Intelligent monitoring and adjusting method, system and equipment based on digital twinning technology

Abstract

The invention belongs to the technical field of digital twinning, particularly relates to an intelligent monitoring and adjusting method, system and equipment based on the digital twinning technology, and aims to solve the problems that in the prior art, efficient and accurate fault prediction cannot be carried out on production equipment, the fault position is difficult to locate accurately, and the maintenance efficiency is affected. The method comprises the following steps: acquiring operation state data, working condition data and environmental condition data of to-be-monitored production equipment as input data; preprocessing the input data, and constructing a digital twin model; real-time monitoring is carried out through the digital twin model, and monitoring data are obtained; based on the monitoring data, fault prediction is carried out through an adaptive threshold fault prediction method, and a fault prediction result is obtained; and the digital twinborn model performs live-action simulation according to a fault prediction result, locates a fault position and gives an alarm to monitoring personnel. According to the invention, efficient and accurate fault prediction of the production equipment is realized, the accuracy of fault location positioning is improved, and the maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of digital twin technology, and specifically relates to an intelligent monitoring and adjustment method, system and equipment based on digital twin technology. Background Art

[0002] In the production process of special labor protection products and labor protection products, the stable operation of production equipment is crucial. These equipment usually include sewing machines, injection molding machines, cutting machines, etc. Their normal operation directly affects the quality and production efficiency of the products. Traditional monitoring methods mainly rely on manual inspections and fixed threshold alarms, which have the following limitations: 1) Slow response: The manual inspection cycle is long and it is impossible to monitor the equipment status in real time, which may lead to failures not being discovered in time, thereby affecting production progress; 2) Low accuracy: The fixed threshold alarm method is based on a pre-set threshold and cannot adapt to changes in the equipment's operating status. It is easy to produce false alarms or missed alarms, which reduces the accuracy of fault detection; 3) Inability to monitor in real time: Traditional methods cannot achieve continuous and real-time monitoring, and it is difficult to capture instantaneous abnormal conditions, increasing the risk of failure; 4) Limited data processing capabilities: Traditional methods have insufficient processing capabilities for large amounts of multi-source data, making it difficult to achieve efficient data preprocessing and feature extraction, limiting the accuracy and reliability of fault prediction; 5) Inaccurate fault location: Traditional methods are difficult to accurately locate the fault location, and maintenance personnel need to spend a lot of time looking for the cause of the fault, affecting maintenance efficiency and production recovery speed;

[0003] Digital twin technology can achieve more accurate monitoring and prediction by building virtual models to reflect the status of physical equipment in real time. Digital twin models can integrate multi-source data, conduct real-time analysis and simulation, and provide new solutions for fault prediction and maintenance. However, the existing digital twin technology still has the following shortcomings in fault prediction: 1) Existing digital twin models often use fixed thresholds or simple rules for fault detection, and cannot be dynamically adjusted according to the actual operating status of the equipment, which is prone to false alarms or missed alarms; 2) Although digital twin technology can simulate the operating status of equipment, it is difficult to accurately identify the location and cause of the fault, which affects maintenance efficiency and production recovery speed.

[0004] Therefore, there is an urgent need for an intelligent monitoring and adjustment method based on digital twin technology that can overcome the above-mentioned shortcomings, improve the accuracy and reliability of fault prediction, and enhance production efficiency and equipment reliability. Summary of the invention

[0005] In order to solve the above problems in the prior art, that is, to solve the problem that the prior art cannot efficiently and accurately predict the faults of production equipment and is difficult to accurately locate the fault location, thus affecting the maintenance efficiency, the first aspect of the present invention proposes an intelligent monitoring and adjustment method based on digital twin technology, which includes:

[0006] S10, obtaining the operating status data, working condition data and environmental condition data of the production equipment to be monitored as input data;

[0007] S20, preprocessing the input data to obtain preprocessed data; and constructing a digital twin model of the production equipment to be monitored according to the preprocessed data;

[0008] S30, monitoring the production equipment to be monitored in real time through the digital twin model to obtain monitoring data;

[0009] S40, performing fault prediction based on the monitoring data by using a pre-built adaptive threshold fault prediction method to obtain a fault prediction result;

[0010] S50, the digital twin model performs real-scene simulation according to the fault prediction result, locates the fault location, and alerts the monitoring personnel.

[0011] In some preferred embodiments, the preprocessing includes: removing outliers, missing values, data conversion and standardization.

[0012] In some preferred embodiments, based on the monitoring data, a pre-built adaptive threshold fault prediction method is used to perform fault prediction to obtain a fault prediction result, and the method is as follows:

[0013] Extracting features of each modal data in the monitoring data; the modal data includes temperature data, vibration data, and current data;

[0014] Calculate the adaptive threshold of each modal data according to the mean and standard deviation of each modal data;

[0015] The features corresponding to different modal data are weighted and fused to obtain multimodal data features;

[0016] The adaptive thresholds of each modality data are weighted and fused to obtain a comprehensive adaptive threshold;

[0017] The comprehensive adaptive threshold is compared with the multimodal data feature. If the multimodal data feature is greater than the comprehensive adaptive threshold, the multimodal data feature is input into a trained fault prediction model to obtain a fault prediction result.

[0018] In some preferred implementations, the adaptive threshold of each modal data is calculated according to the mean and standard deviation of each modal data, and the method is as follows:

[0019]

[0020] Among them, T i(t) represents the adaptive threshold of the i-th modal data at time t, μ i (t) is the mean value of the i-th modal data at time t, σ i (t) is the standard deviation, k represents an adjustable parameter used to control the sensitivity of the threshold, and α and β represent weight coefficients, which are used to consider the first and second derivatives of the mean, respectively.

[0021] In some preferred embodiments, the second features corresponding to different modal data are weightedly fused to obtain multimodal data features, and the method is:

[0022]

[0023] Where F(t) represents the multimodal data features at time t, N represents the number of modalities, and w i is the weight of the i-th mode, f i (t) is the eigenvalue of the i-th mode, γ i ,δ i Represents the weight coefficient, which is used to consider the first-order derivative and second-order derivative of the eigenvalue respectively.

[0024] In some preferred embodiments, the fault prediction model is constructed based on a deep neural network.

[0025] In some preferred embodiments, the fault prediction model includes an input layer, an embedding layer, a residual block, and an output layer connected in sequence;

[0026] The input layer is used to input multimodal data features;

[0027] The embedding layer is used to perform a linear transformation on the multimodal data features;

[0028] The residual block includes two fully connected layers, a residual connection and a layer normalization; the input of the first fully connected layer is the output of the embedding layer; the input of the second fully connected layer is the output of the first fully connected layer;

[0029] The output of the first fully connected layer is residually connected to the output of the second fully connected layer, the summed output is layer normalized, and then input into the output layer;

[0030] The output layer performs linear transformation and activation function processing on the input features to obtain the final fault prediction result.

[0031] In some preferred embodiments, the fault prediction model adopts a cross entropy loss function during training.

[0032] In a second aspect of the present invention, an intelligent monitoring and adjustment system based on digital twin technology is proposed, the system comprising:

[0033] A data acquisition module is configured to acquire operating status data, working condition data, and environmental condition data of the production equipment to be monitored as input data;

[0034] A twin construction module is configured to preprocess the input data to obtain preprocessed data; and construct a digital twin model of the production equipment to be monitored based on the preprocessed data;

[0035] A monitoring module, configured to perform real-time monitoring of the production equipment to be monitored through the digital twin model to obtain monitoring data;

[0036] A fault prediction module is configured to perform fault prediction based on the monitoring data by using a pre-built adaptive threshold fault prediction method to obtain a fault prediction result;

[0037] The alarm module is configured so that the digital twin model performs real-scene simulation according to the fault prediction result, locates the fault location, and alerts the monitoring personnel.

[0038] The third aspect of the present invention provides an intelligent monitoring and regulating device based on digital twin technology, comprising:

[0039] At least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned intelligent monitoring and adjustment method based on digital twin technology.

[0040] Beneficial effects of the present invention:

[0041] The present invention realizes efficient and accurate fault prediction of production equipment, improves the accuracy of fault location, and improves maintenance efficiency.

[0042] 1) The present invention uses a digital twin model to monitor the status of equipment in real time, and combines adaptive thresholds and deep neural networks to improve the accuracy of fault prediction. Through intelligent monitoring and fault prediction, potential faults can be discovered and handled in advance, improving production efficiency and product quality. Accurate fault location and rapid maintenance response reduce maintenance time and costs, and improve equipment reliability.

[0043] 2) The present invention dynamically adjusts the threshold value through an adaptive threshold method to adapt to changes in the operating state of the device and reduce false positives and false negatives;

[0044] 3) The present invention sends fault alarms in real time through a variety of ways to ensure that monitoring personnel can respond in a timely manner and reduce equipment downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings.

[0046] Figure 1 It is a flow chart of an intelligent monitoring and adjustment method based on digital twin technology according to an embodiment of the present invention;

[0047] Figure 2 This is an example diagram of a special labor protection product production workshop according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0050] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other.

[0051] An intelligent monitoring and adjustment method based on digital twin technology in the first embodiment of the present invention is as follows: Figure 1 As shown, the following steps are included:

[0052] S10, obtaining the operating status data, working condition data and environmental condition data of the production equipment to be monitored as input data;

[0053] S20, preprocessing the input data to obtain preprocessed data; and constructing a digital twin model of the production equipment to be monitored according to the preprocessed data;

[0054] S30, monitoring the production equipment to be monitored in real time through the digital twin model to obtain monitoring data;

[0055] S40, performing fault prediction based on the monitoring data by using a pre-built adaptive threshold fault prediction method to obtain a fault prediction result;

[0056] S50, the digital twin model performs real-scene simulation according to the fault prediction result, locates the fault location, and alerts the monitoring personnel.

[0057] In order to more clearly illustrate the intelligent monitoring and adjustment method based on digital twin technology of the present invention, each step in an embodiment of the method of the present invention is described in detail below with reference to the accompanying drawings.

[0058] In the production process of special labor protection products and labor protection products, equipment failure will not only lead to production interruptions and increase maintenance costs, but may also affect product quality and even cause safety hazards. Therefore, it is crucial to predict potential failures in a timely and accurate manner and take preventive measures. Based on this. The present invention proposes an intelligent monitoring and adjustment method based on digital twin technology. By combining digital twin technology with machine learning algorithms, patterns can be learned from a large amount of historical data, and early signs that may lead to failures can be identified, thereby achieving early warning of failures. The specific process is as follows:

[0059] S10, obtaining the operating status data, working condition data and environmental condition data of the production equipment to be monitored as input data;

[0060] In this embodiment, in the special labor protection products production workshop (such as Figure 2 As shown in Figure 1, various sensors are installed to collect operation status data, working condition data, and environmental condition data, and transmit them to the monitoring system in real time. The operation status data includes sensor data such as the temperature, vibration, current, and sound of the equipment; the working condition data includes the working mode, load condition, and operating time of the equipment; and the environmental condition data includes environmental parameters such as the temperature, humidity, and air pressure of the workshop.

[0061] S20, preprocessing the input data to obtain preprocessed data; and constructing a digital twin model of the production equipment to be monitored according to the preprocessed data;

[0062] In this embodiment, the preprocessing includes: removing outliers (preferably by filtering algorithms), missing values ​​(by statistical methods), data conversion and standardization. The preprocessing is prior art and will not be described in detail here.

[0063] Using the preprocessed data, software such as Unity or Simulink is used to build a virtual model of the equipment, namely the digital twin model, which includes the locations of various components and sensors and can reflect the actual operating status of the equipment in real time.

[0064] S30, monitoring the production equipment to be monitored in real time through the digital twin model to obtain monitoring data;

[0065] In this embodiment, the operating status of the equipment (i.e., various indicators such as temperature curve, vibration spectrum, etc.) is simulated in real time by the digital twin model to obtain monitoring data.

[0066] S40, performing fault prediction based on the monitoring data by using a pre-built adaptive threshold fault prediction method to obtain a fault prediction result;

[0067] In traditional fault prediction methods, thresholds are usually set statically, which may lead to a decrease in prediction performance under different working conditions. In addition, data from a single mode often cannot fully reflect the health status of the equipment. This embodiment proposes an adaptive threshold multi-modal fault prediction algorithm, which improves the accuracy and robustness of fault prediction by dynamically adjusting the threshold and fusing multiple sensor data. The details are as follows:

[0068] S41, using a commonly used feature extraction model to extract the features of each modal data in the monitoring data; the modal data includes temperature data, vibration data, and current data; the features of each modal data include: temperature data: average value, standard deviation, peak value; vibration data: spectrum characteristics, root mean square value, kurtosis; current data: peak value, root mean square value, harmonic distortion, etc.

[0069] S42, introduce an adaptive threshold setting mechanism to dynamically adjust the threshold according to the current working condition. By dynamically adjusting the threshold, it can adapt to different working conditions and improve the accuracy of fault prediction. That is, the adaptive threshold of each modal data is calculated according to the mean and standard deviation of each modal data;

[0070]

[0071] Among them, T u (t) represents the adaptive threshold of the i-th modal data at time t, i u (t) is the mean value of the i-th modal data at time t, σ i (t) is the standard deviation, k represents an adjustable parameter used to control the sensitivity of the threshold, and α and β represent weight coefficients, which are used to consider the first and second derivatives of the mean, respectively.

[0072] S43, weighted fusion of features corresponding to different modal data to obtain multimodal data features, and highlighting important features and improving prediction performance by weighted fusion of different modal data;

[0073]

[0074] Where F(t) represents the multimodal data features at time t, D represents the number of modalities, and w i is the weight of the i-th mode, f i (t) is the eigenvalue of the i-th mode, γ i ,δ i Represents the weight coefficient, which is used to consider the first-order derivative and second-order derivative of the eigenvalue, that is, the first-order and second-order derivative terms are introduced in the adaptive threshold and multimodal data fusion to better capture the dynamic changes of the data.

[0075] S44, weighted fusion of the adaptive thresholds of each modality data to obtain a comprehensive adaptive threshold;

[0076] S44, comparing the comprehensive adaptive threshold with the multimodal data feature, if the multimodal data feature is greater than the comprehensive adaptive threshold, inputting the multimodal data feature into a trained fault prediction model to obtain a fault prediction result.

[0077] In the present invention, the comprehensive adaptive threshold is compared with the multimodal data feature. If the multimodal data feature is greater than the comprehensive adaptive threshold, it is preliminarily determined that there may be an abnormality or potential fault. By adaptive threshold comparison, situations where there may be a fault can be preliminarily screened out. However, this is only a preliminary judgment, because the threshold comparison may produce false positives or false negatives. In order to improve the accuracy and reliability of fault prediction, a fault prediction model is introduced. The fault prediction model can more accurately determine whether the equipment really has a fault by learning complex patterns in historical data.

[0078] In the present invention, the fault prediction model is constructed based on a deep neural network; the fault prediction model includes an input layer, an embedding layer, a residual block, and an output layer connected in sequence;

[0079] The input layer is used to input multimodal data features and pass them to the embedding layer;

[0080] Embedding layer, used to perform linear transformation (i.e., weight matrix multiplication) on the multimodal data features, with an output dimension of 128;

[0081] The residual block consists of two fully connected layers, a residual connection and a layer normalization;

[0082] The input of the first fully connected layer is the output of the embedding layer; that is, for the input feature x, a linear transformation y1=W1x+b1 is first performed; then y1 is processed by an activation function, and finally regularized;

[0083] The input of the second fully connected layer is the output of the first fully connected layer; the processing process is similar to that of the first fully connected layer, so it will not be repeated here.

[0084] The output of the first fully connected layer is residually connected to the output of the second fully connected layer, and the summed output is layer normalized and then input into the output layer;

[0085] The output layer performs linear transformation (i.e., fully connected layer) and activation function (converts the output into a probability value, indicating the probability of belonging to a fault category; the final result can be either binary or multi-classification) on the input features to obtain the final fault prediction result.

[0086] In addition, the fault prediction model adopts the cross entropy loss function during the training process.

[0087] S50, the digital twin model performs real-scene simulation according to the fault prediction result, locates the fault location, and alerts the monitoring personnel.

[0088] In this embodiment, based on the fault prediction results, the digital twin model performs real-life simulation to display the location and possible causes of the fault. Through the simulation results, the fault location is accurately located, and the fault information is sent to the monitoring personnel via SMS, email or mobile applications, thereby improving maintenance efficiency.

[0089] In summary, the present invention uses a digital twin model to monitor the status of equipment in real time. Combined with adaptive thresholds and deep neural networks, the factory can significantly improve the accuracy and robustness of fault prediction, thereby improving production efficiency and product quality and reducing maintenance costs and downtime.

[0090] A second embodiment of the present invention provides an intelligent monitoring and adjustment system based on digital twin technology, the system comprising:

[0091] A data acquisition module is configured to acquire operating status data, working condition data, and environmental condition data of the production equipment to be monitored as input data;

[0092] A twin construction module is configured to preprocess the input data to obtain preprocessed data; and construct a digital twin model of the production equipment to be monitored based on the preprocessed data;

[0093] A monitoring module, configured to perform real-time monitoring of the production equipment to be monitored through the digital twin model to obtain monitoring data;

[0094] A fault prediction module is configured to perform fault prediction based on the monitoring data by using a pre-built adaptive threshold fault prediction method to obtain a fault prediction result;

[0095] The alarm module is configured so that the digital twin model performs real-scene simulation according to the fault prediction result, locates the fault location, and alerts the monitoring personnel.

[0096] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process and related instructions of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0097] It should be noted that the intelligent monitoring and adjustment system based on digital twin technology provided in the above embodiment is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be decomposed or combined. For example, the modules in the above embodiments can be combined into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the modules or steps and are not regarded as improper limitations of the present invention.

[0098] A third embodiment of the present invention is an intelligent monitoring and adjustment device based on digital twin technology, comprising: at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned intelligent monitoring and adjustment method based on digital twin technology.

[0099] A computer-readable storage medium according to a fourth embodiment of the present invention stores computer instructions, and the computer instructions are used to be executed by the computer to implement the above-mentioned intelligent monitoring and adjustment method based on digital twin technology.

[0100] Technicians in the technical field can clearly understand that, for the convenience and conciseness of description, the specific working process and related instructions of the above-described intelligent monitoring and adjustment equipment based on digital twin technology and computer-readable storage medium can refer to the corresponding process in the aforementioned method example and will not be repeated here.

[0101] Those skilled in the art should be able to appreciate that the modules and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two, and the programs corresponding to the software modules and method steps can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the technical field. In order to clearly illustrate the interchangeability of electronic hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0102] The terms "first", "second", etc. are used to distinguish similar objects rather than to describe or indicate a particular order or sequence.

[0103] The term "comprise" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus / device.

[0104] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An intelligent monitoring and adjustment method based on digital twin technology, characterized in that: The method comprises the following steps: S10, obtaining the operating status data, working condition data and environmental condition data of the production equipment to be monitored as input data; S20, preprocessing the input data to obtain preprocessed data; and constructing a digital twin model of the production equipment to be monitored according to the preprocessed data; S30, monitoring the production equipment to be monitored in real time through the digital twin model to obtain monitoring data; S40, performing fault prediction based on the monitoring data by using a pre-built adaptive threshold fault prediction method to obtain a fault prediction result; S50, the digital twin model performs real-scene simulation according to the fault prediction result, locates the fault location, and alerts the monitoring personnel.

2. According to claim 1, an intelligent monitoring and adjustment method based on digital twin technology is characterized in that: The preprocessing includes: removing outliers, missing values, data conversion and standardization.

3. According to claim 2, an intelligent monitoring and adjustment method based on digital twin technology is characterized in that: Based on the monitoring data, a pre-built adaptive threshold fault prediction method is used to perform fault prediction to obtain a fault prediction result, and the method is as follows: Extracting features of each modal data in the monitoring data; the modal data includes temperature data, vibration data, and current data; Calculate the adaptive threshold of each modal data according to the mean and standard deviation of each modal data; The features corresponding to different modal data are weighted and fused to obtain multimodal data features; The adaptive thresholds of each modality data are weighted and fused to obtain a comprehensive adaptive threshold; The comprehensive adaptive threshold is compared with the multimodal data feature. If the multimodal data feature is greater than the comprehensive adaptive threshold, the multimodal data feature is input into a trained fault prediction model to obtain a fault prediction result.

4. According to claim 3, an intelligent monitoring and adjustment method based on digital twin technology is characterized in that: According to the mean and standard deviation of each modal data, the adaptive threshold of each modal data is calculated as follows: Among them, T i (t) represents the adaptive threshold of the i-th modal data at time t, μ i (t) is the mean value of the i-th modal data at time t, σ i (t) is the standard deviation, k represents an adjustable parameter used to control the sensitivity of the threshold, and α and β represent weight coefficients, which are used to consider the first and second derivatives of the mean, respectively.

5. According to claim 3, the intelligent monitoring and adjustment method based on digital twin technology is characterized in that: The second features corresponding to different modal data are weighted and fused to obtain multimodal data features. The method is as follows: Where F(t) represents the multimodal data features at time t, N represents the number of modalities, and w i is the weight of the i-th mode, f i (t) is the eigenvalue of the i-th mode, γ i , δ i Represents the weight coefficient, which is used to consider the first-order derivative and second-order derivative of the eigenvalue respectively.

6. According to claim 3, the intelligent monitoring and adjustment method based on digital twin technology is characterized in that: The fault prediction model is built based on a deep neural network.

7. According to claim 6, an intelligent monitoring and adjustment method based on digital twin technology is characterized in that: The fault prediction model includes an input layer, an embedding layer, a residual block, and an output layer connected in sequence; The input layer is used to input multimodal data features; The embedding layer is used to perform a linear transformation on the multimodal data features; The residual block includes two fully connected layers, a residual connection and a layer normalization; the input of the first fully connected layer is the output of the embedding layer; the input of the second fully connected layer is the output of the first fully connected layer; The output of the first fully connected layer is residually connected to the output of the second fully connected layer, the summed output is layer normalized, and then input into the output layer; The output layer performs linear transformation and activation function processing on the input features to obtain the final fault prediction result.

8. The intelligent monitoring and adjustment method based on digital twin technology according to claim 7 is characterized in that: The fault prediction model adopts a cross entropy loss function during the training process.

9. An intelligent monitoring and adjustment system based on digital twin technology, characterized in that: The system includes: A data acquisition module is configured to acquire operating status data, working condition data, and environmental condition data of the production equipment to be monitored as input data; A twin construction module is configured to preprocess the input data to obtain preprocessed data; and construct a digital twin model of the production equipment to be monitored based on the preprocessed data; A monitoring module, configured to perform real-time monitoring of the production equipment to be monitored through the digital twin model to obtain monitoring data; A fault prediction module is configured to perform fault prediction based on the monitoring data by using a pre-built adaptive threshold fault prediction method to obtain a fault prediction result; The alarm module is configured so that the digital twin model performs real-scene simulation according to the fault prediction result, locates the fault location, and alerts the monitoring personnel.

10. An intelligent monitoring and adjustment device based on digital twin technology, characterized in that: include: at least one processor; And a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement an intelligent monitoring and adjustment method based on digital twin technology as described in any one of claims 1-8.

Images