Data analysis method, system and device and electronic equipment

Abstract

The invention provides a data analysis method, system and device and electronic equipment. The method comprises the following steps: acquiring and storing spectral data of goods at a monitoring moment through a quantum dot spectrum device; acquiring environment data at a monitoring moment; inputting the data into a data analysis model to output an analysis result of the stored goods; the analysis result comprises the content of the target component in the stored goods and the freshness of the stored goods at the monitoring moment, and / or the predicted content of the target component and the predicted freshness of the stored goods at the future preset time. According to the mode, a quantum dot spectrum device is adopted to collect and store spectrum data of goods at the monitoring moment in a high-sensitivity mode, and analysis results corresponding to the stored goods are output through a data analysis model. The content of the target component in the stored goods and the freshness of the stored goods can be accurately detected and monitored, and the potential deterioration risk can be identified in time.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a data analysis method, system, device and electronic device. Background Art

[0002] In modern society, with the continuous progress of technology and the improvement of people's living standards, the storage management problems of food safety, drug quality, raw materials, etc. have attracted increasing attention. During the storage process, affected by environmental conditions such as temperature, humidity, and light, food, drugs, raw materials, etc. may deteriorate or be contaminated. Traditional storage monitoring means are difficult to accurately detect and monitor the content of specified components (such as moisture, fatty acids, etc.) in the stored goods and the freshness of the stored goods, thus it is difficult to timely identify potential deterioration risks. Summary of the Invention

[0003] The purpose of the present invention is to provide a data analysis method, system, device and electronic device to accurately detect and monitor the content of specified components in stored goods and the freshness of stored goods, and timely identify potential deterioration risks.

[0004] A data analysis method provided by the present invention includes: collecting spectral data of stored goods at the monitoring moment through a quantum dot spectroscopy device, wherein the quantum dot spectroscopy device includes multiple channels, each channel corresponds to a different quantum dot region, and each quantum dot region has different optical characteristics; obtaining environmental data at the monitoring moment; inputting the spectral data at the monitoring moment and the environmental data at the monitoring moment into a pre-trained data analysis model, preprocessing the spectral data at the monitoring moment through the data analysis model, and extracting spectral feature information of the target band from the processed spectral data, so as to output an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment; wherein, the target band is used to analyze the target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods.

[0005] Further, the data analysis model is trained in the following manner: Obtain a sample data set within a preset time period before a preset time point; wherein, the sample data set includes data samples at multiple time points, each data sample includes historical spectral data corresponding to the time point and corresponding historical environmental data, and is labeled with the standard content and standard freshness of the specified component corresponding to the time point; wherein, the specified component includes at least a part of the target component; Input the sample data set into the initial model to output a prediction result corresponding to the stored goods through the initial model; wherein, the prediction result includes: at the preset time point, the first content of the specified component in the stored goods, the first freshness of the stored goods, and at the first historical time in the future, the predicted second content of the specified component in the stored goods and the predicted second freshness of the stored goods; Calculate a loss value based on the prediction result, the standard content and standard freshness corresponding to the preset time point, and the standard content and standard freshness corresponding to the first historical time, update the initial model based on the loss value, and repeat the step of obtaining the sample data set within a preset time period before a preset time point until the loss value meets the set threshold to obtain a trained data analysis model.

[0006] Further, the step of inputting the sample data set into the initial model to output a prediction result corresponding to the stored goods through the initial model includes: Using the method of principal component analysis to preprocess the historical spectral data at multiple time points in the sample data set to obtain the preprocessed historical spectral data; Extract the specified spectral feature information of the specified band from the preprocessed historical spectral data; wherein, the specified band is used to analyze the specified component; If there is a linear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample data set, use the partial least squares regression algorithm to output a prediction result corresponding to the stored goods; If there is a non-linear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample data set, use the support vector machine algorithm to output a prediction result corresponding to the stored goods.

[0007] Further, the method further includes: Collecting the current environmental temperature of the storage environment through a temperature sensor; Obtaining a preset temperature range corresponding to the analysis result; When the current environmental temperature is lower than the lower limit value of the preset temperature range, controlling the thermostat to turn on to raise the environmental temperature of the storage environment to meet the preset temperature range; When the current environmental temperature is higher than the upper limit value of the preset temperature range, controlling the thermostat to turn on to lower the environmental temperature of the storage environment to meet the preset temperature range.

[0008] Further, the method further includes: collecting the current ambient humidity of the storage environment through a humidity sensor; obtaining a preset humidity range corresponding to the analysis result; when the current ambient humidity is lower than the lower limit value of the preset humidity range, controlling the humidity control device to turn on to increase the ambient humidity of the storage environment to meet the preset humidity range; when the current ambient humidity is higher than the upper limit value of the preset humidity range, controlling the humidity control device to turn on to decrease the ambient humidity of the storage environment to meet the preset humidity range.

[0009] Further, the method further includes: collecting the current air circulation speed of the storage environment through an air flow sensor; obtaining a preset air circulation speed range corresponding to the analysis result; when the current air circulation speed is lower than the lower limit value of the preset air circulation speed range, controlling the air flow regulating device to turn on to increase the air circulation speed; when the current air circulation speed is higher than the upper limit value of the preset air circulation speed range, controlling the air flow regulating device to turn off to decrease the air circulation speed.

[0010] Further, the method further includes: sending and saving the analysis result to a cloud device; receiving a first instruction from the cloud device to perform an operation corresponding to the first instruction according to the first instruction; and / or, the method further includes: displaying the analysis result and the status information of the storage environment through a terminal device; where the status information includes: ambient temperature, ambient humidity, and air circulation speed; receiving a second instruction from the terminal device to perform an operation corresponding to the second instruction according to the second instruction.

[0011] A data analysis system provided by the present invention, the system includes: a bin body, a quantum dot spectroscopy device, a data processing unit, an environmental monitoring sensor, a temperature regulating device, a humidity regulating device, an air flow regulating device, and a communication module, where:

[0012] The quantum dot spectroscopy device includes multiple channels, each corresponding to a different quantum dot region, and each quantum dot region has different optical properties. The quantum dot spectroscopy device is located above the inner side of the bin body. The quantum dot spectroscopy device includes: a light source and a quantum dot spectroscopy sensor. The light source is used to emit light within a preset wavelength range, and the quantum dot spectroscopy sensor is used to obtain spectral data of the light reflected and / or scattered after the light emitted by the light source irradiates the surface of the stored goods; an environmental monitoring sensor is used to collect environmental data at the monitoring time within the bin body; a data processing unit is used to receive the spectral data at the monitoring time and the environmental data at the monitoring time, preprocess the spectral data at the monitoring time through a pre-trained data analysis model, extract spectral feature information of the target wavelength band from the processed spectral data, so as to output an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring time; wherein, the environmental data at the monitoring time includes: environmental temperature, environmental humidity, and air flow velocity; the target wavelength band is used to analyze the target component; the analysis result includes: at the monitoring time, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods. The data processing unit is also used to control the temperature regulating device and / or the humidity regulating device and / or the air flow regulating device according to the analysis result; the communication module is data-connected to the data processing unit and is used to transmit at least the analysis result and the operation instruction.

[0013] Further, the environmental monitoring sensor includes: a temperature sensor, a humidity sensor, and an air flow sensor arranged inside the bin body; and / or, the temperature regulating device includes: a thermostat arranged on the inner side of the bin body; and / or, the humidity regulating device includes: a humidity control device arranged on the inner side of the bin body; and / or, the storage system further includes a terminal device signal-connected to the communication module.

[0014] A data analysis device provided by the present invention includes: a quantum dot spectroscopy device for collecting spectral data of stored goods at the monitoring time; an acquisition module for acquiring environmental data at the monitoring time; an output module for inputting the spectral data at the monitoring time and the environmental data at the monitoring time into a pre-trained data analysis model, preprocessing the spectral data at the monitoring time through the data analysis model, and extracting spectral feature information of the target wavelength band from the processed spectral data, so as to output an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring time; wherein, the target wavelength band is used to analyze the target component; the analysis result includes: at the monitoring time, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods.

[0015] An electronic device provided by the present invention includes a processor and a memory. The memory stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions to implement the data analysis method of any one of the above.

[0016] The data analysis method, system, device and electronic device provided by the present invention collect spectral data of stored goods at the monitoring moment through a quantum dot spectroscopy device; obtain environmental data at the monitoring moment; input the spectral data at the monitoring moment and the environmental data at the monitoring moment into a pre-trained data analysis model, and the data analysis model preprocesses the spectral data at the monitoring moment, extracts spectral feature information of the target band from the processed spectral data, and outputs an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment; wherein, the target band is used to analyze the target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods. This method can collect spectral data of stored goods at the monitoring moment with high sensitivity by using a quantum dot spectroscopy device, and output an analysis result corresponding to the stored goods through a data analysis model. This way of combining a quantum dot spectroscopy device and a data analysis model can accurately detect and monitor the content of the target component in the stored goods and the freshness of the stored goods, and timely identify potential deterioration risks. Brief Description of the Drawings

[0017] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a flowchart of a data analysis method provided by an embodiment of the present invention;

[0019] Figure 2 It is a schematic diagram of a quantum dot spectroscopy storage management system provided by an embodiment of the present invention;

[0020] Figure 3 It is a schematic structural diagram of a data analysis device provided by an embodiment of the present invention;

[0021] Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Detailed Embodiments

[0022] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are some of the embodiments of the present invention, rather than all of them. 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 scope of protection of the present invention.

[0023] In modern society, with the continuous progress of technology and the improvement of people's living standards, issues such as food safety, drug quality, and material management have received increasing attention. The challenges and requirements faced in each related field are described below.

[0024] 1. Challenges in food safety and quality management: Globally, food safety incidents occur frequently, and consumers' attention to food quality is increasing. During the storage, transportation, and sales of food, it is easily affected by environmental factors such as temperature and humidity, leading to spoilage or contamination. Therefore, how to monitor the freshness and safety of food in real time has become an urgent problem to be solved.

[0025] 2. Complexity of drug management: During the production and storage of drugs, they are greatly affected by environmental conditions (such as temperature, humidity, light, etc.), and the efficacy may be reduced or even potential safety hazards may occur. The active ingredients of drugs need to be stored under specific conditions, and traditional monitoring methods often cannot provide timely and accurate data support. There is an urgent need to introduce advanced monitoring technologies.

[0026] 3. Intelligent requirements for raw material management: With the rapid development of manufacturing and industrial production, the management of raw materials is becoming increasingly important. Enterprises hope to improve the quality control and inventory management of raw materials through intelligent means, reduce waste, and improve production efficiency.

[0027] 4. Importance of environmental protection and sustainable development: With the aggravation of global climate change and environmental pollution problems, governments and enterprises around the world are paying more and more attention to environmental monitoring and protection. Traditional monitoring methods often have problems such as slow response and insufficient accuracy. There is an urgent need for new technologies to achieve efficient and accurate environmental monitoring.

[0028] 5. Diversification of market demands: With the increasing demand of consumers for personalized and customized services, the traditional storage management mode cannot meet the diverse needs of the market. Enterprises need more flexible and intelligent management systems to cope with the ever-changing market environment.

[0029] However, traditional storage monitoring means are difficult to accurately detect and monitor the content of specified components (such as moisture, fatty acids, etc.) in stored goods and the freshness of stored goods, thus making it difficult to solve the actual problems faced by the above-mentioned related fields. Based on this, the embodiments of the present invention provide a data analysis method, system, device and electronic device, and this technology can be applied to applications that require intelligent management of stored goods and storage environments.

[0030] For the convenience of understanding this embodiment, first, a data analysis method disclosed in the embodiments of the present invention will be introduced. As Figure 1 shown, the method includes the following steps:

[0031] Step S102, collecting spectral data of stored goods at the monitoring moment through a quantum dot spectral device, where the quantum dot spectral device includes multiple channels, each channel corresponds to a different quantum dot region, and each quantum dot region has different optical characteristics;

[0032] Step S104, obtaining environmental data at the monitoring moment.

[0033] The above-mentioned quantum dot spectral device utilizes the unique optical characteristics of quantum dot materials to accurately identify and analyze the spectral characteristics of different substances within a specific wavelength range; quantum dots have a small volume, high sensitivity, and a wide spectral response range, making them exhibit excellent performance when detecting trace components; the quantum dot spectral device includes a light source and a quantum dot spectral sensor, and the quantum dot spectral sensor usually includes multiple channels, each channel is correspondingly provided with a different quantum dot region, and each quantum dot region has different optical characteristics; the light source can emit light of a specific wavelength (for example, 200 - 5000 nm), which is irradiated onto the surface of the stored goods and then reflected and / or scattered back to the quantum dot spectral sensor to collect the spectral data of the stored goods; the above-mentioned stored goods refer to the goods stored during the storage process, for example, they can be food, medicine, raw materials, etc.; the above-mentioned spectral data reflects the reflection of the stored goods on light of different wavelengths under light irradiation; the environmental data at the monitoring moment can include data such as the temperature, humidity, and air circulation speed of the storage environment. In actual implementation, when it is necessary to detect the components in the stored goods, the stored goods can be scanned non - contactingly through the quantum dot spectral device to collect the spectral data of the stored goods at the monitoring moment, and the environmental data at the monitoring moment can be collected through a temperature sensor, a humidity sensor, etc. The acquisition frequencies of the above - mentioned quantum dot spectral device, temperature sensor, and humidity sensor can be set according to needs, and they can be collected continuously or controlled to start collecting when needed.

[0034] Step S106: Input the spectral data at the monitoring moment and the environmental data at the monitoring moment into a pre-trained data analysis model. The data analysis model preprocesses the spectral data at the monitoring moment, extracts the spectral feature information of the target band from the processed spectral data, and outputs the analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment. Among them, the target band is used to analyze the target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods.

[0035] The above preprocessing methods may include smoothing processing, baseline correction processing, normalization processing, etc. Among them, smoothing processing is mainly used to eliminate noise, errors, etc. in the spectral data; baseline correction processing is mainly used to remove the background signal in the spectral data; normalization processing is mainly used to adjust the spectral data to a specific range or distribution. The above target band is used to analyze the target component, which can also be understood as that the target component will have obvious absorption or reflection in the target band. Usually, the target bands corresponding to different target components may be the same or different. For example, if the target component is moisture, its corresponding target band is in the near-infrared (Near Infrared, NIR) region, and the band range is about 700nm - 2500nm; if the target component is pigment, its corresponding target band is in the visible light region, and the band range is about 400nm - 700nm, etc. The above spectral feature information may include the peak value, peak position, peak area, etc. of the target band, and these indicators usually reflect the change of the component content. For example, some molecular structures will produce significant absorption peaks at specific wavelengths, and tracking these peaks can help identify the dynamic changes of the components.

[0036] The content of the above target component can be expressed as the percentage of the target component in the stored goods. For example, the moisture content in the stored goods is 10%, etc. The freshness of the above stored goods represents the freshness degree of the stored goods. For example, it can be expressed by labels such as "fresh", "critical", "spoiled", etc. The above preset future time can be a time point after the monitoring moment or a time period after the monitoring moment. For example, it can be a three-day time period starting from the monitoring moment, or the time point of 10 o'clock on the third day in the future, etc. If the preset future time is a time point, the above predicted content represents the content of the target component in the stored goods at this future preset time; the predicted freshness represents the freshness of the stored goods at this future preset time. If the preset future time is a time period, the above predicted content can reflect the change trend of the content of the target component in the stored goods during this preset future time period; the predicted freshness can reflect the change trend of the freshness of the stored goods during this preset future time period.

[0037] In actual implementation, after obtaining the spectral data of the stored goods at the monitoring moment and the environmental data at the monitoring moment, the spectral data at the monitoring moment and the environmental data at the monitoring moment can be input into a pre-trained data analysis model. The data analysis model can preprocess the received spectral data, then extract spectral feature information such as the peak value, peak position, and peak area of the target band from it. Finally, the analysis result corresponding to the stored goods can be output according to the extracted spectral feature information and the environmental data at the monitoring moment. For example, at the monitoring moment, the moisture content in the stored goods is 10%, and the freshness of the stored goods is "critical". Three days later, the predicted moisture content in the stored goods drops to 5%, and the predicted freshness of the stored goods becomes "spoiled", etc.

[0038] The above data analysis method collects the spectral data of the stored goods at the monitoring moment through a quantum dot spectroscopy device; obtains the environmental data at the monitoring moment; inputs the spectral data at the monitoring moment and the environmental data at the monitoring moment into a pre-trained data analysis model, preprocesses the spectral data at the monitoring moment through the data analysis model, and extracts spectral feature information of the target band from the processed spectral data, so as to output the analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment; wherein, the target band is used to analyze the target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods. This method can collect the spectral data of the stored goods at the monitoring moment with high sensitivity by using a quantum dot spectroscopy device, and output the analysis result corresponding to the stored goods through the data analysis model. This way of combining the quantum dot spectroscopy device and the data analysis model can accurately detect and monitor the content of the target component in the stored goods and the freshness of the stored goods, and timely identify potential spoilage risks.

[0039] The following is an example to illustrate the main components to be monitored in storage management and the spectral correspondence:

[0040] In the field of food storage:

[0041] 1. Moisture content: The band range is in the near-infrared (NIR) region (about 700 - 2500 nm);

[0042] 2. Acidity (pH value): The band range is usually at 4000 - 400 cm -1 (about 2500 - 25000 nm).

[0043] 3. Volatile compounds: The main band range is at 800 - 4000 cm -1 (about 2500 - 12500 nm).

[0044] 4. Fatty acid composition: The wavelength range is in the near-infrared (NIR) region (about 700 - 2500 nm), which can be used to analyze the oxidation state of oils and fats.

[0045] 5. Pigment change: The wavelength range is in the visible light region (400 - 700 nm), analyzing the absorption characteristics of pigments (such as carotenoids, chlorophyll, etc.).

[0046] 6. Microbial content: It can be detected by fluorescence in combination with specific wavelength bands in the near-infrared (NIR) (about 700 - 2500 nm) and visible light (400 - 700 nm).

[0047] In the field of drug storage:

[0048] 1. Active ingredient content: The wavelength range is in the ultraviolet (UV) region (about 200 - 400 nm), which is commonly used to analyze the active ingredients in drugs.

[0049] 2. Impurities and degradation products: The wavelength ranges of infrared spectroscopy (FTIR) and ultraviolet-visible spectroscopy (200 - 800 nm) can be used to detect impurities.

[0050] 3. Humidity: The wavelength range is in the near-infrared (NIR) region (about 700 - 2500 nm), which can be used to detect moisture.

[0051] 4. pH value: Infrared spectroscopy (FTIR) can be used to analyze the relevant chemical components, and the wavelength range is usually 4000 - 400 cm -1 (about 2500 - 25000 nm).

[0052] In the field of raw material storage:

[0053] 1. Moisture content: The wavelength range is in the near-infrared (NIR) region (about 700 - 2500 nm)

[0054] 2. Chemical composition: The wavelength ranges of infrared spectroscopy (FTIR) and ultraviolet-visible spectroscopy (200 - 800 nm) can be used to identify and quantify chemical components.

[0055] 3. Impurity content: The wavelength ranges of infrared spectroscopy (FTIR) and ultraviolet-visible spectroscopy (200 - 800 nm) can be used to analyze the impurity components.

[0056] 4. Color change: The wavelength range is in the visible light region (400 - 700 nm).

[0057] 5. Microbial content: It can be detected by fluorescence in combination with the near-infrared (NIR) and visible light (400 - 700 nm) in specific wavelength bands.

[0058] 6. Volatile substances: The main wavelength range is in 800 - 4000 cm -1 (approx. 2500 - 12500 nm).

[0059] In addition, in the above applications in various fields, for the water content, there are usually significant absorption peaks in the near-infrared region (1300–2500 nm). The wavelength range related to water can be selected, and the data in this range can be processed. For the fatty acid content, the characteristic peaks of carbon-hydrogen bond vibration can be focused on. For example, the data in the range of 1700–1800 nm in the near-infrared region can be processed, etc.

[0060] The training method of the data analysis model will be described below. Specifically, the data analysis model is trained through the following steps 20 to 22:

[0061] Step 20, obtain a sample data set within a preset time period before a preset time point; where the sample data set includes data samples at multiple time points, each data sample includes the historical spectral data and the corresponding historical environmental data at the corresponding time point, and is marked with the standard content and standard freshness of the specified component at the corresponding time point; where the specified component includes at least a part of the target component;

[0062] The above sample data set can be a data set determined according to the data samples at multiple historical time points; the above preset time point can be a historical time point selected as needed; the above preset time period can be set according to actual needs. For example, it can be three days before the preset time point, etc.; the above standard content can be understood as the actual content of the specified component in the stored goods at the corresponding time point; the above standard freshness can be understood as the actual freshness of the stored goods at the corresponding time point. In actual implementation, a trend model can be established by collecting historical spectral data and historical environmental data (such as temperature, humidity, etc.) of a time series for modeling. Common modeling methods include: (1) Time series analysis: For example, methods such as moving average method, ARIMA (Autoregressive Integrated Moving Average Model) are used to predict the change trend; (2) Machine learning model: A prediction model (such as support vector machine regression, random forest or neural network) is trained to capture the relationship between environmental conditions and spectral features.

[0063] When a data analysis model needs to be trained, a sample data set needs to be constructed first, which can include historical spectral data and historical environmental data at multiple time points, so that during the training process, the relationship between the two can be learned to predict the future spectral change trend, and then the content of the specified component and the freshness of the stored goods can be predicted. In practical applications, in order to ensure the accuracy of the analysis results of the data analysis model obtained by subsequent training, it is usually required that the specified components included in the sample data set cover all the target components that need to be analyzed in practical applications, or cover at least some of the target components.

[0064] Step 21, input the sample data set into the initial model to output the prediction results corresponding to the stored goods through the initial model; wherein, the prediction results include: at a preset time point, the first content of the specified component in the stored goods, the first freshness of the stored goods, and at the first historical time in the future, the predicted second content of the specified component in the stored goods and the predicted second freshness of the stored goods;

[0065] The above-mentioned first historical time in the future can be a time point after the preset time point, or a time period after the preset time point. For example, it can be a three-day time period starting from the preset time point, or the time point of 10 o'clock on the third day in the future, etc.; if the first historical time in the future is a time point, the above-mentioned predicted second content represents the content of the specified component in the stored goods at this first historical time in the future; the predicted second freshness represents the freshness of the stored goods at this first historical time in the future; if the first historical time in the future is a time period, the above-mentioned predicted second content can reflect the change trend of the content of the specified component in the stored goods during this first historical time in the future; the predicted second freshness can reflect the change trend of the freshness of the stored goods during this first historical time in the future.

[0066] This step 21 can be implemented through the following steps 210 to 212:

[0067] Step 210, adopt the method of principal component analysis to preprocess the historical spectral data at multiple time points in the sample data set to obtain the processed historical spectral data; extract the specified spectral feature information of the specified band from the processed historical spectral data; wherein, the specified band is used to analyze the specified component;

[0068] The above-mentioned Principal Component Analysis (PCA) can be used to reduce the dimension of historical spectral data, extract the main components, thereby analyzing the main change trends in the historical spectral data, extracting the most representative features from the high-dimensional spectral data, and reducing data redundancy; the above-mentioned specified bands are used to analyze the specified components, and usually different specified components correspond to different specified bands; the above-mentioned specified spectral feature information may include the peak value, peak position, peak area, etc. of the specified band.

[0069] In actual implementation, the historical spectral data at the above-mentioned multiple time points can be represented in matrix form to obtain a historical spectral data matrix. First, preprocessing such as standardization or normalization can be performed on the historical spectral data at multiple time points in the historical spectral data matrix to eliminate the deviation caused by the amplitude difference between different bands. Then, calculate the covariance matrix of the preprocessed historical spectral data matrix, perform eigen-decomposition on this covariance matrix, extract the main components (eigenvectors), select the main components corresponding to the top k largest eigenvalues, project the historical spectral data onto these main components to obtain a low-dimensional feature matrix; based on the low-dimensional feature matrix, extract the specified spectral feature information of the specified band; the specific processing method for principal component analysis can refer to related technologies and will not be elaborated here.

[0070] Step 211, if there is a linear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample dataset, use the partial least squares regression algorithm to output the prediction result corresponding to the stored goods;

[0071] The above-mentioned partial least squares regression algorithm (PLSR) is a commonly used algorithm in spectral analysis. It can predict the actual content of components by establishing a regression model between band absorption values and component contents. In actual implementation, if the specified component is a simple component (which can be understood as a component affected by single or very few factors, such as moisture content, fatty acid content, etc.), then there is usually a linear relationship between the specified spectral feature information and the content of the specified component. In this case, the partial least squares regression algorithm can be used to output the prediction result corresponding to the stored goods. For example, taking the prediction of the content of the specified component as an example, the component content data can be set as the target variable first. The partial least squares regression algorithm is used to reduce the dimension of the above-mentioned specified spectral feature information and the target variable, extract the most relevant features between the input data and the output variable, and find the regression coefficient by optimizing the objective function to minimize the gap between the target variable and the predicted value; verify the goodness of fit of the model corresponding to the partial least squares regression algorithm and optimize the model parameters (such as the number of features); finally, output the model corresponding to the partial least squares regression algorithm for predicting the specific component content according to the spectral data; the specific processing method of the partial least squares regression algorithm can refer to the related technology and will not be elaborated here.

[0072] Step 212, if there is a non-linear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample dataset, use the support vector machine algorithm to output the prediction result corresponding to the stored goods.

[0073] The above-mentioned support vector machine algorithm (SVM) is more suitable for the cost analysis of complex samples. When the non-linear characteristics of the data are obvious, SVM regression can better capture the relationship between spectral data and component content. In actual implementation, if the specified component is a complex component (which can be understood as a component easily affected by multiple factors), then there is usually a non-linear relationship between the specified spectral feature information and the content of the specified component. In this case, the support vector machine algorithm can be used to output the prediction result corresponding to the stored goods. For example, taking the prediction of the content of the specified component as an example, the component content data can be set as the target variable first, and a suitable kernel function (such as Gaussian kernel, linear kernel or polynomial kernel) is selected to map the original data to a high-dimensional space. When training the model corresponding to the support vector machine algorithm, the support vectors and the regression plane can be found by optimizing the loss function to ensure that the regression error is within a certain tolerance; use cross-validation to determine the penalty parameter and the kernel function parameter; finally, output the model corresponding to the support vector machine algorithm, which is suitable for predicting non-linear component changes.

[0074] Step 22: Calculate the loss value based on the prediction result, the standard content and standard freshness corresponding to the preset time point, and the standard content and standard freshness corresponding to the first historical time. Update the initial model based on the loss value, and repeatedly execute the step of obtaining the sample data set within the preset time period before the preset time point until the loss value meets the set threshold, thereby obtaining the trained data analysis model.

[0075] The above set threshold can be set according to actual needs and will not be limited here. In actual implementation, calculate the loss value based on the prediction result, the standard content and standard freshness corresponding to the preset time point, and the standard content and standard freshness corresponding to the first historical time, so as to train the initial model based on the loss value. Specifically, the models corresponding to the partial least squares regression algorithm and the regression model corresponding to the support vector machine algorithm can be trained respectively according to the description in the above steps, and the hyperparameters can be optimized using cross-validation, and finally the trained data analysis model can be obtained. The above standard content and standard freshness can be the true standard content and standard freshness at the actual preset time point.

[0076] Based on the method of the above embodiment, the data analysis method further includes the following steps:

[0077] Step 1: Collect the current ambient temperature of the storage environment through a temperature sensor;

[0078] Step 2: Obtain the preset temperature range corresponding to the analysis result;

[0079] Step 3: When the current ambient temperature is lower than the lower limit value of the preset temperature range, control the thermostat to turn on to raise the ambient temperature of the storage environment until it meets the preset temperature range;

[0080] Step 4: When the current ambient temperature is higher than the upper limit value of the preset temperature range, control the thermostat to turn on to lower the ambient temperature of the storage environment until it meets the preset temperature range.

[0081] The current ambient temperature mentioned above may be the same as or different from the ambient temperature value at the monitoring moment in the foregoing embodiments. The specific current ambient temperature refers to the ambient temperature collected by the temperature sensor after obtaining the analysis result corresponding to the stored goods. In actual implementation, the optimal storage conditions matched by different analysis results are usually different. A corresponding relationship table can be pre-configured according to actual requirements. Different optimal storage conditions corresponding to different analysis results are configured in this corresponding relationship table. After obtaining the above analysis result, the optimal storage condition corresponding to this analysis result can be queried from the corresponding relationship table, and the optimal preset temperature range can be obtained from this optimal storage condition. When it is detected that the current ambient temperature is lower than the lower limit value of the preset temperature range, a start command can be sent to the temperature controller to control the temperature controller to turn on. After the temperature controller is turned on, the storage environment can be heated until the ambient temperature returns to the preset temperature range, and then the temperature controller can be controlled to turn off to stop further heating. When it is detected that the current ambient temperature is higher than the upper limit value of the preset temperature range, a start command can be sent to the temperature controller to control the temperature controller to turn on. After the temperature controller is turned on, the ambient temperature of the storage environment can be reduced to prevent the stored goods from deteriorating. When the ambient temperature returns to the preset temperature range, the temperature controller can be controlled to turn off to stop further cooling.

[0082] Step Five: Collect the current ambient humidity of the storage environment through the humidity sensor;

[0083] Step Six: Obtain the preset humidity range corresponding to the analysis result;

[0084] Step Seven: When the current ambient humidity is lower than the lower limit value of the preset humidity range, control the humidity control device to turn on to increase the ambient humidity of the storage environment to meet the preset humidity range;

[0085] Step Eight: When the current ambient humidity is higher than the upper limit value of the preset humidity range, control the humidity control device to turn on to reduce the ambient humidity of the storage environment to meet the preset humidity range.

[0086] The current environmental humidity mentioned above may be the same as or different from the environmental humidity value at the monitoring moment in the foregoing embodiments. Specifically, the current environmental humidity refers to the environmental humidity collected by a humidity sensor after obtaining the analysis result corresponding to the stored goods. After obtaining the above analysis result, the optimal storage conditions corresponding to the analysis result can be queried from the above corresponding relationship table, and the optimal preset humidity range can be obtained from the optimal storage conditions. When it is detected that the current environmental humidity is lower than the lower limit value of the preset humidity range, a start command can be sent to the humidity control device to control the opening of the humidity control device. After the humidity control device is turned on, the humidity of the storage environment can be increased by spraying or steaming to maintain the freshness of the stored goods. After the environmental humidity returns to the preset humidity range, the humidity control device can be controlled to turn off to stop further humidification. When it is detected that the current environmental humidity is higher than the upper limit value of the preset humidity range, a start command can be sent to the humidity control device to control the opening of the humidity control device. After the humidity control device is turned on, the environmental humidity of the storage environment can be reduced by the condensation principle to prevent the growth of mold and bacteria. When the environmental humidity returns to the preset humidity range, the humidity control device can be controlled to turn off to stop further dehumidification.

[0087] Step Nine, collect the current air circulation speed of the storage environment through an air flow sensor;

[0088] Step Ten, obtain the preset circulation speed range corresponding to the analysis result;

[0089] Step Eleven, when the current air circulation speed is lower than the lower limit value of the preset circulation speed range, control the air flow regulating device to turn on to increase the air circulation speed;

[0090] Step Twelve, when the current air circulation speed is higher than the upper limit value of the preset circulation speed range, control the air flow regulating device to turn off to reduce the air circulation speed.

[0091] The current air circulation speed mentioned above may be the same as or different from the air circulation speed value at the monitoring moment in the foregoing embodiment. The specific current air circulation speed refers to the air circulation speed collected by the air flow sensor after obtaining the analysis result corresponding to the stored goods. The above-mentioned air flow regulating device may include a fan, a ventilation system, etc., and is mainly used to ensure the air circulation and uniform distribution. In actual implementation, after obtaining the above analysis result, the optimal storage conditions corresponding to the analysis result can be queried from the above corresponding relationship table, and the optimal preset air circulation speed range can be obtained from the optimal storage conditions. When it is detected that the current air circulation speed of the storage environment is lower than the lower limit value of the preset air circulation speed range, a start instruction can be sent to the air flow regulating device, specifically, the fan can be controlled to turn on to promote air flow and ensure uniform temperature and humidity distribution around the stored goods. After the air circulation speed resumes to the preset air circulation speed range, the fan can be controlled to turn off. When it is detected that the current air circulation speed is higher than the upper limit value of the preset air circulation speed range, a close instruction can be sent to the air flow regulating device, specifically, the ventilation system can be controlled to close to slow down the air circulation speed until the air circulation speed resumes to the preset air circulation speed range.

[0092] Step thirteen: Send and save the analysis result to the cloud device;

[0093] Step fourteen: Receive a first instruction from the cloud device to perform an operation corresponding to the first instruction according to the first instruction; and / or,

[0094] Step fifteen: Display the analysis result and the status information of the storage environment through the terminal device; the status information includes: ambient temperature, ambient humidity, air circulation speed.

[0095] Step sixteen: Receive a second instruction from the terminal device to perform an operation corresponding to the second instruction according to the second instruction.

[0096] The cloud device can be understood as a software platform adopting application virtualization technology, which can integrate multiple functions such as search, download, management, backup, etc. The above first instruction can be an adjustment instruction, a detection instruction, etc.; in actual implementation, after obtaining the above analysis result, the analysis result can be sent to the cloud device for recording and analysis through the communication module, and at the same time, the first instruction from the cloud device can be received through the communication module. For example, it can be a temperature adjustment instruction, a humidity adjustment instruction, an active detection instruction, etc. issued by the user, and the corresponding operation can be performed according to the received first instruction, such as adjusting the temperature, adjusting the humidity, etc.; among them, the communication module can support communication methods such as Bluetooth, Wi-Fi or NFC (Near Field Communication).

[0097] The above terminal device generally includes a display screen and a control panel. After obtaining the analysis results, the analysis results and the status information of the storage environment can be displayed in real time through the display screen. Historical spectral data, historical environmental data, etc. can also be displayed. Specifically, the content to be displayed can be set according to actual needs. The user can send a second instruction through the control panel, such as a temperature adjustment instruction, a humidity adjustment instruction, an active detection instruction, etc. After receiving the second instruction sent by the user, corresponding operations can be performed according to the second instruction, such as adjusting the temperature, adjusting the humidity, etc.

[0098] An embodiment of the present invention provides a storage system, which includes: a bin body, a quantum dot spectral device, a data processing unit, an environmental monitoring sensor, a temperature adjustment device, a humidity adjustment device, an air flow adjustment device, and a communication module. The data processing unit is signal-connected to the environmental monitoring sensor, the temperature adjustment device, the humidity adjustment device, and the air flow adjustment device, where:

[0099] The quantum dot spectral device includes multiple channels, each channel corresponds to a different quantum dot region, and each quantum dot region has different optical characteristics. The quantum dot spectral device is located above the inner side of the bin body. Refer to Figure 2 the schematic diagram of a quantum dot spectral storage management system shown. The quantum dot spectral device includes: a light source 1.2 and a quantum dot spectral sensor 1.1. The light source 1.2 can be an LED light source, etc.; the light source 1.2 is used to emit light in a preset wavelength range (for example, 200 - 5000 nm), and the quantum dot spectral sensor 1.1 is used to obtain the spectral data reflected and / or scattered after the light emitted by the light source 1.2 irradiates the surface of the stored goods; the field of view of the quantum dot spectral sensor 1.1 needs to cover most of the area of the stored goods. After the light source 1.2 irradiates the stored goods, the light can be reflected and / or scattered onto the quantum dot spectral sensor 1.1.

[0100] The environmental monitoring sensor is used to collect the environmental data at the monitoring moment inside the bin body;

[0101] The data processing unit 2 is used to receive the spectral data at the monitoring moment and the environmental data at the monitoring moment, preprocess the spectral data at the monitoring moment through a pre-trained data analysis model, extract the spectral feature information of the target band from the processed spectral data, and output the analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment; wherein, the environmental data at the monitoring moment includes: environmental temperature, environmental humidity, and air circulation speed; the target band is used to analyze the target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods. The data processing unit is also used to control the temperature regulating device and / or the humidity regulating device and / or the air flow regulating device according to the analysis result;

[0102] The communication module 4 is data-connected to the data processing unit 2 and is used to transmit at least the analysis result and the operation instruction.

[0103] The above environmental monitoring sensors include: a temperature sensor, a humidity sensor, and an air flow sensor arranged inside the warehouse; and / or, the temperature regulating device 3.1 includes: a thermostat arranged inside the warehouse; and / or, the humidity regulating device 3.2 includes: a humidity control device arranged inside the warehouse; and / or, the storage system further includes a terminal device 5 signal-connected to the communication module 4. The thermostat may include a heater and a cooler, and the humidity control device may include a humidifier and a dehumidifier.

[0104] Specifically, the above are respectively located on two opposite specified sides inside the warehouse to form a convection effect inside the warehouse; the air flow regulating device is located on the specified side of the warehouse;

[0105] The temperature sensor can collect the environmental temperature inside the warehouse; the humidity sensor can collect the environmental humidity inside the warehouse; the air flow sensor can collect the air circulation speed inside the warehouse; the collection frequency of each sensor is not limited, and can be real-time or at a set time interval.

[0106] After obtaining the analysis result, the data processing unit 2 can obtain the preset temperature range, preset humidity range, and preset circulation speed range corresponding to the analysis result;

[0107] The data processing unit 2 is also used to control the thermostat to turn on when the current environmental temperature is lower than the lower limit value of the preset temperature range to increase the environmental temperature of the storage environment until the environmental temperature returns to the preset temperature range; when the current environmental temperature is higher than the upper limit value of the preset temperature range, control the thermostat to turn on to lower the environmental temperature of the storage environment until the environmental temperature returns to the preset temperature range to prevent the stored goods from deteriorating;

[0108] The data processing unit 2 is also used to control the humidity control device to turn on when the current environmental humidity is lower than the lower limit value of the preset humidity range, so as to increase the environmental humidity of the storage environment. The environmental humidity in the storage environment can be increased by spraying or steam methods to maintain the freshness of the stored goods. When the current environmental humidity is higher than the upper limit value of the preset humidity range, it controls the humidity control device to turn on to reduce the environmental humidity of the storage environment. The humidity control device can reduce the humidity in the storage environment through the condensation principle to prevent the growth of mold and bacteria.

[0109] The data processing unit 2 is also used to control the air flow regulating device 3.3 to turn on when the current air flow velocity is lower than the lower limit value of the preset air flow velocity range, so as to increase the current air flow velocity. When the current air flow velocity is higher than the upper limit value of the preset air flow velocity range, it controls the air flow regulating device 3.3 to turn off to reduce the current air flow velocity. Specifically, the air flow regulating device 3.3 is used to ensure the air circulation and uniform distribution. It includes a fan and a ventilation system. When the air circulation is detected to be insufficient, the fan automatically turns on to promote the air flow and ensure the uniform temperature and humidity distribution of the environment around the stored goods. The fan and the ventilation system are located on the side of the warehouse body. The fan controls the internal air flow circulation of the warehouse body, and the ventilation system controls the air flow exchange inside and outside the warehouse body.

[0110] The data processing unit 2 is also used to send the analysis results to the cloud device through the communication module 4. That is, the communication module 4 can realize wireless communication with external devices, be able to send the analysis results to the cloud device for recording and analysis, and at the same time can receive instructions from the cloud device. The communication module 4 can support communication methods such as Bluetooth, Wi-Fi or NFC. The terminal device 5 is used to display the analysis results and the status information of the storage environment. Among them, the status information includes the environmental data at the monitoring moment. The above terminal device 5 can include a display screen and a control panel, and is used to display the analysis results, the status of the storage conditions and the historical spectral data in real time, etc. The user can manually adjust the storage conditions or view the status of the stored goods through the terminal device 5. The system is provided with a friendly user interface, which displays the spectral monitoring results and the status of the storage conditions in real time, etc., facilitating the storage management personnel to monitor and make decisions. In addition, the system can send alarm notifications to remind the storage management personnel to pay attention to potential problems.

[0111] Next, according to Figure 2, the processing process of the entire system is described. First, the light emitted by the light source 1.2 irradiates the surface of the stored goods. The quantum dot spectral sensor 1.1 receives the reflected or scattered spectral signals and converts them into electrical signals, which are then sent to the data processing unit 2. The data processing unit 2 processes and analyzes the input spectral signals. Using a preset data analysis model, it identifies information such as the compositional changes and freshness of the stored goods. The data processing unit 2 issues instructions to the temperature adjustment device and / or humidity adjustment device and / or air flow adjustment device according to the analysis results, automatically adjusting the temperature and / or humidity and / or air circulation speed parameters in the storage environment to ensure the optimal storage conditions for the stored goods. The user can view the real-time monitoring results and the status of the storage conditions through the terminal device 5, and can also select manual adjustment or automatic control mode. The data processing unit 2 also sends the analysis results to the cloud device through the communication module 4 for further recording, analysis, and monitoring.

[0112] The above data analysis method uses a quantum dot spectral device to perform non-contact scanning on the stored goods, obtaining its spectral data in real time. These data contain the spectral characteristic information of the substance, which can reflect the compositional changes and potential deterioration of the stored goods. By analyzing the spectral data at the monitoring moment and the environmental data at the monitoring moment using a pre-trained data analysis model, an analysis result is obtained. According to the analysis result, the optimal storage conditions for the stored goods can be intelligently judged, and the storage environment can be adjusted in real time through an automated control device. For example, when the deterioration risk of the stored goods is detected, the system will automatically lower the temperature or adjust the humidity to extend the shelf life of the stored goods.

[0113] As an emerging spectral analysis method, quantum dot technology has the advantages of high sensitivity, small volume, and easy integration. The quantum dot spectral device can accurately detect the spectral characteristics of materials, providing data support for various applications. At the same time, the development of artificial intelligence and big data technologies has provided new solutions for the analysis and processing of spectral data. This method can solve the practical problems faced in the current fields of food, medicine, raw materials, and environmental monitoring through the application of quantum dot spectral technology. This method can not only improve the efficiency and accuracy of storage management, but also effectively ensure the safety and health of consumers, promote sustainable development, and provide technical support for the intelligent transformation of related industries.

[0114] The above data analysis method can also produce the following beneficial effects:

[0115] 1. Precise identification of changes in material composition: This method uses a highly sensitive quantum dot spectroscopy device, which can demonstrate superior performance in trace component detection, ensuring precise identification of changes in the composition of goods. Through the quantum dot spectroscopy device, spectral data of different stored goods can be detected with high sensitivity, enabling high-precision monitoring of changes in the composition of goods, ensuring real-time monitoring of the status of goods, and promptly identifying potential spoilage risks.

[0116] 2. Intelligent data analysis system: By constructing an intelligent data analysis model and training based on big data, the comprehensive judgment ability for the spoilage of goods is continuously optimized. As the amount of data increases, the judgment accuracy of the model is significantly improved, thus achieving more reliable monitoring of the status of goods.

[0117] 3. Automatic adjustment of storage conditions: According to the analysis results of the data analysis model, storage conditions such as temperature and humidity in the storage environment are automatically adjusted to ensure that goods are stored under optimal conditions, extend the shelf life, prevent spoilage and loss. Through an automated monitoring and adjustment mechanism, human intervention is reduced, and the efficiency, accuracy, and consistency of storage management are improved, reducing economic losses caused by improper storage conditions.

[0118] 4. Reduction of waste, economic loss, and improvement of safety: Effectively detecting changes in the composition of stored goods, promptly discovering spoilage risks, and adjusting storage conditions contribute to reducing economic losses caused by the spoilage of goods and improving the economic benefits of storage operations; effectively monitoring the freshness of food and drugs can ensure the safety of consumers and reduce health risks caused by spoilage.

[0119] 5. Diversified application capabilities: This method has wide applicability and can perform personalized intelligent monitoring and management for different types of goods (such as food, drugs, raw materials, etc.), adapting to diversified storage needs, and enhancing the versatility, flexibility, and application breadth of the system.

[0120] 6. User-friendly interface design: This method is equipped with an intuitive user interface that displays monitoring results and the status of storage conditions in real time, facilitating effective monitoring and decision-making by management personnel.

[0121] 7. Alarm and feedback mechanism: This method also has an alarm function, which can promptly notify management personnel when potential spoilage risks are detected, ensuring the safety and quality of stored goods.

[0122] 8. Promotion of sustainable development: By optimizing storage conditions and reducing waste of food and drugs, the efficient use of resources is promoted, meeting the goals of sustainable development.

[0123] This method effectively solves the technical problems existing in traditional storage management, such as inaccurate component detection and lagging adjustment of storage conditions, and provides a more intelligent and efficient solution for storage management. It should be noted that this method can be applied not only to the storage environment but also to logistics. The system can be linked with other intelligent devices to achieve intelligent management and allocation of goods and improve storage efficiency. By real-time monitoring the status of goods, it provides traceability during the logistics process and ensures the safety of goods during transportation. The system can be deployed in logistics centers or on transportation vehicles, trains, ships, and space equipment. This solution can be applied not only to fields such as food and medicine but also to various fields that require real-time detection, such as chemical industry, drinking water, and organic waste.

[0124] An embodiment of the present invention provides a data analysis device, as Figure 3 shown. The device includes: a quantum dot spectroscopy device 30 for collecting spectral data of stored goods at the monitoring moment. The quantum dot spectroscopy device includes multiple channels, each channel corresponding to a different quantum dot region, and each quantum dot region having different optical characteristics; an acquisition module 31 for acquiring environmental data at the monitoring moment; an output module 32 for inputting the spectral data at the monitoring moment and the environmental data at the monitoring moment into a pre-trained data analysis model. The data analysis model preprocesses the spectral data at the monitoring moment and extracts spectral feature information of the target band from the processed spectral data, so as to output an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment. Among them, the target band is used to analyze the target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods.

[0125] The above data analysis device can collect spectral data of stored goods at the monitoring moment with high sensitivity by using the quantum dot spectroscopy device and output an analysis result corresponding to the stored goods through the data analysis model. This way of combining the quantum dot spectroscopy device and the data analysis model can accurately detect and monitor the content of the target component in the stored goods and the freshness of the stored goods, and timely identify potential spoilage risks.

[0126] Further, the device further includes a data analysis model training module, which is used for: obtaining a sample data set within a preset time period before a preset time point; wherein, the sample data set includes data samples at multiple time points, and each data sample includes historical spectral data and corresponding historical environmental data at the corresponding time point, and is marked with the standard content and standard freshness of the specified component at the corresponding time point; wherein, the specified component includes at least a part of the target component; inputting the sample data set into the initial model to output a prediction result corresponding to the stored goods through the initial model; wherein, the prediction result includes: at the preset time point, the first content of the specified component in the stored goods, the first freshness of the stored goods, and at the first historical time in the future, the predicted second content of the specified component in the stored goods and the predicted second freshness of the stored goods; calculating a loss value based on the prediction result, the standard content and standard freshness corresponding to the preset time point, and the standard content and standard freshness corresponding to the first historical time, updating the initial model based on the loss value, and repeating the step of obtaining the sample data set within a preset time period before the preset time point until the loss value meets the set threshold to obtain a trained data analysis model.

[0127] Further, the data analysis model training module is used for: preprocessing the historical spectral data at multiple time points in the sample data set by using the principal component analysis method to obtain the processed historical spectral data; extracting the specified spectral feature information of the specified band from the processed historical spectral data; wherein, the specified band is used for analyzing the specified component; if there is a linear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample data set, using the partial least squares regression algorithm to output the prediction result corresponding to the stored goods; if there is a nonlinear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample data set, using the support vector machine algorithm to output the prediction result corresponding to the stored goods.

[0128] Further, the device is further used for: collecting the current environmental temperature of the storage environment through a temperature sensor; obtaining a preset temperature range corresponding to the analysis result; when the current environmental temperature is lower than the lower limit value of the preset temperature range, controlling the thermostat to turn on to increase the environmental temperature of the storage environment; when the current environmental temperature is higher than the upper limit value of the preset temperature range, controlling the thermostat to turn on to decrease the environmental temperature of the storage environment.

[0129] Further, the device is also configured to: collect the current ambient humidity of the storage environment through a humidity sensor; obtain a preset humidity range corresponding to the analysis result; when the current ambient humidity is lower than the lower limit of the preset humidity range, control the humidity control device to turn on to increase the ambient humidity of the storage environment; when the current ambient humidity is higher than the upper limit of the preset humidity range, control the humidity control device to turn on to decrease the ambient humidity of the storage environment; and / or, the device is also configured to: collect the current air circulation speed of the storage environment through an air flow sensor; obtain a preset air circulation speed range corresponding to the analysis result; when the current air circulation speed is lower than the lower limit of the preset air circulation speed range, control the air flow regulating device to turn on to increase the air circulation speed; when the current air circulation speed is higher than the upper limit of the preset air circulation speed range, control the air flow regulating device to turn off to decrease the air circulation speed.

[0130] Further, the device is also configured to: send and save the analysis result to a cloud device; receive a first instruction from the cloud device to perform an operation corresponding to the first instruction according to the first instruction; and / or, the device is also configured to: display the analysis result and the status information of the storage environment through a terminal device; where the status information includes: ambient temperature, ambient humidity, and air circulation speed; receive a second instruction from the terminal device to perform an operation corresponding to the second instruction according to the second instruction.

[0131] For the data analysis device provided by the embodiments of the present invention, its implementation principle and the technical effects produced are the same as those of the foregoing embodiments of the data analysis method. For a brief description, for the parts not mentioned in the embodiments of the data analysis device, reference may be made to the corresponding content in the foregoing embodiments of the data analysis method.

[0132] Embodiments of the present invention also provide an electronic device. Refer to Figure 4 As shown, the electronic device includes a processor 130 and a memory 131. The memory 131 stores machine-executable instructions that can be executed by the processor 130. The processor 130 executes the machine-executable instructions to implement the above data analysis method.

[0133] Further, Figure 4 As shown, the electronic device further includes a bus 132 and a communication interface 133. The processor 130, the communication interface 133, and the memory 131 are connected through the bus 132.

[0134] Among them, the memory 131 may include high-speed random access memory (RAM), and may also include non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 133 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 132 can be an ISA bus, a PCI bus, an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 4 only a bidirectional arrow is used in Figure 4 , but it does not mean that there is only one bus or one type of bus.

[0135] The processor 130 may be an integrated circuit chip with signal processing capabilities. In the implementation process, the steps of the above method can be completed by the integrated logic circuit in the hardware of the processor 130 or the instructions in the form of software. The above-mentioned processor 130 may be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it may also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field programmable gate array (FPGA for short), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by the combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, register, etc. This storage medium is located in the memory 131, and the processor 130 reads the information in the memory 131 and combines its hardware to complete the steps of the method in the foregoing embodiments.

[0136] The embodiments of the present invention also provide a machine-readable storage medium. The machine-readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by the processor, the machine-executable instructions prompt the processor to implement the above data analysis method. For the specific implementation, reference can be made to the method embodiments and details are not described herein again.

[0137] The computer program product of the data analysis method, system, device and electronic device provided by the embodiments of the present invention includes a computer-readable storage medium storing program codes. The instructions included in the program codes can be used to execute the methods described in the foregoing method embodiments. For specific implementation, reference can be made to the method embodiments and will not be elaborated here.

[0138] If the above functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data analysis method, characterized in that, The method includes: Collecting spectral data of stored goods at a monitoring moment through a quantum dot spectroscopy device, where the quantum dot spectroscopy device includes multiple channels, each channel corresponding to a different quantum dot region, and each quantum dot region having different optical characteristics; Obtaining environmental data at the monitoring moment; Inputting the spectral data at the monitoring moment and the environmental data at the monitoring moment into a pre-trained data analysis model, preprocessing the spectral data at the monitoring moment through the data analysis model, and extracting spectral feature information of a target band from the processed spectral data, so as to output an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment; Wherein, the target band is used to analyze a target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods.

2. The method according to claim 1, characterized in that, The data analysis model is trained through the following method: Obtaining a sample data set within a preset time period before a preset time point; wherein, the sample data set includes data samples at multiple time points, each data sample including historical spectral data corresponding to the time point and corresponding historical environmental data, and being labeled with the standard content and standard freshness of a specified component corresponding to the time point; wherein, the specified component includes at least a part of the target component; Inputting the sample data set into an initial model to output a prediction result corresponding to the stored goods through the initial model; wherein, the prediction result includes: at the preset time point, the first content of the specified component in the stored goods, the first freshness of the stored goods, and at a first historical time in the future, the predicted second content of the specified component in the stored goods and the predicted second freshness of the stored goods; Calculating a loss value based on the prediction result, the standard content and standard freshness corresponding to the preset time point, and the standard content and standard freshness corresponding to the first historical time, updating the initial model based on the loss value, and repeating the step of obtaining a sample data set within a preset time period before a preset time point until the loss value meets a set threshold to obtain the trained data analysis model.

3. The method according to claim 2, wherein The step of inputting the sample data set into an initial model to output a prediction result corresponding to the stored goods through the initial model includes: Preprocessing the historical spectral data at multiple time points in the sample data set by using principal component analysis to obtain processed historical spectral data; extracting specified spectral feature information of a specified band from the processed historical spectral data; wherein, the specified band is used to analyze a specified component; If there is a linear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample dataset, the partial least squares regression algorithm is used to output the prediction result corresponding to the stored goods; If there is a non-linear relationship between the specified spectral feature information and the content of the specified component, based on the specified spectral feature information and the historical environmental data at multiple time points in the sample dataset, the support vector machine algorithm is used to output the prediction result corresponding to the stored goods.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Collecting the current environmental temperature of the storage environment through a temperature sensor; Obtaining a preset temperature range corresponding to the analysis result; When the current environmental temperature is lower than the lower limit value of the preset temperature range, controlling the thermostat to turn on to raise the environmental temperature of the storage environment to meet the preset temperature range; When the current environmental temperature is higher than the upper limit value of the preset temperature range, controlling the thermostat to turn on to lower the environmental temperature of the storage environment to meet the preset temperature range.

5. The method according to any one of claims 1 to 3, characterized in that, The method further includes: Collecting the current environmental humidity of the storage environment through a humidity sensor; Obtaining a preset humidity range corresponding to the analysis result; When the current environmental humidity is lower than the lower limit value of the preset humidity range, controlling the humidity control device to turn on to raise the environmental humidity of the storage environment to meet the preset humidity range; When the current environmental humidity is higher than the upper limit value of the preset humidity range, controlling the humidity control device to turn on to lower the environmental humidity of the storage environment to meet the preset humidity range; and / or, the method further includes: Collecting the current air circulation speed of the storage environment through an air flow sensor; Obtaining a preset air circulation speed range corresponding to the analysis result; When the current air circulation speed is lower than the lower limit value of the preset air circulation speed range, controlling the air flow regulating device to turn on to increase the air circulation speed; When the current air circulation speed is higher than the upper limit value of the preset air circulation speed range, controlling the air flow regulating device to turn off to decrease the air circulation speed.

6. The method according to any one of claims 1 to 3, characterized in that The method further includes: Sending and saving the analysis result to a cloud device; Receiving a first instruction from the cloud device to perform an operation corresponding to the first instruction according to the first instruction; and / or, the method further includes: Displaying the analysis result and the status information of the storage environment through a terminal device; where the status information includes: environmental temperature, environmental humidity, air circulation speed; Receiving a second instruction from the terminal device to perform an operation corresponding to the second instruction according to the second instruction.

7. A storage system, characterized in that, The system includes: a storage body, a quantum dot spectroscopy device, a data processing unit, an environmental monitoring sensor, a temperature regulating device, a humidity regulating device, an air flow regulating device, and a communication module, where: The quantum dot spectral device includes multiple channels, each channel corresponding to a different quantum dot region, each quantum dot region having different optical properties. The quantum dot spectral device is located above the inner side of the bin body. The quantum dot spectral device includes: a light source and a quantum dot spectral sensor. The light source is used to emit light in a preset wavelength range, and the quantum dot spectral sensor is used to obtain spectral data of the light emitted by the light source after being reflected and / or scattered by the surface of the stored goods; The environmental monitoring sensor is used to collect environmental data at the monitoring moment in the bin body; The data processing unit is used to receive the spectral data at the monitoring moment and the environmental data at the monitoring moment, preprocess the spectral data at the monitoring moment through a pre-trained data analysis model, extract spectral feature information of the target band from the processed spectral data, and output an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment. Among them, the environmental data at the monitoring moment includes: environmental temperature, environmental humidity, and air flow velocity; the target band is used to analyze the target component; the analysis result includes: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods. The data processing unit is also used to control the temperature adjustment device and / or the humidity adjustment device and / or the air flow adjustment device according to the analysis result; The communication module is data-connected to the data processing unit and is used to transmit at least the analysis result and operation instructions.

8. The storage system according to claim 7, wherein The environmental monitoring sensor includes: a temperature sensor, a humidity sensor, and an air flow sensor arranged in the bin body; and / or, the temperature adjustment device includes: a thermostat arranged on the inner side of the bin body; and / or, the humidity adjustment device includes: a humidity control device arranged on the inner side of the bin body; and / or, the storage system further includes a terminal device signal-connected to the communication module.

9. A data analysis device, characterized in that, The device includes: A quantum dot spectral device for collecting spectral data of stored goods at the monitoring moment. The quantum dot spectral device includes multiple channels, each channel corresponding to a different quantum dot region, each quantum dot region having different optical properties; An acquisition module for acquiring environmental data at the monitoring moment; An output module for inputting the spectral data at the monitoring moment and the environmental data at the monitoring moment into a pre-trained data analysis model, preprocessing the spectral data at the monitoring moment through the data analysis model, extracting spectral feature information of the target band from the processed spectral data, and outputting an analysis result corresponding to the stored goods based on the spectral feature information and the environmental data at the monitoring moment; Among them, the target band is used to analyze the target component; the analysis results include: at the monitoring moment, the content of the target component in the stored goods, the freshness of the stored goods, and / or, at a preset future time, the predicted content of the target component in the stored goods and the predicted freshness of the stored goods.

10. An electronic device, characterized in that, It includes a processor and a memory, the memory stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions to implement the data analysis method according to any one of claims 1-6.

Images