Household board excess material data monitoring system based on Internet of Things

Abstract

The invention discloses a household board excess material data monitoring system based on the Internet of Things, and the system comprises an interference feature obtaining module which collects electromagnetic interference signal data of different positions and time periods of a workshop, extracts interference signal features, and constructs an interference feature library based on an optimized clustering operation process; the acquisition array optimization module is used for setting a data acquisition antenna array, calculating an optimal weighting coefficient through an optimization algorithm, adjusting a transmitting beam direction of the antenna array based on the optimal weighting coefficient, obtaining an optimal acquisition array, and receiving initial plate excess material data in real time through the optimal acquisition array; and the excess material data optimization module is used for acquiring the initial excess material data of the plate based on the optimal acquisition array, performing interference signal offset on the initial excess material data of the plate by combining the interference feature library with the support vector machine to obtain the optimal excess material data of the plate, and the system can more accurately reflect the actual excess material condition.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring, and in particular to a household board surplus data monitoring system based on the Internet of Things. Background Art

[0002] With the expansion of the production scale of smart home panels, the monitoring and management of panel waste has become a key link to improve production efficiency and reduce costs. At present, the home panel production workshop is full of various metal equipment and electrical equipment, such as large electric saws, electric planers, high-frequency welders, etc. These equipment will generate strong and complex electromagnetic interference during operation. The existing monitoring system mostly uses wireless signals such as Bluetooth and ZigBee for data transmission, which will be greatly affected in this complex electromagnetic environment; In the prior art, the wireless signal of the monitoring system is prone to signal interruption and increased bit error rate in the complex electromagnetic environment of the home furnishing board production workshop (various metal equipment and electrical equipment, such as large electric saws, electric planers, high-frequency welders, etc.). Taking the sensor nodes near large metal processing equipment as an example, the signal interruption probability is as high as 30%, which seriously affects the data integrity and will cause inaccurate subsequent residual material data monitoring results. Therefore, a home furnishing board residual material data monitoring system based on the Internet of Things is proposed herein. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention proposes the following technical solutions: A household board waste data monitoring system based on the Internet of Things, comprising: Interference feature acquisition module: collects electromagnetic interference signal data at different locations and time periods in the workshop, extracts interference signal features, and builds an interference feature library based on an optimized clustering operation process; Acquisition array optimization module: set the data acquisition antenna array, calculate the optimal weighting coefficient through the optimization algorithm, adjust the transmission beam direction of the antenna array based on the optimal weighting coefficient, obtain the optimal acquisition array, and receive the initial plate residue data in real time through the optimal acquisition array; Residue data optimization module: collects initial sheet material residue data based on the optimal collection array, and cancels the interference signal of the initial sheet material residue data through the interference feature library combined with the support vector machine to obtain the optimal sheet material residue data; Residual material data monitoring module: Based on the real-time optimal plate residual material data, the residual material threshold is set. When the residual material data exceeds the residual material threshold, the reminder mechanism is triggered; The optimized clustering operation is achieved by combining hierarchical clustering with K-Means clustering; The hierarchical clustering calculates the distance of cluster centers by using a distance metric learning model based on deep learning as a clustering metric in the clustering process.

[0004] The interference signal feature acquisition process is as follows: Based on the equipment layout, work area division and electromagnetic interference source distribution of the home panel workshop, different locations in the workshop Arrange high-precision electromagnetic interference signal collector, Indicates the position index; In different time periods , according to the fixed sampling frequency , continuously collect electromagnetic interference signals Form a large amount of raw signal data set ; Set the passband and stopband parameters of the filter to remove interference signals in irrelevant frequency bands and highlight the main target frequency band signals ; By using Fourier transform to analyze the main target frequency band signal Perform feature extraction to obtain interference signal characteristics .

[0005] The interference feature library construction process is as follows: The interference feature vector As input data for hierarchical clustering; In the initial division stage of hierarchical clustering, an agglomerative hierarchical clustering method is used, starting from each feature vector as a separate class. When merging classes, a distance metric learning model based on deep learning is used to calculate the distance of cluster centers to obtain clustering results. The results of hierarchical clustering are used as the initial cluster centers of the K-Means clustering algorithm; The K-Means clustering algorithm starts with the initial cluster center and enters the iterative calculation process; For each category obtained by clustering, traverse the interference signal features in this category , giving a unique identifier Create a database table as an interference signature library .

[0006] The process of building a distance metric learning model based on deep learning is: Build a deep neural network to interfere with the signal characteristics As input; Use contrastive learning loss function to train the network; For a given random three eigenvectors , , , define the contrastive learning loss function:

[0007] in represents the distance metric in the embedding space, It is a preset boundary value, which is used to ensure that there is enough distance between vectors of different classes; After the training is completed, a distance metric learning model based on deep learning is obtained.

[0008] The optimal acquisition array determination process is: Arranged by A uniform linear antenna array consisting of antenna elements; The antenna unit spacing is , and satisfies , is the signal wavelength, let The antenna unit receives or transmits the signal. The input signal is ; Determine an optimization objective function to maximize the signal-to-interference-to-noise ratio of the received signal for the goal; Assume the expected signal is , the interference signal is , the output signal of the antenna array It is expressed as: ,in For the The weighting coefficients of the antenna elements; The weight coefficient vector is ,in, represents transpose; Signal to Interference and Noise Ratio The maximization calculation formula is: ,in represents the mathematical expectation, The expected signal is The components of the antenna elements, The interference signal is The components of the antenna elements, Indicates noise; Select the particle swarm optimization algorithm and combine it with the goal of optimizing the objective function to obtain the best weighting coefficient ; Get the best weighting coefficient Finally, the antenna array directivity function is changed by adjusting the weighting coefficient so that the transmission beam direction avoids the direction of the strong electromagnetic interference source and the optimal acquisition array is obtained.

[0009] The process of combining the particle swarm optimization algorithm with the goal of optimizing the objective function is: Randomly generate equal to the number of antenna units The number of particles, the position vector of each particle is , the initial velocity vector is ; Record the historical optimal position of each particle , initialized to its initial position, the global optimal position of the entire particle swarm Initialize to the initial position of the particle with the largest fitness value; Update the particle's position and velocity, for each particle's position vector , the position vector Substitute the signal-to-interference-noise ratio Calculation formula, the result of the calculation is used as the fitness value ; Compare the current fitness value of each particle With its historical best location The corresponding fitness value is , then update is the current location; Compare the current fitness values ​​of all particles and find the particle position corresponding to the maximum fitness value. When the fitness value of this position is greater than the current global optimal position The corresponding fitness value is updated is the particle position; When the preset number of iterations is reached Or the fitness value of the global optimal position is continuous The change in iterations is less than the minimum value When , the iteration stops, and the global optimal position at this time The corresponding weighted coefficient combination is the optimal weighted coefficient ,in, Indicates transpose.

[0010] The optimal sheet material residual material data acquisition process is as follows: The initial sheet material data collected by the optimal acquisition array includes the expected signal , interference signal and noise , forming the initial sheet material data ; For the collected initial sheet material data , using support vector machine and interference feature library Make a match; According to the interference signal matching results, after determining the interference signal type, obtain the corresponding interference feature vector from the interference feature library Ω Frequency , Amplitude and Phase , according to the formula Reconstructing the interference signal ,in, represents the reconstructed interference signal corresponding to the nth interference feature vector; The interference signal will be reconstructed The reference signal is input into the adaptive filter, and the output is close to the interference signal. Subtract the output of the filter , and get the expected signal after canceling the interference ; The signals after all antenna units cancel out interference constitute the optimal sheet material waste data set ,in, is the number of antenna elements , the number of each antenna element corresponds to a set of desired signals.

[0011] Using support vector machine and interference signature library The matching process is: Use the data in the constructed interference feature library Ω to form a training data set ,in It is The signal of the antenna element The interference feature vector of training samples, is the corresponding category label; Solve the optimization problem of the support vector machine to obtain the optimal weight vector and bias Building a classification function ; The feature vector of the collected initial sheet material data , substitute into the classification function ,in, is the symbolic function, It is a function that maps the input feature vector to a high-dimensional space.

[0012] The present invention has the following beneficial effects: In the present invention, firstly, by combining the improved hierarchical clustering with the K-Means clustering algorithm, an interference feature library is constructed, and a distance metric learning model based on deep learning is introduced in the hierarchical clustering. Compared with the traditional clustering method, the distance metric learning model based on deep learning can more accurately analyze the electromagnetic interference signal characteristics, because the traditional clustering method such as the Euclidean distance only measures the similarity of the feature vector from the spatial position and cannot consider the complex correlation between the features, while the deep learning model can mine the complex nonlinear relationship. When processing interference signals with similar frequencies but different modulation modes, it can accurately distinguish and cluster them, and construct a more accurate interference feature library; Secondly, by taking maximizing the signal-to-interference-to-noise ratio as the goal, the particle swarm optimization algorithm is used to calculate the optimal weighting coefficient, and the transmission beam direction of the antenna array is adjusted by the weighting coefficient. In a complex electromagnetic environment, it is possible to effectively avoid strong electromagnetic interference sources, enhance signal coverage of the target monitoring area, and further ensure the stability of the data acquisition process; Finally, through the combined decision-making of the interference feature library and the support vector machine, the interference of the data collected in real time by the optimal acquisition array is offset. The interference feature library contains rich interference feature information, and the support vector machine can accurately identify the interference type. The combination of the two can effectively remove the interference components in the collected data, thereby obtaining the optimal plate waste data and more accurately reflecting the actual waste situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a system block diagram of a household board waste data monitoring system based on the Internet of Things proposed by the present invention. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] Example: Figure 1 As shown, the present invention proposes a household board surplus material data monitoring system based on the Internet of Things, comprising: Interference feature acquisition module: collects electromagnetic interference signal data at different locations and time periods in the workshop, extracts interference signal features, and builds an interference feature library based on an optimized clustering operation process; Based on the equipment layout, work area division and electromagnetic interference source distribution of the home panel workshop, different locations in the workshop ( , Arrange high-precision electromagnetic interference signal collectors for the number of collection points. Indicates the position index; In different time periods ( , is the number of time periods), according to a fixed sampling frequency , continuously and stably collect electromagnetic interference signals ; The collection time period should cover various working states such as the regular working hours of the workshop, the start and stop moments of the equipment, the switching periods of different production processes, etc., to ensure that comprehensive and representative electromagnetic interference signal data is obtained, and ultimately form a large number of original signal data sets. ; According to the main frequency range of electromagnetic interference in the workshop, set the passband and stopband parameters of the filter to remove interference signals in irrelevant frequency bands and highlight the main target frequency band signals. ; By using Fourier transform to analyze the main target frequency band signal Perform feature extraction to obtain interference signal characteristics ; The optimized clustering operation is achieved by combining the improved hierarchical clustering with K-Means clustering; Specifically, based on hierarchical clustering, the interference feature vector set extracted from the electromagnetic interference signal is used as the input data of the hierarchical clustering. Represents a data point; In the initial division stage of hierarchical clustering, an agglomerative hierarchical clustering method is used, starting with each feature vector as a separate class. When merging classes, a distance metric learning model based on deep learning is used to calculate the distance of the cluster center; The specific process is: build a deep neural network to interfere with the signal characteristics As input, a low-dimensional embedding space is obtained through network learning, in which the distance between similar interference signal feature vectors is closer, and the distance between different ones is farther; The network is trained using the contrastive learning loss function. For each interference signal feature vector , randomly select some similar (belonging to the same type of interference) and dissimilar (belonging to different types of interference) vectors as comparison samples; During the training process, the network parameters are adjusted to make the distance between similar vectors in the embedding space as small as possible and the distance between dissimilar vectors as large as possible. , , , define the contrastive learning loss function:

[0016] in represents the distance metric in the embedding space, It is a preset boundary value, which is used to ensure that there is enough distance between vectors of different classes; For a given feature vector , randomly select some similar vectors belonging to the same type of interference , and dissimilar vectors belonging to different types of interference As comparison samples, the reason for doing so is that through these samples with clear similarities or differences, the model can better learn the differences in the features of the interference signals, and then make a reasonable distance measurement for them in the embedding space; After training, the deep learning-based distance metric learning model is used as the distance metric for hierarchical clustering instead of the traditional distance metric (such as Euclidean distance); Specifically, traditional distance measurement methods (such as Euclidean distance) only calculate the distance from the spatial position of data points, which makes it difficult to consider the complex internal connections between features. The distance measurement learning model based on deep learning can mine the complex nonlinear relationship between features by building a deep neural network to learn the feature vectors of interference signals, thereby more accurately measuring the similarity of feature vectors. For example, when processing interference signals with similar frequencies but different modulation methods, the model can learn the impact of modulation method differences on signal similarity, and distinguish different interference signals more accurately than Euclidean distance. The results of hierarchical clustering are used as the initial cluster centers of the K-Means clustering algorithm; The K-Means clustering algorithm starts with the initial cluster center and enters the iterative calculation process. In each iteration, the Euclidean distance clustering calculation method is replaced by the Mahalanobis distance. By calculating each eigenvector (interference signal feature ) to the Mahalanobis distance of the cluster center , the feature vector (interference signal feature ) is assigned to the category of the nearest cluster center, and then the location of the cluster center is updated according to the new category assignment. This iterative process is repeated until the location of the cluster center no longer changes significantly or the preset number of iterations is reached; Specifically, the distance between the feature vector and the cluster center is calculated by replacing the Euclidean distance with the Mahalanobis distance in the clustering process. The traditional Euclidean distance does not consider the correlation between features. When processing a data set such as an electromagnetic interference signal where there may be complex correlations between features, it is easy to cause clustering bias. The Mahalanobis distance can consider the correlation between features and can more accurately measure the similarity between samples. For each category obtained by clustering, traverse the interference signal features in this category , based on the interference signal characteristics Get detailed parameters and assign a unique identifier Create a database table as an interference signature library ; Among them, the interference feature library The detailed parameters (in database table form) include the following fields: Category ID , Interference signal characteristics And the corresponding collection location and the corresponding time period , establish a composite index based on category identification and collection location; Specifically, by combining the improved hierarchical clustering with K-Means clustering as a whole optimized clustering operation process, the advantages of the two algorithms are brought into play; Hierarchical clustering grasps the structure of the data as a whole, which can avoid falling into the local optimal solution, make a preliminary and reasonable division of the electromagnetic interference signal feature vectors, and form a rough clustering framework. The K-Means clustering algorithm takes the results of hierarchical clustering as the starting point and performs local optimization of the clustering results through iterative calculations. This fusion method is not a simple algorithm superposition, but based on the characteristics of the electromagnetic interference signal characteristics, the two algorithms complement each other, thereby improving the accuracy and stability of clustering.

[0017] Acquisition array optimization module: set the data acquisition antenna array, calculate the optimal weighting coefficient through the optimization algorithm, adjust the transmission beam direction of the antenna array based on the optimal weighting coefficient, obtain the optimal acquisition array, and receive the initial plate residue data in real time through the optimal acquisition array; Arranged by A uniform linear antenna array consisting of antenna elements; The antenna unit spacing is , and satisfies ( is the signal wavelength), The antenna unit receives or transmits the signal, and its input signal is assumed to be ; In order to obtain the best weighting coefficient , we need to first determine an optimization objective function to maximize the signal-to-interference-to-noise ratio ( ) is the target, the signal-to-interference-noise ratio reflects the relative size of the expected signal power and the interference and noise power. The higher the value, the better the signal quality. Specifically, let the expected signal be , the interference signal is , the output signal of the antenna array It is expressed as: ,in For the The weighting coefficients of the antenna elements; The weight coefficient vector is ,in, represents transpose, is the number of antenna units, one antenna unit corresponds to one weighting coefficient; Signal to Interference and Noise Ratio The maximization calculation formula is: ,in represents the mathematical expectation, The expected signal is The components of the antenna elements, The interference signal is The components of the antenna elements, Indicates noise; The goal of optimizing the objective function is to maximize the signal-to-interference-noise ratio , so the signal-to-interference-noise ratio The maximization calculation formula is to optimize the objective function; Considering the complexity and real-time variability of the electromagnetic environment in the furniture panel workshop, the particle swarm optimization algorithm is selected with the goal of optimizing the objective function (maximizing the signal-to-interference-noise ratio ) to obtain the best weighting coefficient ; Random Generation Particles ( is equal to the number of antenna units), the position vector of each particle is , the initial velocity vector is ; Record the historical optimal position of each particle , initialized to its initial position, the global optimal position of the entire particle swarm Initialize to the initial position of the particle with the largest fitness value; Update the position and velocity of the particle. The formula is: Speed ​​Update:

[0018] in: Indicates the next position The speed of the inertia weight is ω, which is set to a larger value (such as 0.9) at the beginning of the iteration for global search, and gradually reduced (such as to 0.4) as the iteration proceeds for local search; and is the learning factor, usually set to 1.5; and is a random number between 0 and 1; Location Updates: = +

[0019] For each particle's position vector , substitute it into the signal-to-interference-noise ratio Calculation formula, the result of the calculation is used as the fitness value ; Compare the current fitness value of each particle With its historical best location The corresponding fitness value is , then update is the current location; Compare the current fitness values ​​of all particles and find the particle position corresponding to the maximum fitness value. When the fitness value of this position is greater than the current global optimal position The corresponding fitness value is updated is the particle position; When the preset number of iterations is reached Or the fitness value of the global optimal position is continuous The change in iterations is less than the minimum value When , the iteration stops, and the global optimal position at this time The corresponding weighted coefficient combination is the optimal weighted coefficient ,in, represents transpose, is the number of weighting coefficients ( equal to the number of antenna elements); Get the best weighting coefficient Finally, the radiation characteristics of the antenna array are calculated by the formula:

[0020] in is the wave number, is the radiation angle of the antenna array; By adjusting the weighting coefficients and changing the directional pattern function of the antenna array, the transmission beam direction avoids the direction of the strong electromagnetic interference source, while enhancing the signal coverage of the target monitoring area to obtain the optimal acquisition array; For example, devices such as phase shifters and attenuators can be used to implement weighted adjustments to the signals from each antenna element; Specifically, the signal superposition of each antenna unit in the antenna array will form a specific radiation pattern, and the weighting coefficient It will affect the amplitude and phase of the antenna unit signal, thus changing the directivity function of the antenna array. Therefore, by adjusting the weighting coefficient, the direction of the transmit beam can be changed according to the specified requirements to avoid the interference source and enhance the signal coverage of the target area. After the antenna array is put into use, the signal data of the initial sheet material waste data is received in real time and transmitted to the waste data optimization module.

[0021] Residue data optimization module: collects initial sheet material residue data based on the optimal collection array, and cancels the interference signal of the initial sheet material residue data through the interference feature library combined with the support vector machine to obtain the optimal sheet material residue data; The initial sheet material remaining data collected by the optimal acquisition array includes the expected signal (optimal sheet material remaining data) , interference signal and noise , forming the initial sheet material data ; For the collected initial sheet material data , and use support vector machine to compare it with the interference feature library Make a match; Use the data in the constructed interference feature library Ω to form a training data set ,in It is The signal of the antenna element The interference feature vector of training samples, is the corresponding category label (representing different types of interference); By solving the optimization problem of support vector machine , the constraints are: ; in, is the weight vector, is the bias, is the slack variable, is the penalty parameter, is a function that maps the input feature vector to a high-dimensional space to obtain the optimal and , thereby determining the classification hyperplane, It is interference features; Specifically, in actual industrial environments such as home furnishing board production workshops, there is complex electromagnetic interference, which will affect the collected board waste data. By collecting initial data containing interference, the support vector machine and the interference feature library are used to match and classify the interference signals, and then corresponding processing measures are taken. The core principle is to achieve the classification of different types of data by finding the optimal hyperplane; By solving the above optimization problem, we can obtain the optimal weight vector and bias , and get the weight vector and bias Then, substitute it into the expression of the hyperplane In combination with the symbolic function , constitutes the classification function; For the feature vector of the collected initial sheet material data , substitute into the classification function ,in, is a symbolic function; The interference type is determined by the classification function, and the elements in the interference feature library Ω that match it are found to complete the classification and matching of the interference signal; By symbolic function The output result can classify the collected feature vectors into different interference types. For example, +1 represents high-frequency electromagnetic interference and -1 represents low-frequency electromagnetic interference. After obtaining the judgment result of the interference type, find the matching element in the interference feature library Ω to obtain the corresponding interference feature vector. Detailed parameters of According to the interference signal matching results, after determining the interference signal type, obtain the corresponding interference feature vector from the interference feature library Ω Detailed parameters of Detailed parameters further include frequency , Amplitude and Phase , according to the formula Reconstructing the interference signal ,in, represents the reconstructed interference signal corresponding to the nth interference feature vector; Specifically, the feature vector of the initial sheet material data is collected After that, it is substituted into the classification function of the support vector machine. The relationship between the feature vector and the optimal hyperplane is calculated through the classification function to obtain the category label, so as to determine the interference type. It can effectively distinguish different types of interference signals. After determining the type of interference signal, the corresponding interference feature vector is obtained from the interference feature library Ω. Detailed parameters (frequency , Amplitude and Phase ), usually the interference signal is approximated by a sine wave, according to the formula Reconstruct the interference signal and provide an accurate reference signal for subsequent interference cancellation; The interference signal will be reconstructed As a reference signal, it is input into the adaptive filter, and the filter weight coefficient is continuously adjusted to make the filter output close to the interference signal. Then, the collected signal Subtract the output of the filter , and get the expected signal after canceling the interference ; The signals after all antenna units cancel out interference constitute the optimal sheet material waste data set ,in, is the number of antenna elements , the number of each antenna element corresponds to a set of desired signals; Specifically, the reconstructed interference signal The reference signal is input into the adaptive filter, which continuously adjusts the weight coefficient according to the input signal and the reference signal, in order to make the filter output as close to the interference signal as possible, because the collected signal It contains the expected signal and the interference signal. When the filter output is close to the interference signal, the interference signal can be offset by subtracting the filter output from the collected signal, thereby obtaining the expected signal after the interference is offset. Then, the signal collected by each antenna unit is subjected to the above interference offset operation, and the signals of all antenna units after the interference is offset are integrated to form the optimal sheet material waste data set. .

[0022] Residual material data monitoring module: Based on the real-time optimal plate residual material data, the residual material threshold is set. When the residual material data exceeds the residual material threshold, the reminder mechanism is triggered; Get the optimal sheet material data set After that, traverse the material data set The remaining material quantity of all remaining material threshold types in the system is recorded, and the corresponding alarm threshold is set. When a certain remaining material data exceeds the threshold, the system will issue a reminder. Set a certain type of board ( The minimum number threshold for , the maximum number threshold is ; Specifically, a certain type of board Highest threshold and the minimum threshold Obtained by referring to historical production data and counting the average number of such plates required for each batch of production under normal production conditions; Compare the extracted real-time data with the set threshold to determine whether an alert needs to be triggered; like , a low quantity reminder is triggered; like , a high quantity reminder is triggered.

[0023] In the application, several formulas involved are calculated by removing dimensions and taking their numerical values. The formulas are established by collecting a large amount of data and performing software simulation to obtain a formula for the most recent actual situation. Some coefficients or weights in the formulas are set by technicians in this field according to actual conditions, so they will not be elaborated here.

[0024] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0025] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A household board material surplus data monitoring system based on the Internet of Things, characterized in that: include: Interference feature acquisition module: collects electromagnetic interference signal data at different locations and time periods in the workshop, extracts interference signal features, and builds an interference feature library based on an optimized clustering operation process; Acquisition array optimization module: set the data acquisition antenna array, calculate the optimal weighting coefficient through the optimization algorithm, adjust the transmission beam direction of the antenna array based on the optimal weighting coefficient, obtain the optimal acquisition array, and receive the initial plate residue data in real time through the optimal acquisition array; Residue data optimization module: collects initial sheet material residue data based on the optimal collection array, and cancels the interference signal of the initial sheet material residue data through the interference feature library combined with the support vector machine to obtain the optimal sheet material residue data; Residual material data monitoring module: Based on the real-time optimal plate residual material data, the residual material threshold is set. When the residual material data exceeds the residual material threshold, the reminder mechanism is triggered; The optimized clustering operation is achieved by combining hierarchical clustering with K-Means clustering; The hierarchical clustering calculates the distance of cluster centers by using a distance metric learning model based on deep learning as a clustering metric in the clustering process.

2. According to the Internet of Things-based home board surplus data monitoring system of claim 1, it is characterized in that: The interference signal feature acquisition process is as follows: Based on the equipment layout, work area division and electromagnetic interference source distribution of the home panel workshop, different locations in the workshop Arrange high-precision electromagnetic interference signal collector, Indicates the position index; In different time periods , according to the fixed sampling frequency , continuously collect electromagnetic interference signals Form a large amount of raw signal data set ; Set the passband and stopband parameters of the filter to remove interference signals in irrelevant frequency bands and highlight the main target frequency band signals ; By using Fourier transform to analyze the main target frequency band signal Perform feature extraction to obtain interference signal characteristics .

3. According to the Internet of Things-based home board waste data monitoring system of claim 1, it is characterized in that: The interference feature library construction process is as follows: The interference feature vector As input data for hierarchical clustering; In the initial division stage of hierarchical clustering, an agglomerative hierarchical clustering method is used, starting from each feature vector as a separate class. When merging classes, a distance metric learning model based on deep learning is used to calculate the distance of cluster centers to obtain clustering results. The results of hierarchical clustering are used as the initial cluster centers of the K-Means clustering algorithm; The K-Means clustering algorithm starts with the initial cluster center and enters the iterative calculation process; For each category obtained by clustering, traverse the interference signal features in this category , giving a unique identifier Create a database table as an interference signature library .

4. The home board material surplus data monitoring system based on the Internet of Things according to claim 3 is characterized in that: The process of building a distance metric learning model based on deep learning is: Build a deep neural network to interfere with the signal characteristics As input; Use contrastive learning loss function to train the network; For a given random three eigenvectors , , , define the contrastive learning loss function: ； in represents the distance metric in the embedding space, It is a preset boundary value, which is used to ensure that there is enough distance between vectors of different classes; After the training is completed, a distance metric learning model based on deep learning is obtained.

5. The home board material surplus data monitoring system based on the Internet of Things according to claim 1 is characterized in that: The optimal acquisition array determination process is: Arranged by A uniform linear antenna array consisting of antenna elements; The antenna unit spacing is , and satisfies , is the signal wavelength, let The antenna unit receives or transmits the signal. The input signal is ; Determine an optimization objective function to maximize the signal-to-interference-to-noise ratio of the received signal for the goal; Assume the expected signal is , the interference signal is , the output signal of the antenna array It is expressed as: ,in For the The weighting coefficients of the antenna elements; The weight coefficient vector is ,in, represents transpose; Signal to Interference and Noise Ratio The maximization calculation formula is: ,in represents the mathematical expectation, The expected signal is The components of the antenna elements, The interference signal is The components of the antenna elements, Indicates noise; Select the particle swarm optimization algorithm and combine it with the goal of optimizing the objective function to obtain the best weighting coefficient ; Get the best weighting coefficient Finally, the antenna array directivity function is changed by adjusting the weighting coefficient so that the transmission beam direction avoids the direction of the strong electromagnetic interference source and the optimal acquisition array is obtained.

6. The home board surplus material data monitoring system based on the Internet of Things according to claim 5 is characterized in that: The process of combining the particle swarm optimization algorithm with the goal of optimizing the objective function is: Randomly generate equal to the number of antenna units The number of particles, the position vector of each particle is , the initial velocity vector is ; Record the historical optimal position of each particle , initialized to its initial position, the global optimal position of the entire particle swarm Initialize to the initial position of the particle with the largest fitness value; Update the particle's position and velocity, for each particle's position vector , the position vector Substitute the signal-to-interference-noise ratio Calculation formula, the result of the calculation is used as the fitness value ; Compare the current fitness value of each particle With its historical best location The corresponding fitness value is , then update is the current location; Compare the current fitness values ​​of all particles and find the particle position corresponding to the maximum fitness value. When the fitness value of this position is greater than the current global optimal position The corresponding fitness value is updated is the particle position; When the preset number of iterations is reached Or the fitness value of the global optimal position is continuous The change in iterations is less than the minimum value When , the iteration stops, and the global optimal position at this time The corresponding weighted coefficient combination is the optimal weighted coefficient ,in, Indicates transpose.

7. The home board surplus material data monitoring system based on the Internet of Things according to claim 1 is characterized in that: The optimal sheet material residual material data acquisition process is as follows: The initial sheet material data collected by the optimal acquisition array includes the expected signal , interference signal and noise , forming the initial sheet material data ; For the collected initial sheet material data , using support vector machine and interference feature library Make a match; According to the interference signal matching results, after determining the interference signal type, obtain the corresponding interference feature vector from the interference feature library Ω Frequency , Amplitude and Phase , according to the formula Reconstructing the interference signal ,in, represents the reconstructed interference signal corresponding to the nth interference feature vector; The interference signal will be reconstructed The reference signal is input into the adaptive filter, and the output is close to the interference signal. Subtract the output of the filter , and get the expected signal after canceling the interference ; The signals after all antenna units cancel out interference constitute the optimal sheet material waste data set ,in, is the number of antenna elements , the number of each antenna element corresponds to a set of desired signals.

8. The home board material surplus data monitoring system based on the Internet of Things according to claim 7 is characterized in that: Using support vector machine and interference signature library The matching process is: Use the data in the constructed interference feature library Ω to form a training data set ,in It is The signal of the antenna element The interference feature vector of training samples, is the corresponding category label; Solve the optimization problem of the support vector machine to obtain the optimal weight vector and bias Building a classification function ; The feature vector of the collected initial sheet material data , substitute into the classification function ,in, is the symbolic function, It is a function that maps the input feature vector to a high-dimensional space.

Images