CN117085431A

CN117085431A - Control system and method for bag-type dust collector

Info

Publication number: CN117085431A
Application number: CN202311168821.1A
Authority: CN
Inventors: 郭维
Original assignee: Deqing Zhongxin Environmental Protection Equipment Co ltd
Current assignee: Deqing Zhongxin Environmental Protection Equipment Co ltd
Priority date: 2023-09-11
Filing date: 2023-09-11
Publication date: 2023-11-21

Abstract

The application relates to the field of intelligent control, and particularly discloses a control system and method of a bag-type dust collector, which are characterized in that an artificial intelligent technology based on a deep neural network model is adopted to obtain the rotating speed of a motor and the blade angle of a fan at a plurality of preset time points in a preset time period, the rotating speed and the blade angle are arranged into vectors, relevant characteristic information is extracted through a characteristic extractor, and after correlation encoding, low-dimensional information is further mapped into a high-dimensional characteristic space through a convolution network, so that a classification result for indicating whether the rotating speed of the fan at the current time point is increased, reduced or unchanged is obtained. Therefore, the automatic adjustment of the rotating speed of the fan can be realized, so that the dust removal requirements under different processes and material treatment conditions can be met.

Description

Control system and method for bag-type dust collector

Technical Field

The application relates to the field of intelligent control, in particular to a control system and method of a bag-type dust collector.

Background

The cloth bag dust collector is a device for separating dust from flue gas, and is a dry type high-efficiency dust collector, which is a dust collector for capturing solid particles in dust-containing gas by using a bag type filter element made of fiber woven materials. The fan is a key device in the bag-type dust collector system and is used for generating air flow and driving the air flow to pass through the filter bag for dust collection. If the rotating speed of the fan is too high, abrasion of the cloth bag can be increased, and vibration or vibration of the surface of the cloth bag can be caused. If the rotating speed is too small, particles cannot be effectively trapped, and the particles accumulate too much on the surface of the cloth bag, so that the dust removal efficiency is affected. However, the rotational speed of the fan in the dust removal is difficult to accurately control in the prior art, so that the dust removal effect is poor.

Therefore, an optimized control scheme for a bag house is desired.

Disclosure of Invention

The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a control system and a control method of a bag-type dust collector, which adopt an artificial intelligence technology based on a deep neural network model to acquire the rotating speed of a motor and the blade angle of a fan at a plurality of preset time points in a preset time period, arrange the rotating speed and the blade angle into vectors, extract relevant characteristic information through a characteristic extractor, and map low-dimensional information into a high-dimensional characteristic space through a convolution network after associated coding so as to obtain a classification result for indicating whether the rotating speed of the fan at the current time point is increased, reduced or unchanged. Therefore, the automatic adjustment of the rotating speed of the fan can be realized, so that the dust removal requirements under different processes and material treatment conditions can be met.

According to one aspect of the present application, there is provided a control system of a bag-type dust collector, comprising:

the data acquisition module is used for acquiring the rotating speeds of motors and the blade angles of the fans at a plurality of preset time points in a preset time period;

the rotating speed feature extraction module is used for arranging the rotating speeds of the motors at a plurality of preset time points into rotating speed time sequence input vectors according to the time dimension and then obtaining rotating speed time sequence feature vectors through a rotating speed feature extractor comprising a first convolution and a second convolution;

The blade angle feature extraction module is used for arranging the blade angles of the fans at a plurality of preset time points into angle time sequence input vectors according to the time dimension and then obtaining angle time sequence feature vectors through an angle feature extractor comprising a first convolution and a second convolution;

the characteristic association module is used for carrying out association coding on the rotating speed time sequence characteristic vector and the angle time sequence characteristic vector to obtain a fan working state association matrix;

the state characteristic enhancement module is used for enabling the fan working state incidence matrix to pass through a convolutional neural network model serving as a characteristic extractor to obtain a fan working state characteristic diagram;

the optimizing module is used for carrying out characteristic manifold modulation on the fan working state characteristic diagram so as to obtain an optimized fan working state characteristic diagram;

and the fan rotating speed adjusting module is used for enabling the optimized fan working state characteristic diagram to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rotating speed of the fan rotating speed at the current time point is increased, reduced or unchanged.

In the control system of the bag-type dust collector, the rotational speed feature extraction module includes: a first scale rotational speed feature extraction unit, configured to input the rotational speed time sequence input vector into a first convolution layer of the rotational speed feature extractor including a first convolution and a second convolution to obtain a first scale rotational speed time sequence feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second scale rotational speed feature extraction unit configured to input the rotational speed time sequence input vector into a second convolution layer including a rotational speed feature extractor of a first convolution and a second convolution to obtain a second scale rotational speed time sequence feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and the rotational speed feature fusion unit is used for cascading the first scale rotational speed time sequence feature vector and the second scale rotational speed time sequence feature vector by using the cascading layer of the rotational speed feature extractor comprising the first convolution and the second convolution so as to obtain the rotational speed time sequence feature vector.

In the control system of the bag-type dust collector, the blade angle feature extraction module comprises: a first scale angle feature extraction unit, configured to input the angle timing input vector into a first convolution layer of the angle feature extractor including a first convolution and a second convolution to obtain a first scale angle timing feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second scale angle feature extraction unit configured to input the angle timing input vector into a second convolution layer including the angle feature extractor of the first convolution and a second convolution to obtain a second scale angle timing feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and an angle feature fusion unit, configured to concatenate the first scale angle timing feature vector and the second scale angle timing feature vector using the concatenation layer including the angle feature extractor of the first convolution and the second convolution to obtain the angle timing feature vector.

In the control system of the bag-type dust collector, the characteristic association module is used for respectively calculating products of the transpose vector of the rotating speed time sequence characteristic vector and the angle time sequence characteristic vector to obtain the fan working state association matrix.

In the control system of the bag-type dust collector, the state characteristic enhancing module is used for: each layer of the convolutional neural network model using the feature extractor performs, in forward transfer of the layer, input data: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on the local feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the convolutional neural network model serving as the feature extractor is the fan working state feature map, and the input of the first layer of the convolutional neural network model serving as the feature extractor is the fan working state association matrix.

In the control system of the bag-type dust collector, the fan rotating speed adjusting module comprises: the unfolding unit is used for unfolding the working state characteristic diagram of the optimized fan into a classification characteristic vector; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a full-connection layer of the classifier so as to obtain coded classification characteristic vectors; and a classification result unit, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

According to another aspect of the present application, there is provided a control method of a bag-type dust collector, comprising:

acquiring the rotating speed of a motor and the blade angle of a fan at a plurality of preset time points in a preset time period;

arranging the rotation speeds of the motors at a plurality of preset time points into rotation speed time sequence input vectors according to time dimensions, and then obtaining rotation speed time sequence feature vectors through a rotation speed feature extractor comprising a first convolution and a second convolution;

the blade angles of the fans at a plurality of preset time points are arranged into angle time sequence input vectors according to the time dimension, and then the angle time sequence input vectors are obtained through an angle feature extractor comprising a first convolution and a second convolution;

performing association coding on the rotating speed time sequence feature vector and the angle time sequence feature vector to obtain a fan working state association matrix;

the fan working state association matrix is processed through a convolutional neural network model serving as a feature extractor to obtain a fan working state feature diagram;

performing characteristic manifold modulation on the fan working state characteristic diagram to obtain an optimized fan working state characteristic diagram;

and the optimized fan working state characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rotating speed of the fan at the current time point is increased, reduced or unchanged.

Compared with the prior art, the control system and the method for the bag-type dust collector provided by the application adopt an artificial intelligence technology based on a deep neural network model, acquire the rotating speed of a motor and the blade angle of a fan at a plurality of preset time points in a preset time period, arrange the rotating speed and the blade angle into vectors, extract relevant characteristic information through a characteristic extractor, and map low-dimensional information to a high-dimensional characteristic space through a convolution network after associated coding so as to obtain a classification result for indicating whether the rotating speed of the fan at the current time point is increased, reduced or unchanged. Therefore, the automatic adjustment of the rotating speed of the fan can be realized, so that the dust removal requirements under different processes and material treatment conditions can be met.

Drawings

The above and other objects, features and advantages of the present application will become more apparent by describing embodiments of the present application in more detail with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and together with the embodiments of the application, and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.

Fig. 1 is a block diagram of a control system of a bag-type dust collector according to an embodiment of the application.

Fig. 2 is a schematic architecture diagram of a control system of a bag-type dust collector according to an embodiment of the application.

Fig. 3 is a block diagram of a rotational speed feature extraction module in a control system of a bag-type dust collector according to an embodiment of the application.

Fig. 4 is a block diagram of an optimization module in a control system of a bag-type dust collector according to an embodiment of the application.

Fig. 5 is a flowchart of a control method of a bag-type dust collector according to an embodiment of the present application.

Fig. 6 is a block diagram of an electronic device according to an embodiment of the application.

Detailed Description

Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

Summary of the application

As described above, in the dust removal process, an excessively large or small rotation speed of the blower may cause some problems and adverse effects. Excessive fan speeds may result in excessive airflow rates and pressures, may cause overload or damage to equipment, exacerbate wear of the cloth bags, and may also result in increased energy consumption and increased operating costs. If the rotating speed of the fan is too small, particles cannot be effectively trapped, dust and pollutants cannot be effectively removed from the treatment area, so that the suspension time of the dust in the air is increased, the time and the distance of dust sedimentation are increased, and the dust removal efficiency is reduced. However, the rotating speed of the fan cannot be accurately controlled in the prior art, so that the dust removal efficiency is reduced, and the dust removal progress is influenced. Therefore, an optimized control scheme for a bag house is desired.

In view of the above technical problems, the applicant of the present application obtains a classification result indicating whether the rotational speed of the fan at the current time point should be increased, decreased or unchanged by obtaining the rotational speeds of the motors and the blade angles of the fans at a plurality of predetermined time points in a predetermined time period, arranging the rotational speeds and the blade angles into vectors, extracting relevant feature information by a feature extractor, and mapping low-dimensional information to a high-dimensional feature space by a convolution network after association encoding.

Accordingly, in the technical scheme of the application, the fan rotating speed is considered to be a key parameter for controlling the output air quantity and the air flow speed of the fan. According to different process requirements and material handling characteristics, the rotational speed of the fan needs to be adjusted to ensure that the dust collector system can effectively capture and collect dust particles. Through obtaining fan rotational speed data, can real-time supervision fan running state to adjust according to the demand, in order to reach best dust removal effect. The fan blade angle is an important parameter for adjusting the output air quantity and the air flow direction of the fan. Through adjusting the blade angle, the air flow direction and speed output by the fan can be changed to adapt to different process conditions and material treatment requirements. For example, for certain processes requiring stronger airflow impingement forces, the blade angle may be increased to increase the airflow speed of the fan output. The adjustment condition of the blade angle can be monitored in real time by acquiring the fan blade angle data, so that the air flow output by the fan can meet the process requirement. Therefore, the working state of the fan can be comprehensively monitored and controlled by acquiring the data of the rotating speed and the blade angle of the fan, the working parameters of the dust remover can be optimized, the dust removal efficiency can be improved, and the stable operation of the system under different process and material treatment conditions can be ensured.

In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.

Specifically, in the technical scheme of the application, firstly, the rotating speed of the motor and the blade angle of the fan at a plurality of preset time points in a preset time period are obtained. Considering that the rotational speed time series data is usually a sequence containing a large number of time points, the problem that the dimension of the data is high and the redundant information is large can be faced when the original data is directly used for analysis and processing. The rotating speed feature extractor can extract representative and distinguishing features from time sequence data, convert the original data into feature vectors which are more compact and rich in information, and learn potential modes and associated information in the rotating speed time sequence data. Thus, useful features can be extracted from the rotational speed time series data by the rotational speed feature extractor.

Next, considering that the blade angle timing data is generally a sequence including a large number of time points, the direct use of the raw data for analysis and processing may face the problems of high data dimension and much redundant information. By means of the angle feature extractor, representative and distinguishing features can be extracted from time sequence data, and original data can be converted into feature vectors which are more compact and rich in information. The angle feature extractor may help to reduce the dimension of the high-dimensional blade angle temporal data to a lower-dimensional feature vector, thereby reducing the complexity of data processing and storage, and may learn the underlying patterns and associated information in the blade angle temporal data. Thus, useful features can be extracted from the blade angle timing data by the angle feature extractor.

Then, considering that rotational speed and angle are two important aspects of the operational state of the fan, they provide information about the performance of the fan and the position of the blades, respectively. The information of the two aspects can be integrated by carrying out associated coding on the rotating speed time sequence feature vector and the angle time sequence feature vector, so that more comprehensive fan working state representation is obtained. Specifically, the rotation speed and the angle time sequence feature vector capture dynamic features in different aspects respectively, and can be combined through associated coding, so that the expression capability of the features is improved. Thus, the comprehensive characteristics of the fan under different working states can be better reflected. Therefore, the rotational speed time sequence feature vector and the angle time sequence feature vector can be combined through the associated codes, and more comprehensive and comprehensive fan working state information is obtained.

Next, consider that the feature extractor may learn gradually the hierarchical features in the fan operating state correlation matrix. These features can gradually evolve from low-level local features (e.g., edges, textures) to high-level global features (e.g., shapes, patterns) to better represent key information of the fan operating state. In addition, convolutional neural networks can better accommodate the characteristics of different data and extract the most relevant and useful features from the data than manually designed features. This ability to automatically learn features makes the feature extractor more expressive and generalizable. In addition, the feature extractor may convert the fan operating state correlation matrix into feature maps that may be used as input for subsequent classification tasks. The feature images extracted by the feature extractor can provide input with more distinguishing degree and expression capability, and are beneficial to improving the accuracy of the classification model. Thus, by means of a feature extractor (convolutional neural network model), a higher level, more abstract representation of features can be learned from the fan operating state correlation matrix. This helps to capture key modes and details of the fan operating state.

Further, the fan working state characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rotating speed of the fan at the current time point is increased, decreased or unchanged. Considering the classifier, the method can judge how the rotating speed of the fan should be adjusted according to the characteristic diagram of the working state of the fan at the current time point. The classifier can learn the association between different feature map modes and the rotation speed adjustment, so that corresponding rotation speed adjustment decisions are made according to the mode of the current feature map. In addition, the classifier can classify the feature images in a real-time environment, and a decision of rotating speed adjustment can be quickly made. The fan control system is very important to a fan control system which needs real-time response and adjustment, and can ensure that the rotating speed of the fan can adapt to different working conditions and processing requirements in time, so that automatic rotating speed control is realized, the real-time performance and performance of the system are improved, and the requirement of manual intervention is reduced.

In particular, it is contemplated that the fan operating state feature map may have a high dimension, including a large number of redundant or uncorrelated features. This can lead to increased computational complexity, prolonged model training time, and may introduce overfitting problems. The high-dimensional features can be mapped to the low-dimensional space through the feature manifold modulation, so that the data dimension is reduced, and the computing efficiency and the generalization capability of the model are improved. Moreover, the fan operating state feature diagram may have complex data structures and internal relationships. The direct use of classifiers may not take full advantage of these structures and relationships, resulting in reduced classification performance. The feature manifold modulation may preserve local structure and relationships between the data, making it easier for the classifier to learn valid patterns and rules from it. Whereas feature manifold modulation may map data to more discriminative and interpretable feature spaces. The method is helpful for understanding and explaining the classification result, and reveals the intrinsic law and characteristic importance of the working state of the fan, so that more information is provided for controlling the rotating speed of the fan.

Specifically, performing feature manifold modulation on the fan working state feature map to obtain an optimized fan working state feature map, including: inputting the fan working state feature map into a Si gmoi d activation function to map feature values of all positions of the fan working state feature map into a probability space so as to obtain a probabilistic fan working state feature map; mapping each feature matrix of the probabilistic fan working state feature diagram along the channel dimension into the same target space to obtain a plurality of feature matrices after radiation mapping; calculating mutual information between any two of the plurality of feature matrices after the similar radiation mapping to obtain a similar affine transformation feature vector consisting of a plurality of mutual information; generating a network from the affine-like transformation feature vector through weights comprising a point convolution layer and a batch normalization processing layer to obtain an affine-like associated probability feature vector; and taking the characteristic values of all positions in the affine-associated probability characteristic vector as weight values, and respectively weighting all characteristic matrixes of the fan working state characteristic diagram along the channel dimension to obtain an optimized fan working state characteristic diagram.

In the technical scheme of the application, the fan working state feature diagram can be regarded as a feature set of the channel feature matrix, so that if the association between the channel feature matrices can be utilized, the feature expression accuracy and certainty of the feature set can be optimized based on the association information between the feature elements in the set. Specifically, in the technical scheme of the application, affine transformation is carried out on each channel feature matrix of the fan working state feature vector so as to map each channel feature matrix to a public pivot feature space, then the mutual information between the two channel feature matrices after the two classes of radiation mapping is used for representing the association information between the two channel feature matrices, the global association information is captured by a point convolution kernel and batch normalization processing so as to obtain an affine association probability feature vector, and finally the feature values of each position in the affine association probability feature vector are used as weight values, and each feature matrix of the fan working state feature vector along the channel dimension is weighted respectively so as to obtain the optimized fan working state feature vector. Therefore, the robustness of the fan working state feature diagram can be enhanced, the expression capability of the feature diagram can be improved, and more details and semantic information can be captured.

Based on this, the application provides a control system of a bag-type dust collector, comprising: the data acquisition module is used for acquiring the rotating speeds of motors and the blade angles of the fans at a plurality of preset time points in a preset time period; the rotating speed feature extraction module is used for arranging the rotating speeds of the motors at a plurality of preset time points into rotating speed time sequence input vectors according to the time dimension and then obtaining rotating speed time sequence feature vectors through a rotating speed feature extractor comprising a first convolution and a second convolution; the blade angle feature extraction module is used for arranging the blade angles of the fans at a plurality of preset time points into angle time sequence input vectors according to the time dimension and then obtaining angle time sequence feature vectors through an angle feature extractor comprising a first convolution and a second convolution; the characteristic association module is used for carrying out association coding on the rotating speed time sequence characteristic vector and the angle time sequence characteristic vector to obtain a fan working state association matrix; the state characteristic enhancement module is used for enabling the fan working state incidence matrix to pass through a convolutional neural network model serving as a characteristic extractor to obtain a fan working state characteristic diagram; the optimizing module is used for carrying out characteristic manifold modulation on the fan working state characteristic diagram so as to obtain an optimized fan working state characteristic diagram; and the fan rotating speed adjusting module is used for enabling the optimized fan working state characteristic diagram to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rotating speed of the fan rotating speed at the current time point is increased, reduced or unchanged.

Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.

Exemplary System

Fig. 1 is a block diagram of a control system of a bag-type dust collector according to an embodiment of the application. As shown in fig. 1, a control system 100 of a bag-type dust collector according to an embodiment of the present application includes: a data acquisition module 110, configured to acquire rotational speeds of motors and blade angles of fans at a plurality of predetermined time points in a predetermined period; the rotational speed feature extraction module 120 is configured to arrange rotational speeds of the motors at the plurality of predetermined time points into rotational speed time sequence input vectors according to a time dimension, and obtain rotational speed time sequence feature vectors through a rotational speed feature extractor including a first convolution and a second convolution; the blade angle feature extraction module 130 is configured to arrange the blade angles of the fans at the plurality of predetermined time points into angle time sequence input vectors according to a time dimension, and then obtain angle time sequence feature vectors through an angle feature extractor including a first convolution and a second convolution; the feature association module 140 is configured to perform association encoding on the rotational speed time sequence feature vector and the angle time sequence feature vector to obtain a fan working state association matrix; the state feature enhancement module 150 is configured to pass the fan working state association matrix through a convolutional neural network model serving as a feature extractor to obtain a fan working state feature map; the optimizing module 160 is configured to perform feature manifold modulation on the fan working state feature map to obtain an optimized fan working state feature map; and a fan rotation speed adjustment module 170, configured to pass the optimized fan operation state feature map through a classifier to obtain a classification result, where the classification result is used to indicate whether the rotation speed of the fan rotation speed at the current time point should be increased, decreased or unchanged.

Fig. 2 is a schematic architecture diagram of a control system of a bag-type dust collector according to an embodiment of the application. As shown in fig. 2, first, the rotational speed of the motor and the blade angle of the blower are acquired at a plurality of predetermined time points for a predetermined period of time. And then, arranging the rotating speeds of the motors at a plurality of preset time points into rotating speed time sequence input vectors according to a time dimension, and then obtaining rotating speed time sequence feature vectors through a rotating speed feature extractor comprising a first convolution and a second convolution. Meanwhile, the blade angles of the fans at a plurality of preset time points are arranged into angle time sequence input vectors according to the time dimension, and then the angle time sequence input vectors are obtained through an angle feature extractor comprising a first convolution and a second convolution. And then, carrying out association coding on the rotating speed time sequence feature vector and the angle time sequence feature vector to obtain a fan working state association matrix. And then, the fan working state association matrix is passed through a convolutional neural network model serving as a feature extractor to obtain a fan working state feature map. And then, carrying out characteristic manifold modulation on the fan working state characteristic diagram to obtain an optimized fan working state characteristic diagram. And finally, the optimized fan working state characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rotating speed of the fan at the current time point is increased, decreased or unchanged.

In the embodiment of the present application, the data acquisition module 110 is configured to acquire the rotational speeds of the motors and the blade angles of the fans at a plurality of predetermined time points in a predetermined period. Considering that the rotating speed of the fan is a key parameter for controlling the output air quantity and the air flow speed of the fan. According to different process requirements and material handling characteristics, the rotational speed of the fan needs to be adjusted to ensure that the dust collector system can effectively capture and collect dust particles. Through obtaining fan rotational speed data, can real-time supervision fan running state to adjust according to the demand, in order to reach best dust removal effect. The fan blade angle is an important parameter for adjusting the output air quantity and the air flow direction of the fan. Through adjusting the blade angle, the air flow direction and speed output by the fan can be changed to adapt to different process conditions and material treatment requirements. For example, for certain processes requiring stronger airflow impingement forces, the blade angle may be increased to increase the airflow speed of the fan output. The adjustment condition of the blade angle can be monitored in real time by acquiring the fan blade angle data, so that the air flow output by the fan can meet the process requirement.

In the embodiment of the present application, the rotational speed feature extraction module 120 is configured to arrange rotational speeds of the motors at the plurality of predetermined time points into rotational speed time sequence input vectors according to a time dimension, and then obtain rotational speed time sequence feature vectors through a rotational speed feature extractor including a first convolution and a second convolution. Considering that the rotational speed time series data is usually a sequence containing a large number of time points, the problem that the dimension of the data is high and the redundant information is large can be faced when the original data is directly used for analysis and processing. The rotating speed feature extractor can extract representative and distinguishing features from time sequence data, convert the original data into feature vectors which are more compact and rich in information, and learn potential modes and associated information in the rotating speed time sequence data.

Fig. 3 is a block diagram of a rotational speed feature extraction module in a control system of a bag-type dust collector according to an embodiment of the application. Specifically, in the embodiment of the present application, as shown in fig. 3, the rotational speed feature extraction module 120 includes: a first scale rotational speed feature extraction unit 121, configured to input the rotational speed time sequence input vector into a first convolution layer of the rotational speed feature extractor including a first convolution and a second convolution to obtain a first scale rotational speed time sequence feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second scale rotational speed feature extraction unit 122, configured to input the rotational speed time sequence input vector into a second convolution layer including a rotational speed feature extractor of a first convolution and a second convolution to obtain a second scale rotational speed time sequence feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and a rotational speed feature fusion unit 123, configured to concatenate the first scale rotational speed time sequence feature vector and the second scale rotational speed time sequence feature vector using the concatenation layer including the rotational speed feature extractor of the first convolution and the second convolution to obtain the rotational speed time sequence feature vector.

In the embodiment of the present application, the blade angle feature extraction module 130 is configured to arrange the blade angles of the fans at the plurality of predetermined time points into the angle time sequence input vector according to the time dimension, and then obtain the angle time sequence feature vector by using an angle feature extractor including a first convolution and a second convolution. Considering that blade angle timing data is generally a sequence containing a large number of time points, the direct use of raw data for analysis and processing may face the problems of high data dimension and much redundant information. By means of the angle feature extractor, representative and distinguishing features can be extracted from time sequence data, and original data can be converted into feature vectors which are more compact and rich in information. The angle feature extractor may help to reduce the dimension of the high-dimensional blade angle temporal data to a lower-dimensional feature vector, thereby reducing the complexity of data processing and storage, and may learn the underlying patterns and associated information in the blade angle temporal data.

Specifically, in an embodiment of the present application, the blade angle feature extraction module includes: a first scale angle feature extraction unit, configured to input the angle timing input vector into a first convolution layer of the angle feature extractor including a first convolution and a second convolution to obtain a first scale angle timing feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length; a second scale angle feature extraction unit configured to input the angle timing input vector into a second convolution layer including the angle feature extractor of the first convolution and a second convolution to obtain a second scale angle timing feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length; and an angle feature fusion unit, configured to concatenate the first scale angle timing feature vector and the second scale angle timing feature vector using the concatenation layer including the angle feature extractor of the first convolution and the second convolution to obtain the angle timing feature vector.

In the embodiment of the present application, the feature association module 140 is configured to perform association encoding on the rotational speed time sequence feature vector and the angle time sequence feature vector to obtain a fan working state association matrix. Considering that rotational speed and angle are two important aspects of the operating condition of the fan, they provide information about the performance of the fan and the position of the blades, respectively. The information of the two aspects can be integrated by carrying out associated coding on the rotating speed time sequence feature vector and the angle time sequence feature vector, so that more comprehensive fan working state representation is obtained. Specifically, the rotation speed and the angle time sequence feature vector capture dynamic features in different aspects respectively, and can be combined through associated coding, so that the expression capability of the features is improved. Thus, the comprehensive characteristics of the fan under different working states can be better reflected.

Specifically, in the embodiment of the application, the characteristic association module is used for respectively calculating products between the transpose vector of the rotating speed time sequence characteristic vector and the angle time sequence characteristic vector to obtain the fan working state association matrix.

In the embodiment of the present application, the state feature enhancement module 150 is configured to obtain a fan working state feature map by passing the fan working state correlation matrix through a convolutional neural network model serving as a feature extractor. The feature extractor can gradually learn the hierarchical features in the fan working state association matrix. These features can gradually evolve from low-level local features (e.g., edges, textures) to high-level global features (e.g., shapes, patterns) to better represent key information of the fan operating state. In addition, convolutional neural networks can better accommodate the characteristics of different data and extract the most relevant and useful features from the data than manually designed features. This ability to automatically learn features makes the feature extractor more expressive and generalizable. In addition, the feature extractor may convert the fan operating state correlation matrix into feature maps that may be used as input for subsequent classification tasks. The feature images extracted by the feature extractor can provide input with more distinguishing degree and expression capability, and are beneficial to improving the accuracy of the classification model.

Specifically, in the embodiment of the present application, the state feature enhancement module is configured to: each layer of the convolutional neural network model using the feature extractor performs, in forward transfer of the layer, input data: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on the local feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the convolutional neural network model serving as the feature extractor is the fan working state feature map, and the input of the first layer of the convolutional neural network model serving as the feature extractor is the fan working state association matrix.

In the embodiment of the present application, the optimizing module 160 is configured to perform feature manifold modulation on the fan working state feature map to obtain an optimized fan working state feature map.

Fig. 4 is a block diagram of an optimization module in a control system of a bag-type dust collector according to an embodiment of the application. Specifically, in the embodiment of the present application, as shown in fig. 4, the optimizing module 160 includes: a feature value mapping unit 161, configured to input the fan operation state feature map into a Sigmoid activation function to map feature values of each position of the fan operation state feature map into a probability space to obtain a probabilistic fan operation state feature map; the feature matrix mapping unit 162 is configured to map each feature matrix of the probabilistic fan working state feature map along the channel dimension to the same target space, so as to obtain a plurality of feature matrices after radiation mapping; a mutual information calculation unit 163, configured to calculate mutual information between any two of the feature matrices after the radiation mapping in the plurality of feature matrices after the radiation mapping to obtain a affine transformation-like feature vector composed of a plurality of mutual information; a weight generating unit 164, configured to generate the affine-like transformation feature vector through a weight generating network including a point convolution layer and a batch normalization processing layer to obtain an affine-like associated probability feature vector; and a weighting unit 165, configured to respectively weight each feature matrix of the fan working state feature map along the channel dimension by using the feature value of each position in the affine-associated probability feature vector as a weight value, so as to obtain an optimized fan working state feature map.

Therefore, the robustness of the fan working state feature diagram can be enhanced, the expression capability of the feature diagram can be improved, and more details and semantic information can be captured.

In the embodiment of the present application, the fan rotation speed adjustment module 170 is configured to pass the optimized fan operation state feature map through a classifier to obtain a classification result, where the classification result is used to indicate whether the rotation speed of the fan rotation speed at the current time point should be increased, decreased or unchanged. Considering the classifier, the method can judge how the rotating speed of the fan should be adjusted according to the characteristic diagram of the working state of the optimized fan at the current time point. The classifier can learn the association between different feature map modes and the rotation speed adjustment, so that corresponding rotation speed adjustment decisions are made according to the mode of the current feature map. In addition, the classifier can classify the feature images in a real-time environment, and a decision of rotating speed adjustment can be quickly made. The fan control system is very important to a fan control system which needs real-time response and adjustment, and can ensure that the rotating speed of the fan can adapt to different working conditions and processing requirements in time, so that automatic rotating speed control is realized, the real-time performance and performance of the system are improved, and the requirement of manual intervention is reduced.

Specifically, in an embodiment of the present application, the fan rotation speed adjustment module includes: the unfolding unit is used for unfolding the working state characteristic diagram of the optimized fan into a classification characteristic vector; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a full-connection layer of the classifier so as to obtain coded classification characteristic vectors; and a classification result unit, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

In summary, the control system 100 of the bag-type dust collector according to the embodiment of the present application is illustrated, which adopts an artificial intelligence technology based on a deep neural network model to obtain the rotational speeds of motors and the blade angles of fans at a plurality of predetermined time points in a predetermined period of time, arranges the rotational speeds and the blade angles into vectors, extracts relevant feature information through a feature extractor, and further maps low-dimensional information to a high-dimensional feature space through a convolution network after associated encoding, so as to obtain a classification result for indicating whether the rotational speeds of the fan rotational speeds at the current time point should be increased, decreased or unchanged. Therefore, the automatic adjustment of the rotating speed of the fan can be realized, so that the dust removal requirements under different processes and material treatment conditions can be met.

Exemplary method

Fig. 5 is a flowchart of a control method of a bag-type dust collector according to an embodiment of the present application. As shown in fig. 5, a control method of a bag-type dust collector according to an embodiment of the present application includes: s110, acquiring the rotating speeds of motors and the blade angles of fans at a plurality of preset time points in a preset time period; s120, arranging the rotation speeds of the motors at a plurality of preset time points into rotation speed time sequence input vectors according to a time dimension, and then obtaining rotation speed time sequence feature vectors through a rotation speed feature extractor comprising a first convolution and a second convolution; s130, arranging the blade angles of the fans at a plurality of preset time points into angle time sequence input vectors according to a time dimension, and then obtaining angle time sequence feature vectors through an angle feature extractor comprising a first convolution and a second convolution; s140, performing association coding on the rotating speed time sequence feature vector and the angle time sequence feature vector to obtain a fan working state association matrix; s150, the fan working state association matrix is passed through a convolutional neural network model serving as a feature extractor to obtain a fan working state feature map; s160, carrying out characteristic manifold modulation on the fan working state characteristic diagram to obtain an optimized fan working state characteristic diagram; and S170, the optimized fan working state characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the rotating speed of the fan at the current time point is increased, decreased or unchanged.

Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described control method of the bag-type dust collector have been described in detail in the above description of the control system of the bag-type dust collector with reference to fig. 1 to 4, and thus, repetitive descriptions thereof will be omitted.

Exemplary electronic device

Next, an electronic device according to an embodiment of the present application is described with reference to fig. 6. Fig. 6 is a block diagram of an electronic device according to an embodiment of the application. As shown in fig. 6, the electronic device 10 includes one or more processors 11 and a memory 12.

The processor 11 may be a central processing module (CPU) or other form of processing module having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, random Access Memory (RAM) and/or cache memory (cache), and the like. The non-volatile memory may include, for example, read Only Memory (ROM), hard disk, flash memory, and the like. One or more computer program instructions may be stored on the computer readable storage medium that may be executed by the processor 11 to perform the functions in the control system and method of a bag house dust collector of the various embodiments of the present application described above and/or other desired functions. Various contents such as the rotational speed of the motor and the blade angle of the blower at a plurality of predetermined time points for a predetermined period of time may also be stored in the computer-readable storage medium.

In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).

The input means 13 may comprise, for example, a keyboard, a mouse, etc.

The output device 14 may output various information including the classification result and the like to the outside. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.

Of course, only some of the components of the electronic device 10 that are relevant to the present application are shown in fig. 6 for simplicity, components such as buses, input/output interfaces, etc. are omitted. In addition, the electronic device 10 may include any other suitable components depending on the particular application.

Exemplary computer program product and computer readable storage Medium

In addition to the methods and apparatus described above, embodiments of the application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform steps in the functions of the control method of a bag-type dust collector according to the various embodiments of the application described in the "exemplary methods" section of this specification.

The computer program product may write program code for performing operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present application may also be a computer-readable storage medium, on which computer program instructions are stored, which, when being executed by a processor, cause the processor to perform steps in the functions of the method of controlling a bag-type dust collector according to the various embodiments of the present application described in the "exemplary method" section of the description above.

The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.

The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.

It is also noted that in the apparatus, devices and systems of the present application, components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.

Claims

1. A control system for a bag-type dust collector, comprising:

2. The control system of a bag-type dust collector according to claim 1, wherein the rotational speed feature extraction module comprises:

a first scale rotational speed feature extraction unit, configured to input the rotational speed time sequence input vector into a first convolution layer of the rotational speed feature extractor including a first convolution and a second convolution to obtain a first scale rotational speed time sequence feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length;

a second scale rotational speed feature extraction unit configured to input the rotational speed time sequence input vector into a second convolution layer including a rotational speed feature extractor of a first convolution and a second convolution to obtain a second scale rotational speed time sequence feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length;

And the rotating speed characteristic fusion unit is used for cascading the first scale rotating speed time sequence characteristic vector and the second scale rotating speed time sequence characteristic vector by using the cascading layer of the rotating speed characteristic extractor comprising the first convolution and the second convolution so as to obtain the rotating speed time sequence characteristic vector.

3. The control system of a bag house dust collector of claim 2, wherein the blade angle feature extraction module comprises:

a first scale angle feature extraction unit, configured to input the angle timing input vector into a first convolution layer of the angle feature extractor including a first convolution and a second convolution to obtain a first scale angle timing feature vector, where the first convolution layer has a first one-dimensional convolution kernel with a first length;

a second scale angle feature extraction unit configured to input the angle timing input vector into a second convolution layer including the angle feature extractor of the first convolution and a second convolution to obtain a second scale angle timing feature vector, where the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length;

and the angle characteristic fusion unit is used for cascading the first scale angle time sequence characteristic vector and the second scale angle time sequence characteristic vector by using the cascading layer comprising the angle characteristic extractor of the first convolution and the second convolution so as to obtain the angle time sequence characteristic vector.

4. A control system for a bag house dust collector according to claim 3, wherein the feature relating module is configured to:

and respectively calculating products between the transpose vector of the rotating speed time sequence feature vector and the angle time sequence feature vector to obtain the fan working state association matrix.

5. The control system of a bag-type dust collector according to claim 4, wherein the status feature enhancement module is configured to:

each layer of the convolutional neural network model using the feature extractor performs, in forward transfer of the layer, input data:

carrying out convolution processing on input data to obtain a convolution characteristic diagram;

pooling the convolution feature images based on the local feature matrix to obtain pooled feature images;

non-linear activation is carried out on the pooled feature map so as to obtain an activated feature map;

the output of the last layer of the convolutional neural network model serving as the feature extractor is the fan working state feature map, and the input of the first layer of the convolutional neural network model serving as the feature extractor is the fan working state association matrix.

6. The control system of a bag house dust collector of claim 5, wherein the optimization module comprises:

The characteristic value mapping unit is used for inputting the fan working state characteristic diagram into a Sigmoid activation function to map the characteristic values of all positions of the fan working state characteristic diagram into a probability space so as to obtain a probabilistic fan working state characteristic diagram;

the feature matrix mapping unit is used for mapping each feature matrix of the probabilistic fan working state feature diagram along the channel dimension into the same target space respectively to obtain a plurality of feature matrices after the radiation mapping;

the mutual information calculation unit is used for calculating mutual information between any two of the plurality of characteristic matrixes after the similar radiation mapping to obtain a similar affine transformation characteristic vector consisting of a plurality of mutual information;

the weight generation unit is used for generating the affine-like transformation feature vector through a weight generation network comprising a point convolution layer and a batch normalization processing layer so as to obtain an affine-like associated probability feature vector;

and the weighting unit is used for respectively weighting each feature matrix of the fan working state feature diagram along the channel dimension by taking the feature value of each position in the affine-associated probability feature vector as a weight value so as to obtain an optimized fan working state feature diagram.

7. The control system of a bag house dust collector of claim 6, wherein the fan speed adjustment module comprises:

the unfolding unit is used for unfolding the working state characteristic diagram of the optimized fan into a classification characteristic vector;

the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a full-connection layer of the classifier so as to obtain coded classification characteristic vectors;

and the classification result unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

8. The control method of the bag-type dust collector is characterized by comprising the following steps:

9. The method for controlling a bag-type dust collector according to claim 8, wherein the steps of arranging the rotational speeds of the motors at the plurality of predetermined time points in a time dimension into rotational speed time sequence input vectors and then passing through a rotational speed feature extractor including a first convolution and a second convolution to obtain rotational speed time sequence feature vectors, include:

inputting the rotational speed time sequence input vector into a first convolution layer of the rotational speed feature extractor comprising a first convolution and a second convolution to obtain a first scale rotational speed time sequence feature vector, wherein the first convolution layer has a first one-dimensional convolution kernel with a first length;

Inputting the rotational speed time sequence input vector into a second convolution layer comprising a rotational speed feature extractor of a first convolution and a second convolution to obtain a second scale rotational speed time sequence feature vector, wherein the second convolution layer has a second one-dimensional convolution kernel of a second length, and the first length is different from the second length;

and cascading the first scale rotational speed time sequence feature vector and the second scale rotational speed time sequence feature vector by using the cascading layer comprising the rotational speed feature extractor of the first convolution and the second convolution to obtain the rotational speed time sequence feature vector.

10. The method for controlling a bag-type dust collector according to claim 9, wherein performing association coding on the rotational speed time sequence feature vector and the angle time sequence feature vector to obtain a fan working state association matrix comprises: