Automatic early warning method for monitoring of Internet of Things sensing equipment based on deep learning

Abstract

The invention provides a deep learning-based IOT (Internet of Things) sensing equipment monitoring automatic early warning method, which comprises the following steps of: collecting sensor data generated by IOT sensing equipment, and realizing data collection through a sensor array, an IOT interface and cloud storage; processing and extracting data features by using a deep neural network, and establishing an equipment state recognition model by learning data modes of normal operation and abnormal states of equipment; equipment operation conditions are monitored in real time according to the model, possible abnormal conditions are predicted, and prediction accuracy is improved; when the equipment runs abnormally or an early warning condition is met, an early warning mechanism is triggered automatically, and maintenance measures are taken in time; feedback and new data of an operator are received in real time, the model performance is continuously optimized, and self-learning is achieved. According to the method, the accuracy of equipment fault prediction is improved, the fault rate and the maintenance cost of the equipment are remarkably reduced, the advantage ensures that the equipment can work more stably in the operation process, and economic losses caused by fault shutdown are reduced.

Description

Technical Field

[0001] The present invention relates to the field of Internet of Things technology and artificial intelligence technology, and in particular to an automatic early warning method for monitoring Internet of Things sensing devices based on deep learning. Background Art

[0002] In modern industrial production, the normal operation and maintenance of equipment are crucial to ensuring production efficiency and product quality. Traditional equipment monitoring methods usually rely on regular manual inspections and preventive maintenance, which not only consumes a lot of manpower and resources, but also makes it difficult to detect potential equipment failures in a timely manner, which can easily lead to sudden failures, resulting in production stagnation and economic losses. With the development of industrial Internet of Things (IoT) technology, more and more equipment is equipped with various sensors that can collect data on the operating status of equipment in real time. This provides a rich source of data and new technical means for equipment status monitoring and fault prediction.

[0003] In recent years, deep learning technology has made significant progress in computer vision, natural language processing, speech recognition and other fields, and has gradually been applied to the field of industrial equipment monitoring. Deep learning models can automatically extract features from massive data, perform complex pattern recognition and predictive analysis, and have strong generalization ability and prediction accuracy. By combining deep learning technology with the Internet of Things, real-time monitoring of equipment status and fault prediction can be achieved, improving the level of intelligent equipment management.

[0004] At present, equipment monitoring and early warning systems based on deep learning have been initially applied in some high-end manufacturing industries. For example, fault prediction and health management of key equipment such as aircraft engines and wind turbines. However, for traditional manufacturing industries such as textile printing and dyeing, although there are many types of equipment, complex operating environments, and equipment failures often lead to serious production interruptions and economic losses, the level of intelligence in equipment monitoring is still relatively low. Summary of the invention

[0005] In order to solve the technical problems in the prior art, the present invention provides an automatic early warning method for monitoring IoT sensing devices based on deep learning.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] First aspect:

[0008] An embodiment of the present invention provides an automatic early warning method for monitoring IoT sensing devices based on deep learning, including:

[0009] S1. Collect sensor data generated by IoT sensing devices. The sensor data is collected through sensor arrays, IoT interfaces and cloud storage. The sensor data includes temperature, humidity, voltage, current and pressure generated by the monitored devices during operation.

[0010] S2. Processing and feature extracting the sensor data using a deep neural network;

[0011] S3. Establishing a device state recognition model by learning the data patterns of the normal operation and abnormal state of the monitored device. The device state recognition model is a model for determining the current state of the device based on the data patterns of the device operation state. The device state recognition model is established by learning and analyzing the data of the device in the normal and abnormal state through a deep learning algorithm;

[0012] S4. Monitor the operating status of the monitored equipment in real time and make predictions according to the equipment status recognition model. Input the sensor data into the equipment status recognition model, monitor the equipment operating status in real time, and make predictions for abnormal situations that may occur in the future. At the same time, further data analysis and processing are performed based on historical data and equipment characteristics to improve the accuracy of equipment predictions.

[0013] S5. When the monitored equipment operates abnormally or reaches the warning condition, the warning mechanism is automatically triggered and maintenance measures are taken in time;

[0014] S6. Receive operator feedback information and newly connected IoT sensing device data in real time, and use the feedback information and IoT sensing device data to continuously optimize model performance and achieve self-learning.

[0015] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0016] (1) In the present invention, through deep learning technologies such as deep neural network (DNN) and time series weighted adaptive deep convolutional neural network (TWA-DCNN), the present invention can effectively process and analyze the sensor data of the equipment. In particular, the TWA-DCNN model combined with the time series feature weighting mechanism can give different weights to the time series data of the equipment at different operation stages, especially paying more attention to important moments and sudden fault events. This method significantly enhances the ability to identify sudden faults, timely monitors the equipment status and makes accurate predictions, thereby achieving early fault warning and reducing the occurrence of equipment failures;

[0017] (2) In the present invention, by introducing an adaptive fusion convolutional recurrent neural network (AF-CRNN), combining the feature fusion mechanism of convolutional neural network (CNN) and recurrent neural network (RNN), and a dynamic gating mechanism, feature selection and timing dependency can be effectively optimized. The dynamic gating mechanism enables the model to adaptively adjust information flow according to changes in the equipment operation mode, thereby improving the processing capability of time series data and enhancing the flexibility and accuracy of equipment status recognition. This innovation optimizes the performance of deep learning models in time series data processing and significantly improves the accuracy and sensitivity of fault prediction.

[0018] (3) In order to reduce overfitting and improve the adaptability of the model to anomalies at different stages, the present invention introduces the time enhancement loss function (TELF). By weighted learning of historical data and real-time data, TELF can help the model better learn the changes in abnormal patterns, thereby avoiding overfitting of the model to outdated data and improving the stability and robustness of the model under different operating environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 A system framework diagram of an automatic early warning method for monitoring IoT sensing devices based on deep learning provided by an embodiment of the present invention;

[0021] Figure 2 A data processing flow chart of an automatic early warning method for monitoring IoT sensing devices based on deep learning provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of a device status recognition model based on a deep learning IoT sensing device monitoring automatic early warning method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0024] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0025] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0026] Reference Manual Attached Figure 1-3 , showing a system framework diagram, a data processing flow chart and a schematic diagram of a device status recognition model for a method for automatic early warning of IoT sensing device monitoring based on deep learning provided by an embodiment of the present invention.

[0027] An embodiment of the present invention provides an automatic early warning method for monitoring IoT sensing devices based on deep learning, the method comprising the following steps:

[0028] S1. Collect sensor data generated by IoT sensing devices. The sensor data collection is achieved through sensor arrays, IoT interfaces and cloud storage. The sensor data includes temperature, humidity, voltage, current and pressure generated by the monitored equipment during operation;

[0029] S2, using deep neural network to process sensor data and extract features;

[0030] S3. By learning the data patterns of the normal operation and abnormal state of the monitored equipment, an equipment state recognition model is established. The equipment state recognition model is a model that determines the current state of the equipment based on the data pattern of the equipment operation state. The data of the equipment in normal and abnormal state are learned and analyzed through a deep learning algorithm to establish an equipment state recognition model.

[0031] S4. Monitor the operating status of the monitored equipment in real time and make predictions based on the equipment status recognition model. Input the sensor data into the equipment status recognition model, monitor the equipment operating status in real time, and predict possible abnormal situations in the future. At the same time, further data analysis and processing are performed based on historical data and equipment characteristics to improve the accuracy of equipment predictions.

[0032] S5. When the monitored equipment operates abnormally or reaches the warning condition, the warning mechanism is automatically triggered and maintenance measures are taken in time;

[0033] S6. Receive feedback from operators and data from newly connected IoT sensing devices in real time, and use the feedback information and data from IoT sensing devices to continuously optimize model performance and achieve self-learning.

[0034] S2 specifically includes:

[0035] S21, data preprocessing, before inputting into the deep neural network, the sensor data needs to be preprocessed, including standardization and normalization, to meet the input requirements of the neural network;

[0036] S22. Construct a deep neural network structure. The deep neural network includes an input layer, a hidden layer, and an output layer. Each layer is composed of a number of neurons, and takes the output of the previous layer as input. It performs nonlinear transformation through the activation function and generates new output. When transmitting information from the input layer to the output layer, the forward propagation simplified formula is used.

[0037]

[0038] In the above formula, σ is the activation function, w i is the weight, x i is the input, b is the bias term, and y is the output;

[0039] S23. Feature learning and extraction. During the training process, the weights and biases in the network are adjusted through the back-propagation algorithm so that the predicted output of the deep neural network is as close to the true label as possible. During the training process, the deep neural network automatically learns from the raw data and extracts features that help complete the task.

[0040] S22 specifically includes:

[0041] The input layer is used to receive raw sensor data, including devices that monitor temperature, humidity, and vibration;

[0042] The hidden layer includes multiple fully connected layers, convolutional layers, and recurrent layers. Each layer transforms the output of the previous layer into the input of the next layer through the weight matrix W and the bias b, using the activation function.

[0043] f(x)=max(0,x)

[0044] For nonlinear mapping, the fully connected layer is a feedforward neural network, the convolutional layer is suitable for image or sequence data, and the recurrent layer is suitable for time series data;

[0045] The output layer includes one or more nodes for classification or regression. Classification uses the softmax function to output probability distribution, and regression directly outputs numerical values.

[0046] S3 specifically includes:

[0047] S31. Data preparation, including data collection, data labeling, and data cleaning and normalization. Data collection is to collect data from sensors installed on the equipment under normal operating conditions and abnormal conditions. Data labeling is to annotate the collected data to distinguish normal data from abnormal data. Data cleaning and integration is to remove noise, fill in missing values, and normalize or standardize the data to facilitate model learning.

[0048] S32. Feature engineering, including feature extraction. Feature extraction is to extract meaningful features from raw data. Meaningful features include vibration amplitude and temperature change at specific frequencies. For image data, features are directly learned by convolutional neural networks (CNNs).

[0049] S33. Construct a deep learning algorithm model. The deep learning algorithm model includes a convolutional neural network (CNN) and a recurrent neural network (RNN). If the device state can be identified by image or spectrum analysis, a convolutional neural network is used. For time series data, a recurrent neural network is more suitable. The simplified formula of the convolution operation of the convolutional neural network is:

[0050]

[0051] In the above formula, c ij Represents an element of the feature map after convolution, f represents the filter, x represents the input image, and * represents the convolution operation. In the construction of RNN, a basic RNN unit is represented as,

[0052] ht=f(Wxt+Uh t-1 +b)

[0053] In the above formula, ht represents the hidden state at time step t, xt represents the input of the current time step, (W, U, b) represents the model parameters, and f represents the activation function. The activation function includes tanh or ReLU. Long short-term memory (LSTM) or gated recurrent unit (GRU) is a variant of RNN and is more suitable for learning long sequence data. The basic unit of the LSTM model is expressed as the following four formulas, which are used to update the unit state c t and the hidden state h t , and output predicted values

[0054] The forget gate determines the information that needs to be forgotten in the previous unit state.

[0055] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0056] The input gate determines the new information that should be updated to the cell state.

[0057] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0058] g t =tanh(W g ·[h t-1 ,x t ]+b g )

[0059] The cell state is updated, combining forget and input operations,

[0060] c t =f t ⊙c t-1 +i t ⊙g t

[0061] The output gate determines the cell state information that should be output as the new hidden state.

[0062] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0063] h t =o t ⊙tanh(c t )

[0064] In the above formula, f t represents the output of the forget gate, σ represents the sigmoid activation function, W f represents the weight matrix, b f represents the bias term, [h t-1 ,x t ] represents the concatenation of the hidden state of the previous time step and the input of the current time step, tanh represents the hyperbolic tangent function, W,b represents the model parameters, and ⊙ represents element-wise multiplication;

[0065] S34, training the model, including dividing the data set, selecting a loss function and an optimizer, and a catenary process, dividing the data set into a training set, a validation set, and a test set, and the training process is to adjust the model parameters through a back propagation algorithm, minimize the loss function, and iterate multiple rounds until the model is no longer significantly improved or reaches a predetermined stop condition;

[0066] S35. Evaluation and tuning, including performance evaluation and model tuning. Performance evaluation is to evaluate the model performance on the test set using accuracy, recall and F1 score. Model tuning is to adjust the model structure and hyperparameters based on the evaluation results or use deep CNN and bidirectional RNN for multiple iterative training to optimize performance.

[0067] S4 specifically includes:

[0068] S41, data collection and preprocessing, collecting parameters of the equipment during operation through sensors installed on the equipment, including temperature sensors, vibration sensors, and current and voltage sensors, and preprocessing the collected data, including noise removal, missing value processing, and data standardization, to facilitate subsequent analysis;

[0069] S42, feature engineering, extracting useful features from raw data. Features include time series trends, frequency components, statistical properties of signals, and complex features. Statistical properties of signals include mean, variance, and kurtosis. Complex features include descriptions of waveform shapes. The goal of feature selection is to find the set of variables that best characterizes the state of the device.

[0070] S43. Select or construct a device status recognition model, including supervised learning models, unsupervised learning models and deep learning models. Supervised learning models include support vector machines (SVM), random forests (RF) and gradient boosted trees (GBT), which are used for classification or regression problems to predict whether the device is about to fail or estimate the remaining service life. Unsupervised learning models include clustering algorithms (K-means, DBSCAN), which are used to discover patterns in data and identify normal and abnormal operating modes of the device. Deep learning models include recurrent neural networks (RNN), long short-term memory networks (LSTM) or gated recurrent units (GRU), which are suitable for processing time series data and capturing the changes in device status over time.

[0071] S44, real-time detection and prediction, input real-time sensor data into the trained model, the model evaluates the operating status of the equipment according to the characteristics of the current data, if the model predicts that the status deviates from the normal range, it can issue an early warning to prompt maintenance personnel to take preventive measures.

[0072] By continuously collecting data and feeding it back into the model, we use online learning or regular retraining to optimize model performance and improve prediction accuracy. At the same time, we combine expert knowledge and actual maintenance feedback to continuously adjust and optimize feature engineering and model parameters.

[0073] When the equipment is operating abnormally or reaches the warning condition, the warning mechanism is automatically triggered and maintenance measures are taken in time. When the equipment status detection model detects that the equipment is operating abnormally or reaches the warning condition, the warning mechanism is automatically triggered to notify relevant personnel to handle it. At the same time, corresponding maintenance measures are designed, such as replacing parts and repairing equipment, to ensure the normal operation of the equipment.

[0074] In a possible implementation, the device state recognition model performs weighted processing on the time series features in the sensor data by introducing a time series weighted adaptive deep convolutional neural network (TWA-DCNN) algorithm, and the weight is calculated by the following formula:

[0075]

[0076] Among them, w(t) is the weight value corresponding to the time point t, T is the current time point, λ is the weighted attenuation coefficient, α is the weight adjustment factor, |tT| represents the time difference, and the weighting function can dynamically adjust the sampling frequency, give higher weights to the sampling data at important moments, and enhance the model's ability to identify sudden faults.

[0077] The weighted attenuation coefficient and weight adjustment factor can dynamically weight the sampled data at different time points. The weighting function can dynamically adjust the sampling frequency to give higher weights to the sampled data at important moments, thereby improving the model's ability to respond to sudden failures. The role of this method is to identify potential failure modes in a timely manner for sudden changes or abnormal conditions in the device state without affecting the accuracy of the model due to interference from ordinary data. During the implementation process, by optimizing the weighting strategy, the model can more effectively process sensor data in a high-noise environment.

[0078] In a possible implementation, the device state recognition model specifically includes: using an adaptive fusion convolutional recurrent neural network (AF-CRNN), combining the feature fusion mechanism in the convolutional neural network (CNN) and the recurrent neural network (RNN), and introducing a self-created dynamic gating mechanism to optimize feature selection and timing dependency, and the dynamic gating mechanism is calculated by the following formula:

[0079] G(t)=σ(W g ·[h(t-1),x(t)]+b g )

[0080] Among them, G(t) is the dynamic gating value, h(t-1) is the hidden state at the previous moment, x(t) is the input feature at the current moment, and W g is the gating weight matrix, b gis the bias term, σ is the Sigmoid activation function, and the mechanism can adaptively adjust the information flow and enhance the model's ability to process time series data by learning changes in the device's operating mode.

[0081] The dynamic gating mechanism can adaptively adjust the flow of information, which is implemented by a gating formula, in which the input features at each moment are adjusted according to the hidden state and gating weight matrix of the previous moment. Through this mechanism, the model can adjust its emphasis on input data of different time periods according to the changes in the operating state of the device, thereby improving the model's ability to process complex time series data, especially when the device state changes suddenly, it can quickly capture anomalies and make accurate predictions.

[0082] In a possible implementation, the deep convolutional neural network specifically includes: the weighted deep convolutional neural network (TWA-DCNN) model is combined with a time enhancement loss function (TELF), and the loss functions of historical data and real-time data are weighted so that the model can perform differentiated learning on historical fault modes and real-time anomalies when learning abnormal modes. The loss function is defined as:

[0083]

[0084] Among them, L(t) is the loss value, y i is the actual value, is the predicted value, w i is the weighting coefficient, The error between model prediction and reality is reduced by weighting, and the adaptability of the model to abnormalities at different stages is improved.

[0085] This loss function weights the losses of historical data and real-time data so that the model can achieve differentiated learning between different time periods when learning abnormal patterns, thereby effectively reducing overfitting problems and improving the generalization ability of the model. By weighting the loss function, the model can better balance the impact of historical data on learning and the sensitivity of real-time data to changes in equipment status. This method ensures that the model has stronger adaptability when facing historical failure modes and real-time anomalies, and can capture potential failures in a timely manner and issue early warnings. During the implementation process, by accurately adjusting the weighting coefficients, the learning process of the model can be further optimized to avoid overfitting of the model during training.

[0086] In a possible implementation, the data preprocessing step specifically includes: introducing a time window weighted outlier detection algorithm (TWA-OUD), and detecting and correcting outliers during the preprocessing process using the following formula:

[0087]

[0088] x new (t) = x(t) - γ·z(t)

[0089] Among them, z(t) is the standardized abnormal measurement value, μ(t-1) is the mean of the data at the previous moment, σ(t-1) is the standard deviation of the data at the previous moment, and x new (t) is the corrected data value, γ is the correction coefficient, and the above algorithm is used to detect sudden changes or anomalies in the equipment status data in real time. By effectively correcting the abnormal values, the accuracy of subsequent model learning is improved.

[0090] Through this algorithm, outliers are detected and corrected in real time during the data preprocessing stage. The correction method is based on data standardization and correction coefficients. The standardized anomaly measurement value is compared with the mean and standard deviation of the data at the previous moment to detect sudden changes or anomalies in the device status data in real time. By correcting the outliers, the noise or errors in the sensor data can be effectively removed, thereby ensuring the accuracy of subsequent model learning. The role of this method is to reduce the errors introduced by factors such as external environmental changes and sensor drift in the equipment monitoring system, and improve the accuracy and reliability of equipment status recognition.

[0091] In a possible implementation, the deep neural network specifically includes: further introducing an ensemble learning-based fault prediction module (EL-FPM), integrating multiple weak classifiers and combining them with a deep learning model for fusion, and using the following weighted fusion formula:

[0092]

[0093] Among them, P fusion is the final prediction probability after fusion, P i is the predicted probability of the i-th classifier, w i is the weight coefficient of the corresponding classifier, and m is the number of classifiers.

[0094] The fault prediction module (EL-FPM) integrates multiple weak classifiers and fuses them with the deep learning model, and uses weighted fusion to improve the prediction accuracy. By weighting the prediction results of multiple classifiers, the robustness of the overall system can be improved and the deviation that may be caused by a single classifier can be reduced. Specifically, the final prediction probability is obtained by weighting the output results of each classifier. The weighting strategy is adjusted based on the prediction accuracy and weight coefficient of each classifier. The role of this method is to effectively improve the prediction accuracy under various complex fault modes, while reducing the prediction errors caused by the limitations of a single model. During the implementation process, by adjusting the number of classifiers and weight coefficients, the performance of the fault prediction system can be further optimized and its adaptability to different types of faults can be enhanced.

[0095] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0096] (1) In the present invention, through deep learning technologies such as deep neural network (DNN) and time series weighted adaptive deep convolutional neural network (TWA-DCNN), the present invention can effectively process and analyze the sensor data of the equipment. In particular, the TWA-DCNN model combined with the time series feature weighting mechanism can give different weights to the time series data of the equipment at different operation stages, especially paying more attention to important moments and sudden fault events. This method significantly enhances the ability to identify sudden faults, timely monitors the equipment status and makes accurate predictions, thereby achieving early fault warning and reducing the occurrence of equipment failures;

[0097] (2) In the present invention, by introducing an adaptive fusion convolutional recurrent neural network (AF-CRNN), combining the feature fusion mechanism of convolutional neural network (CNN) and recurrent neural network (RNN), and a dynamic gating mechanism, feature selection and timing dependency can be effectively optimized. The dynamic gating mechanism enables the model to adaptively adjust information flow according to changes in the equipment operation mode, thereby improving the processing capability of time series data and enhancing the flexibility and accuracy of equipment status recognition. This innovation optimizes the performance of deep learning models in time series data processing and significantly improves the accuracy and sensitivity of fault prediction.

[0098] (3) In order to reduce overfitting and improve the adaptability of the model to anomalies at different stages, the present invention introduces the time enhancement loss function (TELF). By weighted learning of historical data and real-time data, TELF can help the model better learn the changes in abnormal patterns, thereby avoiding overfitting of the model to outdated data and improving the stability and robustness of the model under different operating environments.

[0099] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0100] There are a few points to note:

[0101] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention, and other structures may refer to the general design.

[0102] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0103] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0104] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for automatic early warning monitoring of IoT sensing devices based on deep learning, characterized in that: include: S1. Collect sensor data generated by IoT sensing devices. The sensor data is collected through sensor arrays, IoT interfaces and cloud storage. The sensor data includes temperature, humidity, voltage, current and pressure generated by the monitored devices during operation. S2. Processing and feature extracting the sensor data using a deep neural network; S3. Establishing a device state recognition model by learning the data patterns of the normal operation and abnormal state of the monitored device. The device state recognition model is a model for determining the current state of the device based on the data patterns of the device operation state. The device state recognition model is established by learning and analyzing the data of the device in the normal and abnormal state through a deep learning algorithm; S4. Monitor the operating status of the monitored equipment in real time and make predictions according to the equipment status recognition model. Input the sensor data into the equipment status recognition model, monitor the equipment operating status in real time, and make predictions for abnormal situations that may occur in the future. At the same time, further data analysis and processing are performed based on historical data and equipment characteristics to improve the accuracy of equipment predictions. S5. When the monitored equipment operates abnormally or reaches the warning condition, the warning mechanism is automatically triggered and maintenance measures are taken in time; S6. Receive operator feedback information and newly connected IoT sensing device data in real time, and use the feedback information and IoT sensing device data to continuously optimize model performance and achieve self-learning.

2. According to claim 1, a method for automatic early warning based on deep learning IoT sensing device monitoring is characterized in that: The S2 specifically includes: S21, data preprocessing: before inputting into the deep neural network, the sensor data needs to be preprocessed, and the preprocessing includes standardization and normalization to adapt to the input requirements of the neural network; S22. Construct a deep neural network structure, wherein the deep neural network includes an input layer, a hidden layer, and an output layer. Each layer is composed of a number of neurons, and takes the output of the previous layer as input, performs nonlinear transformation through an activation function, and generates a new output. When information is transmitted from the input layer to the output layer, a simplified forward propagation formula is used. In the above formula, σ is the activation function, w i is the weight, x i is the input, b is the bias term, and y is the output; S23, feature learning and extraction. During the training process, the weights and biases in the network are adjusted through the back propagation algorithm so that the predicted output of the deep neural network is as close to the true label as possible. During the training process, the deep neural network automatically learns from the original data and extracts features that are helpful for task completion.

3. According to claim 2, a method for automatic early warning based on deep learning IoT sensing device monitoring is characterized in that: The S22 specifically includes: The input layer is used to receive raw sensor data, and the raw sensors include devices for monitoring temperature, humidity and vibration; The hidden layer includes multiple fully connected layers, convolutional layers and recurrent layers. Each layer transforms the output of the previous layer into the input of the next layer through the weight matrix W and the bias b, and uses the activation function. f(x)=max(0,x) Perform nonlinear mapping, the fully connected layer is a feedforward neural network, the convolutional layer is suitable for image or sequence data, and the recurrent layer is suitable for time series data; The output layer includes one or more nodes for classification or regression. The classification uses a softmax function to output a probability distribution, and the regression directly outputs a numerical value.

4. According to claim 1, a method for automatic early warning based on deep learning IoT sensing device monitoring is characterized in that: The S3 specifically includes: S31, data preparation, including data collection, data labeling, and data cleaning and normalization. The data collection is to collect data in normal operation and abnormal state from sensors installed on the equipment. The labeling data is to annotate the collected data to distinguish normal data from abnormal data. The data cleaning and integration is to remove noise, fill in missing values, and normalize or standardize the data to facilitate model learning. S32, feature engineering, including feature extraction, wherein the feature extraction is to extract meaningful features from the original data, wherein the meaningful features include vibration amplitude and temperature change of a specific frequency, and for image data, the features are directly learned by a convolutional neural network (CNN); S33, construct a deep learning algorithm model, the deep learning algorithm model includes the deep learning algorithm including the convolutional neural network (CNN) and the recurrent neural network (RNN). If the device state can be identified by image or spectrum analysis, the convolutional neural network is used. For time series data, the recurrent neural network is more suitable. The simplified formula of the convolution operation of the convolutional neural network is: In the above formula, c ij represents an element of the feature map after convolution, f represents the filter, x represents the input image, and * represents the convolution operation. In the construction of the RNN, a basic RNN unit is represented as, ht=f(Wxt+Uh t-1 +b) In the above formula, ht represents the hidden state at time step t, xt represents the input of the current time step, (W, U, b) represents the model parameters, and f represents the activation function, which includes tanh or ReLU. Long short-term memory (LSTM) or gated recurrent unit (GRU) is a variant of the RNN, which is more suitable for long sequence data learning. The basic unit of the LSTM model is expressed as the following four formulas, which are used to update the unit state c t and the hidden state h t , and output predicted values The forget gate determines the information that needs to be forgotten in the previous unit state. f t =σ(W f ·[h t-1 ,x t ]+b f ) The input gate determines the new information that should be updated to the cell state. i t =σ(W i ·[h t-1 ,x t ]+b i ) g t =tanh(W g ·[h t-1 ,x t ]+b g ) The cell state is updated, combining forget and input operations, c t =f t ⊙c t-1 +i t ⊙g t The output gate determines the cell state information that should be output as the new hidden state. the t =σ(W o ·[h t-1 ,x t ]+b o ) h t =o t ⊙tanh(c t ) In the above formula, f t represents the output of the forget gate, σ represents the sigmoid activation function, W f represents the weight matrix, b f represents the bias term, [h t-1 ,x t ] represents the concatenation of the hidden state of the previous time step and the input of the current time step, tanh represents the hyperbolic tangent function, W,b represents the model parameters, and ⊙ represents element-wise multiplication; S34, training the model, including dividing the data set, selecting a loss function and an optimizer, and a catenary process, wherein the data set is divided into a training set, a validation set, and a test set, and the training process is to adjust the model parameters through a back propagation algorithm, minimize the loss function, and iterate multiple rounds until the model is no longer significantly improved or reaches a predetermined stop condition; S35. Evaluation and tuning, including performance evaluation and model tuning. The performance evaluation is to evaluate the model performance on the test set using accuracy, recall and F1 score. The model tuning is to adjust the model structure and hyperparameters according to the evaluation results or use deep CNN and bidirectional RNN for multiple iterative training to optimize performance.

5. According to claim 1, a method for automatic early warning based on deep learning IoT sensing device monitoring is characterized in that: The S4 specifically includes: S41, data collection and preprocessing, collecting parameters of the equipment during operation through sensors installed on the equipment, the sensors including temperature sensors, vibration sensors and current and voltage sensors, and preprocessing the collected data, the preprocessing including noise removal, missing value processing and data standardization, for subsequent analysis; S42, feature engineering, extracting useful features from raw data, the features include time series trends, frequency components, statistical properties of signals and complex features, the statistical properties of signals include mean, variance and kurtosis, the complex features include descriptions of waveform shapes, and the goal of feature selection is to find the set of variables that best characterizes the device status; S43. Select or construct a device state recognition model, wherein the model includes a supervised learning model, an unsupervised learning model, and a deep learning model. The supervised learning model includes a support vector machine (SVM), a random forest (RF), and a gradient boosted tree (GBT), which are used for classification or regression problems, predicting whether the device is about to fail or estimating the remaining service life. The unsupervised learning model includes a clustering algorithm (K-means, DBSCAN), which is used to discover patterns in data and identify normal and abnormal operating modes of the device. The deep learning model includes a recurrent neural network (RNN), a long short-term memory network (LSTM), or a gated recurrent unit (GRU), which is suitable for processing time series data and capturing the law of changes in device state over time. S44, real-time detection and prediction, input real-time sensor data into the trained model, the model evaluates the operating status of the equipment according to the characteristics of the current data, if the model predicts that the status deviates from the normal range, it can issue an early warning to prompt maintenance personnel to take preventive measures.

6. According to claim 5, a method for automatic early warning based on deep learning IoT sensing device monitoring is characterized in that: The device state recognition model specifically includes: By introducing the time series weighted adaptive deep convolutional neural network (TWA-DCNN) algorithm, the time series features in the sensor data are weighted, and the weight is calculated by the following formula: Among them, w(t) is the weight value corresponding to the time point t, T is the current time point, λ is the weighted attenuation coefficient, α is the weight adjustment factor, |tT| represents the time difference, and the weighting function can dynamically adjust the sampling frequency, give higher weights to the sampling data at important moments, and enhance the model's ability to identify sudden faults.

7. The automatic early warning method for monitoring IoT sensing devices based on deep learning according to claim 6 is characterized in that: The device state recognition model specifically includes: Adaptive fusion convolutional recurrent neural network (AF-CRNN) is used to combine the feature fusion mechanism in convolutional neural network (CNN) and recurrent neural network (RNN), and introduce a self-created dynamic gating mechanism to optimize feature selection and timing dependency. The dynamic gating mechanism is calculated by the following formula: G(t)=σ(W g ·[h(t-1),x(t)]+b g ) Among them, G(t) is the dynamic gating value, h(t-1) is the hidden state at the previous moment, x(t) is the input feature at the current moment, and W g is the gating weight matrix, b g is the bias term, σ is the Sigmoid activation function, and the mechanism can adaptively adjust the information flow and enhance the model's ability to process time series data by learning changes in the device's operating mode.

8. The automatic early warning method for monitoring IoT sensing devices based on deep learning according to claim 6 is characterized in that: The deep convolutional neural network specifically includes: The weighted deep convolutional neural network (TWA-DCNN) model combines the time enhancement loss function (TELF) to weight the loss functions of historical data and real-time data, so that the model can differentiate between historical fault modes and real-time anomalies when learning abnormal modes. The loss function is defined as: Among them, L(t) is the loss value, y i is the actual value, is the predicted value, w i is the weighting coefficient, The error between model prediction and reality is reduced by weighting, and the adaptability of the model to abnormalities at different stages is improved.

9. The automatic early warning method for monitoring IoT sensing devices based on deep learning according to claim 2 is characterized in that: The data preprocessing step specifically includes: The time window weighted outlier detection algorithm (TWA-OUD) is introduced to detect and correct outliers in the preprocessing process using the following formula: x new (t)=x(t)-γ·z(t) Among them, z(t) is the standardized abnormal measurement value, μ(t-1) is the mean of the data at the previous moment, σ(t-1) is the standard deviation of the data at the previous moment, and x new (t) is the corrected data value, γ is the correction coefficient, and the above algorithm is used to detect sudden changes or anomalies in the equipment status data in real time. By effectively correcting the abnormal values, the accuracy of subsequent model learning is improved.

10. The automatic early warning method for monitoring IoT sensing devices based on deep learning according to claim 2 is characterized in that: The deep neural network specifically includes: The fault prediction module based on ensemble learning (EL-FPM) is further introduced, which integrates multiple weak classifiers and combines them with the deep learning model to perform fusion, using the following weighted fusion formula: Among them, P fusion is the final prediction probability after fusion, P i is the predicted probability of the i-th classifier, w i is the weight coefficient of the corresponding classifier, and m is the number of classifiers.

Images