Kelp drying moisture dynamic balance control method and system based on multi-source perception

Abstract

The present invention discloses a method and system for controlling the dynamic balance of moisture in kelp drying based on multi-source sensing, comprising: real-time data collection through multi-source sensors inside the drying equipment; real-time identification of the drying stage based on the acquired temperature and humidity time series data, and dynamic adjustment of process parameters according to data differences; spatiotemporal calibration and compensation of kelp surface temperature distribution data and kelp surface image data; construction of a moisture distribution prediction model to obtain kelp moisture distribution gradient data; identification of local areas with moisture content higher than the global average as high-moisture areas, and targeted intensive drying of the high-moisture areas by controlling the zoned heating units; acquisition of moisture distribution gradient data, and completion of drying if the moisture gradient difference of the entire area is less than a threshold and the global average moisture content is less than a threshold. The advantages of the present invention are: real-time monitoring and dynamic adjustment of the drying process by sensors, precise control of kelp moisture distribution, improved drying efficiency and quality, and ensured uniform drying of the kelp.

Description

Technical Field

[0001] The present invention relates to a moisture control technology, and in particular to a method and system for controlling the dynamic balance of moisture in kelp drying based on multi-source perception. Background Art

[0002] Kelp is a nutrient-rich seaweed rich in iodine, calcium, iron, dietary fiber, and various vitamins. Due to its excellent health benefits and edible value, it is widely used in food processing, health supplements, and pharmaceuticals. However, moisture content plays a crucial role in the storage and processing of kelp. Kelp has a high moisture content after harvesting. If not handled promptly, it is prone to rotting and deterioration, affecting its nutritional content and commercial value.

[0003] Current methods and systems for dynamically controlling the moisture balance of kelp drying on the market typically rely on a single temperature and humidity sensor and fixed drying process parameters, lacking real-time monitoring and adjustment of the kelp's surface temperature distribution and morphological changes. These traditional methods struggle to address the uneven moisture distribution of the kelp caused by factors such as deformation and curling during the drying process, and are prone to localized overdrying or overwetting, affecting product quality and production efficiency. Furthermore, due to the lack of spatiotemporal calibration and compensation technology, these methods have a delayed response to temperature and humidity changes and may be unable to adjust process parameters in a timely manner during the different drying stages, resulting in unstable drying results and even energy waste. Compared with dynamic control methods based on multi-source sensing, traditional methods do not fully utilize historical data and intelligent algorithms to predict moisture distribution and accurately identify high-moisture areas, resulting in lower drying accuracy. Furthermore, these traditional methods often overlook the refined control of targeted and intensified drying, failing to effectively optimize the drying process in each area, limiting improvements in drying efficiency. Summary of the Invention

[0004] In order to improve the existing methods and systems, a method and system for dynamic balance control of kelp drying moisture based on multi-source sensing is provided. This method uses multi-source sensors to monitor and dynamically adjust the drying process in real time, accurately control the moisture distribution of kelp, optimize the process at each stage, improve the drying efficiency and quality, ensure that the kelp is evenly dried and reduce energy waste.

[0005] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0006] The method for regulating the dynamic balance of moisture in kelp drying based on multi-source perception includes:

[0007] Multi-source sensors inside the drying equipment collect data in real time, including temperature and humidity time series data, kelp surface temperature distribution data, and kelp surface image data;

[0008] Based on the acquired temperature and humidity time series data, the drying stage is identified in real time, and the process parameters are dynamically adjusted according to the data difference between the real-time data and the standard data of each stage. The drying stage includes a heating stage, a constant speed drying and dehumidification stage, and a speed reduction and shaping stage.

[0009] Based on each drying stage, the kelp surface temperature distribution data and kelp surface image data are temporally and spatially calibrated and compensated;

[0010] Based on historical kelp drying data, a moisture distribution prediction model is constructed, and the calibrated and compensated data is input into the moisture distribution prediction model to generate kelp moisture distribution gradient data;

[0011] Based on the kelp moisture distribution gradient data, local areas with moisture content higher than the global average are identified as high-moisture areas, and targeted enhanced drying is carried out in the high-moisture areas by controlling the zoned heating units;

[0012] Based on the targeted enhanced drying of kelp, its moisture distribution gradient data is obtained. If the moisture gradient difference in the entire area is less than the threshold and the global average moisture content is less than the threshold, the drying is completed.

[0013] Preferably, the drying stage is identified in real time based on the acquired temperature and humidity time series data, and the process parameters are dynamically adjusted according to the data difference between the real-time data and the standard data of each stage. The drying stage includes a heating stage, a constant speed drying and dehumidification stage, and a speed reduction and shaping stage. Specifically, the drying stage includes:

[0014] Based on the acquired real-time temperature and humidity time series data, feature extraction is performed to obtain key change indicator data, including temperature change rate, humidity change rate, and temperature and humidity correlation;

[0015] Based on the key change index data, the drying stage is identified. The temperature in the heating stage continues to rise, the temperature change rate is 0 and the humidity change rate is negative in the constant speed drying and dehumidification stage, and the absolute value of the humidity change rate decreases in the deceleration stage.

[0016] Based on the identified drying stage, the standard temperature and humidity target curve of this stage is obtained according to historical data, and the actual temperature and humidity curve is obtained based on the real-time collected temperature and humidity time series data;

[0017] Based on the standard temperature and humidity target curve and the actual temperature and humidity curve, the deviation values ​​are calculated, including temperature deviation, humidity deviation and change rate deviation;

[0018] Based on the obtained deviation value, a targeted adjustment strategy is implemented to intervene in the drying stage data by adjusting the heating power and the opening of the dehumidification valve.

[0019] Preferably, the spatiotemporal calibration and compensation of the kelp surface temperature distribution data and the kelp surface image data based on each drying stage specifically includes:

[0020] Based on the slowly changing images during the warming phase, the initial positions and rough outline templates of the kelp individuals were established, and the initial kelp surface temperature distribution data were recorded.

[0021] Based on the constant-rate drying and dehumidification stage, the pixel-level displacement field of the kelp surface between adjacent visible light images is calculated using the dense optical flow method. The local deformation field of the kelp surface from the end of the heating stage to the current frame is calculated based on the optical flow field.

[0022] The local deformation field is superimposed on the kelp surface image data that has been spatially calibrated by projective transformation using a basic spatial mapping model to calibrate the actual shape and position of the kelp after deformation;

[0023] Based on the slight deformation of the local curling area of ​​kelp during the speed reduction and shaping stage, the temperature and image features of the curling edge are accurately matched.

[0024] Preferably, the step of constructing a moisture distribution prediction model based on historical kelp drying data, inputting the calibrated and compensated data into the moisture distribution prediction model, and generating kelp moisture distribution gradient data specifically includes:

[0025] Based on historical kelp drying data, we extracted feature data of humidity time series data, kelp surface temperature distribution data, and kelp surface image data, and built a moisture distribution prediction model through training using an LSTM neural network model.

[0026] Based on the trained water distribution prediction model, the temporally and spatially calibrated and compensated data are input into the model to predict the water distribution data on the kelp surface.

[0027] Based on the predicted water distribution data on the kelp surface, the water content gradient is calculated by spatial difference to obtain the kelp water distribution gradient data.

[0028] Preferably, the method of identifying local areas with moisture content higher than the global average as high-moisture areas based on the kelp moisture distribution gradient data, and performing targeted enhanced drying on the high-moisture areas by controlling the zoned heating units specifically includes:

[0029] Based on the acquired kelp water distribution gradient data, the global average water content is calculated;

[0030] By comparing the difference between the moisture content of each location of the kelp and the global average moisture content, the part exceeding the threshold is marked as a high moisture area;

[0031] By adjusting the zoned heating units in high-moisture areas, the heating intensity can be increased, and targeted and enhanced drying can be performed on high-moisture areas.

[0032] Preferably, the step of obtaining the moisture distribution gradient data of the kelp after targeted intensified drying, and completing the drying if the moisture gradient difference of the entire region is less than a threshold value and the global average moisture content is less than a threshold value, specifically includes:

[0033] Based on the targeted and enhanced drying of the kelp, the sensor collects data again and inputs it into the moisture distribution prediction model to obtain the kelp moisture distribution gradient data;

[0034] The global average moisture content is calculated based on the kelp moisture distribution gradient data. If the moisture gradient difference in the entire area is less than the threshold and the global average moisture content is less than the threshold, the drying is completed.

[0035] Furthermore, a kelp drying moisture dynamic balance control system based on multi-source perception is proposed, including:

[0036] Data acquisition module: The data acquisition module is used to collect temperature and humidity time series data, kelp surface temperature distribution data and kelp surface image data in real time during the drying process, providing a basis for subsequent data analysis;

[0037] Drying stage identification module: The drying stage identification module identifies the current drying stage through the real-time acquired temperature and humidity time series data, and dynamically adjusts the process parameters according to the stage differences;

[0038] Spatiotemporal calibration and compensation module: The spatiotemporal calibration and compensation module performs spatiotemporal calibration and compensation on the kelp surface data based on the temperature distribution and image data of each drying stage to ensure accuracy;

[0039] Moisture distribution prediction module: The moisture distribution prediction module uses historical drying data to build a moisture distribution prediction model, and generates kelp moisture distribution gradient data by inputting compensated data;

[0040] High-moisture area identification module: The high-moisture area identification module identifies local high-moisture areas based on moisture distribution gradient data and performs targeted enhanced drying by adjusting the zoned heating units;

[0041] Drying determination module: The drying determination module determines the moisture gradient difference of the entire area and the global average moisture content based on the moisture distribution data after enhanced drying, and completes drying if the standards are met;

[0042] Processor: The processor is used to process the calculation process of each formula and the construction calculation process of each model.

[0043] Compared with the prior art, the advantages of the present invention are:

[0044] By collecting temperature, humidity, temperature distribution, and surface image data in real time through multi-source sensors, and combining it with advanced spatiotemporal calibration and compensation technology, the system can accurately identify changes in different drying stages and dynamically adjust process parameters to ensure the optimal drying effect at each stage. Especially when dealing with high-moisture areas, the system can precisely analyze the moisture distribution gradient and strengthen heating in a targeted manner to avoid overdrying or localized over-wetting, thereby ensuring the overall quality of the kelp. In addition, the moisture distribution prediction model constructed based on historical data can further optimize moisture control, avoid waste, and improve production efficiency. Ultimately, by determining the moisture differences across the entire area in real time, it is ensured that the kelp reaches an ideal, uniformly dried state at the end of drying. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of the method proposed in the present invention;

[0046] Figure 2 This is a schematic diagram of the drying stage identification proposed by the present invention;

[0047] Figure 3 This is a schematic diagram of the time-space calibration compensation proposed by the present invention;

[0048] Figure 4 This is a schematic diagram of the moisture distribution prediction model proposed in the present invention;

[0049] Figure 5 This is a schematic diagram of drying proposed by the present invention;

[0050] Figure 6 This is a schematic diagram of the drying completion determination proposed by the present invention;

[0051] Figure 7 This is a diagram of the architecture of the electronic equipment in this solution;

[0052] Figure 8 This is a schematic diagram of the computer-readable storage medium structure in this solution. DETAILED DESCRIPTION

[0053] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0054] The kelp drying moisture dynamic balance control system based on multi-source perception includes:

[0055] Data acquisition module: The data acquisition module is used to collect temperature and humidity time series data, kelp surface temperature distribution data and kelp surface image data in real time during the drying process, providing a basis for subsequent data analysis;

[0056] Drying stage identification module: The drying stage identification module identifies the current drying stage through the real-time acquired temperature and humidity time series data, and dynamically adjusts the process parameters according to the stage differences;

[0057] Spatiotemporal calibration and compensation module: The spatiotemporal calibration and compensation module performs spatiotemporal calibration and compensation on the kelp surface data based on the temperature distribution and image data of each drying stage to ensure accuracy;

[0058] Moisture distribution prediction module: The moisture distribution prediction module uses historical drying data to build a moisture distribution prediction model, and generates kelp moisture distribution gradient data by inputting compensated data;

[0059] High-moisture area identification module: The high-moisture area identification module identifies local high-moisture areas based on moisture distribution gradient data and performs targeted enhanced drying by adjusting the zoned heating units;

[0060] Drying determination module: The drying determination module determines the moisture gradient difference of the entire area and the global average moisture content based on the moisture distribution data after enhanced drying, and completes drying if the standards are met;

[0061] Processor: The processor is used to process the calculation process of each formula and the construction calculation process of each model.

[0062] See Figure 1 As shown, the method for regulating the dynamic balance of moisture in kelp drying based on multi-source perception includes:

[0063] Step 1: Use multi-source sensors inside the drying equipment to collect data in real time, including temperature and humidity time series data, kelp surface temperature distribution data, and kelp surface image data;

[0064] Step 2: Based on the acquired temperature and humidity time series data, the drying stage is identified in real time, and the process parameters are dynamically adjusted according to the data difference between the real-time data and the standard data of each stage. The drying stage includes a heating stage, a constant speed drying and moisture removal stage, and a speed reduction and shaping stage.

[0065] Step 3: Based on each drying stage, perform spatiotemporal calibration and compensation on the kelp surface temperature distribution data and kelp surface image data;

[0066] Step 4: Based on the historical kelp drying data, a moisture distribution prediction model is constructed, and the calibrated and compensated data is input into the moisture distribution prediction model to generate the kelp moisture distribution gradient data;

[0067] Step 5: Based on the kelp moisture distribution gradient data, identify local areas with moisture content higher than the global average as high-moisture areas, and perform targeted enhanced drying of the high-moisture areas by controlling the zoned heating units;

[0068] Step 6: Based on the targeted enhanced drying of the kelp, its moisture distribution gradient data is obtained. If the moisture gradient difference in the entire area is less than the threshold and the global average moisture content is less than the threshold, the drying is completed.

[0069] See Figure 2 As shown, the drying stage is identified in real time based on the acquired temperature and humidity time series data, and the process parameters are dynamically adjusted according to the data difference between the real-time data and the standard data of each stage. The drying stage includes a heating stage, a constant speed drying and dehumidification stage, and a speed reduction and shaping stage. Specifically, it includes:

[0070] Based on the acquired real-time temperature and humidity time series data, feature extraction is performed to obtain key change indicator data, including temperature change rate, humidity change rate, and temperature and humidity correlation;

[0071] Based on the key change index data, the drying stage is identified. The temperature in the heating stage continues to rise, the temperature change rate is 0 and the humidity change rate is negative in the constant speed drying and dehumidification stage, and the absolute value of the humidity change rate decreases in the deceleration stage.

[0072] Based on the identified drying stage, the standard temperature and humidity target curve of this stage is obtained according to historical data, and the actual temperature and humidity curve is obtained based on the real-time collected temperature and humidity time series data;

[0073] Based on the standard temperature and humidity target curve and the actual temperature and humidity curve, the deviation values ​​are calculated, including temperature deviation, humidity deviation and change rate deviation;

[0074] Based on the obtained deviation value, a targeted adjustment strategy is implemented to intervene in the drying stage data by adjusting the heating power and the opening of the dehumidification valve.

[0075] Specifically, key features are extracted from the real-time collected temperature and humidity time series data, mainly including the temperature change rate, humidity change rate, and temperature-humidity correlation. The temperature-humidity correlation is measured by calculating the correlation coefficient between temperature and humidity. The Pearson correlation coefficient is usually used, and the formula is:

[0076]

[0077] in, is the Pearson correlation coefficient between temperature and humidity, 、 is the temperature and humidity at the i-th time point, 、 is the mean of temperature and humidity, n is the number of data points;

[0078] The main characteristic of the warming stage is that the temperature change rate continues to be significantly positive. The secondary characteristic is that the relative humidity usually rises rapidly or begins to fall. When the temperature is detected to rise continuously and rapidly, and the humidity change is in line with the expected trend, it is determined to have entered the warming stage.

[0079] The main characteristics of the constant-rate drying and dehumidification stage are that the temperature change rate is close to 0 (the temperature is basically stable around the set drying temperature with slight fluctuations), and the humidity change rate is significantly negative. The secondary characteristics are that the material surface temperature tends to be stable and slightly lower than the ambient temperature. When the temperature reaches the preset drying temperature range and tends to be stable, and the humidity shows a continuous and stable downward trend, it is determined to enter the constant-rate drying and dehumidification stage, which is the stage with the most intense water evaporation;

[0080] The main characteristics of the deceleration and finalization stage are that the absolute value of the humidity change rate decreases significantly (close to 0 or becomes a very small negative value or even a slightly positive value), indicating that the water evaporation rate is greatly reduced and the temperature change rate begins to become negative or remains stable. Secondary characteristics are that the surface temperature of the material may begin to approach or equal the ambient temperature. Image data shows that the edge of the kelp begins to curl and the color deepens. When the humidity drop rate is detected to have significantly slowed down, tended to be stable or slightly rebounded (because the evaporation amount is very small, the ambient humidity is difficult to continue to drop), and the temperature begins to drop according to the preset curve or remains stable, it is determined to have entered the deceleration and finalization stage.

[0081] Through historical data, the standard temperature and humidity target curve is calculated and fitted. The interpolation method or curve fitting method can be used to obtain the target temperature and humidity values ​​​​for each stage. The temperature and humidity data of the current drying stage are extracted from the real-time collected temperature and humidity time series data, and the actual temperature and humidity curve is constructed. The deviation between the standard temperature and humidity target curve and the actual temperature and humidity curve is calculated, including temperature deviation, humidity deviation and change rate deviation.

[0082] See Figure 3 As shown, based on each drying stage, the spatiotemporal calibration and compensation of the kelp surface temperature distribution data and the kelp surface image data specifically include:

[0083] Based on the slowly changing images during the warming phase, the initial positions and rough outline templates of the kelp individuals were established, and the initial kelp surface temperature distribution data were recorded.

[0084] Based on the constant-rate drying and dehumidification stage, the pixel-level displacement field of the kelp surface between adjacent visible light images is calculated using the dense optical flow method. The local deformation field of the kelp surface from the end of the heating stage to the current frame is calculated based on the optical flow field.

[0085] The local deformation field is superimposed on the kelp surface image data that has been spatially calibrated by projective transformation using a basic spatial mapping model to calibrate the actual shape and position of the kelp after deformation;

[0086] Based on the slight deformation of the local curling area of ​​kelp during the speed reduction and shaping stage, the temperature and image features of the curling edge are accurately matched.

[0087] Specifically, the outline and position of kelp individuals were captured through the image sequence of the warming stage to obtain the preliminary kelp outline;

[0088] The dense optical flow method estimates the movement direction and speed of each pixel by analyzing the changes in pixel brightness between adjacent frames. The formula is:

[0089]

[0090] in, is the spatial gradient of the image, is the displacement vector of the pixel point, is the derivative of image brightness over time;

[0091] The estimated displacement field can be used to calculate the local deformation field of the kelp surface from the end of the heating stage to the current frame. The image is projected and calibrated using a basic spatial mapping model to ensure that the actual shape and position of the kelp after deformation are consistent with the image data.

[0092] During the deceleration and shaping stage, the local curled area of ​​the kelp may undergo slight deformation, which requires focused compensation, especially the temperature and image characteristics of the curled edge. The deformed shape can be compensated by a small displacement field;

[0093] For areas with minor deformation, the temperature of the curled area is recorded using a temperature sensor or infrared imaging device, and accurately corresponds to the image features. The temperature of each location on the compensated image can be calculated using interpolation methods.

[0094] See Figure 4 As shown in the figure, based on the historical kelp drying data, a moisture distribution prediction model is constructed, and the calibrated and compensated data is input into the moisture distribution prediction model to generate the kelp moisture distribution gradient data. Specifically, the following steps are involved:

[0095] Based on historical kelp drying data, we extracted feature data of humidity time series data, kelp surface temperature distribution data, and kelp surface image data, and built a moisture distribution prediction model through training using an LSTM neural network model.

[0096] Based on the trained water distribution prediction model, the temporally and spatially calibrated and compensated data are input into the model to predict the water distribution data on the kelp surface.

[0097] Based on the predicted water distribution data on the kelp surface, the water content gradient is calculated by spatial difference to obtain the kelp water distribution gradient data.

[0098] Specifically, based on the extracted feature data: humidity time series data, temperature distribution data, and image data, an LSTM neural network is used for training to build a moisture distribution prediction model. The LSTM model learns the relationship between features such as humidity, temperature, and images and moisture distribution through training on time series data, and updates the model weights.

[0099] Based on the trained LSTM model, the spatiotemporally calibrated and compensated data is input to predict the moisture distribution data on the kelp surface. The gradient of the moisture content is calculated through spatial differentiation to obtain the changing trend of the moisture distribution on the kelp surface. The moisture distribution gradient formula is:

[0100]

[0101] in, For the kelp surface in position The moisture distribution gradient, The distribution of water in and Directional gradient.

[0102] See Figure 5 As shown in the figure, based on the kelp moisture distribution gradient data, local areas with moisture content higher than the global average are identified as high-moisture areas. The high-moisture areas are then subjected to targeted and enhanced drying by controlling the zoned heating units. The specific steps include:

[0103] Based on the acquired kelp water distribution gradient data, the global average water content is calculated;

[0104] By comparing the difference between the moisture content of each location of the kelp and the global average moisture content, the part exceeding the threshold is marked as a high moisture area;

[0105] By adjusting the zoned heating units in high-moisture areas, the heating intensity can be increased, and targeted and enhanced drying can be performed on high-moisture areas.

[0106] Specifically, based on the obtained kelp moisture distribution gradient data, the global average moisture content of the kelp surface water is first calculated using the formula:

[0107]

[0108] in, is the global average moisture content, is the total number of kelp surface locations, is the sum of the moisture contents at all locations;

[0109] By comparing the difference between the moisture content of each location and the global average moisture content, the area with large moisture deviation can be identified. The moisture difference can be expressed by the deviation of the moisture content of each location from the global average moisture content;

[0110] Once high-moisture areas are identified, it is necessary to enhance the drying effect in these areas by adjusting the heating units. Adjusting the heating units helps to specifically enhance the drying effect, especially in high-moisture areas. By increasing the heating intensity in these areas, the moisture can be evaporated quickly, reducing the moisture differences in these areas.

[0111] See Figure 6 As shown in the figure, based on the targeted and intensified drying of the kelp, its moisture distribution gradient data is obtained. If the moisture gradient difference of the entire area is less than the threshold and the global average moisture content is less than the threshold, the drying is completed. Specifically, the following steps are performed:

[0112] Based on the targeted and enhanced drying of the kelp, the sensor collects data again and inputs it into the moisture distribution prediction model to obtain the kelp moisture distribution gradient data;

[0113] The global average moisture content is calculated based on the kelp moisture distribution gradient data. If the moisture gradient difference in the entire area is less than the threshold and the global average moisture content is less than the threshold, the drying is completed.

[0114] Furthermore, the method according to the embodiment of the present application can also be used with the aid of Figure 7 The electronic device architecture shown in FIG. Figure 7 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store the method and system for regulating the dynamic balance of moisture in kelp drying based on multi-source perception provided in this application. The electronic device 500 may also include a terminal interface 508. Of course, Figure 7 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 7 One or more components of an electronic device are shown.

[0115] Figure 8 This is a schematic diagram of the computer-readable storage medium structure provided by an embodiment of the present application. Figure 8As shown, a computer-readable storage medium 600 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are executed by the processor, the method and system for regulating the dynamic balance of moisture in kelp drying based on multi-source perception according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0116] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0117] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0118] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for controlling the dynamic balance of moisture in kelp drying based on multi-source perception, characterized in that: include: Multi-source sensors inside the drying equipment collect data in real time, including temperature and humidity time series data, kelp surface temperature distribution data, and kelp surface image data; Based on the acquired temperature and humidity time series data, the drying stage is identified in real time, and the process parameters are dynamically adjusted according to the data difference between the real-time data and the standard data of each stage. The drying stage includes a heating stage, a constant speed drying and dehumidification stage, and a speed reduction and shaping stage. Based on each drying stage, the kelp surface temperature distribution data and kelp surface image data are temporally and spatially calibrated and compensated; Based on historical kelp drying data, a moisture distribution prediction model is constructed, and the calibrated and compensated data is input into the moisture distribution prediction model to generate kelp moisture distribution gradient data; Based on the kelp moisture distribution gradient data, local areas with moisture content higher than the global average are identified as high-moisture areas, and targeted enhanced drying is carried out in the high-moisture areas by controlling the zoned heating units; Based on the targeted and enhanced drying of the kelp, its moisture distribution gradient data is obtained. If the moisture gradient difference of the entire area is less than the threshold and the global average moisture content is less than the threshold, the drying is completed. The spatiotemporal calibration and compensation of the kelp surface temperature distribution data and the kelp surface image data based on each drying stage specifically includes: Based on the slowly changing images during the warming phase, the initial positions and rough outline templates of the kelp individuals were established, and the initial kelp surface temperature distribution data were recorded. Based on the constant-rate drying and dehumidification stage, the pixel-level displacement field of the kelp surface between adjacent visible light images is calculated using the dense optical flow method. The local deformation field of the kelp surface from the end of the heating stage to the current frame is calculated based on the optical flow field. The local deformation field is superimposed on the kelp surface image data that has been spatially calibrated by projective transformation using a basic spatial mapping model to calibrate the actual shape and position of the kelp after deformation; Based on the slight deformation of the local curling area of ​​kelp during the speed reduction and shaping stage, the temperature and image features of the curling edge are accurately matched.

2. The method for controlling the dynamic balance of moisture in kelp drying based on multi-source sensing according to claim 1, characterized in that: The drying stage is identified in real time based on the acquired temperature and humidity time series data, and the process parameters are dynamically adjusted according to the data difference between the real-time data and the standard data of each stage. The drying stage includes a heating stage, a constant speed drying and dehumidification stage, and a speed reduction and shaping stage. Specifically, Based on the acquired real-time temperature and humidity time series data, feature extraction is performed to obtain key change indicator data, including temperature change rate, humidity change rate, and temperature and humidity correlation; Based on the key change index data, the drying stage is identified. The temperature in the heating stage continues to rise, the temperature change rate is 0 and the humidity change rate is negative in the constant speed drying and dehumidification stage, and the absolute value of the humidity change rate decreases in the deceleration stage. Based on the identified drying stage, the standard temperature and humidity target curve of this stage is obtained according to historical data, and the actual temperature and humidity curve is obtained based on the real-time collected temperature and humidity time series data; Based on the standard temperature and humidity target curve and the actual temperature and humidity curve, the deviation values ​​are calculated, including temperature deviation, humidity deviation and change rate deviation; Based on the obtained deviation value, a targeted adjustment strategy is implemented to intervene in the drying stage data by adjusting the heating power and the opening of the dehumidification valve.

3. The method for controlling the dynamic balance of moisture in kelp drying based on multi-source sensing according to claim 1, characterized in that: The method of constructing a moisture distribution prediction model based on historical kelp drying data, inputting the calibrated and compensated data into the moisture distribution prediction model, and generating kelp moisture distribution gradient data specifically includes: Based on historical kelp drying data, we extracted feature data of humidity time series data, kelp surface temperature distribution data, and kelp surface image data, and built a moisture distribution prediction model through training using an LSTM neural network model. Based on the trained water distribution prediction model, the temporally and spatially calibrated and compensated data are input into the water distribution prediction model to predict the water distribution data on the kelp surface. Based on the predicted water distribution data on the kelp surface, the water content gradient is calculated by spatial difference to obtain the kelp water distribution gradient data.

4. The method for controlling the dynamic balance of moisture in kelp drying based on multi-source sensing according to claim 1, characterized in that: The method of identifying local areas with moisture content higher than the global average as high-moisture areas based on the kelp moisture distribution gradient data and performing targeted enhanced drying on the high-moisture areas by controlling the zoned heating units specifically includes: Based on the acquired kelp water distribution gradient data, the global average water content is calculated; By comparing the difference between the moisture content of each location of the kelp and the global average moisture content, the part exceeding the threshold is marked as a high moisture area; By adjusting the zoned heating units in high-moisture areas, the heating intensity can be increased, and targeted and enhanced drying can be performed on high-moisture areas.

5. The method for controlling the dynamic balance of moisture in kelp drying based on multi-source sensing according to claim 1, characterized in that: The method of obtaining the moisture distribution gradient data of the kelp after targeted intensified drying, and completing the drying specifically includes: if the moisture gradient difference of the entire region is less than a threshold value and the global average moisture content is less than a threshold value; and Based on the targeted and enhanced drying of kelp, data is collected again through multi-source sensors and input into the moisture distribution prediction model to obtain the kelp moisture distribution gradient data; The global average moisture content is calculated based on the kelp moisture distribution gradient data. If the moisture gradient difference in the entire area is less than the threshold and the global average moisture content is less than the threshold, the drying is completed.

6. A kelp drying moisture dynamic balance control system based on multi-source sensing, used to implement the kelp drying moisture dynamic balance control method based on multi-source sensing as described in any one of claims 1 to 5, characterized in that: include: Data acquisition module: The data acquisition module is used to collect temperature and humidity time series data, kelp surface temperature distribution data and kelp surface image data in real time during the drying process, providing a basis for subsequent data analysis; Drying stage identification module: The drying stage identification module identifies the current drying stage through the real-time acquired temperature and humidity time series data, and dynamically adjusts the process parameters according to the stage differences; Spatiotemporal calibration and compensation module: The spatiotemporal calibration and compensation module performs spatiotemporal calibration and compensation on the kelp surface data based on the temperature distribution and image data of each drying stage to ensure accuracy; Moisture distribution prediction module: The moisture distribution prediction module uses historical drying data to build a moisture distribution prediction model, and generates kelp moisture distribution gradient data by inputting calibrated and compensated data; High-moisture area identification module: The high-moisture area identification module identifies local high-moisture areas based on moisture distribution gradient data and performs targeted enhanced drying by adjusting the zoned heating units; Drying determination module: The drying determination module determines the moisture gradient difference of the entire area and the global average moisture content based on the moisture distribution data after enhanced drying, and completes drying if the standards are met; Processor: The processor is used to process the calculation process of each formula and the construction calculation process of each model.

7. An electronic device, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the kelp drying moisture dynamic balance control method based on multi-source perception as described in any one of claims 1-5.

8. A computer-readable storage medium storing computer-readable instructions, characterized in that: When the computer-readable instructions are executed by a processor, the method for controlling the dynamic balance of moisture in kelp drying based on multi-source perception according to any one of claims 1 to 5 is implemented.

Images