uses variable frequency control, and the wind speed adjustment range is 2 - 6 m / s. Subsequently, the arrangement spacing and size of the kelp are measured through an infrared induction positioning system, and combined with the dried length L of the kelp, the conveyor belt moving speed (in m / min) is set to ensure that the kelp does not overlap or curl during the shaping process, and realize synchronous operation of cooling and shaping.

[0066] Preferably, the preheating and sweating production optimization in the multi-temperature zone kelp drying production optimization process in the intelligent intervention adjustment module includes:

[0067] Based on the data of the kelp queue to be dried, control the kelp drying unit to enter the first temperature zone in the double-line hanging multi-temperature zone double-layer tunnel drying structure through the hanging conveyor belt;

[0068] Carry out preheating operation control on the first temperature zone according to the reference data of the first-zone preheating operation in the automated kelp production operation parameters, and continuously monitor the surface temperature of the kelp drying unit using a temperature sensor to obtain the kelp surface temperature distribution data;

[0069] Use a humidity sensor to real-time sense the local microenvironment humidity around the kelp drying unit to generate the local humidity on the kelp surface;

[0070] Calculate the average surface temperature and temperature uniformity index of the kelp according to the kelp surface temperature distribution data;

[0071] Evaluate the current sweating state of the local humidity on the kelp surface, the average surface temperature of the kelp, and the temperature uniformity index through the preset ideal sweating state parameters to generate the kelp segmented sweating evaluation data.

[0072] In the embodiment of the present invention, the feeding system of the kelp drying unit adopts a double-track hanging conveyor structure, which is driven by a high-precision servo motor, and the running speed of the conveyor belt is fixed at 0.3 m / s. First, the system automatically batches according to the data of the kelp queue to be dried, and the total weight of each batch is controlled within the range of 75±2 kg. The kelp unit is fixed by hooks made of stainless steel 304, the hook spacing is 250 mm, and the load of each hook is controlled within the range of 200±20 g. The conveyor belt adopts a chain structure, the chain pitch is 25 mm, and the mechanical strength reaches 3500 N. The feeding device is equipped with a high-resolution infrared sensor. When it detects that the position of the kelp unit deviates by more than 10 mm, it automatically triggers the correction mechanism to adjust the posture. The double-line hanging system consists of upper and lower layers, the distance between the layers is 800 mm, and a total of 32 hooks form a standard batch unit. An automatic disinfection spray device is installed at the entrance of the conveyor system, and the spray pressure is controlled at 0.4 MPa to ensure that the surface of the kelp is free of miscellaneous bacteria contamination before entering the drying tunnel. Under the precise scheduling of the conveyor belt control unit, each batch of kelp units smoothly enters the preheating area of the first temperature zone to start the subsequent preheating and sweating process. The preheating operation control in the first temperature zone is executed based on the pre-determined preheating operation reference data in the first zone, including the preheating temperature set value of 57°C, the humidity set value of 65%, the circulating fan speed of 1300 rpm, and the preheating duration of 180 s. The control system adopts a distributed PLC controller, model Siemens S7-1500 series, with an operation cycle of 30 ms / time. The temperature control adopts the proportional-integral-derivative (PID) algorithm, the proportional coefficient is set to 2.5, the integral time is 120 s, the derivative time is 30 s, and the temperature control accuracy reaches ±0.5°C. 16 PT100 platinum resistance temperature sensors are installed in the drying chamber of the first temperature zone, evenly distributed around the drying chamber, and the sampling frequency is 2 s / time. There are also 8 infrared array temperature sensors, model MLX90640, with a resolution of 32×24 pixels, directly aiming at the surface of the kelp for non-contact temperature measurement, the temperature measurement range is 20-120°C, and the accuracy is ±1°C. The surface of each kelp drying unit is divided into 5×4 temperature monitoring areas, and the system continuously collects the real-time temperature data of these 20 areas to form a complete kelp surface temperature distribution data matrix. The temperature data is transmitted to the central control unit through the RS485 bus, the data transmission rate is 19200 bps, and the data acquisition cycle is 5 s. The system deploys 12 digital thermistor humidity sensors, model DHT22, in the first temperature zone, with a measurement range of 0-100%RH and an accuracy of ±2%RH, distributed at different heights around the drying unit. The distance between each sensor and the surface of the kelp is controlled within the range of 50-100 mm to form a three-dimensional monitoring network. The sensor sampling cycle is 3 s, and the data is sent to the regional data concentrator through a micro wireless transmission module.The system also installs a capacitive humidity sensor, model HYT271, above, below and on both sides of each kelp unit, with a measurement range of 0-100%RH, an accuracy of ±1.8%RH and a response time of less than 10 seconds. The sensor is connected to the regional data integration unit through the I2C bus to form a local microenvironment humidity monitoring network. The system performs spatiotemporal fusion processing on the collected humidity data, and uses the cubic spline interpolation algorithm to reconstruct the humidity field around the kelp, with a resolution of 10mm×10mm×10mm. After the humidity data is processed by Gaussian filtering and noise reduction, a local humidity numerical matrix corresponding to 20 areas on the kelp surface is generated, and the data update frequency is 5 seconds / time. The validity of the 20 regional temperature point data obtained is tested to remove abnormal values beyond the normal range (25-80℃). For the valid temperature points, the system assigns different weight coefficients according to the different areas represented by each temperature measurement point. The weight of the central area is 1.2, the weight of the edge area is 0.8, and the weight of the remaining areas is 1.0. The average surface temperature is calculated by dividing the sum of the product of the temperature of each area and the corresponding weight by the total weight. The temperature uniformity index is calculated using an improved coefficient of variation method. First, the standard deviation of all valid temperature points is calculated, then divided by the average temperature value, and then multiplied by 100 to obtain the uniformity index in percentage form. The system sets a three-level uniformity evaluation standard: when the index is less than 5%, it is excellent uniformity, 5%-10% is good uniformity, and greater than 10% is uneven. The system updates the average surface temperature and uniformity index calculation results every 10 seconds, and displays them intuitively in digital and color coding on the operation interface. When it is found that the uniformity index calculated three times in a row is greater than 12%, the system automatically triggers the non-uniform heating alarm and adjusts the heating strategy. The ideal sweating state parameters are determined based on a large amount of historical production data analysis, including three key indicators: the local humidity index range of kelp surface is 75%-85%, the average surface temperature index range of kelp is 53-58℃, and the temperature uniformity index is less than 8%. The system uses a fuzzy evaluation method to evaluate the current sweating state, constructs a three-dimensional state space, and maps each indicator to a scoring range of 0-100. The local humidity scoring function is a bell-shaped function, and the highest score is when the humidity is 80%; the average temperature scoring function is a trapezoidal function, and the highest score is in the range of 54-57°C; the uniformity index scoring function is a decreasing function, and the lower the index, the higher the score. The system scores the sweating state of the 20 areas divided on the surface of the kelp drying unit, forming a complete segmented evaluation data matrix. The scoring weights are distributed as follows: humidity factor weight 0.4, temperature factor weight 0.35, uniformity factor weight 0.25.The system divides the sweating state into four levels according to the comprehensive score of each area: above 90 points is the best sweating state, 75 - 90 points is the good sweating state, 60 - 75 points is the general sweating state, and below 60 points is the poor sweating state, generating kelp segmented sweating evaluation data, including the sweating score of each area, sweating uniformity, percentage of sweating completion, and remaining estimated sweating time.

[0073] Preferably, the balance dehumidification production optimization in the multi - temperature - zone kelp drying production optimization process in the intelligent intervention and regulation module includes:

[0074] Evaluate the sweating difference degree of the first area according to the kelp segmented sweating evaluation data, and generate the sweating difference degree data of the first area;

[0075] Dynamically correct the angle of the micro air vents on the inner wall of the first temperature zone and the local radiation heating power through the sweating difference degree data of the first area, and generate the dynamic fine - tuning execution data of the first area;

[0076] Based on the dynamic fine - tuning execution data of the first area, judge whether the kelp pre - heating is qualified. When the kelp pre - heating is qualified, control the kelp drying unit that has completed pre - heating in the first temperature zone to enter the second temperature zone, and set the initial temperature, air valve opening, and fresh air supply volume of the second temperature zone according to the dehumidification operation benchmark data in the kelp automated production operation parameters to obtain the initial environment setting data of the second area;

[0077] In the second temperature zone, continuously monitor the real - time weight and surface moisture content change of the kelp drying unit using a weighing sensor and a near - infrared moisture sensor, and calculate the water loss rate to generate the real - time water loss rate data of the second area;

[0078] Through the humidity sensor and wind speed sensor in the drying temperature zone monitoring network, real - time monitor the actual local humidity and actual average wind speed around the kelp drying unit to generate the real - time local humidity - speed data of the second area;

[0079] Analyze the abnormal dehumidification state according to the real - time water loss rate data of the second area and the real - time local humidity - speed data of the second area, and generate the abnormal dehumidification state data;

[0080] Based on the abnormal dehumidification state data, intelligently adjust the weak - wind dehumidification mode for the initial environment setting data of the second area, and monitor the discharging state of the kelp drying unit in the second area to generate the discharging state data of the second area.

[0081] In the embodiments of the present invention, the system divides the surface of kelp into 20 regions in the form of a 5×4 matrix, extracts the sweating state score values of each region, and calculates the standard deviation σ and coefficient of variation CV of the scores of all regions. When the standard deviation σ is greater than 8.5 or the coefficient of variation CV exceeds 15%, the system determines it as a high difference degree; when the standard deviation is between 5.5 - 8.5 or the coefficient of variation is between 10% - 15%, it is determined as a medium difference degree; when the standard deviation is less than 5.5 and the coefficient of variation is less than 10%, it is determined as a low difference degree. The system also calculates the temperature gradient distribution on the surface of kelp through thermal imaging analysis, and marks the regions with a temperature gradient exceeding 3.2℃ / cm as key intervention regions. The evaluation algorithm comprehensively considers the two factors of the difference in sweating state scores and the temperature gradient in a weighted manner, with a weight ratio of 7:3, to generate a packet of sweating difference degree in the first region, including the difference degree level, coordinates of the key intervention regions, and the sorting of intervention priorities. The data refresh frequency is 3 seconds / time. 36 independently controlled micro air vents are arranged on the inner wall of the drying chamber. Each air vent is equipped with a deflector blade driven by a high-precision servo motor, with an angle adjustment range of 0 - 85 degrees and an adjustment accuracy of ±0.5 degrees. The diameter of the micro air vent is 25 mm, the maximum air outlet speed is 4 m / s, and the distance between the air vents is 300 mm, evenly distributed in a matrix. The local radiation heating system consists of 24 carbon fiber far-infrared heating units, each with a rated power of 150 W, a power adjustment range of 30% - 100%, and an adjustment step of 1%. The system accurately locates the kelp regions that need to be intervened according to the coordinate information of the key intervention regions in the data of the sweating difference degree in the first region, and correspondingly activates the micro air vents and radiation heating units in specific regions. For low-temperature and high-humidity regions, the system increases the corresponding air vent angle to 55 - 65 degrees and raises the radiation power to 80% - 90%; for high-temperature and low-humidity regions, the system reduces the air vent angle to 15 - 25 degrees and lowers the radiation power to 40% - 50%. The correction parameters are calculated in real time by the PID controller, with an update period of 2 seconds, and a data matrix of dynamic fine-tuning execution in the first region, including 36 air vent angle values and 24 heating power values, is generated and sent to each execution unit through the fieldbus network. The kelp preheating qualification judgment system combines the dynamic fine-tuning execution data in the first region and real-time monitoring feedback, and executes a three-stage judgment process. The first stage checks whether the average temperature on the surface of kelp is stable within the range of 56±2℃ for more than 30 seconds; the second stage verifies whether the sweating state score reaches more than 80 points and the difference degree between regions drops to the low difference degree level; the third stage confirms whether the fine-tuning execution parameters remain within the stable range for 5 consecutive adjustment cycles. After all three stages are met, the system determines that the kelp preheating is qualified, triggers the start signal of the conveyor system, and controls the transfer conveyor with a speed of 0.2 m / s to transport the kelp drying unit to the second temperature zone. At the same time, the system reads the reference data for dehumidification operation in the second region, including the target temperature of 50℃, the initial opening of the air valve of 40%, and the fresh air supply of 25 cubic meters per minute, and sets the environmental parameters of the second temperature zone accordingly.The dehumidification area is equipped with 4 sets of temperature control systems with a temperature control accuracy of ±0.8°C; 8 electric air valve actuators with a control accuracy of ±1.2%; and 6 fresh air inlets with a flow control range of 10 - 60 cubic meters per minute. The system transmits the initial set values to the second temperature zone control unit, forming a two-zone initial environment setting data packet including a temperature zone setting value matrix, an air valve opening array, and a fresh air supply adjustment curve. The second temperature zone real-time monitoring system installs 8 high-precision suspended weighing sensors at the conveyor belt nodes, with the model of HBM C16, a range of 0 - 5 kg, an accuracy of 0.01%, and a sampling frequency of 50 Hz. Each kelp drying unit is connected to the weighing sensor through a special hook to achieve continuous dynamic weighing during the drying process. The system records the initial weight M_initial of the kelp entering the second temperature zone and records the real-time weight M_real every 10 seconds, and calculates the water loss rate through the weight change per unit time. At the same time, the system deploys 16 near-infrared moisture sensors, using dual-wavelength scanning technology of 940 nm and 1450 nm, with a measurement range of 15% - 95%, an accuracy of ±1.5%, a scanning interval of 50 mm, covering all key areas of the kelp surface. The moisture content data is transmitted to the data processing unit through optical fibers and cross-validated and calibrated in combination with the weight data. The system uses the sliding window method to smooth the water loss rate data within 30 consecutive seconds, calculates the average water loss rate after removing outliers, and simultaneously identifies the characteristics of three typical stages: the water loss acceleration period, the stable period, and the slowdown period. All data is integrated into a two-zone real-time water loss rate data packet, including the current water loss rate value, the water loss state determination result, the moisture content distribution map, and the moisture content change trend prediction, with a data update cycle of 5 seconds. The drying temperature zone monitoring network deploys an all-round perception system inside the second temperature zone, including 24 capacitive humidity sensors with a measurement range of 0 - 100%RH, an accuracy of ±1.5%RH, and a response time of less than 8 seconds; and 16 thermal anemometers with a measurement range of 0 - 10 m / s, an accuracy of ±0.2 m / s + 3% of the reading. The sensors are arranged in a matrix layout, covering the entire dehumidification area space. Vertically, one layer is set at 50 mm, 200 mm, and 400 mm away from the kelp surface, and horizontally, they are evenly distributed at an interval of 2 m. The data acquisition unit uses a 24-bit analog-to-digital converter with a sampling rate of 100 Hz and transmits the real-time data to the central processing system through the RS485 bus. The system calculates the local average humidity H_local and the actual average wind speed V_actual within a range of 200 mm around the kelp, and simultaneously records the spatial distribution gradients of humidity and wind speed. Triangular interpolation calculation is performed on the collected raw data to generate three-dimensional models of the humidity field and the wind speed field with a resolution of 100 mm × 100 mm × 100 mm. The data processing algorithm combines time series analysis to identify the change trends of humidity and wind speed, and updates and generates two-zone real-time local humidity and wind speed data every 5 seconds, including the humidity field distribution map, the wind speed field distribution map, the gradient change rate, and the microenvironment state evaluation result.The system sets the normal moisture removal reference range: the water loss rate is 0.8 - 1.5 grams per minute, the local humidity is 45% - 65%, and the average wind speed is 1.8 - 2.5 meters per second. The system constructs a moisture removal state feature vector, matches the current state with a preset standard mode library, and identifies potential anomalies. The abnormal modes include: wet resistance accumulation type (local humidity is greater than 70%, water loss rate is less than 0.6 grams per minute), over-drying type (surface moisture content change is greater than 3% per minute, water loss rate is greater than 1.8 grams per minute), air duct blockage type (local wind speed is less than 1.2 meters per second, and there is an obvious area with uneven wind speed), heat and moisture imbalance type (local humidity and water loss rate are negatively correlated), etc. typical abnormal modes. The system uses a fuzzy inference algorithm to calculate the matching degree between the current state and each abnormal mode, and determines the type and severity of the anomaly. The severity is divided into four levels: minor (matching degree 0.3 - 0.5), moderate (matching degree 0.5 - 0.7), severe (matching degree 0.7 - 0.85), and extreme (matching degree greater than 0.85). The system is equipped with 8 groups of independently controlled moisture removal fans with a power range of 0.5 - 2.5 kilowatts and an air volume range of 800 - 4000 cubic meters per hour; 12 electric air valve controllers with an opening adjustment range of 5% - 100%; 4 groups of variable frequency heaters with a power adjustment range of 3 - 15 kilowatts. For the wet resistance accumulation type anomaly, the system increases the air valve opening by 25% and reduces the temperature in the second zone by 3°C; for the over-drying type anomaly, the system reduces the wind speed by 30% and increases the fresh air supply by 5 cubic meters per minute; for the air duct blockage type anomaly, the system starts the auxiliary air duct and temporarily increases the main fan speed by 20%; for the heat and moisture imbalance type anomaly, the system adjusts the ratio of heating power to wind speed to maintain the best range of 1.8 - 2.2 kilowatt·seconds per meter. The adjusted parameters are updated every 10 seconds, and the adjustment effect is continuously monitored. At the same time, the second zone discharge state monitoring system deploys 4 high-precision infrared moisture meters with a detection accuracy of ±1%, a scanning width of 800 millimeters, covering the entire conveyor belt width; 2 thermal imagers with a temperature resolution of 0.05°C and a pixel resolution of 384×288; and a high-precision weighing platform with an accuracy of ±5 grams. The system comprehensively detects the kelp about to leave the second temperature zone, records the discharge moisture content, temperature distribution, and weight data, and judges whether the dehumidification effect meets the standard. When the moisture content of the kelp drops to 65% ± 3% and the temperature uniformity index is less than 5%, the system determines that the dehumidification is qualified, generates detailed second zone discharge state data, and controls the kelp to enter the next process.

[0082] Preferably, the heating and drying production optimization in the multi-temperature zone kelp drying production optimization process of the intelligent intervention and adjustment module includes:

[0083] Obtain the upper and lower layer load data of the third temperature zone; make upper and lower layer tunnel operation decisions based on the second zone discharge state data and the upper and lower layer load data of the third temperature zone, and control the kelp drying unit that has completed dehumidification in the second temperature zone to enter the third temperature zone;

[0084] Set the target temperature for heating and drying in the third temperature zone and the target air velocity value in the main air duct of the powerful dehumidifying fan according to the reference data of the three-zone drying operation in the automatic kelp production operation parameters to obtain the initial drying environment parameters for the three zones.

[0085] Carry out heating and drying operations on the kelp drying unit based on the initial drying environment parameters for the three zones, and continuously monitor the real-time weight change and surface moisture content distribution of the kelp drying unit in the third temperature zone to generate real-time drying state data for the three zones.

[0086] Use the temperature sensors and air velocity sensors in the drying temperature zone monitoring network to detect the actual temperature gradient and air velocity distribution around the kelp drying unit to generate local drying environment data for the three zones.

[0087] Conduct an assessment of drying non-uniformity based on the real-time drying state data for the three zones and the local drying environment data for the three zones to obtain the drying uniformity index.

[0088] When the drying uniformity index is lower than the preset threshold, conduct an analysis of the drying heat energy utilization gradient based on the local drying environment data for the three zones, and adjust the suspension height of the kelp drying unit to generate drying balance adjustment data for the three zones.

[0089] In the embodiments of the present invention, 4 infrared counters, model TR-52H, with an accuracy of 99.8% and a sampling frequency of 10 Hz, are installed at the upper and lower tunnel entrances respectively to count the current number of kelp drying units on each layer in real time. At the same time, 8 suspended weighing sensors, model LCM-103, with a range of 0-10 kg and an accuracy of 0.05%, are deployed in both the upper and lower tunnels to monitor the total weight load on each layer. The system uses a data fusion algorithm to integrate the unit quantity and the total weight to generate a standardized load index Li, with a range of 0-100. When the difference between the upper load index Lu and the lower load index Ld exceeds 15, the load balancing mechanism is triggered. The operation decision controller for the upper and lower tunnels receives the discharge status data of the second zone, analyzes the moisture content distribution and weight parameters of the kelp, and combines the real-time load conditions in the third temperature zone to execute the optimal distribution strategy. The decision rule is set as follows: when |Lu - Ld| > 25, 100% of the new batch of kelp is allocated to the low-load tunnel; when 15 < |Lu - Ld| ≤ 25, it is allocated in the ratio of 70% to the low-load tunnel and 30% to the high-load tunnel; when |Lu - Ld| ≤ 15, intelligent diversion is carried out according to the principle of allocating 55% of the heavily weighted kelp to the upper tunnel (more favorable for heat rising characteristics) and 45% of the lightly weighted kelp to the lower tunnel. The system realizes the precise guidance of the kelp drying units through a diversion mechanism composed of 25 pneumatic steering baffles, and the unit transfer speed is controlled at 0.25 m / s. The target temperature parameter for heating and drying is read, set at 87°C for the upper tunnel and 85°C for the lower tunnel, and the temperature difference setting is used to offset the natural temperature gradient caused by the rising hot air. The temperature control system consists of 8 groups of high-power electric heaters, with a single-group power of 18 kW, a temperature control accuracy of ±0.8°C, and a heating rate of 3°C / minute. The strong moisture exhaust fan system includes 4 centrifugal main fans, with a single-machine air volume of 15000 m³ / hour, a rotational speed range of 800-2400 revolutions per minute, and a wind speed adjustment accuracy of 0.1 m / s. The target wind speed value for the main air duct is set at 9.5 m / s according to the reference data, and a closed-loop control is formed by the real-time feedback of the wind speed sensor. The system also sets auxiliary parameters, including an upper and lower air duct balance of 95%, a humidity control target of 40%, and a fresh air introduction ratio of 15%. The environmental parameter actuators include 24 automatic regulating dampers, 16 groups of humidity regulating units, and 8 temperature regulating loops, all connected to the central controller through the on-site industrial bus network. The system fine-tunes the reference parameters according to the characteristic parameters of the current batch of kelp in the kelp characteristic database, generates an accurate set of initial drying environment parameters for the three zones, and sends them to each execution unit in real time through the optical fiber network to start the heating and drying process flow. Precise temperature ramp curve control is performed. In the first-stage heating stage, the temperature rises from 65°C to 75°C at a rate of 2°C / minute for 2 minutes; in the second-stage heating stage, it rises to 85°C at a rate of 1.5°C / minute for 5 minutes; in the third-stage heating stage, it slowly rises to 87°C and remains stable. The whole process is monitored by a temperature monitoring network composed of 32 K-type thermocouple temperature sensors, with a temperature measurement range of 0-1000°C and an accuracy of ±0.5°C.Meanwhile, the system deploys 12 high-precision weighing units inside the drying tunnel, with a measuring range of 0 - 5 kg, an accuracy of 0.02%, a sampling rate of 100 Hz, and records the mass change of the kelp drying unit in real time. Near-infrared spectroscopy moisture analyzers (wavelength range 900 - 1700 nm, resolution 2 nm) are installed one every 1.5 meters along the drying tunnel, with a total of 16 units covering the entire drying area, a scanning width of 1 meter, and continuously monitor the moisture content distribution on the surface of the kelp. The system uses a data fusion algorithm to integrate the mass change rate, moisture content gradient, and temperature response curve to generate a three-dimensional state matrix. The data processing unit uses the fourth-order Runge-Kutta method to predict the trend of moisture content change and calculates the drying efficiency index in combination with the mass change rate. The system updates the data every 5 seconds and integrates key parameters such as the mass change curve, moisture content distribution map, drying efficiency index, and predicted remaining drying time into a three-zone real-time drying status data packet. 56 PT100 platinum resistance temperature sensors are deployed inside the tunnel space, with a temperature measurement range of 0 - 150 °C, an accuracy of ±0.1 °C, a response time of less than 5 seconds, and a three-dimensional layout forms an 8×7 sensor matrix, with a probe spacing of 0.8 meters in the horizontal direction and 0.4 meters in the vertical direction. The wind speed monitoring system consists of 32 hot-wire anemometers, with a measurement range of 0 - 15 m / s, an accuracy of ±0.2 m / s, and a response time of 0.2 seconds, covering all key positions in the main air duct and auxiliary air duct. All sensors are connected to the data collector through the RS485 bus, with a sampling period of 1 second and a transmission rate of 19,200 bps. The system uses the cubic spline interpolation algorithm to reconstruct the discrete measurement point data into a continuous three-dimensional temperature field and wind speed field, with a spatial resolution of 0.1 m×0.1 m×0.1 m. The data processing unit calculates the temperature gradient vector field and wind speed divergence field to identify temperature non-uniform areas and air flow dead corners. The system pays special attention to the microenvironment parameters within a range of 200 mm around the kelp drying unit and calculates the distribution of the drying driving force index (the product of the temperature gradient and wind speed). The monitoring data is updated every 3 seconds to generate local drying environment data for the three zones, including the temperature gradient distribution map, wind speed vector map, air flow vortex identification result, and drying driving force distribution map. Multidimensional analysis is performed based on the three-zone real-time drying status data and the three-zone local drying environment data. The system first divides the surface of the kelp into a 6×5 grid, with a total of 30 evaluation units, and calculates three key indicators: moisture content, drying rate, and temperature response for each unit. The calculation of moisture content uniformity uses the standard deviation normalization method, dividing the standard deviation of the moisture content in 30 regions by the average moisture content and then multiplying by 100 to obtain a percentage index. The drying rate uniformity is calculated using a similar method to calculate the coefficient of variation of the drying rate in each region. The temperature response uniformity is calculated from the surface temperature distribution map collected by a thermal imager, with the thermal imager model being FLIR T640, a resolution of 640×480 pixels, a temperature resolution of 0.03 °C, and an imaging frequency of 30 Hz.The system integrates three indicators using a weighted fusion algorithm, with the weight distribution as follows: moisture content uniformity 40%, drying rate uniformity 35%, and temperature response uniformity 25%. The comprehensive drying uniformity index DUI is calculated, with a value range of 0 - 100. The higher the value, the more uniform the drying. The preset threshold is determined through statistical analysis of historical production data. The standard process threshold is set at 75, and the strict process threshold is set at 85. The system calculates the DUI value every 10 seconds and maintains a uniformity trend graph within a 10 - minute sliding window. When the DUI is lower than the preset threshold for three consecutive times, an uneven drying alarm is triggered, and the drying equilibrium adjustment process is started. A three - dimensional computational fluid dynamics analysis is performed on the local drying environment data of the three zones to draw a complete heat energy flow map inside the drying tunnel, identifying high - and low - heat - energy - density areas. The heat energy density calculation comprehensively considers three factors: local temperature, wind speed, and humidity, and is converted into an effective drying energy density value through a heat and mass transfer model. The system calculates the heat energy gradient distribution within the height range of 0 - 2 meters at 5 - centimeter intervals in the vertical direction, generating a vertical - direction heat energy density curve. At the same time, the hanging - height adjustment execution system consists of 16 groups of chain lifting devices driven by precision stepper motors. The single - group load capacity is 50 kg, the adjustment accuracy is 1 mm, the adjustment range is 0 - 500 mm, and the adjustment speed is 5 mm / s. The system calculates the optimal hanging height based on the heat energy gradient analysis results and the current drying state of the kelp, and precisely adjusts the height of the kelp units with uneven drying. For kelp with excessive drying at the upper part, the system moves it down 80 - 120 mm; for kelp with lagging drying at the bottom, the system moves it up 100 - 150 mm; for kelp with uneven drying on the left and right sides, the system adjusts it horizontally by changing the hanging angle. All adjustment parameters are integrated into a three - zone drying equilibrium adjustment data packet, including the adjustment target height, adjustment rate, and adjustment sequence, and are sent to the execution unit through the industrial control network to achieve precise drying equilibrium control.

[0090] Preferably, the cooling and shaping production optimization in the multi - temperature - zone kelp drying production optimization process in the intelligent intervention adjustment module includes:

[0091] When the drying of the kelp drying unit in the third temperature zone meets the preset conditions, the hanging conveyor belt is used to control the kelp drying unit to enter the fourth temperature zone at the corresponding level;

[0092] Based on the four - zone shaping operation reference data, cooling and shaping operation control is performed on the fourth temperature zone. The surface temperature of the kelp and environmental parameters are monitored using the drying temperature zone monitoring network until the preset shaping requirements are met, and the conveyor belt target moving speed in the four - zone shaping operation reference data is processed to control the finished kelp drying unit to pass through the drying discharge station, completing the kelp drying production process;

[0093] Use a load cell and a near-infrared moisture content sensor to evaluate the final moisture content and shaping quality of the kelp drying unit in the fourth temperature zone, and conduct grading of the finished kelp drying products to generate grading data for the finished kelp drying products.

[0094] In the embodiment of the present invention, when the drying of the kelp drying unit in the third temperature zone meets the preset conditions (the moisture content reaches 22±2%, the surface temperature is 60-65°C, and the moisture content uniformity index ≥0.9), the system triggers the transfer operation of the kelp drying unit. The preset conditions are obtained by real-time monitoring of the NIR-100 type near-infrared moisture sensor and the PT100 temperature sensor, and the sampling frequency is 2 minutes / time. When the sampling results of three consecutive times all meet the conditions, the system determines that the transfer standard is reached. The transfer instruction is transmitted to the Siemens S7-400 PLC control unit through the industrial fieldbus to drive the transfer device between the third temperature zone and the fourth temperature zone. The hanging conveyor belt adopts variable frequency speed control, and the initial speed is set to 1.2 m / min. When the hanging unit is close to 5 meters from the entrance of the fourth temperature zone, the conveyor belt speed automatically decreases to 0.8 m / min to ensure a smooth transition. The system identifies the identity of the kelp unit through the RFID reader (model JT-R340, reading distance 3 meters) suspended at the exit of the third temperature zone and the entrance of the fourth temperature zone, and records the accurate interval transfer time. During the transfer process, the kelp drying unit enters the fourth temperature zone through the partition door corresponding to the level (electric rolling shutter type, opening and closing speed 0.5 m / s). The whole process monitoring system tracks the position of the hanging unit in real time, and the deviation does not exceed ±10 mm. Based on the benchmark data of the four-zone shaping operation, the system precisely controls the environmental parameters of the fourth temperature zone, and realizes precise control through the FCU fan coil unit and the frequency converter-driven fan (refrigerating capacity 15 kW). The relative humidity is controlled at 35±3%, and is adjusted by the combination of the ultrasonic humidifier and the dehumidifier. The drying temperature zone monitoring network consists of 16 temperature sensors and 8 humidity sensors to form a monitoring array. The sampling period is 30 seconds, and the monitoring data is transmitted to the central control system through the data acquisition module. The system continuously monitors the surface temperature of the kelp. When the surface temperature drops to 30±2°C, and the temperature uniformity index ≥0.95, and the holding time reaches the specified value in the benchmark data (10 minutes), it is determined that the preset shaping requirement is met. Subsequently, the system controls the operation of the conveyor line according to the target moving speed of the conveyor belt in the four-zone shaping operation benchmark data. When the hanging unit passes through the drying discharge station, the counter is triggered by the photoelectric switch (diffuse reflection type, detection distance 5 meters), and at the same time the infrared thermometer detects the final surface temperature to ensure that it does not exceed 32°C. The automatic unloading mechanism (rotary type, 360° turntable structure) removes the finished kelp from the hanging unit and places it on the discharge conveyor belt to complete the drying production process. The DS-1 type high-precision weighing sensor is used to measure the final weight of the finished kelp, with an accuracy of ±1 g. The near-infrared moisture sensor scans the finished kelp at 9 points to measure the final moisture content. The system comprehensively evaluates the shaping quality of the finished kelp, and the evaluation indicators include: final moisture content (target value 18±2%), color uniformity (collecting images through the RGB camera and calculating the standard deviation), surface hardness (elastic modulus, measured by the pressure sensor), and shape retention (comparing the matching degree of the original shape and the finished shape through the 3D scanner).The system classifies grades according to the comprehensive score, and the classification criteria are: special grade (95 - 100 points), first grade (85 - 94 points), second grade (75 - 84 points), third grade (60 - 74 points), and out-of-grade (<60 points). The grade scoring adopts the weighted calculation method: total score = 30% × moisture content score + 25% × color score + 25% × hardness score + 20% × shape score, and each score is calculated using the normal distribution model.

[0095] Preferably, the intelligent intervention and adjustment module further includes a heat compensation function for waste heat exchange in the temperature zone, specifically:

[0096] Conduct a real-time heat demand assessment on the first temperature zone to generate the current heat load data of the first zone;

[0097] Measure the recoverable waste heat in the humid and hot air discharged from the third temperature zone to obtain the recoverable waste heat value of the third zone;

[0098] Calculate the heat energy utilization efficiency of the temperature zone based on the recoverable waste heat value of the third zone;

[0099] When the heat energy utilization efficiency of the temperature zone is lower than the preset heat energy utilization threshold, activate the waste heat recovery channel to collect the waste heat from the exhaust gas discharged from the third temperature zone and generate waste heat recovery amount data;

[0100] Calculate the waste heat supplement ratio based on the waste heat recovery amount data and the current heat load data of the first zone, and perform waste heat exchange heat compensation treatment on the first temperature zone to generate a waste heat recycling strategy.

[0101] In the embodiments of the present invention, 16 high-precision temperature sensors (model PT100, accuracy ±0.1°C) are evenly arranged in the preheating zone space to construct a complete temperature field. The system also deploys 8 heat flux density sensors (model HFP01, measurement range -2000 to +2000 W / m², accuracy ±3%) to directly measure the heat transfer rate. The sensor data is transmitted to the heat load calculation unit in real time via a data acquisition module (sampling frequency 10 Hz, 24-bit analog-to-digital conversion accuracy). The heat load evaluation algorithm calculates the current required heat power based on five key parameters: the real-time temperature field distribution, the current load in the preheating zone (the quantity and quality data of kelp provided by the feeding detection system), the difference between the preset target temperature (57°C) and the current actual temperature, the time required for temperature rise, and the system heat loss (measured by the wall heat flux density sensor). The system uses the sliding window method to smooth the heat load data for the past 5 minutes and combines it with the kelp feeding prediction model to generate a heat load prediction curve for the next 30 minutes. The heat load data is in kilowatts and is divided into the basic heat load (to maintain the system temperature) and the dynamic heat load (for heating newly fed kelp). Four high-temperature thermocouples (model K, temperature measurement range 0 - 1000°C, accuracy ±1.0°C) and four humidity sensors (model HX85A, measurement range 0 - 100%RH, high-temperature applicable type, accuracy ±2%RH) are installed at the inlet of the main exhaust duct, evenly distributed across the duct cross-section, to collect the temperature and humidity parameters of the exhaust gas in real time. The gas flow rate is measured using a Pitot tube flowmeter (range 0 - 20 m / s, accuracy ±1.5%), installed at the center of the exhaust duct to continuously record the exhaust gas velocity. The cross-sectional area of the duct is 0.5 square meters, and the system calculates the volume flow rate by multiplying the velocity by the cross-sectional area. The residual heat calculation unit comprehensively considers three key parameters: the exhaust gas temperature, humidity, and flow rate, and determines the sensible heat and latent heat carried by unit mass of air based on the enthalpy calculation principle. Considering the heat exchange efficiency limit in the actual recovery process, the system sets the recovery coefficient to 0.7, indicating that 70% of the theoretically recoverable total heat is the actual recoverable heat. The residual heat calculation results are in kilowatts and are updated every 15 seconds, and a complete data packet containing the temperature distribution, humidity distribution, flow rate value, and recoverable heat value is transmitted to the central control system to form a dataset of recoverable waste heat values for three zones. A complete energy input model is established, including the power of the electric heater (directly measured by a power transmitter, accuracy ±0.5%), the heat output of the gas heater (calculated through the gas flow meter and combustion efficiency), and the power consumption of the circulation fan (obtained from the power feedback of the frequency converter). The energy output model includes the effective thermal energy (the heat used for kelp drying, calculated based on the change in kelp moisture content and mass), the exhaust heat loss (through the measured recoverable waste heat values for three zones), and other heat losses (calculated from the temperature of the system outer wall sensor and the ambient temperature). The formula for calculating the thermal energy utilization efficiency of the temperature zone is the ratio of the effective thermal energy to the total input energy, and the result is expressed as a percentage.The system is set with a historical data comparison function that compares the current efficiency with the average efficiency of the past 7 days to identify abnormal fluctuations. The preset thermal energy utilization threshold is determined through big data analysis, set at 65% under standard operating conditions, 62% under high-load conditions, and 70% under low-load conditions. The efficiency calculation is performed once per minute, with the result accurate to one decimal place, and a complete data report containing real-time efficiency values, historical comparison curves, and threshold differences is generated. When the thermal energy utilization efficiency in the temperature zone is detected to be lower than the preset thermal energy utilization threshold three times in a row, the waste heat recovery mechanism is automatically triggered. The activation program first starts the electric air valve actuator (model Honeywell ML7421, torque 20 Nm, running time 90 seconds), turning the three-way air valve in the main exhaust duct from the direct exhaust position to the recovery position, with a turning angle of 90 degrees. At the same time, the system starts the waste heat recovery fan (frequency conversion centrifugal, power 5.5 kW, air volume 12,000 cubic meters per hour) to create the necessary air pressure difference. The waste heat recovery channel is composed of stainless steel pipes, with an aluminum finned tube heat exchanger installed inside, a heat transfer area of 32 square meters, and a designed heat transfer efficiency of 85%. The recovery system uses a dual-loop design. The primary loop uses an ethylene glycol aqueous solution as the heat transfer medium (concentration 30%, flow rate 15 cubic meters per hour), and is driven to circulate by a pipeline circulation pump (power 2.2 kW, head 20 meters). The waste heat recovery monitoring system includes 4 flow meters (accuracy ±0.5%) and 8 pairs of temperature sensors (one installed at each of the inlet and outlet, accuracy ±0.1°C), and calculates the actual recovered heat by measuring the fluid flow rate and the temperature difference between the inlet and outlet. The system also deploys a condensate collection device to calculate the latent heat recovery amount by measuring the condensate flow rate (electromagnetic flow meter, accuracy ±0.2%). All data is integrated into a waste heat recovery amount data packet, including four core indicators: total recovered heat, sensible heat recovery amount, latent heat recovery amount, and recovery efficiency. The data update frequency is 10 seconds. Calculate the waste heat supplement ratio η, which is the ratio of the waste heat recovery amount to the current heat load in Zone 1, and is usually controlled within the range of 10% - 40%. According to the η value, the system adopts different compensation strategies: when η < 15%, the full heat supplement mode is adopted, and all recovered heat is directly used for heating in Zone 1; when 15% ≤ η < 25%, the mixed supplement mode is adopted, 85% of the recovered heat is used for Zone 1, and 15% is used for preheating fresh air; when η ≥ 25%, the optimized distribution mode is adopted, 70% is used for Zone 1, 20% is used for fresh air preheating, and 10% is used for floor radiant heating. The heat transfer system consists of a plate heat exchanger (heat transfer area 18 square meters, designed pressure drop < 0.05 MPa) and a three-way control valve (accuracy ±2%, response time < 8 seconds) to achieve precise heat distribution. The heat supplement system in Zone 1 includes 4 groups of hot water coils (total heat transfer area 25 square meters) and 6 hot air mixers to ensure uniform heat distribution. The control system adopts a feedforward-feedback combined control strategy, and dynamically adjusts the heat distribution ratio by real-time monitoring of the temperature field changes in Zone 1 through 12 temperature sensors.The system automatically generates a 24-hour waste heat recycling strategy according to the kelp drying production plan, including a predicted recovered heat curve, a distribution plan, and an energy-saving benefit assessment. The whole process is optimized and controlled through the central control unit, and the waste heat utilization status is displayed in real time on the control interface, and the strategy data is updated every 5 minutes.

[0102] The beneficial effects of this application are as follows: By means of intelligent perception, the characteristics of kelp are accurately identified and analyzed, and personalized processing is carried out according to the actual conditions of different batches of kelp. According to the multi-temperature zone, precise control is carried out in stages. According to the moisture migration law during the kelp drying process, the most suitable temperature and humidity environment is provided for different drying stages, greatly improving the rehydration and taste of the product, shortening the drying cycle at the same time, and reducing energy consumption. Especially in the four key stages of preheating and sweating, balancing dehumidification, heating and drying, and cooling and shaping, intelligent intervention and adjustment are carried out to ensure the balanced evaporation of moisture inside and outside the kelp, prevent the phenomenon of surface film formation, and form tiny channels conducive to moisture discharge, making the texture of the product more uniform. The implementation of the waste heat recycling strategy significantly reduces the system energy consumption. Through intelligent heat energy allocation, precise compensation of waste heat from high-temperature areas to low-temperature areas is achieved, and the energy utilization efficiency is increased by about 30%, achieving the purpose of energy conservation and environmental protection. The double-line hanging multi-temperature zone double-layer tunnel structure design improves the production capacity by more than 50% under the same floor area, realizing the efficient utilization of space resources. The entire kelp drying system uses multi-temperature zones to control the temperature in stages, with adjustable speed, automatic control, energy conservation and environmental protection, safe operation process, and high degree of continuity, automation, and intelligence. The quality of the dried kelp is uniform, pure, and hygienic.

[0103] Therefore, in any regard, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to embrace all changes that fall within the meaning and scope of the equivalent elements of the application documents within the present invention.

[0104] The above are only the specific implementation manners of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An all - automated management system for the kelp drying production process based on artificial intelligence, characterized in that, The fully automated management system for the kelp drying production process based on artificial intelligence adopts a double - line hanging multi - temperature - zone double - layer tunnel - type drying structure, including the following modules: The kelp characteristic perception module is used to analyze the kelp monitoring characteristics of single - piece / single - cluster fresh kelp units, obtain the target kelp monitoring characteristic data; and generate the queue data of kelp to be dried according to the target kelp monitoring characteristic data for hanging queue layout. The drying parameter planning module is used to perform automated cleaning - air - blowing pre - treatment production operations on single - piece / single - cluster fresh kelp units through the queue data of kelp to be dried, and set the production operation parameters of multiple temperature zones according to the target kelp monitoring characteristic data to obtain the automated production operation parameters of kelp. The intelligent intervention and regulation module is used to optimize the multi - temperature - zone kelp drying production based on the automated production operation parameters of kelp, and classify the grades of the dried kelp products to generate the classification data of the dried kelp products; among them, the optimization of multi - temperature - zone kelp drying production includes pre - heating sweating production optimization, balance dehumidification production optimization, heating and drying production optimization, and cooling and shaping production optimization. The production parameter intelligent optimization module performs transfer learning based on the preset convolutional neural network model through the classification data of the dried kelp products and the automated production operation parameters of kelp, and gives automated production parameter feedback to obtain the intelligent production operation parameters. The production visualization management module is used to monitor the kelp drying process in real - time to achieve comprehensive and transparent management of the kelp drying production process. Among them, the double - line hanging multi - temperature - zone double - layer tunnel - type drying structure includes: Set two independently operating drying lines, each drying line is equipped with a hanging conveyor belt, and a plurality of hanging units are installed on the hanging conveyor belt to obtain a double - line hanging conveyor mechanism. Each hanging unit is provided with a weighing sensor and a near - infrared moisture content sensor, and a three - dimensional laser scanner is deployed at the entrance of the double - line hanging conveyor mechanism. According to the double - line hanging conveyor mechanism, it is set as a tunnel structure, including upper and lower layers, and the kelp carried by the hanging units can pass through both the upper layer and the lower layer for drying to obtain a double - layer reusable tunnel structure. Along the length direction of the double - layer reusable tunnel structure, its internal space is divided into at least four physically isolated temperature zones with multiple independent temperature zones to obtain multiple drying function temperature zones; among them, the division of multiple independent temperature zones includes a first temperature zone for kelp pre - heating sweating, a second temperature zone for balance dehumidification, a third temperature zone for heating and drying, and a fourth temperature zone for cooling and shaping. Independently adjustable temperature sensors, humidity sensors, and wind speed sensors are distributed on the inner walls and tops of the multiple drying function temperature zones to construct a drying temperature zone monitoring network.

2. The fully automated management system for the kelp drying production process based on artificial intelligence according to claim 1, wherein The kelp characteristic perception module is specifically: Use a three - dimensional laser scanner to perform non - contact contour scanning on single - piece / single - cluster fresh kelp units entering the automatic hanging station to obtain kelp three - dimensional morphology data. Perform three - dimensional surface reconstruction based on the kelp three - dimensional morphology data and estimate the macroscopic surface area of the kelp. Perform kelp monitoring characteristic analysis based on the macroscopic surface area of the kelp to obtain the target kelp monitoring characteristic data; among them, the target kelp monitoring characteristic data includes kelp wet weight data, target kelp average thickness data, and multi - point moisture content on the kelp surface. Evaluate the hanging distance value between the front and rear hanging units according to the macroscopic surface area of the kelp, and obtain the target kelp hanging distance value; Control the hanging unit to adjust the distance on the conveyor chain according to the target kelp hanging distance value, and obtain the real-time hanging queue layout data; Match the spraying pressure and spraying action time according to the target kelp average thickness data and the macroscopic surface area of the kelp, and obtain the spraying instruction in the cleaning area; Set the air knife drying parameters according to the target kelp monitoring characteristic data and the macroscopic surface area of the kelp, and obtain the air knife drying control instruction for the kelp; Associate the production instructions of the spraying instruction in the cleaning area and the air knife drying control instruction for the kelp through the real-time hanging queue layout data, and obtain the queue data of the kelp to be dried; 3. The fully automated management system for the kelp drying production process based on artificial intelligence according to claim 2, wherein, The kelp monitoring characteristic analysis based on the macroscopic surface area of the kelp includes: Weigh the single-piece / single-cluster fresh kelp unit using a weighing sensor to generate kelp wet weight data; Evaluate the average thickness of the kelp according to the kelp wet weight data and the macroscopic surface area of the kelp, and obtain the target kelp average thickness data; Use a near-infrared moisture content sensor to perform multi-point near-infrared spectral rapid scanning on the single-piece / single-cluster fresh kelp unit to generate kelp surface spectral data; Perform baseline drift correction and filtering processing on the kelp surface spectral data to generate corrected kelp surface spectral data; Identify the characteristic bands of the corrected kelp surface spectral data through a preset kelp spectral calibration information library, and evaluate the average initial moisture content on the kelp surface to obtain the multi-point moisture content on the kelp surface; 4. The fully automated management system for the whole process of kelp drying production based on artificial intelligence according to claim 3, characterized in that, The automated production operation parameters of the kelp include the reference data for the preheating operation in the first zone, the reference data for the dehumidification operation in the second zone, the reference data for the drying operation in the third zone, and the reference data for the shaping operation in the fourth zone. The drying parameter planning module specifically is: Perform an automated cleaning-air knife pre-treatment production operation on the single-piece / single-cluster fresh kelp unit through the queue data of the kelp to be dried to generate kelp pre-cleaning status data; Perform the target duration processing for preheating in the first zone according to the kelp pre-cleaning status data and the target kelp average thickness data in the target kelp monitoring characteristic data to generate the target duration data for preheating in the first zone; Set the rotational speed of the preheating circulation fan in the first zone according to the target duration data for preheating in the first zone, and integrate the operation control reference according to the target duration data for preheating in the first zone to obtain the reference data for the preheating operation in the first zone; Plan the dehumidification target temperature, the initial air valve target opening percentage of the exhaust fan, and the dehumidification processing duration in the second zone according to the multi-point moisture content on the kelp surface in the target kelp monitoring characteristic data to obtain the reference data for the dehumidification operation in the second zone; Set the target temperature for heating and drying, the target wind speed value of the main air duct of the strong exhaust fan, and the heating and drying processing duration according to the kelp wet weight data in the target kelp monitoring characteristic data to obtain the reference data for the drying operation in the third zone; Set the cooling temperature and the target wind speed of the cooling fan for air supply, and perform the processing of the target moving speed of the conveyor belt according to the target kelp monitoring characteristic data to obtain the reference data for the shaping operation in the fourth zone; 5. The fully automated management system for the whole process of kelp drying production based on artificial intelligence according to claim 4, wherein, The preheating sweating production optimization in the multi-temperature zone kelp drying production optimization process in the intelligent intervention adjustment module includes: Based on the data of the seaweed team to be dried, the seaweed drying unit is controlled to enter the first temperature zone of the double-line hanging multi-temperature zone double-layer tunnel drying structure through the hanging conveyor belt; The first temperature zone is preheated by the reference data of the preheating operation in the first zone of the automatic seaweed production operation parameters, and the surface temperature of the seaweed drying unit is continuously monitored by a temperature sensor to obtain the seaweed surface temperature distribution data; A humidity sensor is used to sense the local microenvironment humidity around the seaweed drying unit in real time to generate the local humidity on the seaweed surface; The average surface temperature of the seaweed and the temperature uniformity index are calculated based on the seaweed surface temperature distribution data; The current sweating state of the local humidity on the seaweed surface, the average surface temperature of the seaweed, and the temperature uniformity index are evaluated through the preset ideal sweating state parameters to generate the seaweed segmented sweating evaluation data.

6. The fully automated management system for the kelp drying production process based on artificial intelligence according to claim 5, characterized in that, The balance dehumidification production optimization in the multi-temperature zone seaweed drying production optimization process in the intelligent intervention and adjustment module includes: The sweating difference degree in the first zone is evaluated according to the seaweed segmented sweating evaluation data to generate the sweating difference degree data in the first zone; The angles of the micro air vents on the inner wall of the first temperature zone and the local radiation heating power are dynamically corrected through the sweating difference degree data in the first zone to generate the dynamic fine-tuning execution data in the first zone; Based on the dynamic fine-tuning execution data in the first zone, a judgment on the qualification of seaweed preheating is made. When the seaweed preheating is qualified, the seaweed drying unit that has completed the preheating in the first temperature zone is controlled to enter the second temperature zone, and the initial temperature, air valve opening, and fresh air supply volume of the second temperature zone are set according to the reference data of the dehumidification operation in the second zone of the automatic seaweed production operation parameters to obtain the initial environment setting data in the second zone; In the second temperature zone, a weighing sensor and a near-infrared moisture content sensor are used to continuously monitor the real-time weight and the change of the surface moisture content of the seaweed drying unit, and the water loss rate is calculated to generate the real-time water loss rate data in the second zone; The actual local humidity and the actual average wind speed around the seaweed drying unit are continuously monitored through the humidity sensor and the wind speed sensor in the drying temperature zone monitoring network to generate the real-time local humidity and wind speed data in the second zone; The abnormal dehumidification state is analyzed according to the real-time water loss rate data in the second zone and the real-time local humidity and wind speed data in the second zone to generate the abnormal dehumidification state data; Based on the abnormal dehumidification state data, the intelligent adjustment of the weak wind dehumidification mode is carried out on the initial environment setting data in the second zone, and the discharging state of the seaweed drying unit in the second zone is monitored to generate the discharging state data in the second zone.

7. The fully automated management system for the whole kelp drying production process based on artificial intelligence according to claim 6, characterized in that, The temperature-raising drying production optimization in the multi-temperature zone seaweed drying production optimization process in the intelligent intervention and adjustment module includes: Obtain the upper and lower layer load data of the third temperature zone; make decisions on the upper and lower layer tunnel operations according to the discharging state data in the second zone and the upper and lower layer load data of the third temperature zone, and control the seaweed drying unit that has completed the dehumidification in the second temperature zone to enter the third temperature zone; Set the target temperature for temperature-raising drying and the target wind speed value of the main air duct of the strong dehumidification fan in the third temperature zone according to the reference data of the drying operation in the third zone of the automatic seaweed production operation parameters to obtain the initial drying environment parameters in the third zone; The temperature rise and drying operation of the kelp drying unit are carried out based on the initial drying environment parameters of the three zones, and the real-time weight change and surface moisture content distribution of the kelp drying unit in the third temperature zone are continuously monitored to generate the real-time drying state data of the three zones; The actual temperature gradient and wind speed distribution around the kelp drying unit are detected by the temperature sensors and wind speed sensors in the drying temperature zone monitoring network to generate the local drying environment data of the three zones; The drying non-uniformity is evaluated according to the real-time drying state data of the three zones and the local drying environment data of the three zones to obtain the drying uniformity index; When the drying uniformity index is lower than the preset threshold, the drying heat energy utilization gradient analysis is carried out according to the local drying environment data of the three zones, and the suspension height of the kelp drying unit is adjusted to generate the drying balance adjustment data of the three zones.

8. The fully automated management system for the kelp drying production process based on artificial intelligence according to claim 7, characterized in that, The cooling and shaping production optimization in the multi-temperature zone kelp drying production optimization process in the intelligent intervention and adjustment module includes: When the drying of the kelp drying unit in the third temperature zone meets the preset conditions, the kelp drying unit is controlled to enter the fourth temperature zone of the corresponding level through the hanging conveyor belt; Based on the four-zone shaping operation reference data, the cooling and shaping operation control of the fourth temperature zone is carried out, and the surface temperature and environmental parameters of the kelp are monitored by the drying temperature zone monitoring network until the preset shaping requirements are met, and the finished kelp drying unit is controlled to pass through the drying and discharging station according to the target moving speed of the conveyor belt in the four-zone shaping operation reference data to complete the kelp drying production process; The terminal moisture content and shaping quality of the kelp drying unit in the fourth temperature zone are evaluated by using the weighing sensor and the near-infrared moisture sensor, and the grading of the dried kelp products is carried out to generate the grading data of the dried kelp products.

9. The fully automated management system for the kelp drying production process based on artificial intelligence according to claim 1, characterized in that The intelligent intervention and adjustment module also includes the temperature zone waste heat exchange and heat compensation function, specifically: The real-time heat demand of the first temperature zone is evaluated to generate the current heat load data of the first zone; Measure the recoverable waste heat in the humid hot air discharged from the third temperature zone to obtain the recoverable waste heat value of the three zones; Calculate the temperature zone heat energy utilization efficiency according to the recoverable waste heat value of the three zones; When the temperature zone heat energy utilization efficiency is lower than the preset heat energy utilization threshold, the waste heat recovery channel is activated to collect the waste heat from the exhaust gas discharged from the third temperature zone to generate the waste heat recovery amount data; Calculate the waste heat supplement ratio according to the waste heat recovery amount data and the current heat load data of the first zone, and carry out the waste heat exchange and heat compensation treatment on the first temperature zone to generate the waste heat recycling strategy.

Images