A cellar temperature monitoring and thermal insulation performance evaluation method and system

Abstract

The embodiments of the present disclosure provide a cellar temperature monitoring and thermal insulation performance evaluation method and system. Applied to the technical field of cellar performance monitoring technology, it includes: placing a number of sensors in the cellar to monitor the temperature parameters in the cellar to obtain environmental cellar temperature monitoring data; collecting infrared images of the cellar electric heating coils, and constructing a temperature model, processing the infrared images of the cellar electric heating coils and environmental monitoring data according to the temperature model to obtain accurate cellar temperature monitoring data; transmitting accurate cellar temperature monitoring data based on ZigBee combined with NB‑IOT and storing it in the cloud; analyzing the stored temperature monitoring data, and reinforcing or rebuilding the cellar according to the analysis results. In this way, the present invention can effectively obtain cellar temperature data and monitor it, and transform the cellar according to the temperature data to increase the thermal insulation of the cellar.

Description

Technical Field

[0001] The present disclosure relates to the technical field of cellar performance monitoring, and in particular to a cellar temperature monitoring and thermal insulation performance evaluation method and system. Background Art

[0002] In my country, winemaking technology has a history of thousands of years. Among many types of wine, the unique aroma of white wine stands out. Chinese white wine is brewed from grain as raw material. During the fermentation process, the environmental changes in the cellar will directly affect the quality of the fermentation process, and then affect the aroma and taste of the wine.

[0003] The traditional cellar temperature monitoring method usually uses manual sampling to measure the cellar temperature and alcohol concentration. This method can easily damage the airtightness of the cellar, and has problems of low efficiency and accuracy, resulting in the brewed wine being unable to produce excellent flavor.

[0004] Since the temperature stability is highly required during the brewing process, it is of great significance to develop a method to monitor the cellar temperature and improve the cellar accordingly for liquor brewing. Summary of the invention

[0005] The present invention provides a cellar temperature monitoring and thermal insulation performance evaluation method and system, which extracts cellar temperature features through a neural network based on a moving window visual self-attention mechanism and obtains accurate temperature data in combination with infrared images, thereby solving the technical problems of limited processing capability and low efficiency of existing methods.

[0006] According to a first aspect of the present disclosure, a method for monitoring cellar temperature and evaluating thermal insulation performance is provided, comprising the following steps:

[0007] Several sensors are arranged in the cellar to monitor the temperature parameters in the cellar and obtain the cellar temperature monitoring data;

[0008] Collect infrared images of the electric heating coils of the cellar, build a temperature model, and process the infrared images and the temperature monitoring data according to the temperature model to obtain accurate temperature monitoring data;

[0009] The precise temperature monitoring data is transmitted based on ZigBee combined with NB-IOT and stored in the cloud;

[0010] Analyzing the stored temperature monitoring data, and reinforcing or rebuilding the cellar according to the analysis results;

[0011] The process of collecting the infrared image of the electric heating coil of the cellar, constructing a temperature model, and processing the infrared image and the temperature monitoring data according to the temperature model to obtain accurate temperature monitoring data is as follows:

[0012] An industrial camera is used to collect infrared images of the electric heating coils of the cellar to obtain infrared image data;

[0013] A temperature model is constructed based on a moving window visual self-attention mechanism network, and the infrared image data is input into the temperature model for processing to obtain infrared image features;

[0014] The temperature monitoring data is integrated with the infrared image features, and accurate temperature monitoring data is obtained based on the temperature model calculation;

[0015] The temperature model includes an input layer, a moving window visual self-attention module and an output layer;

[0016] The input layer includes a pooling layer, and the output layer includes a fully connected layer;

[0017] Among them, based on the moving window visual self-attention module, self-attention is calculated in different windows. Specifically, the window is set to contain M×M image blocks, and the result is:

[0018] ;

[0019] Among them, h is the height of the infrared image feature map, w is the width of the feature map, c is the depth of the feature map, and M is the window size;

[0020] The moving window visual self-attention module includes two consecutive moving window visual self-attention blocks, wherein one moving window visual self-attention block is composed of a moving window-based multi-head self-attention module with a two-layer multi-layer perceptron, and the other moving window visual self-attention block is composed of a window-based multi-head self-attention module with a two-layer multi-layer perceptron;

[0021] The two consecutive moving window visual self-attention blocks are specifically:

[0022] ;

[0023] ;

[0024] ;

[0025] ;

[0026] in, is the module output feature of the lth block, is the output feature of the l-th block multi-layer perceptron module, W-MSA is a window-based multi-head self-attention module, SW-MSA is a moving window-based multi-head self-attention module, and MPL is a multi-layer perceptron module.

[0027] According to the above aspects and any possible implementation, a further implementation is provided, wherein the temperature monitoring data is merged with the infrared image features, and the process of obtaining accurate temperature monitoring data based on the temperature model is as follows:

[0028] Normalizing the temperature monitoring data and fusing it with the infrared image features to obtain fused features;

[0029] A query, key and value vector are obtained according to the fusion features, an attention matrix is ​​calculated according to the query, key and value vectors, and the attention matrix is ​​weighted and processed to output accurate temperature monitoring data.

[0030] According to the above aspects and any possible implementation, an implementation is further provided, wherein the query, key and value vectors are obtained according to the fusion features, the attention matrix is ​​calculated according to the query, key and value vectors, and the calculation process of weighted processing and outputting the attention matrix to obtain the temperature data is:

[0031] ;

[0032] ;

[0033] Among them, d is the embedding dimension, Q is the query vector of the word, K is the key vector to be checked, V is the value vector, and M 2 is the number of windows, B is an M 2 A real matrix of .

[0034] According to the above aspects and any possible implementation, an implementation is further provided, wherein the ZigBee-based NB-IOT includes a ZigBee module and a NB-IOT communication module;

[0035] The ZigBee module is designed with CC2530 chip, including a terminal and a coordinator;

[0036] The NB-IOT communication module adopts MT2625 chip design.

[0037] According to the above aspects and any possible implementation, an implementation is further provided, wherein the process of analyzing the stored temperature monitoring data and reinforcing or rebuilding the cellar according to the analysis results is as follows:

[0038] Collect accurate temperature monitoring data of the cellars on a daily basis within a natural year and calculate the average temperature;

[0039] Setting a temperature judgment standard, if the average temperature is within the temperature judgment standard range, it is judged that the thermal insulation performance of the cellar is good, and the cellar is reinforced;

[0040] If the average temperature is outside the temperature judgment reference range, it is determined that the thermal insulation performance of the cellar is insufficient and the cellar is rebuilt.

[0041] According to the aspects and any possible implementation methods described above, an implementation method is further provided, in which the reconstruction of the cellar adopts a six-sided smooth cellar construction method, and the process includes: masonry layout, masonry layout of the cellar and cellar stone; layered masonry, starting from the cellar foundation and layered masonry until the cellar is completed; secondary grouting, chiseling the mortar joints to a depth of 5 cm, cleaning the residue in the mortar joints and then performing secondary grouting; maintenance operation, immediately after grouting, the surface of the cellar wall on one side of the grouting is completely covered with a film for maintenance, and the maintenance time is 14 days. The film will be removed after the maintenance expires.

[0042] According to a second aspect of the present disclosure, a cellar temperature monitoring and thermal insulation performance evaluation system is provided, comprising a temperature data acquisition module, a temperature data processing module, a data storage module and a cellar thermal insulation performance judgment module connected in sequence;

[0043] The temperature data acquisition module includes several temperature sensors and industrial cameras, which are used to monitor the temperature of the cellar and collect infrared images of the electric heating coils of the cellar to obtain temperature monitoring data and infrared images;

[0044] The temperature data processing module includes a neural network based on a moving window visual self-attention mechanism, which processes and outputs the temperature monitoring data and the infrared image to obtain accurate cellar temperature monitoring data;

[0045] The data storage module includes a ZigBee-based NB-IOT module for storing the precise cellar temperature monitoring data;

[0046] The cellar thermal insulation performance judgment module is used to judge the cellar thermal insulation performance according to the precise cellar temperature monitoring data, and to reinforce or rebuild the cellar.

[0047] Compared with the prior art, the present invention has achieved the following beneficial effects:

[0048] The neural network based on the moving window visual self-attention mechanism provided by the present invention obtains accurate temperature data by extracting the temperature characteristics of the cellar and combining it with the infrared image of the heating coil, and at the same time combines with ZigBee and NB-IOT modules to achieve fast and efficient data transmission, and can accurately monitor the temperature data of the cellar; and the present invention also combines historical data and benchmark data, and judges the thermal insulation performance of the cellar according to the above data, so as to obtain a cellar with better thermal insulation performance, ensure the temperature stability of the liquor brewing process, and has high practical value for the liquor brewing process.

[0049] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, among which:

[0051] Figure 1 A schematic flow chart of a method for monitoring cellar temperature and evaluating thermal insulation performance according to an embodiment of the present disclosure is shown;

[0052] Figure 2 A schematic diagram of the structure of a cellar temperature monitoring and thermal insulation performance evaluation system according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0055] Please refer to Figure 1 As shown, this embodiment provides a method for monitoring cellar temperature and evaluating thermal insulation performance, comprising the following steps:

[0056] S101. Arrange a number of sensors in the cellar to monitor the temperature parameters in the cellar and obtain environmental cellar temperature monitoring data.

[0057] In this embodiment, a plurality of temperature sensors are installed in the cellar for temperature monitoring. In this embodiment, the temperature sensor adopts a DS18B20 sensor, the measurement range of which is -55°C to +125°C, and the working voltage is 3-5V.

[0058] S102, collecting infrared images of the cellar electric heating coils, and constructing a temperature model, processing the infrared images of the cellar electric heating coils and environmental monitoring data according to the temperature model to obtain accurate cellar temperature monitoring data.

[0059] After obtaining the cellar environment monitoring data, this embodiment constructs a cellar temperature model to perform intelligent monitoring of the cellar temperature. Specifically:

[0060] This embodiment uses infrared images based on cellar electric heating coils and a multi-scale feature convolutional network combined with a moving window visual self-attention mechanism to construct a temperature monitoring model to obtain infrared image features, and combines multi-point monitoring data from temperature sensors to obtain accurate cellar temperature values.

[0061] First, a temperature model is constructed. The temperature monitoring model of this embodiment includes a convolutional network model based on a moving window visual self-attention mechanism, which includes a pooling layer, a moving window visual self-attention module and an output layer.

[0062] Specifically, an infrared image is taken by an industrial camera. After the temperature model collects the infrared image, the infrared image is downsampled through a pooling layer. Through the downsampling of the pooling layer, the resolution of the image can be reduced and the amount of computation can be saved.

[0063] After the processing is completed, the image data passes through two consecutive moving window visual self-attention modules, and the infrared image features are obtained after processing and output. In this embodiment, the infrared image passes through two interconnected moving window visual self-attention modules in sequence. The two moving window-based visual self-attention modules adopt a global self-attention mechanism, one of which includes a moving window-based multi-head self-attention module (Shifted Window-based MSA), and the other module includes a window-based multi-head self-attention module (Window-based MSA); two consecutive shifted window visual self-attention modules (Swin Transformer Block), one of which is composed of a shifted window-based multi-head self-attention module (Shifted Window-based MSA) with a two-layer multi-layer perceptron (MLP), and the other is composed of a window-based multi-head self-attention module (Window-based MSA) with a two-layer multi-layer perceptron, and an LN layer is used before each multi-head self-attention module and each multi-layer perceptron, and a residual connection is used after the multi-head self-attention module and the multi-layer perceptron.

[0064] Among them, based on the moving window visual self-attention module, self-attention is calculated in different windows. Specifically, the window is set to contain M×M image blocks, and the result is:

[0065] (1)

[0066] Among them, h is the height of the infrared image feature map, w is the width of the feature map, c is the depth of the feature map, and M is the window size.

[0067] For the window-based multi-head self-attention module (W-MSA), it eliminates the modeling of the relationship between windows while saving the amount of computation, but this will lead to a lack of information exchange and affect the representation of the model. The moving window-based multi-head self-attention module (SW-MSA) can solve the above problem.

[0068] From the above, the expressions of the two modules constructed in this embodiment are:

[0069] (2)

[0070] (3)

[0071] (4)

[0072] (5)

[0073] in, is the module output feature of the lth module, Output features for the lth multilayer perceptron module.

[0074] After the infrared image features and multi-dimensional output features are output, the temperature inside the cellar is measured by the temperature sensor to obtain the temperature data sequence. , and then normalize the data to get:

[0075] (6)

[0076] Among them, max(N) and min(N) are the maximum and minimum data in the sequence.

[0077] In this way, we can get the normalized data sequence: , then the feature information is fused with the data sequence and used as the input of Q, K, V, and fused through the multi-head self-attention mechanism to output the fused output features. Specifically:

[0078] After the output features are fused, the multi-head self-attention mechanism pays corresponding attention to each window periodically. , the input calculated by the self-attention mechanism of the local window is:

[0079] (7)

[0080] Among them, F LW is the fusion feature, P Q, P K , P V are learnable parameters respectively.

[0081] After the parameters are calculated, the parameters are input into the window-based multi-head self-attention module (W-MSA) The computational complexity of the global self-attention mechanism After that, the output temperature value is calculated, which is specifically expressed as:

[0082] (8)

[0083] (9)

[0084] Among them, d is the embedding dimension, Q is the query vector of the word, K is the key vector to be checked, V is the value vector, and M 2 is the number of windows, B is an M 2 The real number matrix B plays the role of adjusting or correcting the attention weight in the calculation of the attention mechanism.

[0085] S103, based on ZigBee combined with NB-IOT method, the precise cellar temperature data is transmitted and stored in the cloud.

[0086] After collecting temperature data and CO 2 After the environmental data is collected, the present embodiment adopts the ZigBee module to collect the environmental data, transmits it to the NB-IOT communication module and performs cloud storage.

[0087] Specifically, the ZigBee module of this embodiment includes a terminal and a coordinator, and is designed using a CC2530 chip. The environmental data is acquired through the terminal, and the data is transmitted to the coordinator through an ad hoc network.

[0088] The specific working process is that the coordinator node initiates network initialization, the terminal actively scans the coordinator, sends an association request to the coordinator after scanning and obtaining information, and then waits for the terminal to successfully connect with the coordinator and establish communication for data acquisition and transmission.

[0089] The NB-IOT communication module uses the MT2625 chip and a 2000mA battery. The module includes three working modes: wake-up mode, in which all functions can be used normally, light sleep mode, but paging information can still be received, and deep sleep mode, in which power consumption is low. Since the communication module in this embodiment needs to send data from the coordinator to the cloud storage platform, the wake-up mode is used for operation.

[0090] S104, analyzing the stored environmental data, and reinforcing or rebuilding the cellar according to the analysis results.

[0091] The cellar temperature is an important process control parameter for liquor fermentation, and when the cellar temperature range is 22°C-23°C, the wine yield and high-quality wine rate can be improved. The environmental data of this embodiment includes temperature data, which combines the ignition temperature and the cellar temperature to more comprehensively reflect the thermal insulation of the cellar.

[0092] Based on this, this embodiment collects daily temperature data and integrates it over a natural year to calculate the average temperature, and uses 20°C-24°C as the temperature judgment standard to judge the thermal insulation of the cellar:

[0093] (1) When the average temperature is within the temperature judgment reference range, the thermal insulation performance of the cellar is judged to be good and the cellar should be reinforced.

[0094] (2) When the average temperature is outside the temperature judgment reference range, it is judged that the thermal insulation performance of the cellar is insufficient and the cellar should be rebuilt.

[0095] When the cellar is rebuilt, this embodiment adopts a six-sided smooth cellar construction method, which can reduce mortar joint cracks during the cellar construction of the winemaking plant, ensure that the cellar is water-proof and airtight, and prevent the cellar from being reworked due to water seepage or air leakage. The specific working process is as follows:

[0096] Step 1: Lay out the cellars and cellar stones. According to the topography and soil environment, determine the number and size of the cellars and lay out the cellars; according to the size of the cellars, determine the size of the cellar stones and lay out the cellars, and process the cellar stones into six smooth sides.

[0097] Step 2: Lay out the stones in layers, starting from the cellar foundation until the cellar is completed. Lay out the stones in layers starting from the cellar foundation. Sprinkle water on the foundation or the cellar stone that has been laid two hours in advance before laying the stones. Spread mortar on the foundation or the cellar stone that has been laid, and then install the next layer of cellar stone. When installing the cellar stone, after the cellar stone is installed and stabilized, the mortar thickness between the cellar stone and the foundation is The thickness of the mortar joints between the cellar stones is 3cm-5cm; the vertical mortar joints between the cellar stones are compacted with steel rods; after laying all the cellar stones in one layer, lay the next layer of cellar stones after an interval of twelve hours; repeat the above steps until the cellar wall reaches the elevation and the laying is completed.

[0098] Step 3: Grout the joints for the second time. Chisel the joints to a depth of 5 cm, clean up the residue in the joints and then grout the joints for the second time.

[0099] Step 4: Maintenance work: immediately after grouting, cover the entire surface of the cellar wall on one side of the grouting with film for maintenance. The maintenance time is 14 days. The film will be removed after the maintenance expires.

[0100] like Figure 2As shown, this embodiment also provides a cellar temperature monitoring and thermal insulation performance evaluation system, comprising a temperature data acquisition module 1, a temperature data processing module 2, a data storage module 3 and a cellar thermal insulation performance judgment module 4 connected in sequence;

[0101] The temperature data acquisition module 1 includes several temperature sensors and industrial cameras, which are used to monitor the temperature of the cellar and collect infrared images of the electric heating coils of the cellar to obtain temperature monitoring data and infrared images;

[0102] The temperature data processing module 2 includes a neural network based on a moving window visual self-attention mechanism, which processes and outputs the temperature monitoring data and infrared images to obtain accurate cellar temperature monitoring data;

[0103] The data storage module 3 includes a ZigBee-based NB-IOT module for storing accurate cellar temperature monitoring data;

[0104] The cellar thermal insulation performance judgment module 4 is used to judge the cellar thermal insulation performance according to the accurate cellar temperature monitoring data, and to reinforce or rebuild the cellar.

[0105] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present disclosure is not limited by the order of the actions described, because according to the present disclosure, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present disclosure.

[0106] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0107] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for monitoring cellar temperature and evaluating thermal insulation performance, characterized in that: The following steps are involved: Several sensors are arranged in the cellar to monitor the temperature parameters in the cellar and obtain the cellar temperature monitoring data; Collect infrared images of the electric heating coils of the cellar, build a temperature model, and process the infrared images and the temperature monitoring data according to the temperature model to obtain accurate temperature monitoring data; The precise temperature monitoring data is transmitted based on ZigBee combined with NB-IOT and stored in the cloud; Analyzing the stored temperature monitoring data, and reinforcing or rebuilding the cellar according to the analysis results; The process of collecting the infrared image of the electric heating coil of the cellar, constructing a temperature model, and processing the infrared image and the temperature monitoring data according to the temperature model to obtain accurate temperature monitoring data is as follows: An industrial camera is used to collect infrared images of the electric heating coils of the cellar to obtain infrared image data; A temperature model is constructed based on a moving window visual self-attention mechanism network, and the infrared image data is input into the temperature model for processing to obtain infrared image features; The temperature monitoring data is integrated with the infrared image features, and accurate temperature monitoring data is obtained based on the temperature model calculation; The temperature model includes an input layer, a moving window visual self-attention module and an output layer; The input layer includes a pooling layer, and the output layer includes a fully connected layer; Among them, based on the moving window visual self-attention module, self-attention is calculated in different windows. Specifically, the window is set to contain M×M image blocks, and the result is: ； Among them, h is the height of the infrared image feature map, w is the width of the feature map, c is the depth of the feature map, and M is the window size; The moving window visual self-attention module includes two consecutive moving window visual self-attention blocks, wherein one moving window visual self-attention block is composed of a moving window-based multi-head self-attention module with a two-layer multi-layer perceptron, and the other moving window visual self-attention block is composed of a window-based multi-head self-attention module with a two-layer multi-layer perceptron; The two consecutive moving window visual self-attention blocks are specifically: ； ； ； ； in, is the module output feature of the lth block, is the output feature of the l-th block multi-layer perceptron module, W-MSA is a window-based multi-head self-attention module, SW-MSA is a moving window-based multi-head self-attention module, and MPL is a multi-layer perceptron module; The process of fusing the temperature monitoring data with the infrared image features and calculating the accurate temperature monitoring data based on the temperature model is as follows: Normalizing the temperature monitoring data and fusing it with the infrared image features to obtain fused features; Obtaining a query, a key, and a value vector according to the fusion feature, calculating an attention matrix according to the query, the key, and the value vector, and performing weighted processing on the attention matrix to output accurate temperature monitoring data; The process of obtaining the query, key and value vectors according to the fusion features, calculating the attention matrix according to the query, key and value vectors, and performing weighted processing on the attention matrix to output the temperature data is as follows: ； ； Among them, d is the embedding dimension, Q is Query, which is the query vector of the word, K is Key, which is the key vector to be checked, V is Value, which is the value vector, and M 2 is the number of windows, B is an M 2 A real matrix of .

2. The method for monitoring cellar temperature and evaluating thermal insulation performance according to claim 1, characterized in that: The ZigBee-based NB-IOT includes a ZigBee module and a NB-IOT communication module; The ZigBee module is designed with CC2530 chip, including a terminal and a coordinator; The NB-IOT communication module adopts MT2625 chip design.

3. The method for monitoring cellar temperature and evaluating thermal insulation performance according to claim 1, characterized in that: The process of analyzing the stored temperature monitoring data and reinforcing or rebuilding the cellar according to the analysis results is as follows: Collect accurate temperature monitoring data of the cellars on a daily basis within a natural year and calculate the average temperature; Setting a temperature judgment standard, if the average temperature is within the temperature judgment standard range, it is judged that the thermal insulation performance of the cellar is good, and the cellar is reinforced; If the average temperature is outside the temperature judgment reference range, it is determined that the thermal insulation performance of the cellar is insufficient and the cellar is rebuilt.

4. The method for monitoring cellar temperature and evaluating thermal insulation performance according to claim 3, characterized in that: The reconstruction of the cellar adopts a six-sided smooth cellar construction method, and the process includes: masonry layout, masonry layout of the cellar and cellar stone; layered masonry, starting from the cellar foundation and layered masonry until the cellar is completed; secondary grouting, chiseling the mortar joints to a depth of 5 cm, cleaning the residue in the mortar joints and then performing secondary grouting; maintenance work, immediately after grouting, the surface of the cellar wall on one side of the grouting is completely covered with film for maintenance, and the maintenance time is 14 days. The film will be removed after the maintenance expires.

5. A cellar temperature monitoring and thermal insulation performance evaluation system, used to implement the cellar temperature monitoring and thermal insulation performance evaluation method as described in any one of claims 1 to 4, characterized in that: It comprises a temperature data acquisition module (1), a temperature data processing module (2), a data storage module (3) and a cellar thermal insulation performance judgment module (4) which are connected in sequence; The temperature data acquisition module (1) comprises a plurality of temperature sensors and industrial cameras, which are used to monitor the temperature of the cellar and collect infrared images of the electric heating coils of the cellar to obtain temperature monitoring data and infrared images; The temperature data processing module (2) includes a neural network based on a moving window visual self-attention mechanism, and processes and outputs the temperature monitoring data and the infrared image to obtain accurate cellar temperature monitoring data; The data storage module (3) includes a ZigBee-based NB-IOT module, which is used to store the precise cellar temperature monitoring data; The cellar thermal insulation performance judgment module (4) is used to judge the cellar thermal insulation performance according to the precise cellar temperature monitoring data, and to reinforce or rebuild the cellar.

Images