Dust removal control system and method for quartz sand production

Abstract

The present disclosure relates to a dust removal control system and method for quartz sand production. The method includes: when using a cyclone dust collector to remove flue gas dust to remove solid particles in the flue gas, a sensor group deployed at the inlet of the cyclone dust collector is used to monitor and collect multi-parameter data of the flue gas in real time, including flue gas temperature value, flue gas humidity value and flue gas particle concentration value, and introduce a data processing and analysis algorithm based on artificial intelligence and deep learning technology at the back end to perform time series collaborative correlation and interaction analysis on these flue gas multi-parameter data, so as to learn and characterize the time series dynamic interaction between flue gas temperature, humidity and particle concentration, so as to adaptively adjust the rotation speed of the cyclone dust collector based on the actual state of the flue gas. In this way, a more intelligent dust removal control for quartz sand production can be achieved, avoiding the problems caused by fixed speed or gear adjustment control mode, so that the dust removal control system can better adapt to changes in flue gas conditions.

Description

Technical Field

[0001] The present disclosure relates to the technical field of quartz sand production, and in particular, to a dust removal control system and method for quartz sand production. Background Art

[0002] In the production process of quartz sand, drying is a key step, which not only affects the quality and output of quartz sand, but also involves environmental protection and energy efficiency issues. Chinese patent CN100453939C discloses a rotary drum quartz sand drying process, which can mix wet quartz sand with high-temperature flue gas in a dryer and then dry it, and then the mixture of flue gas at about 150°C and a large amount of water vapor generated after drying is purified by a cyclone dust collector and a bag dust collector, and then led out by an induced draft fan and discharged into the atmosphere through a chimney.

[0003] In the above-mentioned rotary cylinder quartz sand drying process, although the introduction of cyclone dust collector and bag dust collector can dedust and purify the high-temperature flue gas, specifically, the cyclone dust collector can preliminarily separate the larger solid particles in the flue gas, and the bag dust collector can further purify the flue gas to ensure that the emission standards are met. However, in the cyclone dust collector, a fixed speed or gear adjustment method is usually used to separate and remove larger solid particles in the flue gas. If a fixed speed is used to separate solid particles in the flue gas, it may not be able to adapt to changes in the concentration of particles in the flue gas. When the concentration of particulate matter in the flue gas is high, the fixed speed may not be enough to effectively separate all particles, resulting in an excessive content of solid particles in the discharged flue gas. In addition, the removal effect of particulate matter is also related to the temperature and humidity of the flue gas. High temperature and high humidity flue gas may cause particles to adhere more easily to the inner wall of the cyclone dust collector, reducing the dust removal efficiency.

[0004] Therefore, an optimized dust removal control system for quartz sand production is desired. Summary of the invention

[0005] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present disclosure provides a dust removal control system for quartz sand production, the system comprising: a flue gas multi-parameter data acquisition module, for acquiring a time series of flue gas multi-parameter data collected by a sensor group deployed at an inlet of a cyclone dust collector, wherein the flue gas multi-parameter data comprises a flue gas temperature value, a flue gas humidity value and a flue gas particulate matter concentration value; a flue gas multi-parameter data time series regularization module, for regularizing the time series of the flue gas multi-parameter data according to a time dimension to obtain a flue gas temperature time series vector, a flue gas humidity time series vector and a flue gas particulate matter concentration time series vector; a flue gas parameter time series feature extraction module, for performing time series regularization on the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series vector, respectively. A smoke multi-parameter time series feature enhancement fusion module is used to obtain the smoke temperature time series correlation feature vector, the smoke humidity time series correlation feature vector and the smoke particulate matter concentration time series correlation feature vector through a smoke multi-parameter time series feature energy measurement enhancement module based on the Transformer structure to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector; a cyclone dust collector speed control module is used to determine the control result of the speed of the cyclone dust collector based on the enhanced smoke multi-parameter time series semantic fusion feature vector, and the control result is used to indicate whether the speed of the cyclone dust collector should increase, decrease or remain unchanged.

[0007] Optionally, the flue gas parameter time series feature extraction module is used to: pass the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series vector through a flue gas parameter time series feature extractor based on a time convolution network to obtain the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector.

[0008] Optionally, the smoke multi-parameter time series feature enhanced fusion module includes: a vector combination unit, which is used to combine the smoke temperature time series correlation feature vector, the smoke humidity time series correlation feature vector and the smoke particulate matter concentration time series correlation feature vector into a set of smoke multi-parameter time series correlation features, and calculate the characteristic energy level coefficient of each feature vector in the set of smoke multi-parameter time series correlation features to obtain the smoke temperature time series feature energy level coefficient, the smoke humidity time series feature energy level coefficient and the smoke particulate matter concentration time series feature energy level coefficient; a probabilistic mapping unit, which is used to probabilistically map the smoke temperature time series feature energy level coefficient, the smoke humidity time series feature energy level coefficient and the smoke particulate matter concentration time series feature energy level coefficient respectively to obtain the smoke temperature time series feature energy level weight, the smoke humidity time series feature energy level weight and the smoke particulate matter concentration time series feature energy level weight; a first weight multiplication calculation unit, which is used to combine the The smoke temperature time series correlation feature vector is multiplied by the smoke temperature time series feature energy level weight to obtain the smoke temperature time series energy level modulation feature vector; the second weight multiplication calculation unit is used to multiply the smoke humidity time series correlation feature vector by the smoke humidity time series feature energy level weight to obtain the smoke humidity time series energy level modulation feature vector; the third weight multiplication calculation unit is used to multiply the smoke particulate matter concentration time series correlation feature vector by the smoke particulate matter concentration time series feature energy level weight to obtain the smoke particulate matter concentration time series energy level modulation feature vector; the matrix calculation unit is used to multiply the transposed vector of the smoke temperature time series energy level modulation feature vector by the smoke humidity time series energy level modulation feature vector and then divide it by the square root of the length of the smoke particulate matter concentration time series correlation feature vector by position to obtain the smoke temperature-humidity time series correlation weight matrix; the normalization unit is used to use The function normalizes the characteristic values ​​of each position in the flue gas temperature-humidity time series association weight matrix to obtain the activated flue gas temperature-humidity time series association weight matrix; the matrix multiplication calculation unit is used to calculate the matrix multiplication between the activated flue gas temperature-humidity time series association weight matrix and the transposed vector of the flue gas particulate matter concentration time series energy level modulation feature vector to obtain the enhanced flue gas multi-parameter time series semantic fusion feature vector.

[0009] Optionally, the vector combination unit is used to: for the set of the flue gas multi-parameter time series correlation features, calculate the characteristic maximum value of the flue gas temperature time series correlation feature vector in the set of the flue gas multi-parameter time series correlation features to obtain the maximum value of the flue gas temperature time series feature; respectively calculate the mean and variance of the flue gas temperature time series correlation feature vector to obtain the mean of the flue gas temperature time series feature and the variance of the flue gas temperature time series feature; multiply the value obtained by adding the flue gas temperature time series feature variance and the preset hyperparameter by four to obtain the flue gas temperature time series feature. Characteristic first energy level factor; calculate the square of the difference between the maximum value of the flue gas temperature time series characteristic and the mean value of the flue gas temperature time series characteristic to obtain the flue gas temperature time series characteristic difference value; multiply the flue gas temperature time series characteristic variance by two, and then add it with the preset hyperparameter and the flue gas temperature time series characteristic difference value multiple times to obtain the second energy level factor of the flue gas temperature time series characteristic; divide the first energy level factor of the flue gas temperature time series characteristic by the second energy level factor of the flue gas temperature time series characteristic to obtain the flue gas temperature time series characteristic energy level coefficient.

[0010] Optionally, the probabilistic mapping unit is used to: The function normalizes the flue gas temperature time series characteristic energy level coefficient, the flue gas humidity time series characteristic energy level coefficient and the flue gas particulate matter concentration time series characteristic energy level coefficient respectively to obtain the flue gas temperature time series characteristic energy level weight, the flue gas humidity time series characteristic energy level weight and the flue gas particulate matter concentration time series characteristic energy level weight.

[0011] Optionally, the cyclone dust collector speed control module is used to: pass the enhanced flue gas multi-parameter time-series semantic fusion feature vector through a classifier-based cyclone dust collector speed controller to obtain a control result, and the control result is used to indicate whether the cyclone dust collector speed should increase, decrease or remain unchanged.

[0012] Optionally, the cyclone dust collector speed control module includes: a fully connected encoding unit, used to use multiple fully connected layers of the classifier-based cyclone dust collector speed controller to fully connect encode the enhanced flue gas multi-parameter temporal semantic fusion feature vector to obtain an encoded classification feature vector; a classification unit, used to pass the encoded classification feature vector through the Softmax classification function of the classifier-based cyclone dust collector speed controller to obtain the control result.

[0013] In a second aspect, the present disclosure provides a dust removal control method for quartz sand production, the method comprising: obtaining a time series of flue gas multi-parameter data collected by a sensor group deployed at the inlet of a cyclone dust collector, wherein the flue gas multi-parameter data comprises a flue gas temperature value, a flue gas humidity value and a flue gas particulate matter concentration value; performing data regularization on the time series of the flue gas multi-parameter data according to the time dimension to obtain a flue gas temperature time series vector, a flue gas humidity time series vector and a flue gas particulate matter concentration time series vector; performing time series correlation feature extraction on the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series vector respectively to obtain to the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector; the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector are subjected to the flue gas multi-parameter time series feature energy measurement enhancement module based on the Transformer structure to obtain an enhanced flue gas multi-parameter time series semantic fusion feature vector; based on the enhanced flue gas multi-parameter time series semantic fusion feature vector, the control result of the rotation speed of the cyclone dust collector is determined, and the control result is used to indicate whether the rotation speed of the cyclone dust collector should increase, decrease or remain unchanged.

[0014] Optionally, time series correlation feature extraction is performed on the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series vector respectively to obtain the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector, including: passing the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series correlation feature vector through a flue gas parameter time series feature extractor based on a time convolutional network to obtain the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector.

[0015] Optionally, the smoke temperature time series association feature vector, the smoke humidity time series association feature vector and the smoke particulate matter concentration time series association feature vector are subjected to a smoke multi-parameter time series feature energy measurement enhancement module based on a Transformer structure to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector, including: forming a set of smoke multi-parameter time series association features with the smoke temperature time series association feature vector, the smoke humidity time series association feature vector and the smoke particulate matter concentration time series association feature vector, and calculating the characteristic energy level coefficient of each feature vector in the set of smoke multi-parameter time series association features to obtain the smoke temperature time series feature energy level coefficient, the smoke humidity time series feature energy level coefficient and the smoke particulate matter concentration time series feature energy level coefficient; respectively probabilizing the smoke temperature time series feature energy level coefficient, the smoke humidity time series feature energy level coefficient and the smoke particulate matter concentration time series feature energy level coefficient Mapping to obtain the flue gas temperature time series characteristic energy level weight, the flue gas humidity time series characteristic energy level weight and the flue gas particulate matter concentration time series characteristic energy level weight; multiplying the flue gas temperature time series associated characteristic vector with the flue gas temperature time series characteristic energy level weight to obtain the flue gas temperature time series energy level modulation characteristic vector; multiplying the flue gas humidity time series associated characteristic vector with the flue gas humidity time series characteristic energy level weight to obtain the flue gas humidity time series energy level modulation characteristic vector; multiplying the flue gas particulate matter concentration time series associated characteristic vector with the flue gas particulate matter concentration time series characteristic energy level weight to obtain the flue gas particulate matter concentration time series energy level modulation characteristic vector; multiplying the transposed vector of the flue gas temperature time series energy level modulation characteristic vector with the flue gas humidity time series energy level modulation characteristic vector and then dividing it by the square root of the length of the flue gas particulate matter concentration time series associated characteristic vector by position to obtain the flue gas temperature-humidity time series associated weight matrix; using The function normalizes the characteristic values ​​of each position in the flue gas temperature-humidity time series association weight matrix to obtain the activated flue gas temperature-humidity time series association weight matrix; calculates the matrix multiplication between the activated flue gas temperature-humidity time series association weight matrix and the transposed vector of the flue gas particulate matter concentration time series energy level modulation feature vector to obtain the enhanced flue gas multi-parameter time series semantic fusion feature vector.

[0016] By adopting the above technical solution, when using a cyclone dust collector to remove solid particles in the flue gas, the multi-parameter data of the flue gas, including the flue gas temperature value, the flue gas humidity value and the flue gas particle concentration value, are monitored and collected in real time through a sensor group deployed at the entrance of the cyclone dust collector, and a data processing and analysis algorithm based on artificial intelligence and deep learning technology is introduced at the back end to perform time series collaborative correlation and interaction analysis on these flue gas multi-parameter data, so as to learn and characterize the time series dynamic interaction between the flue gas temperature, humidity and particle concentration, so as to adjust the speed of the cyclone dust collector adaptively based on the actual state of the flue gas. In this way, a more intelligent dust removal control for quartz sand production can be achieved to avoid the problems caused by the fixed speed or gear adjustment control method, and enable the dust removal control system to better adapt to changes in flue gas conditions.

[0017] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0019] In the attached picture: Figure 1 The invention is a block diagram of a dust removal control system for quartz sand production according to an exemplary embodiment.

[0020] Figure 2 The present invention is a flow chart showing a dust removal control method for quartz sand production according to an exemplary embodiment.

[0021] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment.

[0022] Figure 4 It is an application scenario diagram of a dust removal control system for quartz sand production according to an exemplary embodiment. DETAILED DESCRIPTION

[0023] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0024] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0025] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0026] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0027] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0028] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0029] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0030] Figure 1 is a block diagram of a dust removal control system for quartz sand production according to an exemplary embodiment. Figure 1 As shown, the system 100 includes: a flue gas multi-parameter data acquisition module 101, which is used to obtain a time series of flue gas multi-parameter data collected by a sensor group deployed at the inlet of the cyclone dust collector, wherein the flue gas multi-parameter data includes a flue gas temperature value, a flue gas humidity value and a flue gas particulate matter concentration value.

[0031] The flue gas multi-parameter data time series regularization module 102 is used to regularize the time series of the flue gas multi-parameter data according to the time dimension to obtain a flue gas temperature time series vector, a flue gas humidity time series vector and a flue gas particulate matter concentration time series vector.

[0032] The smoke parameter time series feature extraction module 103 is used to extract time series correlation features from the smoke temperature time series vector, the smoke humidity time series vector and the smoke particulate matter concentration time series vector to obtain a smoke temperature time series correlation feature vector, a smoke humidity time series correlation feature vector and a smoke particulate matter concentration time series correlation feature vector.

[0033] The smoke multi-parameter time series feature enhancement fusion module 104 is used to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector by passing the smoke temperature time series associated feature vector, the smoke humidity time series associated feature vector and the smoke particulate matter concentration time series associated feature vector through a smoke multi-parameter time series feature energy measurement enhancement module based on a Transformer structure.

[0034] The cyclone dust collector speed control module 105 is used to determine the control result of the speed of the cyclone dust collector based on the enhanced flue gas multi-parameter time series semantic fusion feature vector, and the control result is used to indicate whether the speed of the cyclone dust collector should increase, decrease or remain unchanged.

[0035] In view of the above technical problems, in the technical solution of the present application, a dust removal control system for quartz sand production is proposed, which can monitor and collect multi-parameter data of flue gas in real time through a sensor group deployed at the inlet of the cyclone dust collector when using a cyclone dust collector to remove solid particles in the flue gas, including flue gas temperature value, flue gas humidity value and flue gas particle concentration value, and introduce a data processing and analysis algorithm based on artificial intelligence and deep learning technology at the back end to perform time series collaborative correlation and interaction analysis on these flue gas multi-parameter data, so as to learn and characterize the time series dynamic interaction between flue gas temperature, humidity and particle concentration, so as to adaptively adjust the speed of the cyclone dust collector based on the actual state of the flue gas. In this way, a more intelligent dust removal control for quartz sand production can be achieved to avoid the problems caused by fixed speed or gear adjustment control methods, and enable the dust removal control system to better adapt to changes in flue gas conditions.

[0036] Specifically, in the technical solution of the present application, first, the time series of flue gas multi-parameter data collected by the sensor group deployed at the inlet of the cyclone dust collector is obtained, wherein the flue gas multi-parameter data includes flue gas temperature value, flue gas humidity value and flue gas particulate matter concentration value. Next, considering that the flue gas temperature, flue gas humidity and flue gas particulate matter concentration have a time series correlation relationship in the time dimension, that is, the multi-parameter data of the flue gas has a dynamic change regularity of the time series. Therefore, in order to be able to analyze the time series dynamic change characteristics of these flue gas parameter data and capture the time series correlation relationship and mutual influence between these flue gas multi-parameter data, in the technical solution of the present application, it is necessary to regularize the time series of the flue gas multi-parameter data according to the time dimension to obtain the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series vector.

[0037] Then, the smoke temperature time series vector, the smoke humidity time series vector and the smoke particulate matter concentration time series vector are passed through a smoke parameter time series feature extractor based on a time convolutional network to obtain a smoke temperature time series correlation feature vector, a smoke humidity time series correlation feature vector and a smoke particulate matter concentration time series correlation feature vector. It should be understood that a time convolutional network is a neural network architecture for processing sequence data. It captures local dependencies in a time series through a convolutional layer and can solve the problem of gradient vanishing or gradient exploding encountered by recurrent neural networks (such as LSTM and GRU) when processing long sequences. Therefore, by applying the smoke parameter time series feature extractor based on a time convolutional network, the time series dynamic characteristics and time series correlation of smoke temperature, smoke humidity and smoke particulate matter concentration in the time dimension can be learned and captured respectively.

[0038] In one embodiment of the present application, the flue gas parameter time series feature extraction module is used to: pass the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series vector through a flue gas parameter time series feature extractor based on a time convolution network to obtain the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector.

[0039] It should be understood that since the flue gas temperature, flue gas humidity and flue gas particle concentration are mutually influential and dynamically related in time series, the correlation and implicit role between these flue gas parameter time series characteristics are important factors affecting the dust removal effect and subsequent speed control. In addition, it is also considered that the time series characteristics of flue gas temperature, flue gas humidity and flue gas particle concentration in the time dimension have different importance and contribution to the subsequent speed control task and dust removal effect. Based on this, in order to capture and characterize the dynamic relationship and interaction between these flue gas parameter time series characteristics, so as to more comprehensively reflect the state of the flue gas, so as to perform the corresponding cyclone dust collector speed control, in the technical solution of the present application, the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particle concentration time series correlation feature vector are further enhanced through the flue gas multi-parameter time series feature energy measurement enhancement module based on the Transformer structure to obtain an enhanced flue gas multi-parameter time series semantic fusion feature vector. In particular, through the processing of the flue gas multi-parameter time series feature energy measurement enhancement module based on the Transformer structure, the self-attention mechanism in the Transformer structure can be used to effectively capture the contextual association relationship and hidden interaction between the flue gas temperature time series features, the flue gas humidity time series features and the flue gas particle concentration time series features, so as to give different importance to different flue gas parameter time series features in the subsequent feature fusion process, thereby improving the system's attention to important flue gas state features in subsequent speed control tasks. In addition, by calculating the energy level coefficients of different flue gas parameter time series features in the self-attention mechanism, the energy distribution of each flue gas parameter time series feature can be quantified, and this energy distribution can be used to infer the semantic information content density of each flue gas parameter time series feature vector itself and its influence on other flue gas parameter time series features, thereby applying different attention in subsequent tasks to improve the generalization ability of the model and improve the performance of the dust removal control system.

[0040] In one embodiment of the present application, the flue gas multi-parameter time series feature enhanced fusion module includes: a vector combination unit, which is used to form a set of flue gas multi-parameter time series correlation features by combining the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector, and calculate the characteristic energy level coefficient of each characteristic vector in the set of flue gas multi-parameter time series correlation features to obtain the flue gas temperature time series feature energy level coefficient, the flue gas humidity time series feature energy level coefficient and the flue gas particulate matter concentration time series feature energy level coefficient; a probabilistic mapping unit, which is used to probabilistically map the flue gas temperature time series feature energy level coefficient, the flue gas humidity time series feature energy level coefficient and the flue gas particulate matter concentration time series feature energy level coefficient respectively to obtain the flue gas temperature time series feature energy level weight, the flue gas humidity time series feature energy level weight and the flue gas particulate matter concentration time series feature energy level weight; a first weight multiplication calculation unit, A second weight multiplication calculation unit is used to multiply the smoke temperature time series correlation feature vector with the smoke temperature time series feature energy level weight to obtain the smoke temperature time series energy level modulation feature vector; a second weight multiplication calculation unit is used to multiply the smoke humidity time series correlation feature vector with the smoke humidity time series feature energy level weight to obtain the smoke humidity time series energy level modulation feature vector; a third weight multiplication calculation unit is used to multiply the smoke particulate matter concentration time series correlation feature vector with the smoke particulate matter concentration time series feature energy level weight to obtain the smoke particulate matter concentration time series energy level modulation feature vector; a matrix calculation unit is used to multiply the transposed vector of the smoke temperature time series energy level modulation feature vector with the smoke humidity time series energy level modulation feature vector and then divide it by the square root of the length of the smoke particulate matter concentration time series correlation feature vector by position to obtain the smoke temperature-humidity time series correlation weight matrix; a normalization unit is used to use The function normalizes the characteristic values ​​of each position in the flue gas temperature-humidity time series association weight matrix to obtain the activated flue gas temperature-humidity time series association weight matrix; the matrix multiplication calculation unit is used to calculate the matrix multiplication between the activated flue gas temperature-humidity time series association weight matrix and the transposed vector of the flue gas particulate matter concentration time series energy level modulation feature vector to obtain the enhanced flue gas multi-parameter time series semantic fusion feature vector.

[0041] Specifically, first, the characteristic energy level coefficients of the smoke temperature time series associated feature vector, the smoke humidity time series associated feature vector, and the smoke particulate matter concentration time series associated feature vector are calculated respectively, which can be regarded as a measure of the importance or energy of each feature in the overall time series data, so as to quantify the energy distribution of each smoke parameter time series feature. This step helps to identify which smoke parameter time series features have a greater impact on smoke state detection and subsequent speed control tasks. In the technical solution of the present application, the characteristic energy level coefficient is related to the scale of the time series feature vector of each smoke parameter and the statistical characteristics of the characteristic distribution, which is used to represent the semantic information content density of each smoke parameter time series feature vector itself and its influence on other smoke parameter time series features.

[0042] Then, the characteristic energy level coefficients of different smoke parameter time series features are used to generate time series feature energy level weights after probabilistic mapping with the sigmoid function, so that each smoke parameter time series feature vector is multiplied by its corresponding energy level weight during energy level modulation to obtain the smoke parameter time series energy level modulation feature vector. This step enhances the influence of important smoke parameter time series features in the subsequent feature fusion process, while suppressing the semantic expression of less important features.

[0043] When using the self-attention mechanism for feature enhancement expression, the smoke temperature-humidity time series correlation weight matrix is ​​constructed by calculating the correlation weight between the transposed vector of the smoke temperature time series energy level modulation feature vector and the smoke humidity time series energy level modulation feature vector. This helps capture the interaction between smoke temperature and humidity features, such as how temperature and humidity jointly affect the removal efficiency of particulate matter. After the function normalizes and activates the interaction, it performs weighted optimization with the temporal energy level modulation characteristics of the smoke particle concentration, so as to self-attentively fuse the temporal characteristics of different smoke parameters and their mutual influence, thereby improving the generalization ability of the model by dynamically adjusting the feature weights and strengthening the feature representation, so that the model can flexibly cope with the diversity and complexity of the temporal changes of different smoke parameters and the correlation relationships.

[0044] In one embodiment of the present application, the vector combination unit is used to: for the set of the flue gas multi-parameter time series correlation features, calculate the characteristic maximum value of the flue gas temperature time series correlation feature vector in the set of the flue gas multi-parameter time series correlation features to obtain the maximum value of the flue gas temperature time series feature; respectively calculate the mean and variance of the flue gas temperature time series correlation feature vector to obtain the mean of the flue gas temperature time series feature and the variance of the flue gas temperature time series feature; multiply the value obtained by adding the flue gas temperature time series feature variance and the preset hyperparameter by four to obtain the flue gas temperature time series feature. The first energy level factor of the temperature time series characteristic; calculating the square of the difference between the maximum value of the flue gas temperature time series characteristic and the mean value of the flue gas temperature time series characteristic to obtain the difference value of the flue gas temperature time series characteristic; multiplying the variance of the flue gas temperature time series characteristic by two, and then adding it with the doubled preset hyperparameter and the difference value of the flue gas temperature time series characteristic to obtain the second energy level factor of the flue gas temperature time series characteristic; dividing the first energy level factor of the flue gas temperature time series characteristic by the second energy level factor of the flue gas temperature time series characteristic to obtain the energy level coefficient of the flue gas temperature time series characteristic.

[0045] Furthermore, in one embodiment of the present application, the probabilistic mapping unit is used to: The function normalizes the flue gas temperature time series characteristic energy level coefficient, the flue gas humidity time series characteristic energy level coefficient and the flue gas particulate matter concentration time series characteristic energy level coefficient respectively to obtain the flue gas temperature time series characteristic energy level weight, the flue gas humidity time series characteristic energy level weight and the flue gas particulate matter concentration time series characteristic energy level weight.

[0046] Specifically, the smoke temperature time series associated feature vector, the smoke humidity time series associated feature vector and the smoke particulate matter concentration time series associated feature vector are processed by the smoke multi-parameter time series feature energy measurement enhancement module based on the Transformer structure with the following feature enhancement aggregation formula to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector; wherein the feature enhancement aggregation formula is: ;in, is a set of flue gas multi-parameter time series correlation features consisting of the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector, is the flue gas temperature time series correlation feature vector, is the flue gas humidity time series correlation feature vector, is the time series correlation feature vector of the flue gas particulate matter concentration, and are respectively the first The mean and variance of the eigenvectors, For the said The maximum value of the feature in the eigenvectors, is a preset hyperparameter, It is the The characteristic energy level coefficients of the eigenvectors, yes function, It is the The characteristic energy level weight of the characteristic vector, yes function, is the length of the time series correlation characteristic vector of the flue gas particulate matter concentration, is the enhanced smoke multi-parameter temporal semantic fusion feature vector, Represents the transposed vector of a vector.

[0047] Furthermore, the enhanced flue gas multi-parameter time-series semantic fusion feature vector is passed through a cyclone dust collector speed controller based on a classifier to obtain a control result, and the control result is used to indicate whether the speed of the cyclone dust collector should increase, decrease, or remain unchanged. In other words, the semantic fusion information between the multi-parameter time-series features of flue gas temperature, humidity, and particulate matter concentration is used for classification processing, so as to adaptively adjust the speed of the cyclone dust collector based on the actual state of the flue gas. In this way, a more intelligent dust removal control for quartz sand production can be achieved to avoid the problems caused by fixed speed or gear adjustment control methods, and enable the dust removal control system to better adapt to changes in flue gas conditions.

[0048] In one embodiment of the present application, the cyclone dust collector speed control module is used to: pass the enhanced flue gas multi-parameter time-series semantic fusion feature vector through a classifier-based cyclone dust collector speed controller to obtain a control result, and the control result is used to indicate whether the speed of the cyclone dust collector should increase, decrease, or remain unchanged.

[0049] Furthermore, in one embodiment of the present application, the cyclone dust collector speed control module includes: a fully connected encoding unit, used to use multiple fully connected layers of the classifier-based cyclone dust collector speed controller to fully connect encode the enhanced flue gas multi-parameter temporal semantic fusion feature vector to obtain an encoded classification feature vector; a classification unit, used to pass the encoded classification feature vector through the Softmax classification function of the classifier-based cyclone dust collector speed controller to obtain the control result.

[0050] The present application takes into account that the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector respectively express the local time series correlation characteristics of the flue gas temperature value, the flue gas humidity value and the flue gas particulate matter concentration value. In this way, when the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector are passed through a flue gas multi-parameter time series feature energy measurement enhancement module based on a Transformer structure, the enhanced flue gas multi-parameter time series semantic fusion feature vector obtained will also cause insufficient multi-parameter semantic context association coverage due to the feature distribution energy measurement difference of the source data based on the respective time series feature distributions, thereby resulting in the loss of outlier class reasoning mapping information of the synthetic feature, affecting the expression effect of the enhanced flue gas multi-parameter time series semantic fusion feature vector, and reducing the accuracy of the control result obtained by the classifier-based cyclone dust collector speed controller.

[0051] Based on this, in a preferred example, the enhanced flue gas multi-parameter time series semantic fusion feature vector is passed through a classifier-based cyclone dust collector speed controller to obtain a control result, including: optimizing the enhanced flue gas multi-parameter time series semantic fusion feature vector to obtain an optimized enhanced flue gas multi-parameter time series semantic fusion feature vector; passing the optimized enhanced flue gas multi-parameter time series semantic fusion feature vector through a classifier-based cyclone dust collector speed controller to obtain a control result.

[0052] Specifically, the enhanced smoke multi-parameter temporal semantic fusion feature vector is optimized to obtain an optimized enhanced smoke multi-parameter temporal semantic fusion feature vector, including the following steps: calculating the mean and variance of all eigenvalues ​​of the enhanced smoke multi-parameter temporal semantic fusion feature vector, adding the enhanced smoke multi-parameter temporal semantic fusion feature vector to the mean and the variance respectively, taking the reciprocal of each eigenvalue and further performing point multiplication with the enhanced smoke multi-parameter temporal semantic fusion feature vector to obtain a first intermediate feature vector and a second intermediate feature vector, then transposing the enhanced smoke multi-parameter temporal semantic fusion feature vector to a column vector and performing matrix multiplication with the first intermediate feature vector as a row vector to obtain a first intermediate feature matrix, and transposing the second intermediate feature vector to a column vector and performing matrix multiplication with the enhanced smoke multi-parameter temporal semantic fusion feature vector to obtain a second intermediate feature matrix, and finally performing matrix multiplication on the transpose of the first intermediate feature matrix, the second intermediate feature matrix and the transpose of the enhanced smoke multi-parameter temporal semantic fusion feature vector to obtain an optimized enhanced smoke multi-parameter temporal semantic fusion feature vector.

[0053] The enhanced smoke multi-parameter temporal semantic fusion feature vector is expressed as , then the optimization process is expressed as: ;in, represents the enhanced smoke multi-parameter temporal semantic fusion feature vector, and represent the mean and variance of all eigenvalues ​​of the enhanced smoke multi-parameter temporal semantic fusion feature vector, respectively, It means adding by position. It means point multiplication by position. It means to calculate the inverse of each eigenvalue of the vector. and are the first intermediate eigenvector and the second intermediate eigenvector respectively, represents matrix multiplication, represents the transpose of a vector, Represents the optimized enhanced smoke multi-parameter temporal semantic fusion feature vector.

[0054] That is, in view of the tail feature distribution collapse caused by the loss of tail feature information in the synthetic feature distribution ecology of the enhanced flue gas multi-parameter time series semantic fusion feature vector, the descriptive ecological discovery of the feature set of the enhanced flue gas multi-parameter time series semantic fusion feature vector is realized by the stable diffusion of the distribution probability of the interaction of the global statistical characteristics of the enhanced flue gas multi-parameter time series semantic fusion feature vector, so as to form a hybrid intuitive feature ecology through the variational autoencoder based on the stable diffusion probability distribution, so as to improve the universality of the feature real interactive value acquisition of the enhanced flue gas multi-parameter time series semantic fusion feature vector as a synthetic feature, improve the overall expression effect of the enhanced flue gas multi-parameter time series semantic fusion feature vector, and improve the accuracy of the control result obtained by the optimized enhanced flue gas multi-parameter time series semantic fusion feature vector through the classifier-based cyclone dust collector speed controller. In this way, the speed of the cyclone dust collector can be adjusted more accurately based on the actual state of the flue gas, thereby realizing more intelligent dust removal control in quartz sand production.

[0055] In summary, when the cyclone dust collector is used for flue gas dust removal to remove solid particles in the flue gas, the multi-parameter data of the flue gas, including the flue gas temperature value, the flue gas humidity value and the flue gas particle concentration value, are monitored and collected in real time by the sensor group deployed at the inlet of the cyclone dust collector, and the data processing and analysis algorithm based on artificial intelligence and deep learning technology is introduced at the back end to perform time series collaborative correlation and interaction analysis on these flue gas multi-parameter data, so as to learn and characterize the time series dynamic interaction between the flue gas temperature, humidity and particle concentration, so as to adjust the speed of the cyclone dust collector adaptively based on the actual state of the flue gas. In this way, more intelligent dust removal control of quartz sand production can be achieved to avoid the problems caused by fixed speed or gear adjustment control mode, and make the dust removal control system better adapt to changes in flue gas conditions.

[0056] Figure 2 FIG. 1 is a flow chart showing a dust removal control method for quartz sand production according to an exemplary embodiment. Figure 2 As shown, the method includes: step S201, obtaining a time series of flue gas multi-parameter data collected by a sensor group deployed at the inlet of a cyclone dust collector, wherein the flue gas multi-parameter data includes a flue gas temperature value, a flue gas humidity value and a flue gas particulate matter concentration value.

[0057] Step S202: regularize the time series of the flue gas multi-parameter data according to the time dimension to obtain a flue gas temperature time series vector, a flue gas humidity time series vector and a flue gas particulate matter concentration time series vector.

[0058] Step S203, respectively extracting time series correlation features from the flue gas temperature time series vector, the flue gas humidity time series vector, and the flue gas particulate matter concentration time series vector to obtain a flue gas temperature time series correlation feature vector, a flue gas humidity time series correlation feature vector, and a flue gas particulate matter concentration time series correlation feature vector.

[0059] Step S204: The smoke temperature time series associated feature vector, the smoke humidity time series associated feature vector and the smoke particulate matter concentration time series associated feature vector are passed through a smoke multi-parameter time series feature energy measurement enhancement module based on a Transformer structure to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector.

[0060] Step S205: determining a control result of the rotation speed of the cyclone dust collector based on the enhanced flue gas multi-parameter time series semantic fusion feature vector, wherein the control result is used to indicate whether the rotation speed of the cyclone dust collector should increase, decrease, or remain unchanged.

[0061] In one embodiment of the present application, time series correlation feature extraction is performed on the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series vector to obtain the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector, including: passing the flue gas temperature time series vector, the flue gas humidity time series vector and the flue gas particulate matter concentration time series correlation feature vector through a flue gas parameter time series feature extractor based on a time convolutional network to obtain the flue gas temperature time series correlation feature vector, the flue gas humidity time series correlation feature vector and the flue gas particulate matter concentration time series correlation feature vector.

[0062] In one embodiment of the present application, the smoke temperature time series association feature vector, the smoke humidity time series association feature vector and the smoke particulate matter concentration time series association feature vector are subjected to a smoke multi-parameter time series feature energy measurement enhancement module based on a Transformer structure to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector, including: the smoke temperature time series association feature vector, the smoke humidity time series association feature vector and the smoke particulate matter concentration time series association feature vector are combined into a set of smoke multi-parameter time series association features, and the characteristic energy level coefficient of each feature vector in the set of smoke multi-parameter time series association features is calculated to obtain the smoke temperature time series feature energy level coefficient, the smoke humidity time series feature energy level coefficient and the smoke particulate matter concentration time series feature energy level coefficient; the smoke temperature time series feature energy level coefficient, the smoke humidity time series feature energy level coefficient and the smoke particulate matter concentration time series feature energy level coefficient are respectively Probabilistic mapping is performed to obtain the time series characteristic energy level weight of the smoke temperature, the time series characteristic energy level weight of the smoke humidity and the time series characteristic energy level weight of the smoke particulate matter concentration; the smoke temperature time series associated feature vector is multiplied by the smoke temperature time series characteristic energy level weight to obtain the smoke temperature time series energy level modulation feature vector; the smoke humidity time series associated feature vector is multiplied by the smoke humidity time series characteristic energy level weight to obtain the smoke humidity time series energy level modulation feature vector; the smoke particulate matter concentration time series associated feature vector is multiplied by the smoke particulate matter concentration time series characteristic energy level weight to obtain the smoke particulate matter concentration time series energy level modulation feature vector; the transposed vector of the smoke temperature time series energy level modulation feature vector is multiplied by the smoke humidity time series energy level modulation feature vector and then divided by the square root of the length of the smoke particulate matter concentration time series associated feature vector by position to obtain the smoke temperature-humidity time series associated weight matrix; and the smoke humidity time series associated feature vector is multiplied by the transposed vector of the smoke temperature time series energy level modulation feature vector and the smoke humidity time series energy level modulation feature vector. The function normalizes the characteristic values ​​of each position in the flue gas temperature-humidity time series association weight matrix to obtain the activated flue gas temperature-humidity time series association weight matrix; calculates the matrix multiplication between the activated flue gas temperature-humidity time series association weight matrix and the transposed vector of the flue gas particulate matter concentration time series energy level modulation feature vector to obtain the enhanced flue gas multi-parameter time series semantic fusion feature vector.

[0063] Reference below Figure 3 , which shows a schematic diagram of the structure of an electronic device 600 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0064] like Figure 3 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0065] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0066] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0067] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0068] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0069] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0070] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0071] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0072] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of a module does not limit the module itself in some cases. For example, a test parameter acquisition module may also be described as a "module for acquiring device test parameters corresponding to a target device".

[0073] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0074] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0075] Figure 4 FIG. 1 is an application scenario diagram of a dust removal control system for quartz sand production according to an exemplary embodiment. Figure 4 As shown, in this application scenario, first, a time series of flue gas multi-parameter data collected by a sensor group deployed at the inlet of a cyclone dust collector is obtained, wherein the flue gas multi-parameter data includes a flue gas temperature value (for example, Figure 4 C1 as indicated in the figure), smoke humidity value (e.g. Figure 4 C2) and flue gas particulate matter concentration values ​​(e.g. Figure 4 Then, the obtained flue gas temperature value, flue gas humidity value and flue gas particulate matter concentration value are input to a server (for example, Figure 4 S) as shown in , wherein the server is capable of processing the flue gas temperature value, the flue gas humidity value and the flue gas particulate matter concentration value based on the dust removal control for quartz sand production to determine the control result of the rotation speed of the cyclone dust collector, and the control result is used to indicate whether the rotation speed of the cyclone dust collector should increase, decrease or remain unchanged.

[0076] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0077] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0078] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims. Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Claims

1. A dust removal control system for quartz sand production, characterized in that: include: A flue gas multi-parameter data acquisition module, used to obtain a time series of flue gas multi-parameter data collected by a sensor group deployed at the inlet of the cyclone dust collector, wherein the flue gas multi-parameter data includes a flue gas temperature value, a flue gas humidity value and a flue gas particulate matter concentration value; A smoke multi-parameter data time series regularization module, used to regularize the time series of the smoke multi-parameter data according to the time dimension to obtain a smoke temperature time series vector, a smoke humidity time series vector and a smoke particulate matter concentration time series vector; a smoke parameter time series feature extraction module, used to extract time series correlation features from the smoke temperature time series vector, the smoke humidity time series vector and the smoke particulate matter concentration time series vector respectively to obtain a smoke temperature time series correlation feature vector, a smoke humidity time series correlation feature vector and a smoke particulate matter concentration time series correlation feature vector; A smoke multi-parameter time series feature enhancement fusion module is used to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector by passing the smoke temperature time series associated feature vector, the smoke humidity time series associated feature vector and the smoke particulate matter concentration time series associated feature vector through a smoke multi-parameter time series feature energy measurement enhancement module based on a Transformer structure; A cyclone dust collector speed control module, used to determine a control result of the speed of the cyclone dust collector based on the enhanced flue gas multi-parameter time series semantic fusion feature vector, wherein the control result is used to indicate whether the speed of the cyclone dust collector should be increased, decreased, or unchanged; The smoke multi-parameter time series feature enhancement fusion module includes: a vector combination unit, used for combining the smoke temperature time series correlation feature vector, the smoke humidity time series correlation feature vector and the smoke particulate matter concentration time series correlation feature vector into a set of smoke multi-parameter time series correlation features, and calculating the characteristic energy level coefficient of each characteristic vector in the set of smoke multi-parameter time series correlation features to obtain the smoke temperature time series characteristic energy level coefficient, the smoke humidity time series characteristic energy level coefficient and the smoke particulate matter concentration time series characteristic energy level coefficient; A probabilistic mapping unit, used for probabilistically mapping the smoke temperature time series characteristic energy level coefficient, the smoke humidity time series characteristic energy level coefficient and the smoke particulate matter concentration time series characteristic energy level coefficient respectively to obtain the smoke temperature time series characteristic energy level weight, the smoke humidity time series characteristic energy level weight and the smoke particulate matter concentration time series characteristic energy level weight; A first weight multiplication calculation unit, used for multiplying the flue gas temperature time series associated feature vector by the flue gas temperature time series feature energy level weight to obtain a flue gas temperature time series energy level modulation feature vector; A second weight multiplication calculation unit, used for multiplying the smoke humidity time series associated feature vector by the smoke humidity time series feature energy level weight to obtain a smoke humidity time series energy level modulation feature vector; A third weight multiplication calculation unit, used for multiplying the smoke particulate matter concentration time series correlation feature vector by the smoke particulate matter concentration time series feature energy level weight to obtain a smoke particulate matter concentration time series energy level modulation feature vector; a matrix calculation unit, for multiplying the transposed vector of the flue gas temperature time series energy level modulation feature vector by the flue gas humidity time series energy level modulation feature vector and then dividing the multiplied vector by the square root of the length of the flue gas particulate matter concentration time series correlation feature vector by position to obtain a flue gas temperature-humidity time series correlation weight matrix; Normalization unit, used to utilize The function normalizes the characteristic values ​​of each position in the flue gas temperature-humidity time series association weight matrix to obtain an activated flue gas temperature-humidity time series association weight matrix; A matrix multiplication calculation unit, used for calculating the matrix multiplication between the activated smoke temperature-humidity time series association weight matrix and the transposed vector of the smoke particulate matter concentration time series energy level modulation feature vector to obtain the enhanced smoke multi-parameter time series semantic fusion feature vector; Wherein, the vector combination unit is used for: For the set of the flue gas multi-parameter time series correlation features, calculating the maximum value of the characteristic of the flue gas temperature time series correlation feature vector in the set of the flue gas multi-parameter time series correlation features to obtain the maximum value of the flue gas temperature time series feature; Calculating the mean and variance of the flue gas temperature time series associated feature vector respectively to obtain the flue gas temperature time series feature mean and the flue gas temperature time series feature variance; The value obtained by adding the flue gas temperature time series characteristic variance and the preset hyperparameter is multiplied by four to obtain the first energy level factor of the flue gas temperature time series characteristic; Calculate the square of the difference between the maximum value of the flue gas temperature time series characteristic and the mean value of the flue gas temperature time series characteristic to obtain a flue gas temperature time series characteristic difference value; After multiplying the flue gas temperature time series characteristic variance by two, the variance is then added to the preset hyperparameter and the flue gas temperature time series characteristic difference value which are doubled to obtain the second energy level factor of the flue gas temperature time series characteristic; Dividing the first energy level factor of the flue gas temperature time series characteristic by the second energy level factor of the flue gas temperature time series characteristic to obtain the flue gas temperature time series characteristic energy level coefficient; Wherein, the probabilistic mapping unit is used to: The function normalizes the flue gas temperature time series characteristic energy level coefficient, the flue gas humidity time series characteristic energy level coefficient and the flue gas particulate matter concentration time series characteristic energy level coefficient respectively to obtain the flue gas temperature time series characteristic energy level weight, the flue gas humidity time series characteristic energy level weight and the flue gas particulate matter concentration time series characteristic energy level weight.

2. The dust removal control system for quartz sand production according to claim 1, characterized in that: The smoke parameter time series feature extraction module is used to: pass the smoke temperature time series vector, the smoke humidity time series vector and the smoke particulate matter concentration time series vector through a smoke parameter time series feature extractor based on a time convolution network to obtain the smoke temperature time series correlation feature vector, the smoke humidity time series correlation feature vector and the smoke particulate matter concentration time series correlation feature vector.

3. The dust removal control system for quartz sand production according to claim 2, characterized in that: The cyclone dust collector speed control module is used to: pass the enhanced flue gas multi-parameter time series semantic fusion feature vector through a classifier-based cyclone dust collector speed controller to obtain a control result, and the control result is used to indicate whether the cyclone dust collector speed should increase, decrease or remain unchanged.

4. The dust removal control system for quartz sand production according to claim 3, characterized in that: The cyclone dust collector speed control module comprises: A fully connected encoding unit, used for performing fully connected encoding on the enhanced flue gas multi-parameter temporal semantic fusion feature vector using multiple fully connected layers of the classifier-based cyclone dust collector speed controller to obtain an encoded classification feature vector; A classification unit is used to pass the encoded classification feature vector through the Softmax classification function of the cyclone dust collector speed controller based on the classifier to obtain the control result.

5. A dust removal control method for quartz sand production, using the dust removal control system for quartz sand production according to claim 1, characterized in that: include: Acquire a time series of flue gas multi-parameter data collected by a sensor group deployed at the inlet of a cyclone dust collector, wherein the flue gas multi-parameter data includes a flue gas temperature value, a flue gas humidity value, and a flue gas particulate matter concentration value; Regularizing the time series of the flue gas multi-parameter data according to the time dimension to obtain a flue gas temperature time series vector, a flue gas humidity time series vector, and a flue gas particulate matter concentration time series vector; Extracting time series correlation features from the smoke temperature time series vector, the smoke humidity time series vector, and the smoke particulate matter concentration time series vector to obtain a smoke temperature time series correlation feature vector, a smoke humidity time series correlation feature vector, and a smoke particulate matter concentration time series correlation feature vector; The smoke temperature time series associated feature vector, the smoke humidity time series associated feature vector and the smoke particulate matter concentration time series associated feature vector are subjected to a smoke multi-parameter time series feature energy measurement enhancement module based on a Transformer structure to obtain an enhanced smoke multi-parameter time series semantic fusion feature vector; Based on the enhanced flue gas multi-parameter time series semantic fusion feature vector, a control result of the rotation speed of the cyclone dust collector is determined, and the control result is used to indicate whether the rotation speed of the cyclone dust collector should increase, decrease, or remain unchanged.

6. The dust removal control method for quartz sand production according to claim 5, characterized in that: The time series correlation feature extraction is performed on the smoke temperature time series vector, the smoke humidity time series vector and the smoke particulate matter concentration time series vector respectively to obtain the smoke temperature time series correlation feature vector, the smoke humidity time series correlation feature vector and the smoke particulate matter concentration time series correlation feature vector, including: passing the smoke temperature time series vector, the smoke humidity time series vector and the smoke particulate matter concentration time series vector through a smoke parameter time series feature extractor based on a time convolution network to obtain the smoke temperature time series correlation feature vector, the smoke humidity time series correlation feature vector and the smoke particulate matter concentration time series correlation feature vector.

Images