Ice maker fault monitoring system and method based on Internet of Things technology

Abstract

The invention discloses an ice maker fault monitoring system and method based on the Internet of Things technology. The method comprises the following steps: firstly, arranging time sequences of acquired operating parameters into an ice maker operating parameter time sequence correlation input matrix according to a time dimension and an operating parameter sample dimension, and then performing feature extraction through an operating parameter time sequence correlation mode feature extractor based on a deep neural network model so as to obtain ice maker operating parameter time sequence correlation features; then, the ice machine operation parameter time sequence correlation characteristic graph passes through a characteristic interaction strengthening module based on a triple interaction attention module to obtain an ice machine operation parameter time sequence strengthening correlation characteristic graph; and performing essential feature extraction on the ice maker operation parameter time sequence enhanced correlation feature map to obtain purification ice maker operation parameter time sequence essential features, and finally determining whether the operation state of the monitored ice maker is abnormal or not based on the purification ice maker operation parameter time sequence essential features.

Description

Technical Field

[0001] The present disclosure relates to the field of ice maker fault monitoring, and in particular to an ice maker fault monitoring system and method based on Internet of Things technology. Background Art

[0002] Ice makers are a type of refrigeration equipment widely used in food processing, catering services, healthcare and other fields. Their main function is to cool water or other liquids below freezing and convert them into solid ice cubes. With the widespread use of ice makers in all walks of life, higher requirements have been placed on the operating efficiency, safety, and reliability of ice makers. However, traditional ice maker fault monitoring systems usually use manual inspections or scheduled maintenance. The manual inspection method is inefficient and can easily miss hidden faults, posing safety hazards. The regular maintenance method is costly and cannot detect faults in a timely manner. In addition, after a fault occurs in the ice maker, the maintenance time is long, which affects the normal use of the ice maker.

[0003] Therefore, an optimized ice machine fault monitoring system is desired. Summary of the invention

[0004] The present disclosure is made in view of the above problems. One object of the present disclosure is to provide an ice maker fault monitoring system and method based on Internet of Things technology.

[0005] The embodiment of the present disclosure provides an ice machine fault monitoring system based on Internet of Things technology, which includes:

[0006] An operating parameter data acquisition module, used for acquiring a time series of operating parameters of the monitored ice-making machine acquired by the sensor module, wherein the operating parameters of the ice-making machine include temperature value, pressure value, current value and voltage value;

[0007] An operating parameter time series data arrangement module, used to arrange the time series of the operating parameters according to the time dimension and the operating parameter sample dimension into an ice machine operating parameter time series associated input matrix;

[0008] An operating parameter time series correlation feature extraction module, used for extracting features from the ice maker operating parameter time series correlation input matrix through an operating parameter time series correlation pattern feature extractor based on a deep neural network model to obtain an ice maker operating parameter time series correlation feature graph;

[0009] An operating parameter time series feature enhancement module, used for obtaining an ice-making machine operating parameter time series enhanced association feature diagram by a feature interaction enhancement module based on a triple interactive attention module;

[0010] An essential feature extraction module is used to extract the essential features of the ice-making machine operating parameter time series intensification correlation feature diagram to obtain the purified ice-making machine operating parameter time series essential features; and

[0011] The operating state detection module is used to determine whether the operating state of the monitored ice maker is abnormal based on the time sequence essential characteristics of the operating parameters of the purified ice maker.

[0012] For example, according to the ice machine fault monitoring system based on Internet of Things technology according to an embodiment of the present disclosure, the deep neural network model is a convolutional neural network model.

[0013] For example, according to the ice-making machine fault monitoring system based on the Internet of Things technology of the embodiment of the present disclosure, the operating parameter time series feature reinforcement module is used to: process the ice-making machine operating parameter time series correlation feature diagram through the feature interaction reinforcement module based on the triple interactive attention module with the following triple interactive attention interaction formula to obtain the ice-making machine operating parameter time series reinforcement correlation feature diagram;

[0014] Among them, the triple interactive attention interaction formula is:

[0015] α1=δ(W 2 *σ(W 1 *X))

[0016]

[0017] α2=δ(W 2 *σ(W 1 *X1))

[0018]

[0019] α3=δ(W 2 *σ(W 1 *X2))

[0020]

[0021] Where X is each feature matrix along the channel dimension in the temporal correlation feature graph of the ice machine operating parameters, W 1 represents a 1×1 convolution, W 2 represents a 7×7 convolution, * represents a convolution operation, σ(·) represents a ReLU function, δ(·) represents a Sigmoid function, α1 is the first weight matrix, represents the Hadamard product, is the first attention feature map, X1 is each feature matrix along the height dimension in the ice-making machine operating parameter time series correlation feature map, α2 is the second weight matrix, Trans(·) represents the transposition transformation, is the second attention feature map, X2 is each feature matrix along the width dimension in the ice machine operating parameter time series correlation feature map, α3 is the third weight matrix, Trans(·) represents the transposition transformation, is the third attention feature map, f(·) represents the sum and average function, It is a time series enhanced correlation characteristic diagram of the ice making machine operating parameters.

[0022] For example, according to the ice machine fault monitoring system based on the Internet of Things technology according to an embodiment of the present disclosure, the essential feature extraction module includes:

[0023] The essential feature expression unit is used to input the ice-making machine operating parameter time series reinforcement association feature map into a feature purifier based on an essential feature extraction network to obtain a purified ice-making machine operating parameter time series essential feature vector as the purified ice-making machine operating parameter time series essential feature.

[0024] For example, according to the ice maker fault monitoring system based on the Internet of Things technology according to the embodiment of the present disclosure, the operating parameter timing enhanced correlation feature optimization unit is used to optimize and fuse the ice maker operating parameter timing enhanced correlation feature diagram and the ice maker operating parameter timing enhanced correlation feature diagram to obtain the ice maker operating parameter timing enhanced correlation feature diagram.

[0025] For example, according to the ice-making machine fault monitoring system based on the Internet of Things technology of the embodiment of the present disclosure, the essential feature expression unit is used to: input the ice-making machine operating parameter time series reinforcement association feature map into the feature purifier based on the essential feature extraction network to process it with the following essential formula to obtain the purified ice-making machine operating parameter time series essential feature vector;

[0026] Among them, the prototype formula is:

[0027]

[0028] Among them, v i (x) and v j (x) are respectively the i-th and j-th ice-maker operating parameter time series enhanced association feature vectors in the sequence of ice-maker operating parameter time series enhanced association feature vectors obtained by quantizing each feature matrix along the channel dimension in the ice-maker operating parameter time series enhanced association feature graph, V k is the sequence of the ice-making machine operating parameter time series intensive association feature vectors, log represents the logarithmic function value with base 2, L is the length of each ice-making machine operating parameter time series intensive association feature vector, M is the number of vectors in the sequence of the ice-making machine operating parameter time series intensive association feature vector - 1, D iis the eigenvalue of each position in the local time series correlation difference eigenvector of the ice machine operating parameters, N s is the length of the local time series associated differential feature vector of the ice-making machine operating parameters, exp(·) is an exponential operation, and V is the time series essential feature vector of the purified ice-making machine operating parameters.

[0029] For example, according to the ice maker fault monitoring system based on the Internet of Things technology according to the embodiment of the present disclosure, the operating status detection module is used to: pass the purified ice maker operating parameter time series essential feature vector through a classifier-based fault monitoring result generator to obtain a fault monitoring result, and the fault monitoring result is used to indicate whether there is any abnormality in the operating status of the monitored ice maker.

[0030] For example, according to the ice machine fault monitoring system based on Internet of Things technology according to the embodiment of the present disclosure, a training module is used to train the operating parameter time series correlation pattern feature extractor based on the deep neural network model, the feature interaction enhancement module based on the triple interaction attention module, the feature purifier based on the essential feature extraction network and the fault monitoring result generator based on the classifier.

[0031] For example, according to the ice maker fault monitoring system based on the Internet of Things technology of the embodiment of the present disclosure, the training module includes: a training data acquisition unit, which is used to obtain training data, and the training data includes: a time series of training operating parameters of the monitored ice maker, wherein the training operating parameters of the ice maker include training temperature values, training pressure values, training current values ​​and training voltage values; a training parameter time series data arrangement unit, which is used to arrange the time series of the training operating parameters according to the time dimension and the operating parameter sample dimension into a training ice maker operating parameter time series associated input matrix; a training operating parameter time series associated feature extraction unit, which is used to extract features from the training ice maker operating parameter time series associated input matrix through an operating parameter time series associated pattern feature extractor based on a deep neural network model to obtain a training ice maker operating parameter time series associated feature map; a training operating parameter time series feature reinforcement unit, which is used to extract features from the training ice maker operating parameter time series associated feature map through a feature interaction reinforcement module based on a triple interactive attention module to obtain a training ice maker; The invention relates to an ice machine operating parameter time series reinforced association feature map; a training essential feature expression unit, which is used to input the training ice machine operating parameter time series reinforced association feature map into a feature purifier based on an essential feature extraction network to obtain a training purified ice machine operating parameter time series essential feature vector; a classification loss unit, which is used to pass the training purified ice machine operating parameter time series essential feature vector through a fault monitoring result generator based on a classifier to obtain a classification loss function value; a loss function value calculation unit, which is used to calculate the purified ice machine operating parameter time series essential loss function value based on the training purified ice machine operating parameter time series essential feature vector; a training unit, which is used to train the operating parameter time series association pattern feature extractor based on a deep neural network model, the feature interaction reinforcement module based on a triple interactive attention module, the feature purifier based on an essential feature extraction network, and the fault monitoring result generator based on a classifier using the weighted sum of the classification loss function value and the purified ice machine operating parameter time series essential loss function value as the final loss function value.

[0032] The embodiment of the present disclosure also provides an ice machine fault monitoring method based on the Internet of Things technology, which includes:

[0033] Acquire a time series of operating parameters of the monitored ice-making machine collected by the sensor module, wherein the operating parameters of the ice-making machine include temperature value, pressure value, current value and voltage value;

[0034] Arranging the time series of the operating parameters according to the time dimension and the operating parameter sample dimension into an ice-making machine operating parameter time series correlation input matrix;

[0035] Extracting features from the ice-making machine operating parameter time series correlation input matrix using an operating parameter time series correlation pattern feature extractor based on a deep neural network model to obtain an ice-making machine operating parameter time series correlation feature graph;

[0036] The ice-making machine operating parameter time series correlation feature graph is passed through a feature interaction reinforcement module based on a triple interaction attention module to obtain an ice-making machine operating parameter time series reinforcement correlation feature graph;

[0037] Extracting essential features from the ice-making machine operating parameter time series intensification correlation feature diagram to obtain purified ice-making machine operating parameter time series essential features; and

[0038] Based on the essential characteristics of the time sequence of the operating parameters of the purified ice-making machine, it is determined whether the operating state of the monitored ice-making machine is abnormal.

[0039] For example, according to the ice maker fault monitoring method based on Internet of Things technology according to an embodiment of the present disclosure, the deep neural network model is a convolutional neural network model.

[0040] According to the ice maker fault monitoring system based on the Internet of Things technology of the embodiment of the present disclosure, the Internet of Things technology is used to monitor and collect the operating parameters of the ice maker in real time through a sensor network, such as temperature, pressure, current and voltage, and transmit the data to a cloud server, so as to use data analysis and processing algorithms in the cloud server to perform time series collaborative analysis of the operating parameters of these ice makers, thereby realizing real-time monitoring of the operating status of the ice maker and intelligent fault diagnosis, so as to improve the operating efficiency and safety of the ice maker, reduce maintenance costs and time, and improve user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings of the embodiments of the present disclosure. Obviously, the drawings described below only relate to some embodiments of the present disclosure, and are not intended to limit the present disclosure.

[0042] Figure 1 A schematic diagram of the application architecture of an ice machine fault monitoring system based on the Internet of Things technology in an embodiment of the present disclosure is shown;

[0043] Figure 2 A schematic diagram of the structure of an ice machine fault monitoring system based on the Internet of Things technology in an embodiment of the present disclosure is shown;

[0044] Figure 3 A flow chart of an ice machine fault monitoring method based on Internet of Things technology in an embodiment of the present disclosure is shown;

[0045] Figure 4An application scenario diagram of an ice machine fault monitoring system based on Internet of Things technology in an embodiment of the present disclosure is shown; and

[0046] Figure 5 A schematic diagram of a storage medium according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present disclosure, rather than all embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present disclosure.

[0048] The terms used in this specification are those general terms currently widely used in the art in consideration of the functions of the present disclosure, but these terms may vary according to the intention of a person of ordinary skill in the art, precedents, or new technologies in the art. In addition, specific terms may be selected by the applicant, and in this case, their detailed meanings will be described in the detailed description of the present disclosure. Therefore, the terms used in the specification should not be understood as simple names, but rather as an overall description based on the meaning of the terms and the present disclosure.

[0049] Although the present disclosure makes various references to certain modules in the system according to embodiments of the present disclosure, any number of different modules can be used and run on a user terminal and / or server. The modules are illustrative only, and different aspects of the systems and methods can use different modules.

[0050] Flowcharts are used in this disclosure to illustrate the operations performed by the system according to the embodiments of the present disclosure. It should be understood that the preceding or following operations are not necessarily performed precisely in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0051] Figure 1 A schematic diagram of the application architecture of an ice machine fault monitoring system based on the Internet of Things technology in an embodiment of the present disclosure is shown, including a server 100 and a terminal device 200.

[0052] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some embodiments, the data exchanged through the network is represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0053] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0054] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present disclosure, the memory 120 may be used to store the corresponding method of the ice maker fault monitoring system based on the Internet of Things technology in the embodiment of the present disclosure.

[0055] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the ice maker fault monitoring methods based on the Internet of Things technology in the embodiments of the present disclosure according to the obtained program instructions.

[0056] In addition, the application architecture diagram in the embodiment of the present disclosure is intended to more clearly illustrate the technical solution in the embodiment of the present disclosure, and does not constitute a limitation on the technical solution provided in the embodiment of the present disclosure. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present disclosure is also applicable to similar problems.

[0057] The following is a non-restrictive description of the ice maker fault monitoring system based on the Internet of Things technology provided according to at least one embodiment of the present disclosure through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present disclosure.

[0058] With the rapid development of Internet of Things technology, the ice machine industry has also ushered in new development opportunities. Based on this, in the technical solution of this application, an ice machine fault monitoring system based on Internet of Things technology is proposed, which can use Internet of Things technology to monitor and collect the operating parameters of the ice machine in real time through a sensor network, such as temperature, pressure, current and voltage, and transmit the data to a cloud server, so as to use data analysis and processing algorithms in the cloud server to perform time series collaborative analysis of the operating parameters of these ice machines, thereby realizing real-time monitoring of the operating status of the ice machine and intelligent fault diagnosis, so as to improve the operating efficiency and safety of the ice machine, reduce maintenance costs and time, and improve user satisfaction.

[0059] Figure 2The schematic diagram of the structure of the ice maker fault monitoring system 800 based on the Internet of Things technology in the embodiment of the present disclosure is shown. The ice maker fault monitoring system 800 based on the Internet of Things technology includes: an operating parameter data acquisition module 810, which is used to obtain the time series of the operating parameters of the monitored ice maker collected by the sensor module, wherein the operating parameters of the ice maker include temperature value, pressure value, current value and voltage value; an operating parameter time series data arrangement module 820, which is used to arrange the time series of the operating parameters according to the time dimension and the operating parameter sample dimension into an ice maker operating parameter time series associated input matrix; an operating parameter time series associated feature extraction module 830, which is used to extract the ice maker operating parameter time series associated pattern feature extractor based on the deep neural network model. The invention relates to a method for extracting features from a sequential correlation input matrix to obtain a time series correlation feature diagram of ice maker operating parameters; an operating parameter time series feature enhancement module 840, which is used to obtain an ice maker operating parameter time series enhanced correlation feature diagram by subjecting the ice maker operating parameter time series associated feature diagram to a feature interaction enhancement module based on a triple interactive attention module; an essential feature extraction module 850, which is used to extract essential features from the ice maker operating parameter time series enhanced correlation feature diagram to obtain a purified ice maker operating parameter time series essential feature; and an operating status detection module 860, which is used to determine whether there is an abnormality in the operating status of the monitored ice maker based on the purified ice maker operating parameter time series essential feature.

[0060] Specifically, in the technical solution of the present application, first, the time series of the operating parameters of the monitored ice maker collected by the sensor module is obtained, wherein the operating parameters of the ice maker include temperature value, pressure value, current value and voltage value. It should be understood that the various operating parameters of the ice maker usually have time series correlation, that is, there are certain correlations and regularities between the parameter values ​​at adjacent time points. At the same time, there are also mutual correlations between the various operating parameters, which can reflect the operating state of the ice maker. Therefore, in order to better understand the operating state of the ice maker, in the technical solution of the present application, it is necessary to arrange the time series of the operating parameters according to the time dimension and the operating parameter sample dimension into an ice maker operating parameter time series correlation input matrix. In this way, the distribution information of each motion parameter of the ice maker in the time series can be integrated according to the time dimension and the operating parameter sample dimension, so that these operating parameter data can be better used for feature extraction and fault monitoring, so as to achieve accurate monitoring and intelligent diagnosis of the operating state of the ice maker.

[0061] Then, the ice-making machine operating parameter temporal correlation input matrix is ​​subjected to feature mining in an operating parameter temporal correlation pattern feature extractor based on a convolutional neural network model to extract the temporal collaborative correlation feature information of each operating parameter of the ice-making machine in the time dimension, thereby obtaining the ice-making machine operating parameter temporal correlation feature graph.

[0062] Correspondingly, in the operating parameter time series correlation feature extraction module 830, the deep neural network model is a convolutional neural network model. It is worth mentioning that the convolutional neural network (CNN) is a deep learning model specifically used to process data with a similar grid structure. The core idea of ​​the convolutional neural network is to extract the features of the input data through the convolution layer and the pooling layer, so as to achieve efficient processing and learning of the data. In the ice maker fault monitoring system of the Internet of Things technology, the convolutional neural network model is used in the operating parameter time series correlation feature extraction module 830 to extract features from the ice maker operating parameter time series correlation input matrix to obtain the ice maker operating parameter time series correlation feature map. The convolutional neural network mainly includes: 1. Convolutional layer: The convolutional layer slides on the input data through a filter to extract local features. These filters learn different features, such as edges, textures, etc. The convolution operation can reduce the number of parameters while retaining spatial structure information. 2. Pooling Layer: The pooling layer is used to reduce the size of the feature map, reduce computational complexity, and enhance the model's ability to learn translation invariance. Common pooling operations include maximum pooling and average pooling. 3. Activation Function: The activation function introduces nonlinear characteristics, allowing the neural network to learn complex patterns and relationships. Common activation functions include ReLU (Rectified Linear Unit) and Sigmoid function. 4. Multi-layer structure: Convolutional neural networks are usually composed of multiple convolutional layers, pooling layers, and fully connected layers. By stacking these layers, high-level features of the data can be extracted. 5. Weight Sharing: In the convolutional layer, the same filter shares weights on the entire input data, which can reduce the number of parameters and improve the generalization ability of the model.

[0063] Further, considering that the three dimensions of the height, width and channel of the ice maker operating parameter time series correlation feature graph respectively contain the correlation features of each operating parameter of the ice maker in different aspects. Therefore, in order to enhance the interactive fusion of feature information in the three dimensions of the height, width and channel of the ice maker operating parameter time series correlation feature graph, so as to more comprehensively capture the correlation and pattern in the operating parameter data, in the technical solution of the present application, the ice maker operating parameter time series correlation feature graph is further subjected to the feature interaction enhancement module based on the triple interactive attention module to obtain the ice maker operating parameter time series enhanced correlation feature graph. It should be understood that this multi-dimensional interaction helps to improve the model's understanding and characterization capabilities of data features, so as to more accurately determine whether there is an abnormality in the working state of the ice maker. Specifically, enhancing feature information in the height dimension can help the model better capture the feature change trends and correlation relationships of each operating parameter at different time points, so as to better understand the time series characteristics of the data. Enhancing feature information in the width dimension can promote feature interaction and fusion between different operating parameters, help the model to more comprehensively consider the correlation relationship between different operating parameters, and improve the representation ability of features. Enhancement of feature information in the channel dimension can help the model better learn the correlation and importance between different features, improve the ability to extract key information, and thus more accurately judge abnormal situations. Therefore, by performing feature enhancement and interactive fusion of information in the three dimensions of height, width, and channel, the model can more comprehensively understand the complexity and diversity of these operating parameter data, thereby better learning the potential patterns and laws between the various operating parameter data, and improving the detection performance of abnormal working conditions of ice machines.

[0064] Correspondingly, the operating parameter time series feature enhancement module 840 is used to: process the ice maker operating parameter time series association feature graph through the feature interaction enhancement module based on the triple interactive attention module with the following triple interactive attention interaction formula to obtain the ice maker operating parameter time series enhancement association feature graph; wherein the triple interactive attention interaction formula is:

[0065] α1=δ(W 2 *σ(W 1 *X))

[0066]

[0067] α2=δ(W 2 *σ(W 1 *X1))

[0068]

[0069] α3=δ(W 2 *σ(W1 *X2))

[0070]

[0071] Where X is each feature matrix along the channel dimension in the temporal correlation feature graph of the ice machine operating parameters, W 1 represents a 1×1 convolution, W 2 represents a 7×7 convolution, * represents a convolution operation, σ·) represents a ReLU function, δ(·) represents a Sigmoid function, α1 is the first weight matrix, represents the Hadamard product, is the first attention feature map, X1 is each feature matrix along the height dimension in the ice-making machine operating parameter time series correlation feature map, α2 is the second weight matrix, Trans(·) represents the transposition transformation, is the second attention feature map, X2 is each feature matrix along the width dimension in the ice machine operating parameter time series correlation feature map, α3 is the third weight matrix, Trans(·) represents the transposition transformation, is the third attention feature map, f(·) represents the sum and average function, It is a time series enhanced correlation characteristic diagram of the ice making machine operating parameters.

[0072] It should be understood that the ice maker operating parameter time series enhanced association feature map is generated through a convolutional neural network model and triple interactive feature enhancement. This feature map contains the time domain correlation feature information between the various motion parameters of the ice maker, which can reflect the dynamic changes and correlations between these operating parameters, thereby reflecting the working state of the ice maker. However, the ice maker operating parameter time series enhanced association feature map may contain a large amount of redundant information and noise, which may not be efficient and accurate enough for subsequent analysis and anomaly detection. Therefore, it is necessary to further process the time series association features between the various operating parameters of the ice maker to extract more essential and useful feature vectors. Based on this, in order to reduce the impact of the deviated essential semantic features of the operating state of the ice maker on the prototype calculation, in the technical solution of the present application, the ice maker operating parameter time series enhanced association feature map is input into a feature purifier based on an essential feature extraction network to obtain a purified ice maker operating parameter time series essential feature vector. By using the feature purifier based on the essential feature extraction network to purify the ice-making machine operating parameter time series enhanced correlation feature map, the time series correlation essence and purified feature information between the various operating parameters of the ice-making machine can be extracted, and these features can be used for subsequent ice-making machine operating state detection to ensure the accuracy of judgment. In this way, the influence of redundant information and noise can be reduced, the expression ability and discrimination ability of the operating state characteristics of the ice-making machine can be improved, and accurate monitoring and intelligent diagnosis of the operating state of the ice-making machine can be achieved.

[0073] Correspondingly, the essential feature extraction module 850 includes: an essential feature expression unit, which is used to input the ice maker operating parameter timing reinforcement association feature map into a feature purifier based on the essential feature extraction network to obtain a purified ice maker operating parameter timing essential feature vector as the purified ice maker operating parameter timing essential feature.

[0074] The essential feature expression unit is used to: input the ice-making machine operating parameter time series reinforcement association feature map into the feature purifier based on the essential feature extraction network to process it with the following essential formula to obtain the purified ice-making machine operating parameter time series essential feature vector; wherein the prototype formula is:

[0075]

[0076] Among them, v i (x) and v j (x) are respectively the i-th and j-th ice-maker operating parameter time series enhanced association feature vectors in the sequence of ice-maker operating parameter time series enhanced association feature vectors obtained by quantizing each feature matrix along the channel dimension in the ice-maker operating parameter time series enhanced association feature graph, V k is the sequence of the ice-making machine operating parameter time series intensive association feature vectors, log represents the logarithmic function value with base 2, L is the length of each ice-making machine operating parameter time series intensive association feature vector, M is the number of vectors in the sequence of the ice-making machine operating parameter time series intensive association feature vector - 1, D i is the eigenvalue of each position in the local time series correlation difference eigenvector of the ice machine operating parameters, N s is the length of the local time series associated differential feature vector of the ice-making machine operating parameters, exp(·) is an exponential operation, and V is the time series essential feature vector of the purified ice-making machine operating parameters.

[0077] Then, the purified ice-making machine operating parameter time series essential feature vector is passed through a classifier-based fault monitoring result generator to obtain a fault monitoring result, and the fault monitoring result is used to indicate whether there is an abnormality in the operating state of the monitored ice-making machine. In other words, the purified and feature-enhanced time series associated essential feature information of each operating parameter of the ice-making machine is used for processing, so as to detect the operating state of the ice-making machine, thereby determining whether there is an abnormality in the operating state of the monitored ice-making machine. In this way, real-time monitoring of the operating state of the ice-making machine and intelligent fault diagnosis can be achieved, so as to improve the operating efficiency and safety of the ice-making machine, reduce maintenance costs and time, and improve user satisfaction.

[0078] Correspondingly, the operating status detection module 860 is used to: pass the purified ice-making machine operating parameter time series essential feature vector through a classifier-based fault monitoring result generator to obtain a fault monitoring result, and the fault monitoring result is used to indicate whether there is an abnormality in the operating status of the monitored ice-making machine.

[0079] Specifically, the operating status detection module 860 includes: a fully connected encoding unit, which uses the fully connected layer of the classifier-based fault monitoring result generator to fully connect the purified ice maker operating parameter time series essential feature vector to obtain a coded classification feature vector; and a classification unit, which is used to input the coded classification feature vector into the Softmax classification function of the classifier-based fault monitoring result generator to obtain the fault monitoring result.

[0080] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

[0081] Furthermore, in the technical solution of the present application, the ice machine fault monitoring system based on the Internet of Things technology also includes: a training module for training the operating parameter time series correlation pattern feature extractor based on the deep neural network model, the feature interaction enhancement module based on the triple interactive attention module, the feature purifier based on the essential feature extraction network and the fault monitoring result generator based on the classifier. The training module includes: a training data acquisition unit for acquiring training data, wherein the training data includes: a time series of training operating parameters of a monitored ice maker, wherein the training operating parameters of the ice maker include training temperature values, training pressure values, training current values, and training voltage values; a training parameter time series data arrangement unit for arranging the time series of the training operating parameters into a training ice maker operating parameter time series associated input matrix according to the time dimension and the operating parameter sample dimension; a training operating parameter time series associated feature extraction unit for extracting features from the training ice maker operating parameter time series associated input matrix through an operating parameter time series associated pattern feature extractor based on a deep neural network model to obtain a training ice maker operating parameter time series associated feature map; a training operating parameter time series feature reinforcement unit for extracting features from the training ice maker operating parameter time series associated feature map through a feature interaction reinforcement module based on a triple interactive attention module to obtain a training ice maker operating parameter time series reinforced association feature map. a training essential feature expression unit, used for inputting the training ice-making machine operating parameter time series reinforcement association feature map into a feature purifier based on an essential feature extraction network to obtain a training purified ice-making machine operating parameter time series essential feature vector; a classification loss unit, used for passing the training purified ice-making machine operating parameter time series essential feature vector through a fault monitoring result generator based on a classifier to obtain a classification loss function value; a loss function value calculation unit, used for calculating the purified ice-making machine operating parameter time series essential loss function value based on the training purified ice-making machine operating parameter time series essential feature vector; a training unit, used for training the operating parameter time series association pattern feature extractor based on a deep neural network model, the feature interaction reinforcement module based on a triple interactive attention module, the feature purifier based on an essential feature extraction network, and the fault monitoring result generator based on a classifier using the weighted sum of the classification loss function value and the purified ice-making machine operating parameter time series essential loss function value as the final loss function value.

[0082] In the technical solution of the present application, each eigenvalue of the training ice-making machine operating parameter time series association feature map expresses the time series-sample dimension cross-local association feature of the training operating parameter, and each feature matrix follows the channel distribution of the convolutional neural network model. In this way, after the training ice-making machine operating parameter time series association feature map passes through the feature interaction reinforcement module based on the triple interactive attention module, interactive attention will be performed on the spatial distribution dimension within the feature matrix and the channel distribution dimension between the feature matrices. However, considering that the significance and importance of the feature distribution on the spatial distribution dimension within the feature matrix are much higher than the channel distribution dimension between the feature matrices, the training ice-making machine operating parameter time series reinforcement association feature map may have a relatively obvious complex feature expression classification understanding problem of insufficient training purification ice-making machine operating parameter time series essential feature vector obtained by inputting the training ice-making machine operating parameter time series reinforcement association feature map into the feature purifier based on the essential feature extraction network due to the distribution expression difference between the eigenvalues ​​on each dimension, which affects the accuracy of the classification result obtained by the classifier of the training purification ice-making machine operating parameter time series essential feature vector.

[0083] Based on this, the present application further introduces a loss function for promoting classification understanding based on the complex feature expression of the training purification ice machine operating parameter time series essential feature vector, that is, based on the training purification ice machine operating parameter time series essential feature vector, the purification ice machine operating parameter time series essential loss function value is calculated. Wherein, the loss function value calculation unit includes: first, by calculating the mean and variance between the eigenvalues ​​of any two positions of the training purification ice machine operating parameter time series essential feature vector to obtain the mean weight matrix and the variance weight matrix, and then after query-multiplying the training purification ice machine operating parameter time series essential feature vector with the mean weight matrix and the variance weight matrix respectively, the Frobenius norm of the autocorrelation matrix of the two eigenvectors is calculated, and the product of the Frobenius norm of the autocorrelation matrix of the training purification ice machine operating parameter time series essential feature vector and the weight as a hyperparameter is subtracted to obtain the purification ice machine operating parameter time series essential loss function value.

[0084] For example, expressed as

[0085]

[0086] Wherein, V represents the time series essential feature vector of the training purification ice machine operation parameters, M μ is the mean weight matrix, M σ is the variance weight matrix, T represents the transpose of the vector, represents the matrix multiplication operation, α is the weight as a hyperparameter, ||·|| Frepresents the Forbenius norm of the matrix, Loss represents the time series essential loss function value of the operating parameters of the purified ice making machine, and as described above:

[0087]

[0088] Among them, v i represents the ith eigenvalue of the training purified ice machine operating parameter time series essential eigenvector, v j represents the jth eigenvalue of the training purified ice machine operating parameter time series essential eigenvector, M μ (i,j) represents the eigenvalue of the (i,j)th position of the mean weight matrix, M σ (i,j) represents the eigenvalue at the (i,j)th position of the variance weight matrix.

[0089] That is, the time series essential loss function value of the operating parameters of the purified ice maker is based on the overall distributed structural bottleneck of the training time series essential feature vector of the operating parameters of the purified ice maker determined by the multi-dimensional interaction of the detail group of the training time series essential feature vector of the operating parameters of the purified ice maker, and the correlation between the structural details and the macro-structured feature representation behavior of the training time series essential feature vector of the operating parameters of the purified ice maker is simulated through the low-rank structured reasoning of the training time series essential feature vector of the operating parameters of the purified ice maker. In this way, by training the classifier with the time series essential loss function value of the operating parameters of the purified ice maker, the structural detail dependency relative to the macro-complexity can be modeled in the classification scenario, thereby promoting the understanding of the macro-complex feature expression based on the structural details of the training time series essential feature vector of the operating parameters of the purified ice maker, and improving the accuracy of the classification results obtained by the classifier. In this way, the real-time monitoring of the operating status of the ice maker and the intelligent diagnosis of faults can be more accurately realized to improve the operating efficiency and safety of the ice maker.

[0090] Based on the above embodiments, see Figure 3 As shown in FIG. 1 , a flow chart of an ice maker fault monitoring method based on the Internet of Things technology in an embodiment of the present disclosure is shown. For example, the ice maker fault monitoring method based on the Internet of Things technology can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 3As shown, according to the ice maker fault monitoring method based on the Internet of Things technology of the embodiment of the present disclosure, the method comprises the following steps: S510, acquiring the time series of the operating parameters of the monitored ice maker collected by the sensor module, wherein the operating parameters of the ice maker include temperature value, pressure value, current value and voltage value; S520, arranging the time series of the operating parameters into an ice maker operating parameter time series correlation input matrix according to the time dimension and the operating parameter sample dimension; S530, extracting the ice maker operating parameters by using the operating parameter time series correlation pattern feature extractor based on the deep neural network model. The time series correlation input matrix is ​​subjected to feature extraction to obtain a time series correlation feature diagram of the ice maker operating parameters; S540, the time series correlation feature diagram of the ice maker operating parameters is subjected to a feature interaction reinforcement module based on a triple interaction attention module to obtain a time series reinforcement correlation feature diagram of the ice maker operating parameters; S550, the ice maker operating parameter time series reinforcement correlation feature diagram is subjected to essential feature extraction to obtain a time series essential feature of the purified ice maker operating parameters; and, S560, based on the time series essential feature of the purified ice maker operating parameters, it is determined whether the operating state of the monitored ice maker is abnormal.

[0091] In one example, the deep neural network model is a convolutional neural network model.

[0092] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned ice maker fault monitoring method based on the Internet of Things technology have been referred to above. Figure 2 The description of the ice maker fault monitoring system 800 based on the Internet of Things technology has been introduced in detail, and therefore, its repeated description will be omitted.

[0093] Figure 4 FIG. 1 is an application scenario diagram of an ice machine fault monitoring system based on Internet of Things technology according to an embodiment of the present disclosure. Figure 4 As shown, in this application scenario, first, the time series of the operating parameters of the monitored ice maker collected by the sensor module is obtained (for example, Figure 4 D), wherein the operating parameters of the ice maker include temperature value, pressure value, current value and voltage value, and then the time series of the operating parameters are input to a server (for example, Figure 4 In S) shown in , the server can use the ice maker fault monitoring algorithm based on the Internet of Things technology to process the time series of the operating parameters to obtain a fault monitoring result indicating whether there is an abnormality in the operating status of the monitored ice maker.

[0094] Based on the above embodiment, another exemplary embodiment of an electronic device is also provided in the embodiments of the present disclosure. In some possible implementations, the electronic device in the embodiments of the present disclosure may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the ice maker fault monitoring method based on the Internet of Things technology in the above embodiment when executing the program.

[0095] For example, the electronic device is disclosed in the present invention Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0096] An embodiment of the present disclosure also provides a computer-readable storage medium. Figure 5 FIG. 1 is a schematic diagram of a computer-readable storage medium 1000 according to an embodiment of the present disclosure. Figure 5 As shown, the computer-readable storage medium 1000 stores computer-executable instructions 1001. When the computer-executable instructions 1001 are executed by the processor, the ice machine fault monitoring method based on the Internet of Things technology according to the embodiment of the present disclosure described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.

[0097] The embodiment of the present disclosure also provides a computer program product or a computer program, which includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the ice maker fault monitoring method based on the Internet of Things technology according to the embodiment of the present disclosure.

[0098] Those skilled in the art will appreciate that the contents disclosed in this disclosure may be subject to various modifications and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0099] In addition, although the present disclosure makes various references to certain units in the system according to embodiments of the present disclosure, any number of different units may be used and run on the client and / or server. The units are illustrative only, and different aspects of the systems and methods may use different units.

[0100] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of a software functional module. The present disclosure is not limited to any particular form of combination of hardware and software.

[0101] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or highly formal sense, unless explicitly defined as such herein.

[0102] The above is an explanation of the present disclosure and should not be considered as a limitation thereof. Although several exemplary embodiments of the present disclosure have been described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present disclosure. Therefore, all such modifications are intended to be included within the scope of the present disclosure as defined in the claims. It should be understood that the above is an explanation of the present disclosure and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present disclosure is defined by the claims and their equivalents.

Claims

1. An ice machine fault monitoring system based on Internet of Things technology, characterized in that: include: An operating parameter data acquisition module, used for acquiring a time series of operating parameters of the monitored ice-making machine acquired by the sensor module, wherein the operating parameters of the ice-making machine include temperature value, pressure value, current value and voltage value; An operating parameter time series data arrangement module, used to arrange the time series of the operating parameters according to the time dimension and the operating parameter sample dimension into an ice machine operating parameter time series associated input matrix; An operating parameter time series correlation feature extraction module, used for extracting features from the ice maker operating parameter time series correlation input matrix through an operating parameter time series correlation pattern feature extractor based on a deep neural network model to obtain an ice maker operating parameter time series correlation feature graph; An operating parameter time series feature enhancement module, used for obtaining an ice-making machine operating parameter time series enhanced association feature diagram by a feature interaction enhancement module based on a triple interactive attention module; An essential feature extraction module is used to extract the essential features of the ice-making machine operating parameter time series intensification correlation feature diagram to obtain the purified ice-making machine operating parameter time series essential features; and The operating state detection module is used to determine whether the operating state of the monitored ice maker is abnormal based on the time sequence essential characteristics of the operating parameters of the purified ice maker.

2. The ice machine fault monitoring system based on Internet of Things technology according to claim 1 is characterized in that: The deep neural network model is a convolutional neural network model.

3. The ice machine fault monitoring system based on Internet of Things technology according to claim 2 is characterized in that: The operating parameter time series feature enhancement module is used to: process the ice-making machine operating parameter time series correlation feature graph through the feature interaction enhancement module based on the triple interactive attention module according to the following triple interactive attention interaction formula to obtain the ice-making machine operating parameter time series enhancement correlation feature graph; Among them, the triple interactive attention interaction formula is: α1=δ(W 2 *σ(W 1 *X)) α2=δ(W 2 *σ(W 1 *X1) α3=δ(W 2 *σ(W1*X2)) Where X is each feature matrix along the channel dimension in the temporal correlation feature graph of the ice machine operating parameters, W 1 represents a 1×1 convolution, W 2 represents a 7×7 convolution, * represents a convolution operation, σ(·) represents a ReLU function, δ(·) represents a Sigmoid function, α1 is the first weight matrix, represents the Hadamard product, is the first attention feature map, X1 is each feature matrix along the height dimension in the ice-making machine operating parameter time series correlation feature map, α2 is the second weight matrix, Trans(·) represents the transposition transformation, is the second attention feature map, X2 is each feature matrix along the width dimension in the ice machine operating parameter time series correlation feature map, α3 is the third weight matrix, Trans(·) represents the transposition transformation, is the third attention feature map, f(·) represents the sum and average function, It is a time series enhanced correlation characteristic diagram of the ice making machine operating parameters.

4. The ice machine fault monitoring system based on Internet of Things technology according to claim 3 is characterized in that: The essential feature extraction module comprises: The essential feature expression unit is used to input the ice-making machine operating parameter time series reinforcement association feature map into a feature purifier based on an essential feature extraction network to obtain a purified ice-making machine operating parameter time series essential feature vector as the purified ice-making machine operating parameter time series essential feature.

5. The ice machine fault monitoring system based on Internet of Things technology according to claim 4 is characterized in that: The essential feature expression unit is used to: input the ice-making machine operating parameter time series enhanced correlation feature map into the feature purifier based on the essential feature extraction network to process it with the following essential formula to obtain the purified ice-making machine operating parameter time series essential feature vector; Among them, the prototype formula is: Among them, v i (x) and v j (x) are respectively the i-th and j-th ice-maker operating parameter time series enhanced association feature vectors in the sequence of ice-maker operating parameter time series enhanced association feature vectors obtained by quantizing each feature matrix along the channel dimension in the ice-maker operating parameter time series enhanced association feature graph, V k is the sequence of the ice-making machine operating parameter time series intensive association feature vectors, log represents the logarithmic function value with base 2, L is the length of each ice-making machine operating parameter time series intensive association feature vector, M is the number of vectors in the sequence of the ice-making machine operating parameter time series intensive association feature vector - 1, D i is the eigenvalue of each position in the local time series correlation difference eigenvector of the ice machine operating parameters, N s is the length of the local time series associated differential feature vector of the ice-making machine operating parameters, exp(·) is an exponential operation, and V is the time series essential feature vector of the purified ice-making machine operating parameters.

6. The ice machine fault monitoring system based on Internet of Things technology according to claim 5 is characterized in that: The operation status detection module is used to: pass the time series essential feature vector of the operation parameter of the purified ice maker through a fault monitoring result generator based on a classifier to obtain a fault monitoring result, and the fault monitoring result is used to indicate whether there is an abnormality in the operation status of the monitored ice maker.

7. The ice machine fault monitoring system based on Internet of Things technology according to claim 6 is characterized in that: Also includes: A training module is used to train the operating parameter timing correlation pattern feature extractor based on the deep neural network model, the feature interaction enhancement module based on the triple interaction attention module, the feature purifier based on the essential feature extraction network and the fault monitoring result generator based on the classifier.

8. The method for preparing human placenta-derived hematopoietic stem cells according to claim 7, characterized in that: The training module comprises: A training data acquisition unit, used to acquire training data, the training data comprising: a time series of training operating parameters of the monitored ice-making machine, wherein the training operating parameters of the ice-making machine comprise training temperature values, training pressure values, training current values, and training voltage values; A training parameter time series data arrangement unit, used to arrange the time series of the training operation parameters according to the time dimension and the operation parameter sample dimension into a training ice machine operation parameter time series correlation input matrix; A training operating parameter time series correlation feature extraction unit, used for extracting features from the training ice-making machine operating parameter time series correlation input matrix through an operating parameter time series correlation pattern feature extractor based on a deep neural network model to obtain a training ice-making machine operating parameter time series correlation feature graph; A training operating parameter time series feature reinforcement unit, used for obtaining a training ice maker operating parameter time series reinforcement correlation feature graph through a feature interaction reinforcement module based on a triple interaction attention module; A training essential feature expression unit, used for inputting the training ice-making machine operating parameter time series reinforcement association feature map into a feature purifier based on an essential feature extraction network to obtain a training purified ice-making machine operating parameter time series essential feature vector; A classification loss unit, used for passing the training purified ice machine operation parameter time series essential feature vector through a classifier-based fault monitoring result generator to obtain a classification loss function value; A loss function value calculation unit, used to calculate the time series essential loss function value of the purified ice-making machine operating parameter based on the training purified ice-making machine operating parameter time series essential feature vector; A training unit is used to train the operating parameter time series association pattern feature extractor based on the deep neural network model, the feature interaction reinforcement module based on the triple interaction attention module, the feature purifier based on the essential feature extraction network and the fault monitoring result generator based on the classifier by taking the weighted sum of the classification loss function value and the purification ice machine operating parameter time series essential loss function value as the final loss function value.

9. A method for monitoring ice machine faults based on Internet of Things technology, characterized in that: include: Acquire a time series of operating parameters of the monitored ice-making machine collected by the sensor module, wherein the operating parameters of the ice-making machine include temperature value, pressure value, current value and voltage value; Arranging the time series of the operating parameters according to the time dimension and the operating parameter sample dimension into an ice-making machine operating parameter time series correlation input matrix; Extracting features from the ice-making machine operating parameter time series correlation input matrix using an operating parameter time series correlation pattern feature extractor based on a deep neural network model to obtain an ice-making machine operating parameter time series correlation feature graph; The ice-making machine operating parameter time series correlation feature graph is passed through a feature interaction reinforcement module based on a triple interaction attention module to obtain an ice-making machine operating parameter time series reinforcement correlation feature graph; Extracting essential features from the ice-making machine operating parameter time series intensification correlation feature diagram to obtain purified ice-making machine operating parameter time series essential features; and Based on the essential characteristics of the time sequence of the operating parameters of the purified ice-making machine, it is determined whether the operating state of the monitored ice-making machine is abnormal.

10. The ice machine fault monitoring method based on Internet of Things technology according to claim 9, characterized in that: The deep neural network model is a convolutional neural network model.

Images