A method, system, device and storage medium for automated equipment detection
The method enhances automation equipment detection by integrating image and parameter processing with ResNet-101 and Kalman filtering to provide real-time, accurate feedback on equipment status, reducing manual errors and improving efficiency.
Patent Information
- Application Number
- CN202411218326.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-02
- Publication Date
- 2024-12-10
- Estimated Expiration
- 2044-09-02
AI Technical Summary
Traditional manual data acquisition and processing methods for automation equipment detection result in low efficiency and accuracy due to large errors and the complexity of control objects, which cannot be accurately reflected by data alone.
A method involving image and parameter acquisition, preprocessing, feature extraction, scoring, and anomaly detection using ResNet-101 and Kalman filtering to provide real-time feedback on equipment status through colored lights.
Improves detection efficiency and accuracy by reducing manual intervention and data errors, enabling precise monitoring and timely maintenance of automation equipment.
Smart Images
Figure CN118761030B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of automation equipment, and in particular to a method, system, equipment and storage medium for detecting automation equipment. Background Art
[0002] At present, the manufacturing industry has a huge market. Traditional manual production is not only costly and inefficient, but also some manufacturing projects cannot meet the minimum manufacturing standards. Automation equipment, as a substitute for manual labor, not only improves production efficiency and product quality, but also improves the working environment, reduces labor costs, and enhances the flexibility and controllability of the production process. However, with the improvement of the mechanization and automation level of production equipment, the control objects are becoming increasingly complex, and the requirements for detection technology are also constantly increasing.
[0003] In traditional detection methods, data acquisition and processing of detection objects mainly rely on manual labor, which has many limitations. Manual data acquisition and processing have large errors. In addition, since there are many types and large numbers of working parameters of automation equipment, manual acquisition and processing alone result in low efficiency and detection accuracy. At the same time, in some special cases, data alone cannot directly reflect the status of automation equipment, affecting detection accuracy.
[0004] The above detection methods result in low detection accuracy and efficiency of automated equipment. Summary of the invention
[0005] The present invention provides a method, system, device and storage medium for detecting automated equipment, so as to realize the detection of automated equipment and improve the detection efficiency and accuracy.
[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a method for automated equipment detection, comprising:
[0007] During the operation of the automation equipment, obtaining working images and equipment operation parameters collected from the automation equipment;
[0008] According to the working image, a recognition processing operation is performed to obtain working state characteristics;
[0009] According to the working state characteristics, a feature vector conversion operation is performed to obtain a working state feature vector;
[0010] Performing a device scoring operation according to the working status feature vector to obtain a performance scoring level;
[0011] According to the equipment operation parameters, a parameter inspection operation is performed to obtain equipment abnormality data;
[0012] According to the equipment abnormality data and the performance score level, a light reminder operation is performed to feedback the detection results of the automation equipment to the user.
[0013] As an optional implementation manner, performing recognition processing operations according to the working image to obtain working state features includes:
[0014] Performing a preprocessing operation on the working image to obtain a preprocessed image, wherein the preprocessing operation includes image scaling and image cropping;
[0015] According to the preprocessed image, a recognition processing operation is performed to obtain working state characteristics, wherein the recognition processing operation is to input the preprocessed image into the residual network Res-Net 101, and the working state characteristics include mechanical parts, indicator lights and braking systems.
[0016] As an optional implementation manner, performing a device scoring operation according to the working state feature vector to obtain a performance scoring level includes:
[0017] The equipment scoring operation formula is:
[0018] ,
[0019] Among them, a, b, are pre-configured coefficients, To rate the performance, Rating index for mechanical parts, The indicator rating index is Braking system rating index;
[0020] The mechanical component rating index The formula is:
[0021] ,
[0022] The indicator rating index The formula is:
[0023] ,
[0024] The brake system rating index The formula is:
[0025] ,
[0026] in, is the feature set of mechanical parts in normal state, It is the indicator light feature set in normal state. is the set of braking system characteristics under normal conditions, is the collected feature vector of mechanical parts, is the collected indicator light feature vector, is the collected braking system feature vector, The pre-configured adjustment weight coefficient.
[0027] As an optional implementation manner, before performing a parameter inspection operation according to the device operating parameters to obtain device abnormality data, the method further includes:
[0028] A denoising operation is performed on the device operating parameters to obtain correction data, wherein the denoising operation is denoising using Kalman filtering, and the device operating parameters include temperature, voltage and vibration.
[0029] As an optional implementation manner, performing a parameter inspection operation according to the device operating parameters to obtain device abnormality data includes:
[0030] According to the correction data, a data point distance calculation operation is performed to obtain a data reachable distance, wherein the data point distance calculation is to calculate the Euclidean distance between the operating parameters of the equipment;
[0031] According to the data reachable distance, a local reachable density calculation operation is performed to obtain a local reachable density, wherein the local reachable density calculation operation is to calculate the inverse of the average value of the data reachable distance;
[0032] Performing a local outlier factor calculation operation according to the local reachable density to obtain a local outlier factor;
[0033] According to the local outlier factor, an outlier judgment operation is performed to obtain device abnormal data.
[0034] As an optional implementation manner, performing local reachable density calculation according to the data reachable distance to obtain the local reachable density includes:
[0035] The formula for the local reachable density calculation operation is:
[0036] ,
[0037] Wherein, k is the distance from the device operating parameter to the kth device operating parameter, The data of the i-th point of the equipment operation parameters, The data of the rth point of the equipment operation parameters, reach_dist k ( ) is the reachable distance of the kth data, is the local reachable density of the data at the i-th point of the device operation parameter, is a neighborhood with a distance k from the operating parameters of the device.
[0038] As an optional implementation manner, performing a local outlier factor calculation operation according to the local reachable density to obtain the local outlier factor includes:
[0039] The formula for the local outlier factor calculation operation is:
[0040] ,
[0041] Wherein, k is the distance from the device operating parameter to the kth device operating parameter, is the local outlier factor of the data at the ith point of the equipment operation parameter, is the data of the rth point of the equipment operation parameters, The distance k is the distance in which the device operates. is the local reachable density of the data at the i-th point of the device operation parameter, is the local reachable density of the data at the rth point of the device operating parameter.
[0042] In a second aspect, the present invention provides a system for automated equipment detection, comprising:
[0043] A parameter acquisition module, used to acquire working images and equipment operating parameters collected from the automation equipment during its operation;
[0044] An image feature extraction module is used to perform recognition processing operations according to the working image to obtain working state features;
[0045] A feature conversion module, used to perform a feature vector conversion operation according to the working state feature to obtain a working state feature vector;
[0046] A system scoring module, used to perform a device scoring operation according to the working state feature vector to obtain a performance scoring level;
[0047] A parameter inspection module, used to perform parameter inspection operations according to the equipment operation parameters to obtain equipment abnormality data;
[0048] The user reminder module is used to light up a reminder operation according to the abnormal data of the equipment and the performance score level, so as to provide the user with feedback on the detection results of the automation equipment.
[0049] In a third aspect, the present invention provides an electronic device, characterized in that it includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the method of automated equipment detection as described above.
[0050] In a fourth aspect, the present invention also provides a computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the method of automated equipment detection as described above.
[0051] Compared with the prior art, the present invention has the following beneficial effects:
[0052] The present invention provides a method for detecting automated equipment, comprising: acquiring a working image and equipment operating parameters collected from the automated equipment during operation of the automated equipment; performing an identification and processing operation based on the working image to obtain a working state feature; performing a feature vector conversion operation based on the working state feature to obtain a working state feature vector; performing an equipment scoring operation based on the working state feature vector to obtain a performance scoring grade; performing a parameter verification operation based on the equipment operating parameters to obtain equipment abnormality data; performing a light reminder operation based on the equipment abnormality data and the performance scoring grade to feed back the detection result of the automated equipment to the user.
[0053] In the present invention, the electronic device extracts feature values based on the working image acquired by the camera device and performs scoring operations based on the extracted feature values, and then performs abnormal state detection on the operating state of the device based on relevant parameters. After combining the two, the electronic device reminds the user by lighting lights of different colors. The color of the reminder light is pre-configured based on the score and the abnormal state. According to the above method, the detection of automated equipment is realized, and the whole process does not require manual measurement and analysis of data, which improves the detection efficiency. Kalman filtering and an unsupervised anomaly detection method are used to reduce data errors. On this basis, a camera device is added for auxiliary detection, which reduces the error of detection based on data alone, thereby significantly improving the detection efficiency and accuracy of automated equipment. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 It is a flowchart of a method for automated equipment detection provided by an embodiment of the present invention;
[0055] Figure 2 It is a schematic diagram of the structure of a system for automated equipment detection provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of 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.
[0057] It should be noted that the manufacturing industry has a huge market. Traditional manual production is not only costly and inefficient, but also some manufacturing projects cannot meet the minimum manufacturing standard level. Automation equipment, as a device that replaces manual labor, not only improves production efficiency and product quality, but also improves the working environment, reduces labor costs, and enhances the flexibility and controllability of the production process. However, with the improvement of the mechanization and automation level of production equipment, the control objects are becoming increasingly complex, and the requirements for detection technology are also constantly increasing. Because the data acquisition and processing of the detection object in the traditional detection method mainly rely on manual labor, this method has many limitations. There are large errors in manual data acquisition and processing. In addition, due to the large number of types of working parameters of automation equipment, manual acquisition and processing alone result in low efficiency and detection accuracy. At the same time, in some special cases, data alone cannot directly reflect the status of automation equipment, affecting detection accuracy. The above detection methods result in low detection accuracy and efficiency of automation equipment.
[0058] For the above problems, refer to Figure 1 The first embodiment of the present invention provides a method for automatic equipment detection, comprising the following steps:
[0059] S11, during the operation of the automation equipment, obtaining a working image and equipment operation parameters collected from the automation equipment;
[0060] S12, performing recognition processing operations according to the working image to obtain working state characteristics;
[0061] S13, performing a feature vector conversion operation according to the working state feature to obtain a working state feature vector;
[0062] S14, performing a device scoring operation according to the working status feature vector to obtain a performance scoring level;
[0063] S15, performing parameter inspection operation according to the equipment operation parameters to obtain equipment abnormality data;
[0064] S16, performing a light reminder operation according to the equipment abnormality data and the performance score level to provide the user with feedback on the detection result of the automation equipment.
[0065] In step S11, during the operation of the automation equipment, working images and equipment operation parameters collected from the automation equipment are acquired.
[0066] In the above embodiment, the electronic device performs image acquisition based on its own camera device. This process requires real-time image acquisition to provide an image basis for automated equipment detection. At the same time, it can also provide an information source for subsequent processing of image information to ensure that the acquired image can correctly reflect the working status of the automated equipment. In addition, the electronic device also collects equipment operating parameters in real time based on the sensor device to provide a data basis for subsequent parameter inspection operations.
[0067] In step S12, a recognition processing operation is performed based on the working image to obtain working state characteristics.
[0068] In the above embodiment, after the electronic device acquires the working image, it must first perform a preprocessing operation, wherein the preprocessing operation includes image scaling and image cropping. The purpose of this is to unify the image specifications to facilitate subsequent model processing. Only after the preprocessing operation can the recognition processing operation be performed. The working status characteristics processed by the recognition processing operation include mechanical parts, indicator lights, and braking systems. These three aspects can effectively complete the detection of the entire automated equipment. The fundamental purpose of performing the recognition processing operation is to make the subsequent scoring operation more accurate, thereby ensuring smooth and accurate detection from the working image to the entire automated equipment.
[0069] To further illustrate the recognition processing operation, the recognition processing operation is performed according to the working image to obtain the working state characteristics. Step S12 may include the following sub-steps:
[0070] S121, performing a preprocessing operation according to the working image to obtain a preprocessed image, wherein the preprocessing operation includes image scaling processing and image cropping processing;
[0071] S122. Perform recognition processing operation according to the preprocessed image to obtain working status features, wherein the recognition processing operation is to input the preprocessed image into the residual network Res-Net 101, and the working status features include mechanical parts, indicator lights and braking systems.
[0072] Exemplarily, in step S121, a preprocessing operation is performed according to the working image to obtain a preprocessed image, wherein the preprocessing operation includes image scaling and image cropping, wherein the preprocessing operation is to perform normalization preprocessing on the working image, perform scaling on each image, and crop it into a square of uniform size with a specification of 800 pixels on the shortest side and 1024 pixels on the longest side to facilitate model processing.
[0073] Exemplarily, in step S122, a recognition processing operation is performed based on the preprocessed image to obtain working status features, wherein the recognition processing operation specifically inputs the clothing image into a pre-trained residual network ResNet 101, extracts basic working status features of the working image, processes the corresponding feature image, and inputs the FPN feature pyramid to further extract working status features.
[0074] It should be noted that the working status characteristics include mechanical parts, indicator lights and braking systems.
[0075] It should also be specified that ResNet-101 is a deep convolutional neural network designed to solve the degradation problem in deep network training. Proposed by Microsoft Research, it increases the network depth and maintains the stability of training by introducing residual blocks. Each residual block includes a skip connection, which directly adds the input to the output of the block, thereby effectively alleviating the gradient vanishing problem. The network structure of ResNet-101 contains 101 layers, mainly composed of convolutional layers, batch normalization layers, ReLU activation layers, and fully connected layers. It performs well in tasks such as image classification and target detection, and has strong feature extraction and representation capabilities. Among them, the steps for ResNet-101 to extract working image features are as follows: First, the preprocessed working image is input into the ResNet-101 network. Then, the working image is processed by a series of convolutional layers, residual blocks, pooling layers, etc. These layers gradually extract low-level and high-level features in the image. Finally, at different levels of the network, feature maps of different levels can be extracted. Usually, the features extracted in the last few layers of the network are the richest.
[0076] In step S13, a feature vector conversion operation is performed according to the working state characteristics to obtain a working state feature vector.
[0077] In the above embodiment, it is necessary to first pre-set the Mask R-CNN model in the electronic device to provide a basis for feature vector conversion, and perform feature vector conversion on the pre-processed working image. The obtained working state feature vector can be calculated in the scoring process, ensuring that the scoring can be carried out accurately.
[0078] To further explain sub-step S13, it is necessary to specify that the overall architecture of Mask R-CNN can be divided into several main parts: Backbone: ResNet or VGG is usually used to extract basic convolutional features. Feature Pyramid Network (FPN): Used to process multi-scale features. Region Proposal Network (RPN): Generate candidate regions. RoIAlign: Accurately crop and align candidate regions. Classification and regression head: Used for classification and bounding box regression. Segmentation head: Generates segmentation masks of the target.
[0079] In addition, the specific steps for Mask R-CNN to extract the working state feature vector are as follows: First, send the input image to the Mask R-CNN model for forward propagation. The model will output the bounding box, category label, segmentation mask, etc. of the detected target. This process is called forward propagation. Second, during the forward propagation process, you can extract the intermediate layer features in the model. These features are usually located in the convolutional layer of the backbone network or the output layer of the feature pyramid network. By accessing the output of these layers, you can obtain the intermediate feature representation of the image. Third, in order to extract the global working state feature vector, features can be extracted from the output of the backbone network (such as ResNet) or FPN. These feature vectors are usually high-dimensional and contain global information of the image.
[0080] In step S14, a device scoring operation is performed according to the working status feature vector to obtain a performance scoring level.
[0081] In the above embodiment, the electronic device performs a scoring operation based on the working status feature vector that has been converted from the working picture. The scoring operation is performed by the microcontroller. The performance score level obtained by the microcontroller can intuitively display the performance of the device at this time. However, since the equipment operating parameters have not yet been detected for outliers, user feedback operations cannot be performed directly based on this score alone. The performance score level obtained here is the most important part of automated equipment detection. It avoids detection errors caused by the working status of automated equipment determined only by data, improves detection accuracy, saves human resources for automated equipment detection, and improves detection efficiency.
[0082] In one implementation, a device scoring operation is performed according to the working state feature vector to obtain a performance scoring level, including: the device scoring operation formula is:
[0083] ,
[0084] Among them, a, b, are pre-configured coefficients, To rate the performance, Rating index for mechanical parts, The indicator rating index is Braking system rating index;
[0085] The mechanical component rating index formula:
[0086] ,
[0087] The indicator rating index The formula is:
[0088] ,
[0089] The brake system rating index The formula is:
[0090] ,
[0091] in, is the feature set of mechanical parts in normal state, It is the indicator light feature set in normal state. is the set of braking system characteristics under normal conditions, is the collected feature vector of mechanical parts, is the collected indicator light feature vector, is the collected braking system feature vector, The pre-configured adjustment weight coefficient.
[0092] For example, a, b, are pre-configured coefficients, is a pre-configured adjustment weight coefficient, where a is set to 0.4, b is set to 0.4, Set the coefficient to 0.2, The coefficient is set to 0.05. The equipment scoring operation formula is:
[0093] ,
[0094] in, The score is divided into three levels: less than 0.4 is C, greater than 0.4 and less than 0.7 is B, and greater than 0.7 is A.
[0095] In step S15, a parameter checking operation is performed according to the equipment operation parameters to obtain equipment abnormality data.
[0096] In one embodiment, before performing a parameter verification operation based on the device operating parameters to obtain device abnormality data, it also includes: performing a denoising operation on the device operating parameters to obtain correction data, wherein the denoising operation is denoising using Kalman filtering, and the device operating parameters include temperature, voltage and vibration.
[0097] In the above embodiment, the electronic device obtains the device operating parameters from the sensor device, and after denoising, the obtained correction data is subjected to parameter verification, wherein the parameter verification method is an outlier detection based on the LOF algorithm. The algorithm can effectively detect outliers, and the microcontroller can promptly process and upload them to the electronic device, thereby ensuring real-time monitoring and preprocessing of data, avoiding gross errors caused by artificial data types and quantities, improving measurement accuracy and efficiency, and using the LOF algorithm to more promptly and accurately identify outliers, providing an accurate data basis for user feedback.
[0098] To further illustrate the parameter inspection operation, wherein the parameter inspection operation is performed according to the equipment operation parameters to obtain equipment abnormality data, step S15 may include the following sub-steps:
[0099] S151, performing a data point distance calculation operation according to the correction data to obtain a data reachable distance, wherein the data point distance calculation is to calculate the Euclidean distance between the equipment operation parameters;
[0100] S152, performing a local reachable density calculation operation according to the data reachable distance to obtain a local reachable density, wherein the local reachable density calculation operation is to calculate the inverse of the average value of the data reachable distance;
[0101] S153, performing a local outlier factor calculation operation according to the local reachable density to obtain a local outlier factor;
[0102] S154. Perform an outlier judgment operation according to the local outlier factor to obtain device abnormal data.
[0103] Exemplarily, in step S152, local reachable density calculation is performed according to the data reachable distance to obtain the local reachable density, including: the formula for the local reachable density calculation operation is:
[0104] ,
[0105] Wherein, k is the distance from the device operating parameter to the kth device operating parameter, The data of the i-th point of the equipment operation parameters, The data of the rth point of the equipment operation parameter, reach_distk ( ) is the reachable distance of the kth data, is the local reachable density of the data at the i-th point of the device operation parameter, is a neighborhood with a distance k from the operating parameters of the device.
[0106] Exemplarily, in step S153, a local outlier factor calculation operation is performed according to the local reachable density to obtain a local outlier factor, including: the formula for the local outlier factor calculation operation is:
[0107] ,
[0108] Wherein, k is the distance from the device operating parameter to the kth device operating parameter, is the local outlier factor of the data at the ith point of the equipment operation parameter, is the data of the rth point of the equipment operation parameters, The distance k is the distance in which the device operates. is the local reachable density of the data at the i-th point of the device operation parameter, is the local reachable density of the data at the rth point of the device operating parameter.
[0109] Exemplarily, in step S154, an outlier judgment operation is performed according to the local outlier factor to obtain device abnormal data, wherein the outlier judgment operation is performed when the device operating parameters are When it is less than 1, it is considered an abnormal value. When it is greater than or equal to 1, it is considered a non-abnormal value.
[0110] In step S16, a light reminder operation is performed according to the equipment abnormality data and the performance score level to feed back the detection result of the automation equipment to the user.
[0111] In the above embodiment, when the electronic device receives the device abnormality data and performance rating, it transmits this information to the microcontroller for further processing. The microcontroller first analyzes the abnormal data in detail to determine whether the device currently has a fault or performance abnormality. By analyzing the abnormal data, the microcontroller can identify specific problems in the operation of the device, such as overheating, abnormal vibration, or unstable current. This information is the basis for evaluating the status of the device and can help the system quickly locate the root cause of the problem. Subsequently, the microcontroller will analyze the performance rating. The performance rating is usually based on indicators such as the operating efficiency, reliability, and stability of the device to evaluate whether the device is within the normal range. By comparing the current rating with the preset standard value, the microcontroller can determine the overall health status of the device and possible performance degradation. Finally, the microcontroller controls the indicator lights of different colors according to the analysis results to remind the user of the detection results of the device. Different colors of lights represent different states, such as green indicates that the device is normal, yellow indicates that the device may have a small fault problem that needs to be promptly and carefully checked, and red indicates that the device has a serious fault that needs to be repaired in time. Such a visual indication system allows users to quickly understand the status of the device, so that they can take timely actions to shut down or repair it. In this way, the detection efficiency and accuracy of electronic equipment are significantly improved, while also reducing errors in manual detection and improving the reliability and maintenance efficiency of automated equipment.
[0112] For example, the lighting of different colors of lights is first based on the number of abnormal data. If the number of abnormal data exceeds the preset threshold of 5%, a red light reminder is directly turned on without considering the performance score level. When the number of abnormal data does not exceed the preset threshold of 5%, different colors of lights are turned on according to the performance score level. The light color selection is as follows: The red light is on when the rating is C. The yellow light is on when the rating is B and The green light will be on when the rating is A. The red light means that the automation equipment has a major fault and needs to be repaired by a dedicated person. The yellow light means that the automation equipment has a minor fault and needs to be stopped for repair and inspection. The green light means that the automation equipment is working normally and can be used normally.
[0113] The following describes the working process of the present invention using a common scenario as an example. Figure 1 , Figure 1 The invention is a flow chart of a method for automated equipment detection.
[0114] Step 1: The electronic device obtains a working image according to the camera device carried by the electronic device and obtains an operating parameter of the detection device according to the sensor device carried by the electronic device;
[0115] Step 2: The electronic device first performs image preprocessing on the working image, and the residual network Res-Net 101 extracts the working state features;
[0116] Step 3: The electronic device converts the working state features into working state feature vectors according to the Mask R-CNN model;
[0117] Step 4: The electronic device uses a scoring formula to score the working state feature vector to obtain a performance score level;
[0118] Step 5: The electronic device first performs denoising processing according to the acquired operating parameters of the detection equipment, and then performs parameter verification operation on the obtained correction data to obtain equipment abnormality data;
[0119] Step 6: The electronic device analyzes and processes the obtained performance rating and device abnormality data, and then provides feedback to the user, and reminds the user in real time by lighting a reminder light.
[0120] In summary, the method for automated equipment detection provided by an embodiment of the present invention is executed by an electronic device, and includes: during the operation of the automated equipment, obtaining a working image and equipment operating parameters collected from the automated equipment; performing an identification and processing operation based on the working image to obtain a working state feature; performing a feature vector conversion operation based on the working state feature to obtain a working state feature vector; performing an equipment scoring operation based on the working state feature vector to obtain a performance score grade; performing a parameter verification operation based on the equipment operating parameters to obtain equipment abnormality data; performing a light reminder operation based on the equipment abnormality data and the performance score grade to provide feedback to the user on the detection result of the automated equipment.
[0121] The embodiment of the present invention also provides a system for automatic equipment detection, such as Figure 2 As shown, it is a structural block diagram of a system for automated equipment detection provided by an embodiment of the present invention, including:
[0122] The parameter acquisition module 1 is used to acquire working images and equipment operation parameters collected from the automation equipment during its operation;
[0123] An image feature extraction module 2 is used to perform recognition processing operations according to the working image to obtain working state features;
[0124] A feature conversion module 3 is used to perform a feature vector conversion operation according to the working state feature to obtain a working state feature vector;
[0125] The system scoring module 4 is used to perform a device scoring operation according to the working state feature vector to obtain a performance scoring level;
[0126] A parameter inspection module 5 is used to perform parameter inspection operations according to the equipment operation parameters to obtain equipment abnormality data;
[0127] The user reminder module 6 is used to light up a reminder operation according to the equipment abnormality data and the performance score level, so as to provide the user with feedback on the detection result of the automation equipment.
[0128] In an optional implementation, the image feature extraction module 2 is further used for:
[0129] According to the working image, an identification processing operation is performed to obtain working status characteristics, including: according to the working image, a preprocessing operation is performed to obtain a preprocessed image, wherein the preprocessing operation includes image scaling processing and image cropping processing; according to the preprocessed image, an identification processing operation is performed to obtain working status characteristics, wherein the identification processing operation is to input the preprocessed image into the residual network Res-Net 101, and the working status characteristics include mechanical parts, indicator lights and braking systems.
[0130] In an optional implementation, the image feature extraction module 4 is further used for:
[0131] The device scoring operation is performed according to the working state feature vector to obtain a performance scoring level, including: the device scoring operation formula is:
[0132] ,
[0133] Among them, a, b, are pre-configured coefficients, To rate the performance, Rating index for mechanical parts, The indicator rating index is Braking system scoring index; Mechanical component scoring index The formula is:
[0134] ,
[0135] The indicator rating index The formula is:
[0136] ,
[0137] The brake system rating index The formula is:
[0138] ,
[0139] in, is the feature set of mechanical parts in normal state, It is the indicator light feature set in normal state. is the set of braking system characteristics under normal conditions, is the collected feature vector of mechanical parts, is the collected indicator light feature vector, is the collected braking system feature vector, The pre-configured adjustment weight coefficient.
[0140] In an optional embodiment, the system is further used for:
[0141] Before performing parameter verification operation according to the equipment operating parameters to obtain equipment abnormality data, it also includes: performing denoising operation on the equipment operating parameters to obtain correction data, wherein the denoising operation is denoising using Kalman filtering, and the equipment operating parameters include temperature, voltage and vibration.
[0142] In an optional implementation, the parameter verification module 5 is further used for:
[0143] According to the equipment operating parameters, a parameter verification operation is performed to obtain equipment abnormal data, including: according to the correction data, a data point distance calculation operation is performed to obtain a data reachable distance, wherein the data point distance calculation is to calculate the Euclidean distance between the equipment operating parameters; according to the data reachable distance, a local reachable density calculation operation is performed to obtain a local reachable density, wherein the local reachable density calculation operation is to calculate the inverse of the average value of the data reachable distance; according to the local reachable density, a local outlier factor calculation operation is performed to obtain a local outlier factor; according to the local outlier factor, an outlier judgment operation is performed to obtain equipment abnormal data.
[0144] In an optional embodiment, the system is further used for:
[0145] According to the data reachable distance, local reachable density calculation is performed to obtain local reachable density, including: the formula for the local reachable density calculation operation is:
[0146] ,
[0147] Wherein, k is the distance from the device operating parameter to the kth device operating parameter, The data of the i-th point of the equipment operation parameters, The data of the rth point of the equipment operation parameters, reach_dist k ( ) is the reachable distance of the kth data, is the local reachable density of the data at the i-th point of the device operation parameter, is a neighborhood with a distance k from the operating parameters of the device.
[0148] In an optional embodiment, the system is further used for:
[0149] According to the local reachable density, a local outlier factor calculation operation is performed to obtain a local outlier factor, including: the formula of the local outlier factor calculation operation is:
[0150] ,
[0151] Wherein, k is the distance from the device operating parameter to the kth device operating parameter, is the local outlier factor of the data at the ith point of the equipment operation parameter, is the data of the rth point of the equipment operation parameters, The distance k is the distance in which the device operates. is the local reachable density of the data at the i-th point of the device operation parameter, is the local reachable density of the data at the rth point of the device operating parameter.
[0152] An embodiment of the present invention further provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for automatic device detection as described in the above embodiment when executing the computer program.
[0153] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.
[0154] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.
[0155] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.
[0156] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0157] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.
[0158] It should be noted that the system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the system embodiment provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.
[0159] Compared with the prior art, the beneficial effect of the embodiments of the present invention is that: in the present invention, the electronic device extracts feature values based on the working image obtained by the camera device and performs scoring operations based on the extracted feature values, and then performs abnormal state detection on the operating state of the device based on relevant parameters. After combining the two, the electronic device reminds the user by lighting lights of different colors. The color of the reminder light is pre-configured based on the score and the abnormal state. According to the above method, the detection of automated equipment is realized, and the whole process does not require manual measurement and analysis of data, which improves the detection efficiency. Kalman filtering and an unsupervised anomaly detection method are used to reduce data errors. On this basis, a camera device is added for auxiliary detection, which reduces the error of detection based on data alone, thereby significantly improving the detection efficiency and accuracy of automated equipment.
[0160] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. A method for automated equipment detection, It is characterized in that include: During the operation of the automation equipment, obtaining working images and equipment operation parameters collected from the automation equipment; According to the working image, a recognition processing operation is performed to obtain working state characteristics; According to the working state characteristics, a feature vector conversion operation is performed to obtain a working state feature vector; Performing a device scoring operation according to the working status feature vector to obtain a performance scoring level; According to the equipment operation parameters, a parameter inspection operation is performed to obtain equipment abnormality data; According to the abnormal data of the equipment and the performance score level, a light reminder operation is performed to feedback the detection result of the automation equipment to the user; Among them, according to the number of abnormal data of the equipment, if the number of abnormal data of the equipment exceeds the preset threshold of 5%, a red light will be turned on to remind you directly; when the number of abnormal data of the equipment does not exceed the preset threshold of 5%, different colors of lights will be turned on to remind you according to the performance score level; The step of performing a device scoring operation according to the working state feature vector to obtain a performance scoring level includes: The equipment scoring operation formula is: ,in, a, b, are pre-configured coefficients, To rate the performance, Rating index for mechanical parts, The indicator rating index is Braking system rating index; The mechanical component rating index The formula is: , the indicator light rating index The formula is: , the braking system rating index The formula is: ,in, is the feature set of mechanical parts in normal state, It is the indicator light feature set in normal state. is the set of braking system characteristics under normal conditions, is the collected feature vector of mechanical parts, is the collected indicator light feature vector, is the collected braking system feature vector, The pre-configured adjustment weight coefficient.
2. The method for automated equipment detection according to claim 1, It is characterized in that The step of performing recognition processing operations according to the working image to obtain working state characteristics includes: Performing a preprocessing operation on the working image to obtain a preprocessed image, wherein the preprocessing operation includes image scaling and image cropping; According to the preprocessed image, a recognition processing operation is performed to obtain working state characteristics, wherein the recognition processing operation is to input the preprocessed image into the residual network Res-Net 101, and the working state characteristics include mechanical parts, indicator lights and braking systems.
3. The method for automated equipment detection according to claim 1, It is characterized in that Before performing parameter inspection operation according to the equipment operation parameters to obtain equipment abnormality data, the method further includes: A denoising operation is performed on the device operating parameters to obtain correction data, wherein the denoising operation is denoising using Kalman filtering, and the device operating parameters include temperature, voltage and vibration.
4. The method for automated equipment detection according to claim 3, It is characterized in that The performing of parameter inspection operation according to the equipment operation parameters to obtain equipment abnormality data includes: According to the correction data, a data point distance calculation operation is performed to obtain a data reachable distance, wherein the data point distance calculation is to calculate the Euclidean distance between the operating parameters of the equipment; According to the data reachable distance, a local reachable density calculation operation is performed to obtain a local reachable density, wherein the local reachable density calculation operation is to calculate the inverse of the average value of the data reachable distance; Performing a local outlier factor calculation operation according to the local reachable density to obtain a local outlier factor; According to the local outlier factor, an outlier judgment operation is performed to obtain device abnormal data.
5. The method for automated equipment detection according to claim 4, It is characterized in that The calculating the local reachable density according to the data reachable distance to obtain the local reachable density includes: The formula for the local reachable density calculation operation is: ,in, The operating parameters of the device are i Points of data, The operating parameters of the device are r Points of data, is the data reachable distance, The operating parameters of the device are i The local reachable density of the data at each point.
6. The method for automated equipment detection according to claim 4, It is characterized in that The performing a local outlier factor calculation operation according to the local reachable density to obtain the local outlier factor includes: The formula for the local outlier factor calculation operation is: ,in, The operating parameters of the device are i The local outlier factor of the data points, The operating parameters of the device are r Points of data, The operating parameters of the device are i The local reachable density of the data at each point, The operating parameters of the device are r The local reachable density of the data at each point.
7. A system for automated equipment detection, It is characterized in that A method for implementing automated equipment detection as claimed in any one of claims 1 to 6, comprising: A parameter acquisition module, used to acquire working images and equipment operating parameters collected from the automation equipment during its operation; An image feature extraction module is used to perform recognition processing operations according to the working image to obtain working state features; A feature conversion module, used to perform a feature vector conversion operation according to the working state feature to obtain a working state feature vector; A system scoring module, used to perform a device scoring operation according to the working state feature vector to obtain a performance scoring level; A parameter inspection module, used to perform parameter inspection operations according to the equipment operation parameters to obtain equipment abnormality data; The user reminder module is used to light up a reminder operation according to the abnormal data of the equipment and the performance score level, so as to provide the user with feedback on the detection results of the automation equipment.
8. An electronic device, It is characterized in that The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the method for detecting an automated device according to any one of claims 1 to 6 when executing the computer program.
9. A computer-readable storage medium, It is characterized in that The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for automated equipment detection according to any one of claims 1 to 6.