Preparation method and equipment of instant Chinese yam powder

Abstract

The invention belongs to the technical field of preparation of Chinese yam powder, particularly relates to a preparation method and equipment of instant Chinese yam powder, and aims to solve the problem that the percent of pass is low due to the fact that drying parameters are inconvenient to adjust in real time in existing preparation of the instant Chinese yam powder, the following scheme is provided: the preparation method comprises the following steps: S1, pretreating raw materials, including cleaning, peeling and slicing; s2, performing color protection and curing treatment on the raw materials; s3, the processed raw materials are dried, and drying parameters are adjusted in real time during drying, specifically, multi-source sensor fusion deployment, data acquisition and edge preprocessing, an algorithm core module, a closed-loop control execution system and quality tracing and continuous optimization are included; and S4, crushing the dried raw materials, sieving, and packaging. According to the method, drying parameters can be adjusted in real time, and the product percent of pass can be increased.

Description

Technical Field

[0001] The present invention relates to the technical field of yam powder preparation, in particular to a method and equipment for preparing instant yam powder. Background Art

[0002] Yam flour is a well-known nutritional tonic. It contains protein, carbohydrates, vitamins, fat, choline, amylase, and other essential minerals and trace elements, including iodine, calcium, iron, and phosphorus. Its fiber content creates a feeling of fullness, helping to curb cravings. Furthermore, yam itself is a highly nutritious, low-calorie food, making it safe to consume in large quantities without the risk of weight gain.

[0003] In the prior art, it is not convenient to adjust the drying parameters in real time when preparing instant yam powder, resulting in a low pass rate. Therefore, we propose a method and equipment for preparing instant yam powder to solve the above problem. Summary of the Invention

[0004] The purpose of the present invention is to solve the disadvantage that it is inconvenient to adjust the drying parameters in real time during the preparation of instant yam powder, resulting in a low qualified rate, and to propose a method and equipment for preparing instant yam powder.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for preparing instant yam powder comprises the following steps:

[0007] S1. Pre-process the raw materials, including washing, peeling and slicing;

[0008] S2. Perform color protection and aging treatment on the raw materials;

[0009] S3: Dry the processed raw materials and adjust the drying parameters in real time during drying. This includes: multi-source sensor fusion deployment, data acquisition and edge preprocessing, algorithm core modules, closed-loop control execution system, quality traceability and continuous optimization;

[0010] S4, the dried raw materials are crushed, sieved and packaged;

[0011] S5. Utilization of by-products: Extraction of polysaccharides from yam peel by enzymatic hydrolysis.

[0012] Preferably, the multi-source sensor fusion deployment includes: installing distributed capacitive humidity sensors and infrared moisture sensors to cover all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high temperature and high humidity environments, operating temperatures of -20 to 150°C, deploying differential pressure sensors to monitor ventilation efficiency, with a range of 0 to 10 kPa, and cooperating with laser particle counters to evaluate particle distribution uniformity.

[0013] Preferably, the data acquisition and edge preprocessing include: integrating a Modbus / OPC UA protocol conversion module to achieve unified access to data from heterogeneous devices such as PLCs and temperature controllers, performing data cleaning and feature extraction, the features include moisture fluctuation frequency and temperature gradient, uploading raw data to the cloud through a 5G private network, and deploying mobile edge computing nodes for real-time response.

[0014] Preferably, the core module of the algorithm includes: constructing an LSTM neural network model, the input layer contains 16-dimensional parameters, including temperature, humidity, and air flow velocity, outputs a moisture content prediction value, uses a genetic algorithm to optimize the drying curve, generates PID control parameters every 10 seconds, and establishes a multi-objective constraint function: moisture deviation ≤±0.5% and energy consumption increase <5%.

[0015] Preferably, the closed-loop control execution system includes: when it is detected that the humidity deviation in the local area is greater than 0.3%, it automatically adjusts the speed of the hot air circulation fan and the power of the heating tube, and corrects the air supply angle in real time according to the change in the thickness of the material layer. When the moisture exceeds the standard, it triggers a three-level response, including:

[0016] Level 1: sound and light alarm + automatic air supply;

[0017] Level 2: shutdown self-check + cloud push diagnosis report;

[0018] Level 3: Isolate the problem batch and start MES traceability.

[0019] Preferably, the quality traceability and continuous optimization include: recording 200+ dimensional process data, including timestamps, equipment status, adjustment records, building SPC statistical process control charts, and aggregating multi-production line data to continuously optimize algorithm models.

[0020] Preferably, in said S1, the raw material is pretreated, specifically: after removing the mud and sand, soaking in salt water or white vinegar water to prevent oxidation and blackening, scraping off the skin and digging out the spots, and cutting into 0.2-0.3 cm thin slices.

[0021] Preferably, in S2, the color protection and ripening treatment of the raw materials includes: using a composite color protection liquid: sodium chloride 140mmol / L + vitamin C 3mmol / L + citric acid 125mmol / L, soaking for 2-3 hours to inhibit browning, then rinsing with clean water to remove residues, boiling the color-protected yam slices in boiling water for 6-8 minutes to destroy oxidase and remove mucus, and quickly rinsing with cold water after blanching to terminate the heating process and maintain the color.

[0022] Preferably, in said S5, the enzymatic hydrolysis and extraction of polysaccharides from yam peel comprises: washing fresh yam peel with a high-pressure water jet to remove surface sand and impurities, drying with hot air at 55°C to a moisture content of ≤8%, and processing with an ultrafine grinder to a fineness of 80 mesh, compounding cellulase: pectinase in a ratio of 3:1, adding an enzyme totaling 8% of the substrate mass, a pH 5.0 buffer system, a material-liquid ratio of 1:35, treating with a constant temperature oscillation at 55°C for 120 min, ultrasonic power of 350W, treating in a pulse mode for 60 min, and controlling the temperature of a water bath at 70±2°C to avoid local overheating leading to enzyme failure. The mixture was activated, quickly heated to 95°C and maintained for 10 minutes to inactivate residual enzyme activity, centrifuged at 4000r / min for 15 minutes, and the supernatant was collected. The precipitate was extracted repeatedly twice, 5% activated carbon was added, and the mixture was decolorized with stirring at 70°C for 30 minutes. The Sevage method was used for repeated treatment for 4 times, and the protein removal rate was >81%. 4 times the volume of 95% ethanol was added to the concentrate, and the mixture was allowed to stand at 4°C for 12 hours. The precipitate was collected and treated with a dialysis bag with a molecular weight cutoff of 3500Da for 48 hours to remove small molecular impurities. The mixture was freeze-dried and passed through a 200-mesh sieve to obtain polysaccharide powder.

[0023] The present invention also proposes an instant yam powder preparation device, including an AIoT monitoring platform, which includes a multi-source sensor fusion deployment module, a data acquisition and edge preprocessing module, an algorithm core module, a closed-loop control execution system, and a quality traceability and continuous optimization module:

[0024] Multi-source sensor fusion deployment module, used to install distributed capacitive humidity sensors and infrared moisture sensors, covering all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high temperature and high humidity environments;

[0025] The data acquisition and edge pre-processing module is used to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, perform data cleaning and feature extraction, upload raw data to the cloud via a 5G private network, and deploy mobile edge computing nodes for real-time response.

[0026] The core module of the algorithm is used to build an LSTM neural network model. The input layer contains 16-dimensional parameters, including temperature, humidity, and airflow velocity, and outputs the predicted value of moisture content;

[0027] Closed-loop control execution system, used to automatically adjust the hot air circulation fan speed and heating tube power, and correct the air supply angle in real time according to the thickness of the material layer;

[0028] The quality traceability and continuous optimization module is used to record 200+ dimensions of process data, including timestamps, equipment status, and adjustment records, build SPC statistical process control charts, and aggregate data from multiple production lines to continuously optimize algorithm models.

[0029] Compared with the prior art, the advantages of the present invention are:

[0030] Distributed capacitive humidity sensors and infrared moisture sensors were installed to cover all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high-temperature and high-humidity environments. Data from heterogeneous devices such as PLCs and thermostats was unified, data cleaning and feature extraction were performed, and raw data was uploaded to the cloud via a dedicated 5G network. Mobile edge computing nodes were deployed for real-time response. An LSTM neural network model was constructed, with the input layer containing 16-dimensional parameters, including temperature, humidity, and airflow velocity, and outputting a predicted moisture content value. The model automatically adjusted the speed of the hot air circulation fan and the power of the heating tube, and corrected the air supply angle in real time based on changes in the material layer thickness. The system recorded over 200 dimensions of process data, including timestamps, equipment status, and adjustment records, constructed SPC statistical process control charts, and aggregated data from multiple production lines to continuously optimize the algorithm model.

[0031] The present invention can adjust drying parameters in real time and improve the product qualification rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The present invention provides a flow chart of a method for preparing instant yam powder. DETAILED DESCRIPTION

[0033] The technical solution of this embodiment will be clearly and completely described below in conjunction with the drawings in this embodiment. Obviously, the described embodiment is only a part of this embodiment, rather than all the embodiments.

[0034] Example 1

[0035] Reference Figure 1 A method for preparing instant yam powder comprises the following steps:

[0036] S1. Pre-process the raw materials, including washing, peeling and slicing;

[0037] S2. Perform color protection and aging treatment on the raw materials;

[0038] S3: Dry the processed raw materials and adjust the drying parameters in real time during drying. This includes: multi-source sensor fusion deployment, data acquisition and edge preprocessing, algorithm core modules, closed-loop control execution system, quality traceability and continuous optimization;

[0039] S4, the dried raw materials are crushed, sieved and packaged;

[0040] S5. Utilization of by-products: Extraction of polysaccharides from yam peel by enzymatic hydrolysis.

[0041] In this embodiment, the multi-source sensor fusion deployment includes: installing distributed capacitive humidity sensors and infrared moisture sensors to cover all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high-temperature and high-humidity environments, and operating at a temperature of -20°C. A differential pressure sensor with a range of 1 kPa is deployed to monitor ventilation efficiency and is used in conjunction with a laser particle counter to assess particle distribution uniformity.

[0042] In this embodiment, data acquisition and edge preprocessing include: integrating a Modbus / OPC UA protocol conversion module to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, performing data cleaning and feature extraction, including moisture fluctuation frequency and temperature gradient, uploading raw data to the cloud via a 5G private network, and deploying mobile edge computing nodes for real-time response.

[0043] In this embodiment, the core module of the algorithm includes: building an LSTM neural network model, the input layer contains 16-dimensional parameters, including temperature, humidity, and air flow velocity, outputs a moisture content prediction value, uses a genetic algorithm to optimize the drying curve, generates PID control parameters every 10 seconds, and establishes a multi-objective constraint function: moisture deviation ≤±0.5% and energy consumption increase <5%.

[0044] In this embodiment, the closed-loop control execution system includes: when it detects that the humidity deviation in the local area is greater than 0.3%, it automatically adjusts the speed of the hot air circulation fan and the power of the heating tube, and corrects the air supply angle in real time according to the thickness of the material layer. Excessive moisture triggers a three-level response, including:

[0045] Level 1: sound and light alarm + automatic air supply;

[0046] Level 2: shutdown self-check + cloud push diagnosis report;

[0047] Level 3: Isolate the problem batch and start MES traceability.

[0048] In this embodiment, quality traceability and continuous optimization include: recording 200+ dimensional process data, including timestamps, equipment status, and adjustment records, building SPC statistical process control charts, and aggregating multi-production line data to continuously optimize algorithm models.

[0049] In this embodiment, in S1, the raw material is pretreated, specifically: after removing the mud and sand, soaking it in salt water or white vinegar water to prevent oxidation and blackening, scraping off the skin and digging out the spots, and cutting it into 0.2 cm thin slices.

[0050] In this embodiment, in S2, the raw materials are subjected to color protection and aging treatments including: using a composite color protection solution consisting of 140 mmol / L sodium chloride, 3 mmol / L vitamin C, and 125 mmol / L citric acid, soaking for 2 hours to inhibit browning, followed by rinsing with clean water to remove residues, boiling the color-protected yam slices in boiling water for 6 minutes to destroy oxidases and remove mucus, and quickly rinsing with cold water after blanching to terminate the heating process and maintain the color.

[0051] In this embodiment, in S5, the enzymatic hydrolysis and extraction of polysaccharides from yam peel includes: using a high-pressure water jet to rinse fresh yam peel to remove surface mud and impurities, hot air drying at 55°C to a moisture content of ≤8%, and processing it with an ultrafine grinder to a fineness of 80 mesh, cellulase: pectinase = 3:1 compounding, the total amount of enzyme added is 8% of the substrate mass, pH 5.0 buffer system, material-liquid ratio 1:35, constant temperature oscillation treatment at 55°C for 120min, ultrasonic power 350W, pulse mode treatment for 60min, water bath temperature control 70±2°C, to avoid local overheating leading to enzyme failure The mixture was activated, quickly heated to 95°C and maintained for 10 minutes to inactivate residual enzyme activity, centrifuged at 4000r / min for 15 minutes, and the supernatant was collected. The precipitate was extracted repeatedly twice, 5% activated carbon was added, and the mixture was decolorized with stirring at 70°C for 30 minutes. The Sevage method was used for repeated treatment for 4 times, and the protein removal rate was >81%. 4 times the volume of 95% ethanol was added to the concentrate, and the mixture was allowed to stand at 4°C for 12 hours. The precipitate was collected and treated with a dialysis bag with a molecular weight cutoff of 3500Da for 48 hours to remove small molecular impurities. The mixture was freeze-dried and passed through a 200-mesh sieve to obtain polysaccharide powder.

[0052] This embodiment also proposes an instant yam powder preparation device, including an AIoT monitoring platform. The AIoT monitoring platform includes a multi-source sensor fusion deployment module, a data acquisition and edge preprocessing module, an algorithm core module, a closed-loop control execution system, and a quality traceability and continuous optimization module:

[0053] Multi-source sensor fusion deployment module, used to install distributed capacitive humidity sensors and infrared moisture sensors, covering all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high temperature and high humidity environments;

[0054] The data acquisition and edge pre-processing module is used to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, perform data cleaning and feature extraction, upload raw data to the cloud via a 5G private network, and deploy mobile edge computing nodes for real-time response.

[0055] The core module of the algorithm is used to build an LSTM neural network model. The input layer contains 16-dimensional parameters, including temperature, humidity, and airflow velocity, and outputs the predicted value of moisture content;

[0056] Closed-loop control execution system, used to automatically adjust the hot air circulation fan speed and heating tube power, and correct the air supply angle in real time according to the thickness of the material layer;

[0057] The quality traceability and continuous optimization module is used to record 200+ dimensions of process data, including timestamps, equipment status, and adjustment records, build SPC statistical process control charts, and aggregate data from multiple production lines to continuously optimize algorithm models.

[0058] Example 2

[0059] A method for preparing instant yam powder comprises the following steps:

[0060] S1. Pre-process the raw materials, including washing, peeling and slicing;

[0061] S2. Perform color protection and aging treatment on the raw materials;

[0062] S3: Dry the processed raw materials and adjust the drying parameters in real time during drying. This includes: multi-source sensor fusion deployment, data acquisition and edge preprocessing, algorithm core modules, closed-loop control execution system, quality traceability and continuous optimization;

[0063] S4, the dried raw materials are crushed, sieved and packaged;

[0064] S5. Utilization of by-products: Extraction of polysaccharides from yam peel by enzymatic hydrolysis.

[0065] In this embodiment, the multi-source sensor fusion deployment includes: installing distributed capacitive humidity sensors and infrared moisture sensors to cover all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high-temperature and high-humidity environments, and operating at a temperature of 100°C. A pressure differential sensor with a range of 5 kPa is deployed to monitor ventilation efficiency and is used in conjunction with a laser particle counter to assess particle distribution uniformity.

[0066] In this embodiment, data acquisition and edge preprocessing include: integrating a Modbus / OPC UA protocol conversion module to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, performing data cleaning and feature extraction, including moisture fluctuation frequency and temperature gradient, uploading raw data to the cloud via a 5G private network, and deploying mobile edge computing nodes for real-time response.

[0067] In this embodiment, the core module of the algorithm includes: building an LSTM neural network model, the input layer contains 16-dimensional parameters, including temperature, humidity, and air flow velocity, outputs a moisture content prediction value, uses a genetic algorithm to optimize the drying curve, generates PID control parameters every 10 seconds, and establishes a multi-objective constraint function: moisture deviation ≤±0.5% and energy consumption increase <5%.

[0068] In this embodiment, the closed-loop control execution system includes: when it detects that the humidity deviation in the local area is greater than 0.3%, it automatically adjusts the speed of the hot air circulation fan and the power of the heating tube, and corrects the air supply angle in real time according to the thickness of the material layer. Excessive moisture triggers a three-level response, including:

[0069] Level 1: sound and light alarm + automatic air supply;

[0070] Level 2: shutdown self-check + cloud push diagnosis report;

[0071] Level 3: Isolate the problem batch and start MES traceability.

[0072] In this embodiment, quality traceability and continuous optimization include: recording 200+ dimensional process data, including timestamps, equipment status, and adjustment records, building SPC statistical process control charts, and aggregating multi-production line data to continuously optimize algorithm models.

[0073] In this embodiment, in S1, the raw material is pretreated, specifically: after removing the mud and sand, soaking it in salt water or white vinegar water to prevent oxidation and blackening, scraping off the outer skin and digging out the spots, and cutting it into 0.25 cm thin slices.

[0074] In this embodiment, in S2, the raw materials are subjected to color protection and aging treatments including: using a composite color protection solution consisting of 140 mmol / L sodium chloride, 3 mmol / L vitamin C, and 125 mmol / L citric acid, soaking for 2.5 hours to inhibit browning, followed by rinsing with clean water to remove residues, boiling the color-protected yam slices in boiling water for 7 minutes to destroy oxidases and remove mucus, and quickly rinsing with cold water after blanching to terminate the heating process and maintain the color.

[0075] In this embodiment, in S5, the enzymatic hydrolysis and extraction of polysaccharides from yam peel includes: using a high-pressure water jet to rinse fresh yam peel to remove surface mud and impurities, hot air drying at 55°C to a moisture content of ≤8%, and processing it with an ultrafine grinder to a fineness of 80 mesh, cellulase: pectinase = 3:1 compounding, the total amount of enzyme added is 8% of the substrate mass, pH 5.0 buffer system, material-liquid ratio 1:35, constant temperature oscillation treatment at 55°C for 120min, ultrasonic power 350W, pulse mode treatment for 60min, water bath temperature control 70±2°C, to avoid local overheating leading to enzyme failure The mixture was activated, quickly heated to 95°C and maintained for 10 minutes to inactivate residual enzyme activity, centrifuged at 4000r / min for 15 minutes, and the supernatant was collected. The precipitate was extracted repeatedly twice, 5% activated carbon was added, and the mixture was decolorized with stirring at 70°C for 30 minutes. The Sevage method was used for repeated treatment for 4 times, and the protein removal rate was >81%. 4 times the volume of 95% ethanol was added to the concentrate, and the mixture was allowed to stand at 4°C for 12 hours. The precipitate was collected and treated with a dialysis bag with a molecular weight cutoff of 3500Da for 48 hours to remove small molecular impurities. The mixture was freeze-dried and passed through a 200-mesh sieve to obtain polysaccharide powder.

[0076] This embodiment also proposes an instant yam powder preparation device, including an AIoT monitoring platform. The AIoT monitoring platform includes a multi-source sensor fusion deployment module, a data acquisition and edge preprocessing module, an algorithm core module, a closed-loop control execution system, and a quality traceability and continuous optimization module:

[0077] Multi-source sensor fusion deployment module, used to install distributed capacitive humidity sensors and infrared moisture sensors, covering all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high temperature and high humidity environments;

[0078] The data acquisition and edge pre-processing module is used to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, perform data cleaning and feature extraction, upload raw data to the cloud via a 5G private network, and deploy mobile edge computing nodes for real-time response.

[0079] The core module of the algorithm is used to build an LSTM neural network model. The input layer contains 16-dimensional parameters, including temperature, humidity, and airflow velocity, and outputs the predicted value of moisture content;

[0080] Closed-loop control execution system, used to automatically adjust the hot air circulation fan speed and heating tube power, and correct the air supply angle in real time according to the thickness of the material layer;

[0081] The quality traceability and continuous optimization module is used to record 200+ dimensions of process data, including timestamps, equipment status, and adjustment records, build SPC statistical process control charts, and aggregate data from multiple production lines to continuously optimize algorithm models.

[0082] Example 3

[0083] A method for preparing instant yam powder comprises the following steps:

[0084] S1. Pre-process the raw materials, including washing, peeling and slicing;

[0085] S2. Perform color protection and aging treatment on the raw materials;

[0086] S3: Dry the processed raw materials and adjust the drying parameters in real time during drying. This includes: multi-source sensor fusion deployment, data acquisition and edge preprocessing, algorithm core modules, closed-loop control execution system, quality traceability and continuous optimization;

[0087] S4, the dried raw materials are crushed, sieved and packaged;

[0088] S5. Utilization of by-products: Extraction of polysaccharides from yam peel by enzymatic hydrolysis.

[0089] In this embodiment, the multi-source sensor fusion deployment includes: installing distributed capacitive humidity sensors and infrared moisture sensors to cover all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high-temperature and high-humidity environments, and operating at a temperature of 150°C. A differential pressure sensor with a range of 10 kPa is deployed to monitor ventilation efficiency and is used in conjunction with a laser particle counter to assess particle distribution uniformity.

[0090] In this embodiment, data acquisition and edge preprocessing include: integrating a Modbus / OPC UA protocol conversion module to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, performing data cleaning and feature extraction, including moisture fluctuation frequency and temperature gradient, uploading raw data to the cloud via a 5G private network, and deploying mobile edge computing nodes for real-time response.

[0091] In this embodiment, the core module of the algorithm includes: building an LSTM neural network model, the input layer contains 16-dimensional parameters, including temperature, humidity, and air flow velocity, outputs a moisture content prediction value, uses a genetic algorithm to optimize the drying curve, generates PID control parameters every 10 seconds, and establishes a multi-objective constraint function: moisture deviation ≤±0.5% and energy consumption increase <5%.

[0092] In this embodiment, the closed-loop control execution system includes: when it detects that the humidity deviation in the local area is greater than 0.3%, it automatically adjusts the speed of the hot air circulation fan and the power of the heating tube, and corrects the air supply angle in real time according to the thickness of the material layer. Excessive moisture triggers a three-level response, including:

[0093] Level 1: sound and light alarm + automatic air supply;

[0094] Level 2: shutdown self-check + cloud push diagnosis report;

[0095] Level 3: Isolate the problem batch and start MES traceability.

[0096] In this embodiment, quality traceability and continuous optimization include: recording 200+ dimensional process data, including timestamps, equipment status, and adjustment records, building SPC statistical process control charts, and aggregating multi-production line data to continuously optimize algorithm models.

[0097] In this embodiment, in S1, the raw material is pretreated, specifically: after removing the mud and sand, soaking it in salt water or white vinegar water to prevent oxidation and blackening, scraping off the outer skin and digging out the spots, and cutting it into 0.3 cm thin slices.

[0098] In this embodiment, in S2, the raw materials are subjected to color protection and aging treatments including: using a composite color protection solution consisting of 140 mmol / L sodium chloride, 3 mmol / L vitamin C, and 125 mmol / L citric acid, soaking for 3 hours to inhibit browning, followed by rinsing with clean water to remove residues, boiling the color-protected yam slices in boiling water for 8 minutes to destroy oxidase and remove mucus, and quickly rinsing with cold water after blanching to terminate the heating process and maintain the color.

[0099] In this embodiment, in S5, the enzymatic hydrolysis and extraction of polysaccharides from yam peel includes: using a high-pressure water jet to rinse fresh yam peel to remove surface mud and impurities, hot air drying at 55°C to a moisture content of ≤8%, and processing it with an ultrafine grinder to a fineness of 80 mesh, cellulase: pectinase = 3:1 compounding, the total amount of enzyme added is 8% of the substrate mass, pH 5.0 buffer system, material-liquid ratio 1:35, constant temperature oscillation treatment at 55°C for 120min, ultrasonic power 350W, pulse mode treatment for 60min, water bath temperature control 70±2°C, to avoid local overheating leading to enzyme failure The mixture was activated, quickly heated to 95°C and maintained for 10 minutes to inactivate residual enzyme activity, centrifuged at 4000r / min for 15 minutes, and the supernatant was collected. The precipitate was extracted repeatedly twice, 5% activated carbon was added, and the mixture was decolorized with stirring at 70°C for 30 minutes. The Sevage method was used for repeated treatment for 4 times, and the protein removal rate was >81%. 4 times the volume of 95% ethanol was added to the concentrate, and the mixture was allowed to stand at 4°C for 12 hours. The precipitate was collected and treated with a dialysis bag with a molecular weight cutoff of 3500Da for 48 hours to remove small molecular impurities. The mixture was freeze-dried and passed through a 200-mesh sieve to obtain polysaccharide powder.

[0100] This embodiment also proposes an instant yam powder preparation device, including an AIoT monitoring platform. The AIoT monitoring platform includes a multi-source sensor fusion deployment module, a data acquisition and edge preprocessing module, an algorithm core module, a closed-loop control execution system, and a quality traceability and continuous optimization module:

[0101] Multi-source sensor fusion deployment module, used to install distributed capacitive humidity sensors and infrared moisture sensors, covering all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high temperature and high humidity environments;

[0102] The data acquisition and edge pre-processing module is used to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, perform data cleaning and feature extraction, upload raw data to the cloud via a 5G private network, and deploy mobile edge computing nodes for real-time response.

[0103] The core module of the algorithm is used to build an LSTM neural network model. The input layer contains 16-dimensional parameters, including temperature, humidity, and airflow velocity, and outputs the predicted value of moisture content;

[0104] Closed-loop control execution system, used to automatically adjust the hot air circulation fan speed and heating tube power, and correct the air supply angle in real time according to the thickness of the material layer;

[0105] The quality traceability and continuous optimization module is used to record 200+ dimensions of process data, including timestamps, equipment status, and adjustment records, build SPC statistical process control charts, and aggregate data from multiple production lines to continuously optimize algorithm models.

[0106] Test example

[0107] The instant yam powder preparation method of Example 1 was compared with a traditional method lacking real-time adjustment of drying parameters. The results are shown in the following table:

[0108]

[0109]

[0110] The above is only a preferred specific implementation method of this embodiment, but the protection scope of this embodiment is not limited to this. Any technician familiar with this technical field can make equivalent replacements or changes based on the technical solution and inventive concept of this embodiment within the technical scope disclosed in this embodiment, and they should be covered by the protection scope of this embodiment.

Claims

1. A method for preparing instant yam powder, characterized in that: The following steps are involved: S1. Pre-process the raw materials, including washing, peeling and slicing; S2. Perform color protection and aging treatment on the raw materials; S3: Dry the processed raw materials and adjust the drying parameters in real time during drying. This includes: multi-source sensor fusion deployment, data acquisition and edge preprocessing, algorithm core modules, closed-loop control execution system, quality traceability and continuous optimization; S4, the dried raw materials are crushed, sieved and packaged; S5. Utilization of by-products: Extraction of polysaccharides from yam peel by enzymatic hydrolysis.

2. The method for preparing instant yam powder according to claim 1, wherein: The multi-source sensor fusion deployment includes: installing distributed capacitive humidity sensors and infrared moisture sensors to cover all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high-temperature and high-humidity environments, and operating temperatures between -20 and 150°C. A differential pressure sensor with a range of 0 to 10 kPa is deployed to monitor ventilation efficiency and is used in conjunction with a laser particle counter to assess particle distribution uniformity.

3. The method for preparing instant yam powder according to claim 2, wherein: The data collection and edge preprocessing include: integrating a Modbus / OPC UA protocol conversion module to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, performing data cleaning and feature extraction, including moisture fluctuation frequency and temperature gradient, uploading raw data to the cloud via a 5G private network, and deploying mobile edge computing nodes for real-time response.

4. The method for preparing instant yam powder according to claim 3, wherein: The algorithm's core modules include: building an LSTM neural network model with an input layer containing 16-dimensional parameters, including temperature, humidity, and airflow velocity, outputting a moisture content prediction value, using a genetic algorithm to optimize the drying curve, generating PID control parameters every 10 seconds, and establishing a multi-objective constraint function: moisture deviation ≤±0.5% and energy consumption increase <5%.

5. The method for preparing instant yam powder according to claim 4, wherein: The closed-loop control execution system includes: when it detects that the humidity deviation in the local area is greater than 0.3%, it automatically adjusts the speed of the hot air circulation fan and the power of the heating tube, and corrects the air supply angle in real time according to the thickness change of the material layer. The moisture exceeding the standard triggers a three-level response, including: Level 1: sound and light alarm + automatic air supply; Level 2: shutdown self-check + cloud push diagnosis report; Level 3: Isolate the problem batch and start MES traceability.

6. The method for preparing instant yam powder according to claim 5, wherein: The quality traceability and continuous optimization include: recording 200+ dimensional process data, including timestamps, equipment status, adjustment records, building SPC statistical process control charts, and aggregating multi-production line data to continuously optimize algorithm models.

7. The method for preparing instant yam powder according to claim 6, wherein: In S1, the raw material is pretreated, specifically: after removing the mud and sand, soaking it in salt water or white vinegar water to prevent oxidation and blackening, scraping off the skin and digging out the spots, and cutting it into 0.2-0.3 cm thin slices.

8. The method for preparing instant yam powder according to claim 7, wherein: In S2, the raw material is subjected to color protection and ripening treatment, which includes: using a composite color protection solution: sodium chloride 140mmol / L + vitamin C 3mmol / L + citric acid 125mmol / L, soaking for 2-3 hours to inhibit browning, then rinsing with clean water to remove residues, boiling the color-protected yam slices in boiling water for 6-8 minutes to destroy oxidase and remove mucus, and quickly rinsing with cold water after blanching to terminate the heating process and maintain the color.

9. The method for preparing instant yam powder according to claim 8, wherein: In S5, the enzymatic hydrolysis and polysaccharide extraction of yam peel includes: using a high-pressure water jet to rinse the fresh yam peel to remove surface sand and impurities, hot air drying at 55°C to a moisture content of ≤8%, and ultrafine grinding to 80 mesh fineness, mixing cellulase and pectinase in a ratio of 3:1, with the total amount of enzyme added being 8% of the substrate mass, using a pH 5.0 buffer system and a material-liquid ratio of 1:35, oscillating at a constant temperature of 55°C for 120 minutes, ultrasonic power of 350W, pulse mode treatment for 60 minutes, and controlling the water bath temperature at 70±2°C to avoid local overheating that may cause enzyme inactivation. Rapidly heat to 95°C and maintain for 10 minutes to inactivate residual enzyme activity, centrifuge at 4000r / min for 15 minutes, collect the supernatant, repeat the extraction of the precipitate twice, add 5% activated carbon, stir and decolorize at 70°C for 30 minutes, and repeatedly treat with the Sevage method for 4 times. The protein removal rate is >81%. Add 4 times the volume of 95% ethanol to the concentrate, let it stand at 4°C for 12 hours, collect the precipitate, and treat it with a dialysis bag with a molecular weight cutoff of 3500Da for 48 hours to remove small molecular impurities. After freeze-drying, pass through a 200-mesh sieve to obtain polysaccharide powder.

10. An instant yam powder preparation device, characterized in that: Including AIoT monitoring platform, the AIoT monitoring platform includes multi-source sensor fusion deployment module, data acquisition and edge pre-processing module, algorithm core module, closed-loop control execution system, quality traceability and continuous optimization module: Multi-source sensor fusion deployment module, used to install distributed capacitive humidity sensors and infrared moisture sensors, covering all temperature zones of the drying oven, using dust-proof and corrosion-resistant packaging technology to adapt to high temperature and high humidity environments; The data acquisition and edge pre-processing module is used to achieve unified access to data from heterogeneous devices such as PLCs and thermostats, perform data cleaning and feature extraction, upload raw data to the cloud via a 5G private network, and deploy mobile edge computing nodes for real-time response. The core module of the algorithm is used to build an LSTM neural network model. The input layer contains 16-dimensional parameters, including temperature, humidity, and airflow velocity, and outputs the predicted value of moisture content; Closed-loop control execution system, used to automatically adjust the hot air circulation fan speed and heating tube power, and correct the air supply angle in real time according to the thickness of the material layer; The quality traceability and continuous optimization module is used to record 200+ dimensions of process data, including timestamps, equipment status, and adjustment records, build SPC statistical process control charts, and aggregate data from multiple production lines to continuously optimize algorithm models.

Images