Dried beancurd stick drying method, device, equipment and medium applied to dried beancurd stick drying oven

Abstract

The invention relates to a dried beancurd stick drying method, device, equipment and medium applied to a dried beancurd stick drying oven, and the method comprises the following steps: obtaining real-time temperature and humidity data of a plurality of monitoring areas in the oven at the initial drying stage of dried beancurd sticks, and generating a first hot air circulation parameter adjustment instruction; and controlling the baking oven to execute a first hot air circulation mode, collecting real-time temperature fluctuation and humidity gradient data and generating an evaluation index. And if the evaluation index does not reach the threshold value, multi-stage parameter readjustment operation is triggered to generate a second hot air circulation parameter adjustment instruction. Controlling the baking box to switch to a second hot air circulation mode, monitoring the surface morphology change data of the dried beancurd sticks, and generating a drying termination instruction to end the drying process according to the matching degree with the standard curve, thereby improving the drying quality and efficiency of the dried beancurd sticks.

Description

Technical Field

[0001] The present disclosure relates to the technical field of yuba processing, and particularly to a yuba drying method, device, equipment and medium applied to a yuba drying and baking oven. Background Art

[0002] Yuba, also known as bean curd skin, is formed by heating and boiling soybean milk, keeping it warm for a period of time, forming a film on the surface, picking it out and hanging it down into a branch shape, and then drying it. Its shape is similar to that of bamboo branches, so it is called yuba. Yuba is a product with a certain structure composed of a soybean protein film and fat. Yuba processing is a complex process, and a baking oven is required during the process, and the baking oven can dry yuba. The dried yuba is not only convenient for storage and prevents spoilage, but also has a better taste after baking. However, in the prior art, it is difficult to accurately control the temperature of yuba drying. It is not only difficult to completely volatilize the bad smell, but also the internal and external moisture is consistent during the baking process, resulting in local mildew or cracking of yuba. Summary of the Invention

[0003] The object of the present invention is to provide a yuba drying method, device, equipment and medium applied to a yuba drying and baking oven, aiming to accurately control the temperature of yuba drying, so that the bad smell can be completely volatilized, and the internal and external moisture is kept consistent during the baking process, avoiding local mildew or cracking of yuba and affecting the taste of yuba.

[0004] To achieve the above object, in the first aspect of the embodiments of the present disclosure, a yuba drying method applied to a yuba drying and baking oven is provided. The method is applied to a yuba drying and baking oven, and the method includes:

[0005] Obtain the real-time temperature data and real-time humidity data of multiple monitoring areas in the baking oven during the initial drying stage of yuba, and dynamically generate a first hot air circulation parameter adjustment instruction based on the real-time temperature data and the real-time humidity data;

[0006] Based on the first hot air circulation parameter adjustment instruction, control the baking oven to execute the first hot air circulation mode, and continuously collect the real-time temperature fluctuation data and real-time humidity gradient data of the first optimization stage during the execution process. Generate a temperature stability evaluation index according to the real-time temperature fluctuation data, and generate a humidity balance evaluation index according to the real-time humidity gradient data;

[0007] When the temperature stability evaluation index does not reach the preset first stability threshold or the humidity balance evaluation index does not reach the preset first balance threshold, trigger a multi-stage parameter readjustment operation, and generate a second hot air circulation parameter adjustment instruction based on the first difference parameter between the temperature stability evaluation index and the first stability threshold and the second difference parameter between the humidity balance evaluation index and the first balance threshold;

[0008] According to the second hot air circulation parameter adjustment instruction, control the baking oven to switch to the second hot air circulation mode, and synchronously monitor the data of the morphological changes on the surface of the dried bean curd sticks during the execution process. Based on the morphological matching degree between the surface morphological change data and the preset standard morphological change curve, generate a drying termination instruction to end the drying process.

[0009] In a second aspect of the embodiments of the present disclosure, there is provided a dried bean curd stick drying device applied to a dried bean curd stick baking oven. The device includes:

[0010] An acquisition and generation module, configured to acquire the real-time temperature data and real-time humidity data of multiple monitoring areas in the baking oven during the initial drying stage of the dried bean curd sticks, and dynamically generate a first hot air circulation parameter adjustment instruction based on the real-time temperature data and the real-time humidity data;

[0011] A control and generation module, configured to control the baking oven to execute the first hot air circulation mode based on the first hot air circulation parameter adjustment instruction, and continuously collect the real-time temperature fluctuation data and real-time humidity gradient data of the first optimization stage during the execution process. Generate a temperature stability evaluation index according to the real-time temperature fluctuation data, and generate a humidity balance evaluation index according to the real-time humidity gradient data;

[0012] A trigger and generation module, configured to trigger a multi-stage parameter readjustment operation when the temperature stability evaluation index does not reach the preset first stability threshold or the humidity balance evaluation index does not reach the preset first balance threshold, and generate a second hot air circulation parameter adjustment instruction based on a first difference parameter between the temperature stability evaluation index and the first stability threshold and a second difference parameter between the humidity balance evaluation index and the first balance threshold;

[0013] An adjustment and control module, configured to control the baking oven to switch to the second hot air circulation mode according to the second hot air circulation parameter adjustment instruction, and synchronously monitor the data of the morphological changes on the surface of the dried bean curd sticks during the execution process. Based on the morphological matching degree between the surface morphological change data and the preset standard morphological change curve, generate a drying termination instruction to end the drying process.

[0014] In a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including:

[0015] A memory, on which a computer program is stored;

[0016] A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of the first aspect.

[0017] In a fourth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of the method described in any one of the first aspects are implemented.

[0018] The present invention provides a method, device, equipment and medium for drying yuba in a yuba drying oven. Compared with the prior art, the following beneficial effects are achieved:

[0019] By obtaining the real-time temperature and real-time humidity data of multiple monitoring areas in the oven during the initial drying stage, generating temperature distribution characteristics and humidity change rate characteristics, the drying environment conditions in the initial stage can be comprehensively understood. Based on the matching degree deviation from the preset reference, the hot air circulation parameter adjustment instruction is dynamically generated, realizing the precise control of the drying process and avoiding the drying quality problems caused by uneven environmental parameters. During the execution of the hot air circulation mode, the temperature fluctuation and humidity gradient data are continuously collected and evaluation indexes are generated, which can evaluate the stability and balance of drying in real time. When the evaluation index does not reach the threshold, a multi-stage parameter re-adjustment operation is triggered to further optimize the drying effect. Finally, according to the matching degree between the yuba surface morphology change data and the standard curve, the drying termination instruction is generated, ensuring the accuracy of the drying end, improving the drying quality and production efficiency of yuba, and significantly improving the intelligent and refined level of the yuba drying process as a whole.

[0020] Other features and advantages of the present disclosure will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings are used to provide a further understanding of the present disclosure, and constitute a part of the specification, and are used together with the following specific implementation to explain the present disclosure, but do not constitute a limitation to the present disclosure. In the drawings:

[0022] Figure 1 is a flowchart of a method for drying yuba in a yuba drying oven shown according to an embodiment of the specification.

[0023] Figure 2 is a block diagram of a device for drying yuba in a yuba drying oven shown according to an embodiment of the specification.

[0024] Figure 3 is a block diagram of a device for drying yuba in a yuba drying oven shown according to an embodiment of the specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all 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.

[0026] The following will detail the specific embodiments of the present disclosure in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only for the purpose of illustration and explanation, and are not intended to limit the present disclosure.

[0027] The present disclosure provides a method for drying bean curd sheets applied to a bean curd sheet drying and baking oven. The method is applied to a bean curd sheet drying and baking oven, where the bean curd sheet drying and baking oven includes a box body. A heating pipe is fixed to the upper inner wall of the box body, a circulation pipe is fixed to the left end of the box body, a main fan and an auxiliary fan are symmetrically fixed on the circulation pipe, a controller and an audible and visual alarm device are successively fixed to the front end of the box body from top to bottom. The output end of the controller is electrically connected to the input ends of the heating pipe, the main fan, the auxiliary fan, and the audible and visual alarm device, and the input end of the controller is electrically connected to the output end of a liquid level sensor. The controller is used to execute the method for drying bean curd sheets applied to the bean curd sheet drying and baking oven provided in the embodiments of the present disclosure. Figure 1 It is a flowchart of a method for drying bean curd sheets applied to a bean curd sheet drying and baking oven shown according to an embodiment. Specifically, the method includes:

[0028] In step S11, real-time temperature data and real-time humidity data of multiple monitoring areas in the baking oven during the initial drying stage of the bean curd sheets are obtained. Based on the real-time temperature data and the real-time humidity data, a first hot air circulation parameter adjustment instruction is dynamically generated;

[0029] Among them, the monitoring areas are multiple specific areas divided inside the baking oven, used to respectively install sensors to collect data such as temperature and humidity in the area in real time, so as to comprehensively understand the environmental conditions at different positions inside the baking oven. The real-time temperature data is the temperature value of the monitoring area in the baking oven collected by a temperature sensor at a specific moment, reflecting the temperature situation of the monitoring area at that moment. The real-time humidity data is the humidity value of the monitoring area in the baking oven collected by a humidity sensor at a specific moment, reflecting the humidity situation of the monitoring area at that moment.

[0030] Among them, the first hot air circulation parameter adjustment instruction is an instruction generated after certain algorithms or logical processing according to the real-time temperature data and real-time humidity data obtained during the initial drying stage of the baking oven, and is used to adjust the hot air circulation parameters of the baking oven, aiming to optimize the hot air circulation effect during the initial drying stage.

[0031] In the embodiments of the present disclosure, in the initial drying stage of the yuba, multiple sensors are installed in the drying oven to monitor the real-time temperature and real-time humidity in different areas. For example, sensors are set at different positions such as the top, middle, bottom, left, right, front, and back of the drying oven to obtain the real-time data of multiple monitoring areas. These sensors will transmit the collected real-time temperature data and real-time humidity data to the control system.

[0032] In this embodiment, temperature sensors and humidity sensors are installed at different positions in the drying oven. These sensors collect the temperature and humidity data of the monitoring area where they are located in real time and transmit the data to the control system. The control system collects data from multiple monitoring areas to comprehensively understand the environmental conditions in the drying oven.

[0033] Dynamically generate the first hot air circulation parameter adjustment instruction: The control system calculates and analyzes according to the obtained real-time temperature data and real-time humidity data by using a preset algorithm or logic. For example, if the temperature in some areas is too high or the humidity is too low, the system will comprehensively consider these data and generate corresponding adjustment instructions, such as adjusting parameters such as the wind speed, wind direction, and temperature of the hot air, to optimize the hot air circulation effect in the initial drying stage and enable the yuba to be dried in a more suitable environment.

[0034] In step S12, based on the first hot air circulation parameter adjustment instruction, control the drying oven to execute the first hot air circulation mode, and continuously collect the real-time temperature fluctuation data and real-time humidity gradient data in the first optimization stage during the execution process. According to the real-time temperature fluctuation data, generate a temperature stability evaluation index, and according to the real-time humidity gradient data, generate a humidity balance evaluation index;

[0035] Among them, the first hot air circulation mode is a specific hot air circulation method executed by the drying oven according to the parameters set in the first hot air circulation parameter adjustment instruction, and is used to dry the yuba in the initial drying stage. The real-time temperature fluctuation data is the data of the temperature change over time in the monitoring area inside the drying oven continuously collected during the execution of the first hot air circulation mode by the drying oven, which reflects the temperature fluctuation within a certain period of time. The real-time humidity gradient data is the data of the humidity change with the spatial position in the monitoring area inside the drying oven continuously collected during the execution of the first hot air circulation mode by the drying oven, which reflects the distribution difference of humidity between different monitoring areas.

[0036] Among them, the temperature stability evaluation index is a quantitative index obtained through a specific calculation method according to the real-time temperature fluctuation data, and is used to evaluate the stability degree of the temperature in the drying oven within a certain time and space range. The humidity balance evaluation index is a quantitative index obtained through a specific calculation method according to the real-time humidity gradient data, and is used to evaluate the balance degree of the humidity between different monitoring areas in the drying oven.

[0037] In this embodiment, relevant devices of the baking oven are adjusted according to the first hot air circulation parameter adjustment instruction, so as to execute the first hot air circulation mode. During the execution process, real-time temperature fluctuation data and real-time humidity gradient data in the first optimization stage are continuously collected. For the real-time temperature fluctuation data, a temperature stability evaluation index is generated through conventional calculation methods in related technologies; for the real-time humidity gradient data, a humidity uniformity evaluation index is generated through corresponding calculation methods.

[0038] In this embodiment, the generated first hot air circulation parameter adjustment instruction is sent to the actuators of the baking oven, such as the fan, heater, etc., so that the baking oven executes the first hot air circulation mode according to the parameters set in the instruction.

[0039] Continuous data collection: During the execution of the first hot air circulation mode, the temperature sensor and the humidity sensor continuously collect real-time temperature fluctuation data and real-time humidity gradient data in the monitored area inside the baking oven. The temperature fluctuation data reflects the change of temperature over time, and the humidity gradient data reflects the difference in humidity between different monitored areas.

[0040] Generate evaluation indexes: According to the collected real-time temperature fluctuation data, a temperature stability evaluation index is generated by using specific calculation methods (such as calculating the standard deviation, variance, etc. of the temperature). The lower this index is, the better the temperature stability. According to the real-time humidity gradient data, a humidity uniformity evaluation index is generated by calculating the average value of the humidity difference between different monitored areas or other relevant indexes. The closer this index is to the ideal value, the better the humidity uniformity.

[0041] In the embodiment of the present disclosure, in step S12, the controlling the baking oven to execute the first hot air circulation mode based on the first hot air circulation parameter adjustment instruction includes:

[0042] In step S121, the operating power of the main fan inside the baking oven is adjusted according to the hot air circulation intensity adjustment value, so that the output wind speed of the main fan matches the first target wind speed in the hot air circulation intensity adjustment value;

[0043] In this embodiment, the operating power of the main fan inside the baking oven is adjusted according to the hot air circulation intensity adjustment value. For example, the first target wind speed in the hot air circulation intensity adjustment value is V1. By adjusting parameters such as the voltage or current of the main fan, the output wind speed of the main fan is made to reach V1. During the adjustment process, the output wind speed of the main fan is monitored in real time, and it is ensured to match the first target wind speed through feedback control.

[0044] In step S122, the start-stop cycle of the auxiliary fan inside the baking oven is adjusted according to the hot air circulation frequency adjustment value, so that the operating interval time of the auxiliary fan matches the second target cycle in the hot air circulation frequency adjustment value;

[0045] In this embodiment, the start-stop cycle of the auxiliary fan in the baking oven is adjusted according to the adjusted value of the hot air circulation frequency. Assume that the second target cycle in the adjusted value of the hot air circulation frequency is T2, then control the auxiliary fan to perform start-stop operations at intervals of T2. For example, the auxiliary fan runs for a period of time and then stops, and starts again after stopping for T2 time, and so on in a cycle.

[0046] In step S123, the angle offset of the deflector in the baking oven is adjusted according to the adjusted value of the hot air flow direction angle, so that the actual deflection angle of the deflector matches the third target angle in the adjusted value of the hot air flow direction angle;

[0047] In this embodiment, the angle offset of the deflector in the baking oven is adjusted according to the adjusted value of the hot air flow direction angle. For example, the third target angle in the adjusted value of the hot air flow direction angle is θ3, and devices such as a stepper motor are controlled to adjust the angle of the deflector so that its actual deflection angle reaches θ3. During the adjustment process, the actual deflection angle feedback data of the deflector is collected in real time, and it is ensured to match the third target angle through a closed-loop control method.

[0048] In step S124, under the coordinated operation of the adjusted main fan, auxiliary fan and deflector, the hot air flow trajectory data in the baking oven is monitored in real time, and a hot air coverage uniformity index is generated based on the hot air flow trajectory data;

[0049] In this embodiment, when the main fan, auxiliary fan and deflector are adjusted and operate in coordination, the hot air flow trajectory data in the baking oven is monitored in real time. The speed and direction information of the hot air can be obtained through devices such as a wind speed sensor and a wind direction sensor installed in the baking oven, so as to determine the flow trajectory of the hot air. Then, a hot air coverage uniformity index is generated based on these hot air flow trajectory data. For example, by calculating parameters such as the hot air speed and coverage time in different regions, the coverage uniformity degree of the hot air in the baking oven is evaluated.

[0050] In step S125, when the hot air coverage uniformity index is lower than a preset uniformity threshold, a fan parameter compensation operation is triggered, and the adjusted value of the hot air circulation intensity and the adjusted value of the hot air flow direction angle are dynamically corrected based on the compensation difference between the hot air coverage uniformity index and the uniformity threshold.

[0051] In this embodiment, when the hot air coverage uniformity index is lower than the preset uniformity threshold, it indicates that the hot air coverage in the baking oven is uneven and a compensation operation is required. Calculate the compensation difference between the hot air coverage uniformity index and the uniformity threshold, and dynamically correct the hot air circulation intensity adjustment value and the hot air flow direction angle adjustment value according to this compensation difference. For example, if the hot air coverage uniformity index is low, it may be necessary to increase the hot air circulation intensity adjustment value or adjust the hot air flow direction angle adjustment value so that the hot air can cover each area in the baking oven more evenly.

[0052] In step S13, when the temperature stability evaluation index does not reach the preset first stability threshold or the humidity balance evaluation index does not reach the preset first balance threshold, trigger a multi-stage parameter readjustment operation, and generate a second hot air circulation parameter adjustment instruction based on the first difference parameter between the temperature stability evaluation index and the first stability threshold and the second difference parameter between the humidity balance evaluation index and the first balance threshold.

[0053] Among them, the first stability threshold is a preset threshold for the temperature stability evaluation index. When the temperature stability evaluation index reaches or exceeds this threshold, it is considered that the temperature stability in the baking oven meets the requirements. The first balance threshold is a preset threshold for the humidity balance evaluation index. When the humidity balance evaluation index reaches or exceeds this threshold, it is considered that the humidity balance in the baking oven meets the requirements. The multi-stage parameter readjustment operation is an operation to readjust the hot air circulation parameters of the baking oven triggered when the temperature stability evaluation index or the humidity balance evaluation index does not reach the corresponding threshold, aiming to further optimize the drying environment.

[0054] Among them, the first difference parameter is the difference between the temperature stability evaluation index and the first stability threshold, reflecting the gap between the current temperature stability and the desired stability. The second difference parameter is the difference between the humidity balance evaluation index and the first balance threshold, reflecting the gap between the current humidity balance and the desired balance. The second hot air circulation parameter adjustment instruction can be an instruction for adjusting the hot air circulation parameters of the baking oven generated after certain algorithms or logical processing based on the first difference parameter and the second difference parameter, aiming to make the temperature stability and humidity balance in the baking oven meet the requirements.

[0055] In this embodiment, compare the generated temperature stability evaluation index with the preset first stability threshold, and compare the humidity balance evaluation index with the preset first balance threshold. If the temperature stability evaluation index does not reach the first stability threshold or the humidity balance evaluation index does not reach the first balance threshold, it indicates that the current temperature or humidity environment in the baking oven does not reach the ideal state and further parameter adjustment is required. At this time, trigger a multi-stage parameter readjustment operation.

[0056] Generate the second hot air circulation parameter adjustment instruction: Calculate the first difference parameter between the temperature stability evaluation index and the first stability threshold, and the second difference parameter between the humidity balance evaluation index and the first balance threshold. Based on these two difference parameters, use a specific algorithm or logic, such as determining the adjustment amplitude and direction according to the magnitude of the difference, to generate the second hot air circulation parameter adjustment instruction to further optimize the temperature and humidity environment in the baking oven.

[0057] In this embodiment, the temperature stability evaluation index is compared with the preset first stability threshold, and the humidity balance evaluation index is compared with the preset first balance threshold. If the temperature stability evaluation index does not reach the first stability threshold or the humidity balance evaluation index does not reach the first balance threshold, it indicates that the current drying parameters need to be further adjusted, triggering a multi-stage parameter readjustment operation.

[0058] In step S14, according to the second hot air circulation parameter adjustment instruction, control the baking oven to switch to the second hot air circulation mode, and synchronously monitor the data of the surface morphology change of the dried bean curd sticks during the execution process. Based on the morphological matching degree between the surface morphology change data and the preset standard morphology change curve, generate a drying termination instruction to end the drying process.

[0059] Among them, the second hot air circulation mode is a new specific hot air circulation method executed by the baking oven according to the parameters set by the second hot air circulation parameter adjustment instruction, which is used to further optimize the drying process. The surface morphology change data is the data of the change of the surface morphology of the dried bean curd sticks over time obtained by image acquisition or other means during the execution of the second hot air circulation mode by the baking oven, which reflects the morphological change of the dried bean curd sticks during the drying process. The standard morphology change curve is a curve preset to describe the change of the surface morphology of the dried bean curd sticks over time in the ideal drying process, which is used as a reference basis for judging the drying degree of the dried bean curd sticks. The morphological matching degree is a quantitative index calculated based on the similarity between the surface morphology change data and the standard morphology change curve, which is used to evaluate the closeness of the current surface morphology of the dried bean curd sticks to the surface morphology in the ideal drying state. The drying termination instruction is an instruction generated when the morphological matching degree reaches a certain requirement to end the drying process, indicating that the dried bean curd sticks have been dried to a state that meets the requirements.

[0060] In this embodiment, according to the second hot air circulation parameter adjustment instruction, adjust the relevant equipment of the baking oven to make it switch to the second hot air circulation mode. During the execution of the second hot air circulation mode, synchronously monitor the data of the surface morphology change of the dried bean curd sticks. Then compare the surface morphology change data with the preset standard morphology change curve, calculate the morphological matching degree, and generate a drying termination instruction according to the morphological matching degree.

[0061] In this embodiment, the generated second hot air circulation parameter adjustment instruction is sent to the actuator of the baking oven, so that the baking oven switches to the second hot air circulation mode and performs hot air circulation according to the new parameters. Synchronously monitor the surface morphology change data: During the execution of the second hot air circulation mode, the surface morphology change data of the yuba is monitored in real time through an image acquisition device (such as a camera) or other means. For example, images of the yuba surface can be collected, and the drying degree, morphological changes, etc. of the yuba are analyzed through image processing techniques. Generate a drying termination instruction: Compare the collected surface morphology change data with a preset standard morphology change curve, and calculate the morphology matching degree. When the morphology matching degree reaches the preset requirement, it indicates that the yuba has been dried to a standard state. At this time, the control system generates a drying termination instruction to end the drying process.

[0062] In the embodiment of the present disclosure, in step S14, the controlling the baking oven to switch to the second hot air circulation mode according to the second hot air circulation parameter adjustment instruction includes:

[0063] In step S141, adjust the operating power of the main fan according to the optimized hot air circulation intensity adjustment value, and collect the current fluctuation data and vibration amplitude data of the main fan in real time during the adjustment process;

[0064] In this embodiment, the operating power of the main fan is adjusted according to the optimized hot air circulation intensity adjustment value. For example, the optimized hot air circulation intensity adjustment value is V_adj1_optimized. By adjusting parameters such as the voltage or current of the main fan, the output wind speed of the main fan is made to reach the wind speed corresponding to V_adj1_optimized. During the adjustment process, the current fluctuation data and vibration amplitude data of the main fan are collected in real time to detect abnormal conditions of the main fan in a timely manner.

[0065] In step S142, when the current fluctuation data exceeds the preset safe current threshold or the vibration amplitude data exceeds the preset safe vibration threshold, trigger the main fan protection operation, gradually reduce the operating power of the main fan to the standby power threshold, and start the standby fan to maintain the hot air circulation intensity;

[0066] In this embodiment, when the collected current fluctuation data of the main fan exceeds the preset safe current threshold or the vibration amplitude data exceeds the preset safe vibration threshold, it indicates that the main fan may be in an abnormal operating state. To avoid equipment damage and ensure the smooth progress of the drying process, the main fan protection operation will be triggered. Specifically, the operating power of the main fan will be gradually reduced to the standby power threshold.

[0067] For example, by adjusting the supply voltage or current of the main blower, its power output is reduced in a smooth manner to prevent the equipment from being impacted due to a sudden large reduction in power. Meanwhile, the standby blower is started. The function of the standby blower is to maintain the intensity of the hot air circulation inside the baking oven when the power of the main blower is reduced, ensuring the continuity and stability of the drying process. After the standby blower is started, it will automatically adjust its operating parameters according to the hot air circulation situation after the main blower reduces its power to achieve coordinated operation with the main blower and ensure that the hot air inside the baking oven can circulate continuously and effectively.

[0068] In step S143, adjust the start-stop cycle of the auxiliary blower according to the optimized hot air circulation frequency adjustment value, and collect the response delay time data of the auxiliary blower each time there is a start-stop switch.

[0069] In this embodiment, the start-stop cycle of the auxiliary blower is adjusted according to the optimized hot air circulation frequency adjustment value. For example, the optimized hot air circulation frequency adjustment value stipulates a new start-stop time interval for the auxiliary blower, and the control system will send start-stop commands to the auxiliary blower according to this new time interval. Each time the auxiliary blower performs a start-stop switching operation, its response delay time data will be collected. The response delay time refers to the time difference from when the control system issues a start-stop command to when the auxiliary blower actually starts to execute the corresponding action. The response delay time data can be accurately collected by setting a time monitoring module in the control circuit of the auxiliary blower for subsequent evaluation and adjustment of the operating state of the auxiliary blower.

[0070] In step S144, when the response delay time data exceeds a preset delay tolerance threshold, trigger the synchronous control operation of the auxiliary blower, and force the start-stop action of the auxiliary blower to be synchronized with the operating cycle of the main blower through an increased pulse signal synchronization module.

[0071] In this embodiment, when the collected response delay time data of the auxiliary blower exceeds the preset delay tolerance threshold, it indicates that the start-stop response speed of the auxiliary blower is relatively slow, which may affect the coordination of the entire hot air circulation system. At this time, the synchronous control operation of the auxiliary blower will be triggered. The specific approach is to add a pulse signal synchronization module, which generates specific pulse signals. The operating cycle of the main blower is input into the pulse signal synchronization module as a reference signal, and the module adjusts the output frequency and time point of the pulse signal according to the operating cycle of the main blower.

[0072] Then, send these adjusted pulse signals to the auxiliary blower to force the start-stop action of the auxiliary blower to be synchronized with the operating cycle of the main blower. This can improve the coordination between the auxiliary blower and the main blower and ensure that the hot air circulation system operates more stably and efficiently.

[0073] In step S145, adjust the deflection angle of the deflector according to the optimized hot air flow direction angle adjustment value, and collect the actual deflection angle feedback data of the deflector in real time during the adjustment process;

[0074] In this embodiment, the deflection angle of the deflector is adjusted according to the optimized hot air flow direction angle adjustment value. The control system sends a control signal to the device (such as a stepper motor) that drives the deflector to rotate, so that the deflector rotates according to the optimized hot air flow direction angle adjustment value.

[0075] During the adjustment process, the actual deflection angle feedback data of the deflector will be collected in real time. The actual deflection angle information can be obtained through an angle sensor installed on the deflector, and then these data are fed back to the control system. The control system will compare the actual deflection angle with the optimized hot air flow direction angle adjustment value to promptly detect deviations and make adjustments.

[0076] In step S146, when the angle deviation between the actual deflection angle feedback data and the optimized hot air flow direction angle adjustment value exceeds a preset deviation threshold, trigger the deflector calibration operation, and generate a stepper motor compensation pulse signal based on the angle deviation to correct the actual deflection angle of the deflector.

[0077] In this embodiment, when the angle deviation between the collected actual deflection angle feedback data of the deflector and the optimized hot air flow direction angle adjustment value exceeds the preset deviation threshold, it indicates that there is a large difference between the actual deflection angle of the deflector and the expected angle, and a calibration operation is required. At this time, a stepper motor compensation pulse signal will be generated based on this angle deviation. Specifically, the control system will calculate the additional pulse quantity and pulse direction that need to be sent to the stepper motor according to the magnitude and direction of the angle deviation. These compensation pulse signals will be sent to the stepper motor, and the stepper motor further adjusts the deflection angle of the deflector according to these signals, thereby correcting the actual deflection angle of the deflector to make it as close as possible to the optimized hot air flow direction angle adjustment value.

[0078] The above technical solution can comprehensively understand the drying environment conditions in the initial stage by obtaining the real-time temperature and real-time humidity data of multiple monitoring areas in the oven during the initial drying stage, generating temperature distribution characteristics and humidity change rate characteristics. Based on the deviation of the matching degree with the preset reference, a hot air circulation parameter adjustment instruction is dynamically generated, realizing precise control of the drying process and avoiding drying quality problems caused by uneven environmental parameters. During the execution of the hot air circulation mode, the temperature fluctuation and humidity gradient data are continuously collected and an evaluation index is generated to be able to evaluate the stability and balance of drying in real time. When the evaluation index does not reach the threshold, a multi-stage parameter re-adjustment operation is triggered to further optimize the drying effect. Finally, a drying termination instruction is generated according to the matching degree between the surface morphology change data of the dried bean curd sticks and the standard curve, ensuring the accuracy of the end of drying, improving the drying quality and production efficiency of the dried bean curd sticks, and significantly enhancing the intelligent and refined level of the dried bean curd stick drying process as a whole.

[0079] In a possible implementation manner, in step S11, the dynamically generating a first hot air circulation parameter adjustment instruction based on the real-time temperature data and the real-time humidity data includes:

[0080] In step S111, based on the real-time temperature data and the real-time humidity data, an initial stage temperature distribution characteristic and an initial stage humidity change rate characteristic are generated;

[0081] In the embodiments of the present disclosure, an initial stage temperature distribution characteristic and an initial stage humidity change rate characteristic are generated based on these real-time data. Specifically, when generating the initial stage temperature distribution characteristic and the initial stage humidity change rate characteristic, multiple monitoring areas are first divided into a central area group, an edge area group, and a transition area group. Exemplarily, several monitoring points at the center position of the oven are divided into the central area group, the monitoring points close to the oven wall are divided into the edge area group, and the monitoring points between the central area group and the edge area group are divided into the transition area group.

[0082] In the embodiments of the present disclosure, temperature sensors arranged in different monitoring areas in the oven continuously collect real-time temperature data. There may be noise or outliers in these data. Therefore, before generating the temperature distribution characteristic, it is necessary to preprocess the collected temperature data. For example, a filtering algorithm (such as mean filtering, median filtering, etc.) is used to remove noise, and outliers are identified and corrected (such as replacing the outliers with the average value of the data of adjacent monitoring areas).

[0083] Further, the preprocessed temperature data is used to generate temperature distribution characteristics. Common methods include calculating statistics such as the mean, variance, and standard deviation of the temperatures in each monitoring area to reflect the overall level and fluctuation degree of the temperature. In addition, spatial interpolation algorithms (such as Kriging interpolation, inverse distance weighted interpolation, etc.) can be used to estimate the temperature values at other positions in the oven based on the temperature data of the limited monitoring areas, so as to obtain a more comprehensive temperature distribution. For example, through Kriging interpolation, the temperature at any position in the oven can be predicted based on the temperature values and spatial position relationships of the known monitoring areas, and then a temperature distribution cloud map can be drawn to intuitively display the temperature distribution characteristics in the initial stage.

[0084] In the embodiments of the present disclosure, a humidity sensor collects humidity data of each monitoring area in the oven in real time. Similarly, these data also need to be preprocessed to remove noise and outliers. For humidity data, problems such as sensor response delay or data drift may exist and need to be corrected accordingly.

[0085] Further, the numerical differentiation method can be used to calculate the humidity change rate in the initial stage. For example, for each monitoring area, at two adjacent sampling time points t1 and t2 (t2>t1), the corresponding humidity values are H1 and H2 respectively, then the humidity change rate v during this time period can be approximately expressed as v=(H2 - H1) / (t2 - t1).

[0086] By statistically analyzing the humidity change rates in multiple time periods, such as calculating the average change rate, maximum change rate, minimum change rate, etc., the humidity change rate characteristics in the initial stage are obtained. In addition, a curve of humidity changing with time can be drawn, and the humidity change rate characteristics can be intuitively reflected through the slope change of the curve.

[0087] In step S112, according to the matching degree deviation between the initial stage temperature distribution characteristics and a preset reference temperature distribution interval, a first matching degree deviation is determined, and according to the matching degree deviation between the initial stage humidity change rate characteristics and a preset reference humidity change rate, a second matching degree deviation is determined;

[0088] In the embodiments of the present disclosure, a reference temperature distribution interval is preset according to the technological requirements and empirical data of dried bean curd stick drying. This interval reflects the ideal distribution range of the temperatures in each monitoring area in the oven during the initial drying stage. For example, for certain varieties of dried bean curd sticks, during the initial drying stage, the temperatures at different positions in the oven should be maintained within a certain range to ensure that the dried bean curd sticks are evenly heated and avoid local overheating or overcooling.

[0089] Further, compare the generated initial-stage temperature distribution characteristics (such as temperature values in each monitoring area, temperature distribution nephogram, etc.) with a preset reference temperature distribution range. A similarity algorithm (such as cosine similarity, Euclidean distance, etc.) can be used to calculate the matching degree between the two. For example, if the Euclidean distance is used, calculate the Euclidean distance between the temperature distribution feature vector and the center vector of the reference temperature distribution range. The smaller the distance, the higher the matching degree. Then, according to the preset conversion relationship between the matching degree and the deviation, convert the calculated matching degree into a first matching degree deviation. The larger the deviation value, the greater the difference between the initial-stage temperature distribution and the reference temperature distribution range.

[0090] In the embodiments of the present disclosure, according to the technological requirements and actual experience of dried beancurd stick drying, a preset reference humidity change rate is set in advance. This rate indicates at what speed the humidity in the drying oven should change during the initial drying stage to ensure that the dried beancurd sticks can lose moisture evenly and avoid problems such as uneven drying or surface cracking.

[0091] Further, compare the generated initial-stage humidity change rate characteristics (such as average change rate, change rate curve, etc.) with the preset reference humidity change rate. Similarly, a similarity algorithm can be used to calculate the matching degree. For example, for the change rate curve, the dynamic time warping (DTW) algorithm can be used to measure the similarity between the two curves. The higher the similarity degree, the higher the matching degree. Then, according to the preset conversion relationship between the matching degree and the deviation, convert the matching degree into a second matching degree deviation. The larger the deviation value, the greater the difference between the initial-stage humidity change rate and the reference humidity change rate.

[0092] In step S113, according to the first matching degree deviation between the initial-stage temperature distribution characteristics and the preset reference temperature distribution range, and the second matching degree deviation between the initial-stage humidity change rate characteristics and the preset reference humidity change rate, dynamically generate a first hot air circulation parameter adjustment instruction.

[0093] In this embodiment, the preset reference temperature distribution range is a preset reasonable temperature distribution range, and the preset reference humidity change rate is also a preset reasonable humidity change rate. Compare the initial-stage temperature distribution characteristics with the reference temperature distribution range and calculate the first matching degree deviation; compare the initial-stage humidity change rate characteristics with the reference humidity change rate and calculate the second matching degree deviation. Then, dynamically generate a first hot air circulation parameter adjustment instruction according to these two deviations.

[0094] In this embodiment, the relationship between the hot air circulation parameters (such as wind speed, wind direction, temperature, etc.) and the temperature distribution and the humidity change rate is analyzed. For example, increasing the wind speed can accelerate the circulation of hot air in the baking oven, which helps to improve the uniformity of the temperature distribution, but may increase the humidity change rate; adjusting the wind direction can change the flow path of the hot air in the baking oven, thereby affecting the temperature and humidity in different regions. Through experiments or simulations, a mathematical model or empirical formula between the hot air circulation parameters and the temperature distribution and the humidity change rate is established.

[0095] Furthermore, according to the magnitudes and directions of the first matching degree deviation and the second matching degree deviation, corresponding adjustment rules for the hot air circulation parameters are formulated. For example, if the first matching degree deviation indicates that the temperature in some regions is too high, it may be necessary to reduce the wind speed of the hot air corresponding to these regions or adjust the wind direction so that the hot air flows more to the regions with lower temperature; if the second matching degree deviation indicates that the humidity change rate is too fast, it may be necessary to reduce the temperature of the hot air or decrease the wind speed to slow down the evaporation rate of moisture.

[0096] Furthermore, the first matching degree deviation and the second matching degree deviation are used as inputs and substituted into a pre-established parameter adjustment logic for comprehensive calculation. For example, by using the method of weighted summation, different weights are assigned to the first matching degree deviation and the second matching degree deviation according to the importance of the temperature distribution and the humidity change rate during the drying process, and a comprehensive deviation value is calculated.

[0097] Furthermore, according to the deviation value obtained from the comprehensive calculation, the specific values and adjustment directions (increase or decrease) of the hot air circulation parameters (such as wind speed, wind direction, temperature, etc.) to be adjusted are determined. Then, this parameter adjustment information is encapsulated into a first hot air circulation parameter adjustment instruction and sent to the control system of the baking oven to perform the corresponding parameter adjustment operation.

[0098] The above technical solution can dynamically generate a suitable first hot air circulation parameter adjustment instruction according to the real-time temperature and humidity data in the initial drying stage of the baking oven, thereby optimizing the environmental parameters in the baking oven and improving the drying quality of the yuba.

[0099] In a possible implementation manner, in step S111, generating the initial stage temperature distribution feature and the initial stage humidity change rate feature based on the real-time temperature data and the real-time humidity data includes:

[0100] In step S1111, the multiple monitoring regions are divided into a central region group, an edge region group, and a transition region group. The average value of the real-time temperature data of all monitoring points in the central region group is calculated to obtain the central region temperature average value; the variance of the real-time temperature data of all monitoring points in the edge region group is calculated to obtain the edge region temperature variance;

[0101] In this step, the average value of the real-time temperature data of all monitoring points within the central region group is calculated. For example, if there are three monitoring points A, B, and C in the central region group, and their real-time temperature data are Ta, Tb, and Tc respectively, then the central region temperature average value T_center_mean = (Ta + Tb + Tc) / 3. The variance of the real-time temperature data of all monitoring points within the edge region group is calculated. Assuming there are three monitoring points D, E, and F in the edge region group, and their real-time temperature data are Td, Te, and Tf respectively, first calculate the edge region temperature average value:

[0102] T_edge_mean = (Td + Te + Tf) / 3;

[0103] Then calculate the variance:

[0104] S_edge = [(Td - T_edge_mean) 2 +(Te - T_edge_mean) 2 +(Tf - T_edge_mean) 2 / 3.

[0105] In step S1112, a temperature distribution difference coefficient is generated based on the ratio between the central region temperature average value and the edge region temperature variance, and a temperature dynamic association feature is generated in combination with the real-time temperature gradient change trend of adjacent monitoring points within the transition region group;

[0106] In the embodiment of the present disclosure, after obtaining the central region temperature average value and the edge region temperature variance, their ratio is calculated to obtain the temperature distribution difference coefficient. For example, the temperature distribution difference coefficient K = T_center_mean / S_edge. At the same time, observe the real-time temperature gradient change trend of adjacent monitoring points within the transition region group. For example, if there are two adjacent monitoring points G and H in the transition region group, and their temperatures are Tg and Th respectively, the temperature gradient is (Th - Tg) / distance(G, H). By analyzing the temperature gradient change trends of multiple adjacent monitoring points, a temperature dynamic association feature is generated.

[0107] In step S1113, the temperature distribution difference coefficient and the temperature dynamic association feature are weighted and fused to generate the initial stage temperature distribution feature;

[0108] In the embodiment of the present disclosure, different weights are assigned to the temperature distribution difference coefficient and the temperature dynamic association feature. For example, the weight of the temperature distribution difference coefficient is w1, and the weight of the temperature dynamic association feature is w2, and w1 + w2 = 1. They are weighted and fused, and the initial stage temperature distribution feature = w1 × temperature distribution difference coefficient + w2 × temperature dynamic association feature.

[0109] In step S1114, the real-time humidity data of the multiple monitoring areas is synchronously extracted, the absolute value of the humidity change of each monitoring area within a preset time window is calculated, and the maximum humidity change rate and the minimum humidity change rate are generated based on the absolute values of the humidity changes of all monitoring areas;

[0110] In the embodiment of the present disclosure, the real-time humidity data of multiple monitoring areas is synchronously extracted. Assuming that the preset time window is t, for each monitoring area, the absolute value of the humidity change within the time window t is calculated. For example, the humidity of a certain monitoring area at time t1 is H1, and the humidity at time t2 (t2 - t1 = t) is H2, and the absolute value of the humidity change is |H2 - H1|. Then, the maximum and minimum values among the absolute values of the humidity changes of all monitoring areas are found, and the maximum humidity change rate V_max = maximum absolute humidity change / t, and the minimum humidity change rate V_min = minimum absolute humidity change / t are calculated respectively.

[0111] In step S1115, a humidity change fluctuation range is generated according to the rate difference between the maximum humidity change rate and the minimum humidity change rate, and the initial stage humidity change rate feature is generated in combination with the position information of the monitoring area corresponding to the maximum humidity change rate.

[0112] In the embodiment of the present disclosure, the humidity change fluctuation range R = V_max - V_min is calculated according to the maximum humidity change rate and the minimum humidity change rate. At the same time, the position information of the monitoring area corresponding to the maximum humidity change rate is recorded, and the humidity change fluctuation range and this position information are combined to generate the initial stage humidity change rate feature.

[0113] In a possible implementation manner, in step S113, a first hot air circulation parameter adjustment instruction is dynamically generated according to the first matching degree deviation between the initial stage temperature distribution feature and a preset reference temperature distribution interval, and the second matching degree deviation between the initial stage humidity change rate feature and a preset reference humidity change rate, including:

[0114] In step S1131, the first matching degree deviation is decomposed into a temperature distribution deviation component and a temperature gradient deviation component, where the temperature distribution deviation component represents the absolute value of the difference between the average temperature of the central region and the central reference temperature of the reference temperature distribution interval, and the temperature gradient deviation component represents the difference ratio between the temperature variance of the edge region and the edge temperature variance threshold of the reference temperature distribution interval;

[0115] In the embodiments of the present disclosure, when calculating the first matching degree deviation, it is decomposed into a temperature distribution deviation component and a temperature gradient deviation component. Assume that the central reference temperature of the reference temperature distribution interval is T_base_center, and the edge temperature variance threshold is S_base_edge.

[0116] Among them, the temperature distribution deviation component = |T_center_mean - T_base_center|, and the temperature gradient deviation component = (S_edge - S_base_edge) / S_base_edge.

[0117] In step S1132, the second matching degree deviation is decomposed into a maximum rate deviation component and a fluctuation range deviation component, where the maximum rate deviation component represents the rate ratio between the maximum humidity change rate and the reference humidity change rate, and the fluctuation range deviation component represents the range difference between the humidity change fluctuation range and a preset reference fluctuation range;

[0118] In the embodiments of the present disclosure, for the second matching degree deviation, it is decomposed into a maximum rate deviation component and a fluctuation range deviation component. Assume that the reference humidity change rate is V_base, and the reference fluctuation range is R_base. The maximum rate deviation component = V_max / V_base, and the fluctuation range deviation component = R - R_base.

[0119] In step S1133, a multi-dimensional adjustment coefficient matrix is established among the temperature distribution deviation component, the temperature gradient deviation component, the maximum rate deviation component, and the fluctuation range deviation component, and the multi-dimensional adjustment coefficient matrix is mapped to an initial hot air circulation parameter adjustment coefficient set through a pre-trained drying parameter mapping model;

[0120] In the embodiments of the present disclosure, a multi-dimensional adjustment coefficient matrix is established, and the elements of the matrix are respectively the temperature distribution deviation component, the temperature gradient deviation component, the maximum rate deviation component, and the fluctuation range deviation component. For example, the matrix M = [temperature distribution deviation component, temperature gradient deviation component, maximum rate deviation component, fluctuation range deviation component]. Then, the multi-dimensional adjustment coefficient matrix is input into a pre-trained drying parameter mapping model. The drying parameter mapping model is an artificial intelligence model trained with a large amount of data. Its input is the multi-dimensional adjustment coefficient matrix, and its output is an initial hot air circulation parameter adjustment coefficient set. The specific calculation logic inside the model is a black box, but after training, it can accurately map the input to the output.

[0121] Regarding the pre-trained drying parameter mapping model and the pre-constructed multi-stage optimization model, the construction and training processes are described in detail below.

[0122] Step S210: Collect training data, including temperature distribution characteristics in different initial stages, humidity change rate characteristics in the initial stage, corresponding first matching degree deviation, second matching degree deviation, and the set of actual required hot air circulation parameter adjustment coefficients.

[0123] To train the drying parameter mapping model, a large amount of training data needs to be collected. During the actual drying process of yuba, record the temperature distribution characteristics and humidity change rate characteristics in the initial drying stage of different batches of yuba. At the same time, calculate the first matching degree deviation and the second matching degree deviation between these characteristics and the preset reference temperature distribution range and reference humidity change rate. In addition, record the set of hot air circulation parameter adjustment coefficients actually used for these deviations. Organize these data into a training data set, where each data sample includes temperature distribution characteristics in the initial stage, humidity change rate characteristics in the initial stage, the first matching degree deviation, the second matching degree deviation, and the corresponding set of hot air circulation parameter adjustment coefficients.

[0124] Step S211: Preprocess the training data, including data cleaning and normalization, to ensure the quality and dimensional consistency of the data.

[0125] Preprocess the collected training data. First, perform data cleaning to check for problems such as missing values and outliers in the data. For missing values, interpolation methods or the method of deleting data samples containing missing values can be used for processing. For outliers, by setting a reasonable threshold range, data exceeding the threshold can be regarded as outliers and corrected or deleted. Then perform normalization processing. Since the dimensions of different features may be different, to ensure the quality and dimensional consistency of the data, it is necessary to normalize the data of all features. For example, the min-max normalization method can be used to map the data of each feature to the [0, 1] interval. For feature X, its normalization formula is: Normalized value = (Original value - Minimum value of the feature) / (Maximum value of the feature - Minimum value of the feature). Through such preprocessing, the training effect and stability of the model can be improved.

[0126] Step S212: Construct a drying parameter mapping model, which includes an input layer, a hidden layer, and an output layer. The input layer receives the preprocessed training data, the hidden layer uses multiple neurons for feature extraction and nonlinear transformation, and the output layer outputs the set of initial hot air circulation parameter adjustment coefficients.

[0127] The drying parameter mapping model adopts a multi-layer neural network structure, mainly including an input layer, a hidden layer, and an output layer. The number of neurons in the input layer is the same as the feature dimension of the preprocessed training data. Its function is to receive the preprocessed training data, including information such as the temperature distribution characteristics in the initial stage, the humidity change rate characteristics in the initial stage, the first matching degree deviation, and the second matching degree deviation. Multiple neurons are set in the hidden layer, and these neurons will perform feature extraction and non-linear transformation on the input data. For example, each neuron will perform a weighted sum on the input data and then process it through a non-linear activation function (such as the ReLU function), and pass the processed result to the next layer. Multiple layers can be set in the hidden layer, and the number of neurons in each layer can be adjusted according to the actual situation to improve the expression ability of the model. The number of neurons in the output layer is the same as the dimension of the initial hot air circulation parameter adjustment coefficient set, and its output result is the initial hot air circulation parameter adjustment coefficient set predicted by the model.

[0128] Step S213: Define a loss function to measure the difference between the initial hot air circulation parameter adjustment coefficient set predicted by the model and the actual initial hot air circulation parameter adjustment coefficient set.

[0129] To train the drying parameter mapping model, a loss function needs to be defined to measure the difference between the model prediction result and the actual result. A common loss function can adopt the mean square error loss function. Specifically, for each training data sample, the model will output a predicted initial hot air circulation parameter adjustment coefficient set, and compare it with the actual initial hot air circulation parameter adjustment coefficient set. Calculate the square of the difference between each corresponding element, then sum and average the squared differences of all elements to obtain the loss value of this sample. For the entire training data set, add and average the loss values of all samples to obtain the total loss value. The smaller the value of the loss function, the closer the model prediction result is to the actual result.

[0130] Step S214: Use an optimization algorithm to train the model. By continuously adjusting the weight parameters of the model, the value of the loss function is gradually reduced until the preset training stop condition is reached.

[0131] The drying parameter mapping model is trained using an optimization algorithm (such as the stochastic gradient descent algorithm). In each training iteration, a small batch of data samples is randomly selected from the training dataset and input into the model for forward propagation calculation to obtain the prediction results of the model. Then, the loss value of this small batch of samples is calculated according to the loss function, and the gradient of the loss value with respect to the model weight parameters is calculated through the backpropagation algorithm. According to the gradient information, the optimization algorithm is used to update the weight parameters of the model, so that the value of the loss function changes in the decreasing direction. This process is continuously repeated until the preset training stop condition is met. The preset training stop condition can be reaching the maximum number of training iterations or the value of the loss function converging below a smaller threshold.

[0132] The construction and training process of the pre-constructed multi-stage optimization model is as follows:

[0133] Step S310: Collect multi-stage optimization training data, including the first difference parameter between the temperature stability evaluation index and the first stability threshold, the second difference parameter between the humidity balance evaluation index and the first balance threshold, the surface shrinkage rate data, the surface crack distribution data, and the corresponding hot air circulation intensity compensation coefficient, hot air circulation frequency compensation coefficient, and hot air flow direction angle compensation coefficient.

[0134] To train the multi-stage optimization model, multi-stage optimization training data needs to be collected. During the drying process of yuba, record the first difference parameter between the temperature stability evaluation index and the first stability threshold, and the second difference parameter between the humidity balance evaluation index and the first balance threshold. At the same time, collect the surface shrinkage rate data and the surface crack distribution data. Also record the actually used hot air circulation intensity compensation coefficient, hot air circulation frequency compensation coefficient, and hot air flow direction angle compensation coefficient for these data. Organize these data into a training dataset, and each data sample includes the first difference parameter, the second difference parameter, the surface shrinkage rate data, the surface crack distribution data, and the corresponding compensation coefficients.

[0135] Step S311: Preprocess the multi-stage optimization training data, and also perform data cleaning and normalization to ensure data quality and consistent dimensions.

[0136] Preprocess the collected multi-stage optimization training data. Similar to the preprocessing of the training data of the drying parameter mapping model, first perform data cleaning to check whether there are missing values, outliers, etc. in the data. Appropriate methods can be used to handle missing values, and outliers are corrected or deleted. Then perform normalization to map the data of different features to the same scale range to ensure consistent dimensions of the data. For example, for each feature, use the min-max normalization method to scale its data values to the interval [0, 1] so that the model can better handle these data.

[0137] Step S312: Construct a multi-stage optimization model, which includes three optimization branches, corresponding to generating a hot air circulation intensity compensation coefficient, a hot air circulation frequency compensation coefficient, and a hot air flow direction angle compensation coefficient respectively. Each optimization branch has an independent input layer, hidden layer, and output layer.

[0138] The multi-stage optimization model consists of three independent optimization branches. The first optimization branch is used to generate the hot air circulation intensity compensation coefficient, the second optimization branch is used to generate the hot air circulation frequency compensation coefficient, and the third optimization branch is used to generate the hot air flow direction angle compensation coefficient. Each optimization branch has an independent input layer, hidden layer, and output layer. The input layer receives the corresponding input data. For example, the input layer of the first optimization branch receives the first difference parameter between the temperature stability evaluation index and the first stability threshold, and the second difference parameter between the humidity balance evaluation index and the first balance threshold; the second optimization branch also receives these two difference parameters; the third optimization branch receives the surface shrinkage rate data and the surface crack distribution data. The hidden layer extracts features and performs non-linear transformation on the input data, and the number of neurons and the number of layers can be adjusted according to the actual situation to improve the performance of the model. The output layer outputs the corresponding compensation coefficients respectively.

[0139] Step S313: Define a loss function for each optimization branch to measure the difference between the predicted compensation coefficient and the actual compensation coefficient of each branch.

[0140] Define an independent loss function for each optimization branch. For example, for the first optimization branch, the mean square error loss function is used to measure the difference between the predicted hot air circulation intensity compensation coefficient and the actual hot air circulation intensity compensation coefficient. The specific calculation method is that for each training data sample, calculate the square of the difference between the predicted compensation coefficient and the actual compensation coefficient, then sum and average the squared differences of all samples to obtain the loss value of this branch. Similarly, define loss functions for the second optimization branch and the third optimization branch respectively to measure the difference between their predicted compensation coefficients and the actual compensation coefficients.

[0141] Step S314: Use an optimization algorithm to train the multi-stage optimization model, and adjust the weight parameters of each optimization branch respectively, so that the loss function values of each branch gradually decrease until the preset training stop condition is reached.

[0142] The multi-stage optimization model is trained using an optimization algorithm (such as the stochastic gradient descent algorithm). In each training iteration, a small batch of data samples is randomly selected from the multi-stage optimization training dataset and input into three optimization branches respectively for forward propagation calculation to obtain the prediction results of each branch. Then, according to the loss functions of each branch, the loss values of this small batch of samples on each branch are calculated, and the gradients of each branch's loss value with respect to its weight parameters are calculated respectively through the backpropagation algorithm. According to the gradient information, the optimization algorithm is used to update the weight parameters of each optimization branch respectively, so that the loss function values of each branch change in the decreasing direction. This process is continuously repeated until the preset training stop condition is met, such as reaching the maximum number of training iterations or the loss function value converges below a small threshold.

[0143] In step S1134, based on the set of initial hot air circulation parameter adjustment coefficients, a hot air circulation intensity adjustment value, a hot air circulation frequency adjustment value, and a hot air flow direction angle adjustment value are generated, and the hot air circulation intensity adjustment value, the hot air circulation frequency adjustment value, and the hot air flow direction angle adjustment value are combined into the first hot air circulation parameter adjustment instruction according to a preset priority order.

[0144] In the embodiments of the present disclosure, according to the set of initial hot air circulation parameter adjustment coefficients, the hot air circulation intensity adjustment value, the hot air circulation frequency adjustment value, and the hot air flow direction angle adjustment value are calculated respectively. For example, assuming there are three coefficients k1, k2, k3 in the set of initial hot air circulation parameter adjustment coefficients, the hot air circulation intensity adjustment value = initial hot air circulation intensity × k1, the hot air circulation frequency adjustment value = initial hot air circulation frequency × k2, and the hot air flow direction angle adjustment value = initial hot air flow direction angle × k3. Then, according to the preset priority order, for example, the hot air circulation intensity adjustment value has the highest priority, the hot air circulation frequency adjustment value has the second highest priority, and the hot air flow direction angle adjustment value has the lowest priority, and they are combined into the first hot air circulation parameter adjustment instruction.

[0145] In a possible implementation manner, in step S14, generating the drying termination instruction based on the morphological matching degree between the surface morphology change data and the preset standard morphology change curve includes:

[0146] In step S141, the real-time surface color data and the real-time surface texture data in the surface morphology change data are extracted, and the real-time surface color data is matched with the preset standard color range to generate a color matching degree score;

[0147] In the embodiments of the present disclosure, during the execution of the second hot air circulation mode, the surface morphology change data of the yuba is continuously monitored, and the real-time surface color data and real-time surface texture data are extracted therefrom. For the real-time surface color data, it will be matched with a preset standard color range. The preset standard color range is preset according to the ideal color range after the yuba is dried. By comparing each color parameter in the real-time surface color data (such as the brightness, saturation, hue, etc. of the color) with the corresponding parameters in the standard color range, the matching degree of each parameter is calculated, and then the color matching degree score is generated by integrating these matching degrees.

[0148] For example, different weights can be assigned to each color parameter, and the matching degree of each parameter is multiplied by the corresponding weight and then added together to obtain the final color matching degree score. For the real-time surface texture data, the similarity calculation is performed between it and a preset standard texture template. The standard texture template is the surface texture image of the yuba when it is dried to the ideal state collected in advance.

[0149] In step S142, the similarity calculation is performed between the real-time surface texture data and a preset standard texture template to generate a texture similarity score;

[0150] In the embodiments of the present disclosure, through image analysis technology, the similarity degree between the texture image corresponding to the real-time surface texture data and the standard texture template is compared to generate a texture similarity score. The method of feature extraction and matching can be adopted to extract the feature points of the texture image, and then the number of matches and the matching quality between these feature points are calculated, and the texture similarity score is generated according to the matching result.

[0151] In step S143, a comprehensive morphology evaluation index is generated according to the color matching degree score and the texture similarity score. When the comprehensive morphology evaluation index reaches a preset morphology compliance threshold, a primary drying termination signal is generated;

[0152] In the embodiments of the present disclosure, a comprehensive morphology evaluation index is generated according to the color matching degree score and the texture similarity score. Different weights can be assigned to the color matching degree score and the texture similarity score respectively, and they are weighted and fused to obtain the comprehensive morphology evaluation index.

[0153] For example, assuming that the weight of the color matching degree score is w1, the weight of the texture similarity score is w2, and w1 + w2 = 1, then the comprehensive morphology evaluation index = w1 × color matching degree score + w2 × texture similarity score. When the comprehensive morphology evaluation index reaches a preset morphology compliance threshold, it indicates that the surface morphology of the yuba has approached or reached the ideal state after drying. At this time, a primary drying termination signal is generated. This signal indicates that the surface morphology of the yuba has met the preliminary drying requirements, but the internal moisture distribution still needs to be further checked.

[0154] In step S144, the internal moisture distribution data of the yuba is synchronously obtained, and the moisture difference coefficient between the moisture content in the central region and the moisture content in the edge region is calculated based on the internal moisture distribution data;

[0155] In the embodiments of the present disclosure, while generating the primary drying termination signal, the internal moisture distribution data of the yuba is synchronously obtained. The moisture content information of different regions inside the yuba can be obtained through non-destructive testing techniques (such as microwave detection, near-infrared detection, etc.).

[0156] Based on this internal moisture distribution data, the moisture difference coefficient between the moisture content in the central region and the moisture content in the edge region is calculated. For example, first calculate the average moisture content in the central region and the edge region respectively, and then divide the difference between the average moisture content in the central region and the average moisture content in the edge region by the average value of the two to obtain the moisture difference coefficient.

[0157] In step S145, when the moisture difference coefficient is lower than the preset moisture balance threshold and the comprehensive morphology evaluation index reaches the morphology compliance threshold, a final drying termination instruction is generated;

[0158] In the embodiments of the present disclosure, when the moisture difference coefficient is lower than the preset moisture balance threshold and the comprehensive morphology evaluation index reaches the morphology compliance threshold, it indicates that not only the surface morphology of the yuba has reached the ideal state, but also the internal moisture distribution is relatively uniform. At this time, a final drying termination instruction will be generated to end the entire drying process.

[0159] In step S146, if the moisture difference coefficient is higher than the moisture balance threshold but the comprehensive morphology evaluation index reaches the morphology compliance threshold, a local drying compensation operation for the corresponding target moisture region is triggered, and a local hot air enhancement instruction is generated based on the moisture difference coefficient to perform directional drying on the target moisture region.

[0160] In the embodiments of the present disclosure, if the moisture difference coefficient is higher than the moisture balance threshold but the comprehensive morphology evaluation index reaches the morphology compliance threshold, it indicates that the surface morphology of the yuba has met the requirements, but the internal moisture distribution is uneven, and there is a target moisture region with a higher moisture content. At this time, a local drying compensation operation for the corresponding target moisture region will be triggered, and a local hot air enhancement instruction is generated according to the moisture difference coefficient. The local hot air enhancement instruction will control the hot air system in the drying oven to make the hot air flow concentratedly towards the target moisture region, perform directional drying on this region to reduce the moisture difference, and make the internal moisture distribution of the yuba more uniform.

[0161] In a possible implementation manner, in step S146, the triggering of the local drying compensation operation for the corresponding target moisture region includes:

[0162] In step S1461, based on the internal moisture distribution data, identify the position coordinates of the target moisture area, and generate a hot air coverage strategy for the target moisture area based on the position coordinates.

[0163] In the embodiments of the present disclosure, based on the obtained internal moisture distribution data of the yuba, identify the position coordinates of the target moisture area with a higher moisture content. The specific position of the target moisture area within the yuba can be determined by performing a spatial analysis on the internal moisture distribution data. Based on these position coordinates, generate a hot air coverage strategy for the target moisture area. For example, according to the position and size of the target moisture area, determine parameters such as the flow direction, wind speed, and coverage range of the hot air, so that the hot air can accurately cover the target moisture area and improve the drying efficiency.

[0164] In step S1462, adjust the angle of the partition flow guiding device in the baking oven so that the hot air flows concentratedly towards the target moisture area, and simultaneously reduce the hot air circulation intensity in other areas.

[0165] In the embodiments of the present disclosure, according to the hot air coverage strategy for the target moisture area, adjust the angle of the partition flow guiding device in the baking oven. The partition flow guiding device can guide the hot air in the baking oven to different areas. By adjusting its angle, the hot air can be concentrated to flow towards the target moisture area. At the same time, in order to avoid over-drying in other areas, the hot air circulation intensity in other areas will be reduced synchronously. For example, by adjusting the fan power in other areas or changing the angle of the flow guiding plate, reduce the hot air flow rate flowing to other areas.

[0166] In step S1463, during the directional drying process, continuously monitor the real-time moisture decline rate of the target moisture area. When the real-time moisture decline rate is lower than a preset compensation rate threshold, trigger a compensation enhancement operation, and generate an additional hot air pulse command based on the rate difference between the real-time moisture decline rate and the compensation rate threshold.

[0167] In the embodiments of the present disclosure, during the process of directional drying of the target moisture area, continuously monitor the real-time moisture decline rate of the target moisture area. The real-time moisture decline rate can be obtained by continuously collecting the moisture content data of the target moisture area, calculating the moisture content difference between adjacent time points, and then dividing by the time interval. When the real-time moisture decline rate is lower than the preset compensation rate threshold, it indicates that the current drying intensity may not be sufficient to quickly reduce the moisture in the target moisture area to an equilibrium level. At this time, a compensation enhancement operation will be triggered. An additional hot air pulse command is generated based on the rate difference between the real-time moisture decline rate and the compensation rate threshold. For example, the larger the rate difference, the higher the hot air pulse intensity and frequency required by the additional hot air pulse command to increase the drying intensity of the target moisture area.

[0168] In step S1464, according to the additional hot air pulse instruction, control the pulse blower to emit a short-time and high-intensity hot air flow to the target moisture area until the moisture content in the target moisture area reaches the moisture balance threshold.

[0169] In the embodiments of the present disclosure, according to the additional hot air pulse instruction, control the pulse blower to emit a short-time and high-intensity hot air flow to the target moisture area. The pulse blower can generate high-intensity hot air in a short time. By adjusting the operating parameters of the pulse blower (such as pulse width, pulse frequency, etc.), the hot air flow is emitted to the target moisture area according to the requirements of the additional hot air pulse instruction. During the process of emitting the hot air flow, continuously monitor the moisture content in the target moisture area until the moisture content in the target moisture area reaches the moisture balance threshold. At this time, stop the operation of the pulse blower to complete the local drying compensation operation.

[0170] In a possible way, after the step of generating the additional hot air pulse instruction based on the rate difference between the real-time moisture decrease rate and the compensation rate threshold, the method further includes:

[0171] Collect the instantaneous energy consumption data during the operation of the pulse blower, and generate a pulse energy consumption evaluation index based on the instantaneous energy consumption data;

[0172] In the embodiments of the present disclosure, during the operation of the pulse blower, collect its instantaneous energy consumption data. The instantaneous energy consumption data can be obtained through a power sensor installed in the pulse blower circuit, and this sensor can measure the power consumption of the pulse blower in real time. Generate a pulse energy consumption evaluation index based on these instantaneous energy consumption data. For example, the average power consumption of the pulse blower over a period of time can be calculated, or the total energy consumption of the pulse blower in each pulse cycle can be calculated, and this is used as the pulse energy consumption evaluation index. This index can reflect the energy consumption situation of the pulse blower during the local drying compensation operation, so as to evaluate and control the energy consumption.

[0173] When the pulse energy consumption evaluation index exceeds the preset energy consumption safety threshold, trigger a pulse interval optimization operation, and generate a pulse interval adjustment coefficient based on the energy consumption difference between the pulse energy consumption evaluation index and the energy consumption safety threshold;

[0174] In the embodiments of the present disclosure, when the pulse energy consumption evaluation index exceeds the preset energy consumption safety threshold, it indicates that the energy consumption of the pulse blower is too high, which may cause damage to the equipment or increase the operating cost. At this time, a pulse interval optimization operation will be triggered. Generate a pulse interval adjustment coefficient based on the energy consumption difference between the pulse energy consumption evaluation index and the energy consumption safety threshold. For example, the greater the energy consumption difference, the greater the pulse interval adjustment coefficient, which means that the interval time between adjacent pulses needs to be increased to reduce the energy consumption of the pulse blower. A linear or non-linear function relationship can be used to calculate the pulse interval adjustment coefficient according to the energy consumption difference.

[0175] Dynamically extend the interval time between adjacent pulses according to the pulse interval adjustment coefficient, and synchronously reduce the single-pulse duration of the pulse blower;

[0176] In the embodiments of the present disclosure, the interval time between adjacent pulses is dynamically extended according to the generated pulse interval adjustment coefficient. For example, the original interval time between adjacent pulses is T1, and according to the pulse interval adjustment coefficient k, the new pulse interval time T2 = T1 × k. At the same time, in order to further reduce energy consumption, the single-pulse duration of the pulse blower is synchronously reduced. The appropriate adjustment ratio of the single-pulse duration can be determined according to the pulse interval adjustment coefficient and the energy consumption situation, and the single-pulse duration is adjusted. In this way, the energy consumption of the pulse blower is reduced on the premise that the target moisture area can continue to be dried.

[0177] Under the combination of the adjusted pulse interval and pulse duration, recalculate the real-time moisture reduction rate. If the real-time moisture reduction rate is still lower than the compensation rate threshold, terminate the local drying compensation operation and generate an artificial intervention request signal.

[0178] In the embodiments of the present disclosure, after adjusting the pulse interval and the single-pulse duration, the real-time moisture reduction rate of the target moisture area is recalculated. According to the method of calculating the real-time moisture reduction rate before, collect the moisture content data of the target moisture area after adjustment, calculate the difference in moisture content between adjacent time points and divide it by the time interval. If the recalculated real-time moisture reduction rate is still lower than the compensation rate threshold, it means that the current adjustment measures still cannot meet the drying requirements of the target moisture area. At this time, the local drying compensation operation will be terminated, and an artificial intervention request signal will be generated. This signal will remind the operator to check and adjust the operating state of the oven, and it may be necessary to further analyze the cause of the problem and take corresponding solutions.

[0179] In a possible way, after generating the artificial intervention request signal, the method further includes:

[0180] Generate a visual alarm interface with the position coordinates, real-time moisture content data of the target moisture area, and the records of the executed compensation operations, and prompt the operator to check the calibration status of the humidity sensor and temperature sensor in the oven through an audible and visual alarm device;

[0181] In the embodiments of the present disclosure, after generating the artificial intervention request signal, the position coordinates, real-time moisture content data of the target moisture area, and the records of the executed compensation operations are integrated to generate a visual alarm interface. The visual alarm interface can display this information in the form of graphs, tables, etc., enabling the operator to intuitively understand the situation of the target moisture area and the compensation operations taken before.

[0182] Meanwhile, the operator is prompted by the acoustic-optic alarm device to check the calibration status of the humidity sensor and the temperature sensor in the baking oven. The acoustic-optic alarm device can emit flashing lights and loud sounds to attract the operator's attention. Since the calibration status of the humidity sensor and the temperature sensor may affect the control and monitoring of the drying process, if the sensors are not accurately calibrated, it may lead to unreasonable setting of drying parameters, thus affecting the drying effect.

[0183] When receiving the feedback signal that the operator confirms the normal calibration of the sensors, restart the local drying compensation operation and add the infrared thermal imaging auxiliary monitoring module;

[0184] In the embodiment of the present disclosure, after the operator checks the calibration status of the humidity sensor and the temperature sensor, if it is confirmed that the sensors are calibrated normally and a confirmation feedback signal is sent to the control system, the local drying compensation operation will be restarted at this time. In order to more accurately monitor the drying situation of the target moisture area, an infrared thermal imaging auxiliary monitoring module will be added. The infrared thermal imaging auxiliary monitoring module can obtain the thermal distribution image of the target moisture area in real time. By analyzing the thermal distribution image, the temperature distribution of the target moisture area can be understood, and thus the evaporation situation of moisture and the drying progress can be indirectly inferred.

[0185] Obtain the real-time thermal distribution image of the target moisture area through the infrared thermal imaging auxiliary monitoring module, and generate the heat flow coverage uniformity correction parameter based on the real-time thermal distribution image;

[0186] In the embodiment of the present disclosure, the real-time thermal distribution image of the target moisture area is obtained through the infrared thermal imaging auxiliary monitoring module. This module uses the principle of infrared radiation to detect the infrared radiation intensity on the surface of the target moisture area and converts it into a thermal distribution image. The heat flow coverage uniformity correction parameter is generated based on these real-time thermal distribution images. For example, analyze the temperature differences in different areas of the thermal distribution image, calculate the standard deviation or variance of the temperature, so as to evaluate the uniformity of the heat flow coverage. Generate the heat flow coverage uniformity correction parameter according to the evaluation result. This parameter can be used to adjust the hot air circulation system in the baking oven to make the heat flow cover the target moisture area more evenly.

[0187] Adjust the angular offset of the partition diversion device and the injection angle of the pulse fan according to the heat flow coverage uniformity correction parameter until the moisture content of the target moisture area reaches the preset termination standard.

[0188] Adjust the angular offset of the partitioned flow guiding device and the injection angle of the pulse fan according to the generated correction parameter for the uniformity of the heat flow coverage. By adjusting the angular offset of the partitioned flow guiding device, the flow direction of the hot air can be changed to make the heat flow more concentrated or dispersed, so as to improve the uniformity of the heat flow coverage. Meanwhile, adjust the injection angle of the pulse fan so that the hot air flow emitted by the pulse fan can more accurately cover the target moisture area. During the adjustment process, continuously monitor the moisture content of the target moisture area until the moisture content of the target moisture area reaches the preset termination standard, at which time the entire drying process ends.

[0189] In a possible implementation manner, in step S13, generating a second hot air circulation parameter adjustment instruction based on the first difference parameter between the temperature stability evaluation index and the first stability threshold, and the second difference parameter between the humidity balance evaluation index and the first balance threshold includes:

[0190] In step S131, input the first difference parameter and the second difference parameter into a pre-constructed multi-stage optimization model, generate a hot air circulation intensity compensation coefficient through the first optimization branch of the multi-stage optimization model, and generate a hot air circulation frequency compensation coefficient through the second optimization branch of the multi-stage optimization model;

[0191] In this embodiment, input the first difference parameter between the temperature stability evaluation index and the first stability threshold and the second difference parameter between the humidity balance evaluation index and the first balance threshold into a pre-constructed multi-stage optimization model. The multi-stage optimization model has multiple optimization branches. The first optimization branch generates a hot air circulation intensity compensation coefficient according to the input data, and the second optimization branch generates a hot air circulation frequency compensation coefficient. The specific calculation logic inside the model is a black box, but after training, it can accurately generate the corresponding compensation coefficients according to the input.

[0192] In step S132, perform a non-linear correction on the hot air circulation intensity adjustment value based on the hot air circulation intensity compensation coefficient to obtain an optimized hot air circulation intensity adjustment value;

[0193] In this embodiment, perform a non-linear correction on the hot air circulation intensity adjustment value based on the hot air circulation intensity compensation coefficient. For example, assume that the hot air circulation intensity compensation coefficient is k4, the hot air circulation intensity adjustment value is V_adj1, and the optimized hot air circulation intensity adjustment value V_adj1_optimized = f(k4, V_adj1), where f is a non-linear function, and the hot air circulation intensity adjustment value is corrected through this non-linear function to achieve a better drying effect.

[0194] In step S133, the adjusted hot air circulation frequency value is segmentally corrected based on the hot air circulation frequency compensation coefficient to obtain an optimized adjusted hot air circulation frequency value;

[0195] In this embodiment, the adjusted hot air circulation frequency value is segmentally corrected based on the hot air circulation frequency compensation coefficient. For example, the adjusted hot air circulation frequency value is divided into different intervals, and different correction methods are adopted according to the hot air circulation frequency compensation coefficient in different intervals. Assume that the adjusted hot air circulation frequency value is T_adj2 and the hot air circulation frequency compensation coefficient is k5. According to the interval where T_adj2 is located, different formulas are used to correct it to obtain an optimized adjusted hot air circulation frequency value T_adj2_optimized.

[0196] In step S134, the surface shrinkage rate data and the surface crack distribution data in the surface morphology change data are synchronously extracted. A shrinkage stability evaluation index is generated according to the surface shrinkage rate data, and a crack density evaluation index is generated according to the surface crack distribution data;

[0197] In this embodiment, the surface shrinkage rate data and the surface crack distribution data in the surface morphology change data are synchronously extracted. For the surface shrinkage rate data, by analyzing the size change of the yuba surface within a certain time, the surface shrinkage rate is calculated, and then a shrinkage stability evaluation index is generated according to the surface shrinkage rate. For the surface crack distribution data, information such as the number and length of cracks on the yuba surface is counted, and the crack density is calculated to generate a crack density evaluation index.

[0198] In step S135, the shrinkage stability evaluation index and the crack density evaluation index are input into the third optimization branch of the multi-stage optimization model to generate a hot air flow direction angle compensation coefficient. Based on the hot air flow direction angle compensation coefficient, the adjusted hot air flow direction angle value is dynamically corrected to obtain an optimized adjusted hot air flow direction angle value;

[0199] In this embodiment, the shrinkage stability evaluation index and the crack density evaluation index are input into the third optimization branch of the multi-stage optimization model, and this branch generates a hot air flow direction angle compensation coefficient according to the input data. For example, assume that the generated hot air flow direction angle compensation coefficient is k6 and the adjusted hot air flow direction angle value is θ_adj3. The optimized adjusted hot air flow direction angle value θ_adj3_optimized = θ_adj3 × k6, and the adjusted hot air flow direction angle value is dynamically corrected in this way.

[0200] In step S136, the optimized adjusted hot air circulation intensity value, the optimized adjusted hot air circulation frequency value, and the optimized adjusted hot air flow direction angle value are combined into the second hot air circulation parameter adjustment instruction according to a preset compensation priority.

[0201] In this embodiment, the optimized hot air circulation intensity adjustment value, the optimized hot air circulation frequency adjustment value, and the optimized hot air flow direction angle adjustment value are combined according to a preset compensation priority to form a second hot air circulation parameter adjustment instruction. For example, the preset compensation priority of the hot air circulation intensity adjustment value is the highest, the hot air circulation frequency adjustment value is the second, and the hot air flow direction angle adjustment value is the lowest, and they are combined together in this priority order.

[0202] In practical applications, this dried bean curd stick drying method can be combined with specific dried bean curd stick production scenarios. For example, on a dried bean curd stick production line, the baking oven usually runs continuously. In such a scenario, the temperature, humidity and other data in multiple baking ovens can be monitored in real time, and these data are input into the above-mentioned drying parameter mapping model and multi-stage optimization model, and the hot air circulation parameters of each baking oven are dynamically adjusted according to the output of the models. For the setting of input data, the real-time temperature data, real-time humidity data, temperature stability evaluation index, humidity balance evaluation index, surface morphology change data, etc. of multiple monitoring areas are used as the input of the model. The outputs of the model are the hot air circulation intensity adjustment value, the hot air circulation frequency adjustment value, the hot air flow direction angle adjustment value, etc., and these output data are directly used to control equipment such as the main fan, auxiliary fan and deflector of the baking oven, so as to achieve precise control of the drying process. In this way, the model and algorithm are closely combined with the specific dried bean curd stick production scenario, improving the quality and efficiency of dried bean curd stick drying.

[0203] In summary, this dried bean curd stick drying method can achieve precise control of the dried bean curd stick drying process by obtaining the real-time data of multiple monitoring areas in the baking oven, generating relevant features and evaluation indexes, and dynamically adjusting the hot air circulation parameters. At the same time, by constructing and training the drying parameter mapping model and the multi-stage optimization model, the accuracy and intelligent level of parameter adjustment are further improved. In practical applications, combined with specific production scenarios, it can effectively improve the quality and efficiency of dried bean curd stick drying.

[0204] This embodiment of the present disclosure also provides a dried bean curd stick drying device applied to a dried bean curd stick drying baking oven. Applied to a dried bean curd stick drying baking oven, see Figure 2 As shown, the device includes:

[0205] An acquisition and generation module 210, configured to acquire the real-time temperature data and real-time humidity data of multiple monitoring areas in the baking oven during the initial drying stage of the dried bean curd stick, and dynamically generate a first hot air circulation parameter adjustment instruction based on the real-time temperature data and the real-time humidity data;

[0206] A control and generation module 220, configured to control the baking oven to execute a first hot air circulation mode based on the first hot air circulation parameter adjustment instruction, continuously collect real-time temperature fluctuation data and real-time humidity gradient data in the first optimization stage during the execution process, generate a temperature stability evaluation index according to the real-time temperature fluctuation data, and generate a humidity balance evaluation index according to the real-time humidity gradient data;

[0207] A trigger and generation module 230, configured to trigger a multi-stage parameter readjustment operation when the temperature stability evaluation index does not reach a preset first stability threshold or the humidity balance evaluation index does not reach a preset first balance threshold, and generate a second hot air circulation parameter adjustment instruction based on a first difference parameter between the temperature stability evaluation index and the first stability threshold and a second difference parameter between the humidity balance evaluation index and the first balance threshold;

[0208] An adjustment and control module 240, configured to control the baking oven to switch to a second hot air circulation mode according to the second hot air circulation parameter adjustment instruction, synchronously monitor the surface morphology change data of the dried bean curd sticks during the execution process, and generate a drying termination instruction to end the drying process based on the morphology matching degree between the surface morphology change data and a preset standard morphology change curve.

[0209] In a possible implementation manner, the acquisition and generation module 210 is configured to:

[0210] Generate an initial stage temperature distribution feature and an initial stage humidity change rate feature based on the real-time temperature data and the real-time humidity data;

[0211] Determine a first matching degree deviation according to the matching degree deviation between the initial stage temperature distribution feature and a preset reference temperature distribution interval, and determine a second matching degree deviation according to the matching degree deviation between the initial stage humidity change rate feature and a preset reference humidity change rate;

[0212] Dynamically generate a first hot air circulation parameter adjustment instruction according to the first matching degree deviation between the initial stage temperature distribution feature and a preset reference temperature distribution interval and the second matching degree deviation between the initial stage humidity change rate feature and a preset reference humidity change rate.

[0213] In a possible implementation manner, the acquisition and generation module 210 is configured to:

[0214] Divide the multiple monitoring areas into a central area group, an edge area group, and a transition area group. Calculate the average value of the real-time temperature data of all monitoring points in the central area group to obtain the central area temperature average value; calculate the variance of the real-time temperature data of all monitoring points in the edge area group to obtain the edge area temperature variance;

[0215] Generate a temperature distribution difference coefficient based on the ratio between the central area temperature average value and the edge area temperature variance, and generate a temperature dynamic association feature by combining the real-time temperature gradient change trend of adjacent monitoring points in the transition area group;

[0216] Perform weighted fusion on the temperature distribution difference coefficient and the temperature dynamic association feature to generate the temperature distribution feature in the initial stage;

[0217] Synchronously extract the real-time humidity data of the multiple monitoring areas, calculate the absolute value of the humidity change within a preset time window for each monitoring area, and generate a maximum humidity change rate and a minimum humidity change rate based on the absolute values of the humidity changes of all monitoring areas;

[0218] Generate a humidity change fluctuation range according to the rate difference between the maximum humidity change rate and the minimum humidity change rate, and generate the initial stage humidity change rate feature by combining the location information of the monitoring area corresponding to the maximum humidity change rate.

[0219] In a possible implementation manner, the obtaining and generating module 210 is configured to:

[0220] Decompose the first matching degree deviation into a temperature distribution deviation component and a temperature gradient deviation component, where the temperature distribution deviation component represents the absolute value of the difference between the central area temperature average value and the central reference temperature in the reference temperature distribution interval, and the temperature gradient deviation component represents the difference ratio between the edge area temperature variance and the edge temperature variance threshold in the reference temperature distribution interval;

[0221] Decompose the second matching degree deviation into a maximum rate deviation component and a fluctuation range deviation component, where the maximum rate deviation component represents the rate ratio between the maximum humidity change rate and the reference humidity change rate, and the fluctuation range deviation component represents the range difference between the humidity change fluctuation range and the preset reference fluctuation range;

[0222] Establish a multi-dimensional adjustment coefficient matrix among the temperature distribution deviation component, the temperature gradient deviation component, the maximum rate deviation component, and the fluctuation range deviation component, and map the multi-dimensional adjustment coefficient matrix to an initial hot air circulation parameter adjustment coefficient set through a pre-trained drying parameter mapping model;

[0223] Generate a hot air circulation intensity adjustment value, a hot air circulation frequency adjustment value, and a hot air flow direction angle adjustment value based on the set of initial hot air circulation parameter adjustment coefficients, and combine the hot air circulation intensity adjustment value, the hot air circulation frequency adjustment value, and the hot air flow direction angle adjustment value in a preset priority order to form the first hot air circulation parameter adjustment instruction.

[0224] In a possible implementation manner, the adjustment and control module 240 is configured to:

[0225] Extract the real-time surface color data and real-time surface texture data from the surface morphology change data, match the real-time surface color data with a preset standard color range, and generate a color matching degree score;

[0226] Calculate the similarity between the real-time surface texture data and a preset standard texture template to generate a texture similarity score;

[0227] Generate a comprehensive morphology evaluation index based on the color matching degree score and the texture similarity score. When the comprehensive morphology evaluation index reaches a preset morphology compliance threshold, generate a primary drying termination signal;

[0228] Synchronously obtain the internal moisture distribution data of the yuba, and calculate the moisture difference coefficient between the moisture content in the central region and the moisture content in the edge region based on the internal moisture distribution data;

[0229] When the moisture difference coefficient is lower than a preset moisture balance threshold and the comprehensive morphology evaluation index reaches the morphology compliance threshold, generate a final drying termination instruction;

[0230] If the moisture difference coefficient is higher than the moisture balance threshold but the comprehensive morphology evaluation index reaches the morphology compliance threshold, trigger a local drying compensation operation for the corresponding target moisture area, and generate a local hot air enhancement instruction based on the moisture difference coefficient to perform directional drying on the target moisture area.

[0231] In a possible implementation manner, the adjustment and control module 240 is configured to:

[0232] Identify the position coordinates of the target moisture area based on the internal moisture distribution data, and generate a hot air coverage strategy for the target moisture area based on the position coordinates;

[0233] Adjust the angle of the partition diversion device in the drying oven so that the hot air is concentrated and flows towards the target moisture area, and simultaneously reduce the hot air circulation intensity in other areas;

[0234] During the directional drying process, the real-time moisture reduction rate of the target moisture area is monitored in real time. When the real-time moisture reduction rate is lower than the preset compensation rate threshold, a compensation enhancement operation is triggered, and an additional hot air pulse instruction is generated based on the rate difference between the real-time moisture reduction rate and the compensation rate threshold;

[0235] According to the additional hot air pulse instruction, the pulse blower is controlled to emit a short-time high-intensity hot air flow to the target moisture area until the moisture content of the target moisture area reaches the moisture balance threshold.

[0236] In a possible implementation manner, the triggering and generating module 230 is configured to:

[0237] Input the first difference parameter and the second difference parameter into a pre-constructed multi-stage optimization model. Generate a hot air circulation intensity compensation coefficient through the first optimization branch of the multi-stage optimization model, and generate a hot air circulation frequency compensation coefficient through the second optimization branch of the multi-stage optimization model;

[0238] Based on the hot air circulation intensity compensation coefficient, perform non-linear correction on the hot air circulation intensity adjustment value to obtain an optimized hot air circulation intensity adjustment value;

[0239] Based on the hot air circulation frequency compensation coefficient, perform segmented correction on the hot air circulation frequency adjustment value to obtain an optimized hot air circulation frequency adjustment value;

[0240] Synchronously extract the surface shrinkage rate data and the surface crack distribution data in the surface morphology change data, generate a shrinkage stability evaluation index according to the surface shrinkage rate data, and generate a crack density evaluation index according to the surface crack distribution data;

[0241] Input the shrinkage stability evaluation index and the crack density evaluation index into the third optimization branch of the multi-stage optimization model to generate a hot air flow direction angle compensation coefficient, and based on the hot air flow direction angle compensation coefficient, perform dynamic correction on the hot air flow direction angle adjustment value to obtain an optimized hot air flow direction angle adjustment value;

[0242] Combine the optimized hot air circulation intensity adjustment value, the optimized hot air circulation frequency adjustment value, and the optimized hot air flow direction angle adjustment value according to the preset compensation priority to form the second hot air circulation parameter adjustment instruction.

[0243] The embodiments of the present disclosure also provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method described in any one of the foregoing embodiments are implemented.

[0244] The embodiments of the present disclosure also provide an electronic device, including:

[0245] A memory, on which a computer program is stored;

[0246] A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of the foregoing embodiments.

[0247] Figure 3 The dried bean curd stick drying device 100 applied to a dried bean curd stick drying and baking oven shown in the figure includes: a processor 1001 and a memory 1003. Among them, the processor 1001 and the memory 1003 are connected, such as being connected through a bus 1002. Optionally, the dried bean curd stick drying device 100 applied to a dried bean curd stick drying and baking oven may further include a communication component 1004, and the communication component 1004 may be used for data interaction between the device 100 and other devices, such as sending and / or receiving data, etc. It should be noted that in actual scheduling, the communication component 1004 is not limited to one, and the structure of the dried bean curd stick drying device 100 applied to a dried bean curd stick drying and baking oven does not constitute a limitation on the embodiments of the present application.

[0248] The processor 1001 may be a CPU (Central Processing Unit, central processor), a general-purpose processor, a DSP (Digital Signal Processor, digital signal processor), an ASIC (Application Specific Integrated Circuit, application-specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute various exemplary logic blocks, modules and circuits described in combination with the disclosure of the present application. The processor 1001 may also be a combination that realizes computing functions, such as a combination including one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0249] The bus 1002 may include a path for transmitting information between the above components. The bus 1002 may be a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus or an EISA (Extended Industry Standard Architecture, extended industry standard architecture) bus, etc. The bus 1002 may be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 3 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0250] The memory 1003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage medium, other magnetic storage devices, or any other medium that can be used to carry or store program code and can be read by a computer, without limitation herein.

[0251] The memory 1003 is used to store program codes for executing the embodiments of the present disclosure, and is controlled to be executed by the processor 1001. The processor 1001 is used to execute the program codes stored in the memory 1003 to implement the steps shown in the embodiment of the bean curd drying method applied to the bean curd drying oven.

[0252] The embodiment of the present disclosure also provides a computer-readable storage medium having program code stored thereon. When the program code is executed by a processor, the steps and corresponding contents of the above-mentioned embodiment of the method for drying bean curd sticks in a bean curd stick drying and baking oven can be implemented.

[0253] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments; within the technical concept of the present disclosure, various changes, modifications, substitutions and variations may be made to these embodiments, and these changes, modifications, substitutions and variations all fall within the protection scope of the present disclosure.

[0254] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any appropriate manner without contradiction, and they should also be regarded as the contents disclosed in this disclosure. In order to avoid unnecessary repetition, this disclosure will not further describe various possible combinations. The technical scope of this application is not limited to the contents in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for drying yuba applied to a yuba drying oven, characterized in that, Applied to a dried bean curd stick drying oven, the method includes: Obtaining real-time temperature data and real-time humidity data of multiple monitoring areas in the drying oven during the initial drying stage of the dried bean curd stick, and dynamically generating a first hot air circulation parameter adjustment instruction based on the real-time temperature data and the real-time humidity data; Based on the first hot air circulation parameter adjustment instruction, controlling the drying oven to execute a first hot air circulation mode, and continuously collecting real-time temperature fluctuation data and real-time humidity gradient data in the first optimization stage during the execution process. According to the real-time temperature fluctuation data, generating a temperature stability evaluation index, and according to the real-time humidity gradient data, generating a humidity balance evaluation index; When the temperature stability evaluation index does not reach a preset first stability threshold or the humidity balance evaluation index does not reach a preset first balance threshold, triggering a multi-stage parameter readjustment operation, and generating a second hot air circulation parameter adjustment instruction based on a first difference parameter between the temperature stability evaluation index and the first stability threshold and a second difference parameter between the humidity balance evaluation index and the first balance threshold; According to the second hot air circulation parameter adjustment instruction, controlling the drying oven to switch to a second hot air circulation mode, and synchronously monitoring the surface morphology change data of the dried bean curd stick during the execution process. Based on the morphological matching degree between the surface morphology change data and a preset standard morphology change curve, generating a drying termination instruction to end the drying process.

2. The method according to claim 1, wherein The dynamically generating a first hot air circulation parameter adjustment instruction based on the real-time temperature data and the real-time humidity data includes: Generating an initial stage temperature distribution feature and an initial stage humidity change rate feature based on the real-time temperature data and the real-time humidity data; Determining a first matching degree deviation according to the matching degree deviation between the initial stage temperature distribution feature and a preset reference temperature distribution range, and determining a second matching degree deviation according to the matching degree deviation between the initial stage humidity change rate feature and a preset reference humidity change rate; Dynamically generating a first hot air circulation parameter adjustment instruction according to the first matching degree deviation between the initial stage temperature distribution feature and the preset reference temperature distribution range and the second matching degree deviation between the initial stage humidity change rate feature and the preset reference humidity change rate.

3. The method according to claim 2, wherein The generating an initial stage temperature distribution feature and an initial stage humidity change rate feature based on the real-time temperature data and the real-time humidity data includes: Dividing the multiple monitoring areas into a central area group, an edge area group, and a transition area group, respectively calculating the mean value of the real-time temperature data of all monitoring points in the central area group to obtain the central area temperature mean value; calculating the variance of the real-time temperature data of all monitoring points in the edge area group to obtain the edge area temperature variance; Generating a temperature distribution difference coefficient based on the ratio between the central area temperature mean value and the edge area temperature variance, and generating a temperature dynamic association feature in combination with the real-time temperature gradient change trend of adjacent monitoring points in the transition area group; Weightedly fuse the temperature distribution difference coefficient and the temperature dynamic association feature to generate the temperature distribution feature in the initial stage; Synchronously extract the real-time humidity data of the multiple monitoring areas, calculate the absolute value of the humidity change in each monitoring area within a preset time window, and generate the maximum humidity change rate and the minimum humidity change rate based on the absolute values of the humidity changes in all monitoring areas; Generate a humidity change fluctuation range according to the rate difference between the maximum humidity change rate and the minimum humidity change rate, and generate the humidity change rate feature in the initial stage in combination with the position information of the monitoring area corresponding to the maximum humidity change rate.

4. The method according to claim 2, wherein Dynamically generate a first hot air circulation parameter adjustment instruction according to the first matching degree deviation between the temperature distribution feature in the initial stage and a preset reference temperature distribution interval, and the second matching degree deviation between the humidity change rate feature in the initial stage and a preset reference humidity change rate, including: Decompose the first matching degree deviation into a temperature distribution deviation component and a temperature gradient deviation component, where the temperature distribution deviation component represents the absolute value of the difference between the average temperature of the central area and the central reference temperature of the reference temperature distribution interval, and the temperature gradient deviation component represents the difference ratio between the temperature variance of the edge area and the edge temperature variance threshold of the reference temperature distribution interval; Decompose the second matching degree deviation into a maximum rate deviation component and a fluctuation range deviation component, where the maximum rate deviation component represents the rate ratio between the maximum humidity change rate and the reference humidity change rate, and the fluctuation range deviation component represents the range difference between the humidity change fluctuation range and a preset reference fluctuation range; Establish a multi-dimensional adjustment coefficient matrix among the temperature distribution deviation component, the temperature gradient deviation component, the maximum rate deviation component, and the fluctuation range deviation component, and map the multi-dimensional adjustment coefficient matrix into an initial hot air circulation parameter adjustment coefficient set through a pre-trained drying parameter mapping model; Generate a hot air circulation intensity adjustment value, a hot air circulation frequency adjustment value, and a hot air flow direction angle adjustment value based on the initial hot air circulation parameter adjustment coefficient set, and combine the hot air circulation intensity adjustment value, the hot air circulation frequency adjustment value, and the hot air flow direction angle adjustment value in a preset priority order to form the first hot air circulation parameter adjustment instruction.

5. The method according to claim 1, wherein Generate a drying termination instruction based on the morphological matching degree between the surface morphology change data and a preset standard morphology change curve, including: Extract the real-time surface color data and real-time surface texture data from the surface morphology change data, match the real-time surface color data with a preset standard color interval, and generate a color matching degree score; Calculate the similarity between the real-time surface texture data and a preset standard texture template to generate a texture similarity score; Generate a comprehensive morphology evaluation index according to the color matching degree score and the texture similarity score, and generate a primary drying termination signal when the comprehensive morphology evaluation index reaches a preset morphology compliance threshold. Synchronously obtain the internal moisture distribution data of the dried bean curd sticks, and calculate the moisture difference coefficient between the moisture content in the central region and the moisture content in the edge region based on the internal moisture distribution data; When the moisture difference coefficient is lower than the preset moisture balance threshold and the comprehensive morphology evaluation index reaches the morphology compliance threshold, generate a final drying termination instruction; If the moisture difference coefficient is higher than the moisture balance threshold but the comprehensive morphology evaluation index reaches the morphology compliance threshold, trigger a local drying compensation operation for the corresponding target moisture area, and generate a local hot air enhancement instruction based on the moisture difference coefficient to perform directional drying on the target moisture area.

6. The method according to claim 5, wherein The triggering of the local drying compensation operation for the corresponding target moisture area includes: Identify the position coordinates of the target moisture area according to the internal moisture distribution data, and generate a hot air coverage strategy for the target moisture area based on the position coordinates; Adjust the angle of the partition diversion device in the drying oven so that the hot air flows concentratedly to the target moisture area, and simultaneously reduce the hot air circulation intensity in other areas; During the directional drying process, continuously monitor the real-time moisture decline rate of the target moisture area. When the real-time moisture decline rate is lower than the preset compensation rate threshold, trigger a compensation enhancement operation, and generate an additional hot air pulse instruction based on the rate difference between the real-time moisture decline rate and the compensation rate threshold; Control the pulse blower to emit a short-time high-intensity hot air flow to the target moisture area according to the additional hot air pulse instruction until the moisture content of the target moisture area reaches the moisture balance threshold.

7. The method according to any one of claims 1-6, characterized in that, The generating of the second hot air circulation parameter adjustment instruction based on the first difference parameter between the temperature stability evaluation index and the first stability threshold and the second difference parameter between the humidity balance evaluation index and the first balance threshold includes: Input the first difference parameter and the second difference parameter into a pre-constructed multi-stage optimization model, generate a hot air circulation intensity compensation coefficient through the first optimization branch of the multi-stage optimization model, and generate a hot air circulation frequency compensation coefficient through the second optimization branch of the multi-stage optimization model; Non-linearly correct the hot air circulation intensity adjustment value based on the hot air circulation intensity compensation coefficient to obtain an optimized hot air circulation intensity adjustment value; Segmentally correct the hot air circulation frequency adjustment value based on the hot air circulation frequency compensation coefficient to obtain an optimized hot air circulation frequency adjustment value; Simultaneously extract the surface shrinkage rate data and the surface crack distribution data in the surface morphology change data, generate a shrinkage stability evaluation index according to the surface shrinkage rate data, and generate a crack density evaluation index according to the surface crack distribution data; Input the shrinkage stability evaluation index and the crack density evaluation index into the third optimization branch of the multi-stage optimization model to generate a hot air flow direction angle compensation coefficient, and dynamically correct the hot air flow direction angle adjustment value based on the hot air flow direction angle compensation coefficient to obtain an optimized hot air flow direction angle adjustment value; Combine the optimized hot air circulation intensity adjustment value, the optimized hot air circulation frequency adjustment value, and the optimized hot air flow direction angle adjustment value into the second hot air circulation parameter adjustment instruction according to a preset compensation priority.

8. A dried yuba device applied to a yuba drying oven, characterized in that, Applied to a dried beancurd stick drying oven, the device includes: An acquisition and generation module, configured to acquire real-time temperature data and real-time humidity data of multiple monitoring areas in the drying oven during the initial drying stage of the dried beancurd stick, and dynamically generate a first hot air circulation parameter adjustment instruction based on the real-time temperature data and the real-time humidity data; A control and generation module, configured to control the drying oven to execute a first hot air circulation mode based on the first hot air circulation parameter adjustment instruction, continuously collect real-time temperature fluctuation data and real-time humidity gradient data in the first optimization stage during the execution process, generate a temperature stability evaluation index according to the real-time temperature fluctuation data, and generate a humidity balance evaluation index according to the real-time humidity gradient data; A trigger and generation module, configured to trigger a multi-stage parameter readjustment operation when the temperature stability evaluation index does not reach a preset first stability threshold or the humidity balance evaluation index does not reach a preset first balance threshold, and generate a second hot air circulation parameter adjustment instruction based on a first difference parameter between the temperature stability evaluation index and the first stability threshold and a second difference parameter between the humidity balance evaluation index and the first balance threshold; An adjustment and control module, configured to control the drying oven to switch to a second hot air circulation mode according to the second hot air circulation parameter adjustment instruction, synchronously monitor the surface morphology change data of the dried beancurd stick during the execution process, and generate a drying termination instruction based on the morphology matching degree between the surface morphology change data and a preset standard morphology change curve to end the drying process.

9. An electronic device, characterized in that, Comprising: A memory, on which a computer program is stored; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method according to any one of claims 1-7.

Images