A production control method and system of LED track lights based on the Internet of Things

Abstract

The present invention discloses a production control method and system for LED track lights based on the Internet of Things, including: obtaining historical production information and production process information, performing optimal process analysis on each production process, and forming a production knowledge graph; obtaining production demand information, constructing a production plan formulation model based on a particle swarm algorithm, formulating the optimal production plan for the current desired production product in combination with the production knowledge graph, performing production control and obtaining real-time operation monitoring information; constructing an abnormal operation detection model based on VAE, performing abnormal operation detection on production equipment according to the real-time operation monitoring information; performing real-time status evaluation on abnormally operating equipment, and formulating a control plan to control the operation of the equipment. The production process is divided, and the optimal production operation parameters of the production process subsequence are selected to generate a production control plan for the entire production process, and real-time operation monitoring and operation abnormality detection are performed to control the abnormal operation, thereby improving production efficiency and quality.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a production control method and system of LED track lights based on the Internet of Things. Background Art

[0002] In the current LED lighting industry, track lights, as an important lighting equipment, are widely used in commercial lighting, home lighting, display lighting and other fields. The traditional production of LED track lights mainly relies on manual operation and traditional automated production lines. This production method has many shortcomings, including low production efficiency, unstable quality control and inconvenient production data management.

[0003] With the rapid development of Internet of Things (IoT) technology, applying it to the production control of LED track lights can effectively overcome the drawbacks of traditional production models. IoT technology connects various production equipment, sensors and information systems to achieve comprehensive perception, real-time monitoring and intelligent control of the production process. Through real-time data collection and analysis, the operating status of the production line can be accurately grasped, potential problems can be discovered and handled in a timely manner, and production efficiency and product quality can be significantly improved. Summary of the invention

[0004] The present invention overcomes the defects of the prior art and provides an Internet of Things-based LED track light production control method and system, an important purpose of which is to improve production efficiency and product quality.

[0005] To achieve the above-mentioned purpose, the first aspect of the present invention provides a production control method of LED track lights based on the Internet of Things, comprising:

[0006] Obtain historical production information and production process information, extract the process characteristics of each production process, conduct optimal process analysis for each production process, and construct a production knowledge graph;

[0007] Obtain production demand information, build a production plan formulation model based on the particle swarm algorithm, formulate the optimal production plan for the current desired production product in combination with the production knowledge graph, perform production control and obtain real-time operation monitoring information;

[0008] Building an abnormal operation detection model based on VAE, and performing abnormal operation detection on the production equipment according to the real-time operation monitoring information to obtain abnormal operation detection information;

[0009] Conduct real-time status assessment of abnormally operating equipment and formulate control plans to regulate equipment operation based on the assessment results.

[0010] In this solution, the acquisition of historical production information and production process information, extraction of process characteristics of each production process, and optimal process analysis of each production process to form a production knowledge graph specifically include:

[0011] Obtain historical production information, including historical production specification information, historical production benefit information, historical production equipment status information and historical production operation parameter information;

[0012] Acquire production process information based on the production line of the LED track light, perform feature extraction on the production process information, and obtain production process feature information;

[0013] De-redundancy and segmentation operations are performed according to the production process characteristic information to form a production process subsequence, and the production process category is set through the production process subsequence to obtain production process category information;

[0014] Introducing the K-means clustering algorithm, initializing the number of cluster centers based on the production process category information, constructing sample data based on the historical production information, and calculating the Euclidean distance between each sample data and each cluster center;

[0015] Obtain the Euclidean distance between each sample data and each cluster center, and sort them according to the Euclidean distance to obtain the sorting result of the Euclidean distance between each sample data and each cluster center, and obtain the cluster center corresponding to the minimum Euclidean distance based on the sorting result;

[0016] The cluster center corresponding to the minimum Euclidean distance is used as the target cluster of the corresponding sample data, all sample data are classified, and a production knowledge graph is constructed in combination with the production process feature information.

[0017] In this solution, the production demand information is obtained, a production plan formulation model is constructed based on the particle swarm algorithm, the optimal production plan for the current desired product is formulated in combination with the production knowledge graph, production control is performed and real-time operation monitoring information is obtained, specifically including:

[0018] Acquire production demand information, extract production specification features according to the production demand information, and divide the production process flow to obtain first analysis information;

[0019] Calculate the similarity between the first analysis information and the constructed production knowledge graph to obtain a similarity value, judge the calculated similarity value with a preset threshold, obtain historical production information of each production process according to the judgment result, and obtain second analysis information;

[0020] Extracting the production efficiency characteristics corresponding to each historical production operation parameter based on the second analysis information, sorting each historical production operation parameter, and selecting candidate production operation parameters for each production process according to the sorting result to obtain candidate production operation parameter information;

[0021] Extracting historical production equipment status information corresponding to each candidate production operation parameter according to the candidate production operation parameter information, and associating it with the corresponding candidate production operation parameter to obtain third analysis information;

[0022] Constructing a production plan formulation model based on a particle swarm algorithm, obtaining equipment status information, combining the third analysis information with the input into the production plan formulation model, formulating a production plan for the desired product, and obtaining optimal production plan information;

[0023] Production control is performed according to the optimal production plan information, and real-time operation monitoring of production equipment is performed to obtain real-time operation monitoring information.

[0024] In this solution, the production plan formulation model is constructed based on the particle swarm algorithm to formulate the production plan for the desired product, specifically including:

[0025] Building a production plan formulation model based on a particle swarm algorithm, introducing a genetic algorithm to optimize the model, and generating an initial particle swarm and an initial population according to the input third analysis information;

[0026] A preset objective function is used to obtain candidate production operation parameters according to the third analysis information, and the production equipment state corresponding to each candidate production operation parameter is used as an auxiliary feature to generate an initial particle swarm for each production process flow.

[0027] Calculate the fitness value of each particle in the initial particle swarm of each production process, compare the fitness value of each particle in each initial particle swarm with the preset threshold, and obtain the global optimal solution based on the judgment result;

[0028] Randomly select several particle individuals from the initial particle group to perform crossover operations to obtain newly generated particle individuals, and perform mutation operations on the newly generated particle individuals to obtain a mutant particle group;

[0029] The speed and position of the mutant particle swarm are updated according to the global optimal solution, a penalty function is constructed based on the input device state information, the global optimal solution is penalized, and the above steps are repeated for iterative optimization until the maximum number of iterations is reached to obtain the optimal solution;

[0030] An initial production plan is generated according to the optimal solution, energy consumption of each initial production plan is calculated to obtain the energy consumption of each initial production plan, the initial production plans are sorted in descending order according to the energy consumption, and the optimal production plan is selected based on the sorting result.

[0031] In this solution, the abnormal operation detection model is constructed based on VAE, and abnormal operation detection of production equipment is performed according to the real-time operation monitoring information to obtain abnormal operation detection information, which specifically includes:

[0032] Obtaining optimal production plan information, generating an optimal operation working curve of the production equipment according to the optimal production plan information, presetting a normal operation fluctuation range, adding random fluctuations to the optimal operation working curve of the production equipment, performing sample enhancement and forming a training data set;

[0033] An abnormal operation detection model is constructed based on VAE, the model is trained according to the training data set, a Markov chain is introduced to optimize the model, and the input training data is mapped to a low-dimensional space to obtain potential features of the training data;

[0034] A normal operating data benchmark is set based on the optimal working curve, a state space is formed by combining the acquired potential features, and the Euclidean distance between each training data and the normal operating data benchmark is calculated as the state transition probability and a state transition matrix is ​​constructed;

[0035] A reconstructed sequence is generated using a state transfer matrix, a reconstruction error between the sequence and the original sequence is calculated, an error threshold is preset, the calculated reconstruction error is judged against the error threshold, model parameters are adjusted according to the judgment result, and an abnormal operation detection model that meets expectations is obtained;

[0036] Real-time operation monitoring information is obtained and input into the abnormal operation monitoring model to obtain a reconstruction sequence of real-time operation data and calculate a reconstruction error, and abnormal operation data is determined according to the reconstruction error to obtain abnormal operation detection information.

[0037] In this solution, the real-time status evaluation of abnormally operating equipment is performed, and a control plan is formulated according to the evaluation results to control the operation of the equipment, specifically including:

[0038] Acquire abnormal operation detection information and normal operation fluctuation range, calculate the deviation value between abnormal operation data and normal operation fluctuation, build a state assessment model, input the calculated deviation value into the state assessment model for assessment, and obtain real-time state assessment information;

[0039] Extract abnormal operation data according to the abnormal operation detection information, obtain expected operation data and preset input control parameters, perform input parameter regulation analysis, calculate the regulated input parameters based on the abnormal operation data, the expected operation data and the preset control parameters, and obtain the regulated input control parameters;

[0040] Calculate the adjusted input control parameters and the preset input control parameters to obtain compensation parameters, and generate an initial equipment control plan based on the compensation parameters;

[0041] Based on the initial equipment control plan combined with the real-time status evaluation information, it is evaluated whether the abnormal equipment can withstand the input parameters after control, and first evaluation information is obtained. According to the first evaluation information, the final equipment control plan is selected to perform equipment operation control.

[0042] A second aspect of the present invention provides an LED track light production control system based on the Internet of Things, the system comprising: a memory, a processor, the memory containing an LED track light production control method program based on the Internet of Things, the LED track light production control method program based on the Internet of Things being executed by the processor to implement the following steps:

[0043] Obtain historical production information and production process information, extract the process characteristics of each production process, conduct optimal process analysis for each production process, and construct a production knowledge graph;

[0044] Obtain production demand information, build a production plan formulation model based on the particle swarm algorithm, formulate the optimal production plan for the current desired production product in combination with the production knowledge graph, perform production control and obtain real-time operation monitoring information;

[0045] Building an abnormal operation detection model based on VAE, and performing abnormal operation detection on the production equipment according to the real-time operation monitoring information to obtain abnormal operation detection information;

[0046] Conduct real-time status assessment of abnormally operating equipment and formulate control plans to regulate equipment operation based on the assessment results.

[0047] The present invention discloses a production control method and system for LED track lights based on the Internet of Things, including: obtaining historical production information and production process information, performing optimal process analysis on each production process, and forming a production knowledge graph; obtaining production demand information, constructing a production plan formulation model based on a particle swarm algorithm, formulating the optimal production plan for the current desired production product in combination with the production knowledge graph, performing production control and obtaining real-time operation monitoring information; constructing an abnormal operation detection model based on VAE, performing abnormal operation detection on production equipment according to the real-time operation monitoring information; performing real-time status evaluation on abnormally operating equipment, and formulating a control plan to control the operation of the equipment. The production process is divided, and the optimal production operation parameters of the production process subsequence are selected to generate a production control plan for the entire production process, and real-time operation monitoring and operation abnormality detection are performed to control the abnormal operation, thereby improving production efficiency and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments or exemplary embodiments of the present invention, the drawings required for use in the embodiments or exemplary descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained according to the drawings without paying creative work.

[0049] Figure 1 A flow chart of a method for controlling production of LED track lights based on the Internet of Things provided by one embodiment of the present invention;

[0050] Figure 2 A flowchart of abnormal operation detection of production equipment provided by an embodiment of the present invention;

[0051] Figure 3 A block diagram of an LED track light production control system based on the Internet of Things provided by one embodiment of the present invention;

[0052] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0053] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0055] Figure 1 A flow chart of a method for controlling production of LED track lights based on the Internet of Things provided by one embodiment of the present invention;

[0056] like Figure 1 As shown, the present invention provides a flow chart of a production control method of LED track lights based on the Internet of Things, including:

[0057] Obtain historical production information and production process information, extract the process characteristics of each production process, conduct optimal process analysis for each production process, and construct a production knowledge graph;

[0058] Obtain production demand information, build a production plan formulation model based on the particle swarm algorithm, formulate the optimal production plan for the current desired production product in combination with the production knowledge graph, perform production control and obtain real-time operation monitoring information;

[0059] Building an abnormal operation detection model based on VAE, and performing abnormal operation detection on the production equipment according to the real-time operation monitoring information to obtain abnormal operation detection information;

[0060] Conduct real-time status assessment of abnormally operating equipment and formulate control plans to regulate equipment operation based on the assessment results.

[0061] It should be noted that the present invention provides a production control method and system for LED track lights based on the Internet of Things, which divides the production process of LED track lights into multiple production process subsequences, selects the best production control parameters in each production process subsequence based on historical production data, and considers the status of existing equipment at the same time, thereby formulating the optimal production control plan for the desired production product. In addition, the equipment in production is monitored in real time to detect abnormal operation data, and the equipment status is evaluated based on the abnormal operation data, and a corresponding equipment operation control plan is formulated, thereby improving production efficiency and production quality.

[0062] Furthermore, in a preferred embodiment of the present invention, the acquisition of historical production information and production process information, extraction of process characteristics of each production process, and optimal process analysis of each production process to form a production knowledge graph specifically include:

[0063] Obtain historical production information, including historical production specification information, historical production benefit information, historical production equipment status information and historical production operation parameter information;

[0064] Acquire production process information based on the production line of the LED track light, perform feature extraction on the production process information, and obtain production process feature information;

[0065] De-redundancy and segmentation operations are performed according to the production process characteristic information to form a production process subsequence, and the production process category is set through the production process subsequence to obtain production process category information;

[0066] Introducing the K-means clustering algorithm, initializing the number of cluster centers based on the production process category information, constructing sample data based on the historical production information, and calculating the Euclidean distance between each sample data and each cluster center;

[0067] Obtain the Euclidean distance between each sample data and each cluster center, and sort them according to the Euclidean distance to obtain the sorting result of the Euclidean distance between each sample data and each cluster center, and obtain the cluster center corresponding to the minimum Euclidean distance based on the sorting result;

[0068] The cluster center corresponding to the minimum Euclidean distance is used as the target cluster of the corresponding sample data, all sample data are classified, and a production knowledge graph is constructed in combination with the production process feature information.

[0069] It should be noted that, first, historical production information is obtained and analyzed, including historical production specification information, historical production benefit information, historical production equipment status information, and historical production operation parameter information. Next, the production process information is obtained through the production line of the LED track light, and the features of this information are extracted to obtain the production process feature information. After obtaining the production process feature information, redundancy removal and segmentation operations are performed to form a more concise and representative production process subsequence. Through these subsequences, different production process categories are set. The K-means clustering algorithm is introduced, the number of cluster centers is initialized based on the production process category information, the sample data is constructed using the historical production information obtained previously, and the Euclidean distance between each sample data and each cluster center is calculated. Then, the Euclidean distance between each sample data and each cluster center is obtained, and the distances are sorted according to the distances to obtain the Euclidean distance sorting results of each sample data and each cluster center. According to the sorting results, the minimum Euclidean distance is determined, and the corresponding cluster center is used as the target cluster of the sample data. In this way, all sample data are classified and combined with the production process characteristic information to finally construct a production knowledge graph. Therefore, through the production knowledge graph, we can understand the various production operation parameters under each production process and the equipment status, production efficiency and production specifications corresponding to the production operation parameters.

[0070] Furthermore, in a preferred embodiment of the present invention, the acquisition of production demand information, the construction of a production plan formulation model based on a particle swarm algorithm, the formulation of an optimal production plan for the currently desired product in combination with the production knowledge graph, the production control and the acquisition of real-time operation monitoring information specifically include:

[0071] Acquire production demand information, extract production specification features according to the production demand information, and divide the production process flow to obtain first analysis information;

[0072] Calculate the similarity between the first analysis information and the constructed production knowledge graph to obtain a similarity value, judge the calculated similarity value with a preset threshold, obtain historical production information of each production process according to the judgment result, and obtain second analysis information;

[0073] Extracting the production efficiency characteristics corresponding to each historical production operation parameter based on the second analysis information, sorting each historical production operation parameter, and selecting candidate production operation parameters for each production process according to the sorting result to obtain candidate production operation parameter information;

[0074] Extracting historical production equipment status information corresponding to each candidate production operation parameter according to the candidate production operation parameter information, and associating it with the corresponding candidate production operation parameter to obtain third analysis information;

[0075] Constructing a production plan formulation model based on a particle swarm algorithm, obtaining equipment status information, combining the third analysis information with the input into the production plan formulation model, formulating a production plan for the desired product, and obtaining optimal production plan information;

[0076] Production control is performed according to the optimal production plan information, and real-time operation monitoring of production equipment is performed to obtain real-time operation monitoring information.

[0077] It should be noted that, first, the production demand information is obtained, and the production specification features are extracted, and the production process flow is divided to obtain the first analysis information. Next, the first analysis information is calculated with the already constructed production knowledge graph to obtain the similarity value. The calculated similarity values ​​are compared with the preset threshold value, and according to the comparison result, the historical production information that best matches the current production demand is screened out to obtain the second analysis information. Based on the second analysis information, the production benefit features corresponding to each historical production operation parameter are extracted, and these operation parameters are sorted. According to the sorting result, the candidate production operation parameters of each production process are selected to obtain the candidate production operation parameter information. Then, according to the candidate production operation parameter information, the historical production equipment status information corresponding to each candidate production operation parameter is extracted, and these status information is associated with the candidate production operation parameter to obtain the third analysis information. Next, a production plan formulation model is constructed based on the particle swarm algorithm, the equipment status information is obtained, and the third analysis information is input into the production plan formulation model. Through this model, the optimal production plan information of the desired production product is formulated. Finally, production control is performed according to the optimal production plan information, and real-time operation monitoring of the production equipment is performed to obtain real-time operation monitoring information. Discover and handle any possible abnormalities in a timely manner to ensure the smooth progress of the production process.

[0078] Further, in a preferred embodiment of the present invention, the production plan formulation model is constructed based on the particle swarm algorithm to formulate a production plan for the desired product, specifically including:

[0079] Building a production plan formulation model based on a particle swarm algorithm, introducing a genetic algorithm to optimize the model, and generating an initial particle swarm and an initial population according to the input third analysis information;

[0080] A preset objective function is used to obtain candidate production operation parameters according to the third analysis information, and the production equipment state corresponding to each candidate production operation parameter is used as an auxiliary feature to generate an initial particle swarm for each production process flow.

[0081] Calculate the fitness value of each particle in the initial particle swarm of each production process, compare the fitness value of each particle in each initial particle swarm with the preset threshold, and obtain the global optimal solution based on the judgment result;

[0082] Randomly select several particle individuals from the initial particle group to perform crossover operations to obtain newly generated particle individuals, and perform mutation operations on the newly generated particle individuals to obtain a mutant particle group;

[0083] The speed and position of the mutant particle swarm are updated according to the global optimal solution, a penalty function is constructed based on the input device state information, the global optimal solution is penalized, and the above steps are repeated for iterative optimization until the maximum number of iterations is reached to obtain the optimal solution;

[0084] An initial production plan is generated according to the optimal solution, energy consumption of each initial production plan is calculated to obtain the energy consumption of each initial production plan, the initial production plans are sorted in descending order according to the energy consumption, and the optimal production plan is selected based on the sorting result.

[0085] It should be noted that the production plan formulation model is constructed based on the particle swarm algorithm. Since the particle swarm algorithm may sometimes fall into the problem of local optimal solution, the genetic algorithm is introduced for model optimization. First, the initial particle swarm and the initial population are generated according to the input third analysis information. The objective function is preset, and the candidate production operation parameters are obtained according to the third analysis information. The production equipment status corresponding to these parameters is used as an auxiliary feature to generate the initial particle swarm of each production process. The production equipment status corresponding to the production operation parameters is associated to avoid the problem that the equipment of the current production line cannot use the corresponding parameters when the result is output, thereby reducing the problem of outputting invalid solutions. After the initial particle swarm is generated, the fitness value of each particle is calculated and compared with the preset threshold to obtain the global optimal solution. Determine the best solution in the particle swarm. Then, several particle individuals of the initial particle swarm are randomly selected for crossover operation to obtain newly generated particle individuals. These newly generated particle individuals are mutated to obtain a mutant particle swarm. Then, according to the global optimal solution, the speed and position of the mutant particle swarm are updated. At this time, the updated particle swarm has individuals in the genetic algorithm, further expanding the population size and improving the search range. At the same time, a penalty function is constructed based on the input device status information to penalize the global optimal solution. Repeat the above steps for iterative optimization until the maximum number of iterations is reached to obtain the optimal solution. Finally, an initial production plan is generated based on the optimal solution, and the energy consumption of each initial production plan is calculated to obtain the energy consumption of each plan. Subsequently, the energy consumption of each initial production plan is sorted in descending order, and the optimal production plan is selected based on the sorting results. Ensure that while meeting production needs, energy consumption is reduced as much as possible, production efficiency and economic benefits are improved, and the production process is intelligent and efficient.

[0086] Further, in a preferred embodiment of the present invention, the abnormal operation detection model is constructed based on VAE, and abnormal operation detection of production equipment is performed according to the real-time operation monitoring information to obtain abnormal operation detection information, specifically including:

[0087] Obtaining optimal production plan information, generating an optimal operation working curve of the production equipment according to the optimal production plan information, presetting a normal operation fluctuation range, adding random fluctuations to the optimal operation working curve of the production equipment, performing sample enhancement and forming a training data set;

[0088] An abnormal operation detection model is constructed based on VAE, the model is trained according to the training data set, a Markov chain is introduced to optimize the model, and the input training data is mapped to a low-dimensional space to obtain potential features of the training data;

[0089] A normal operating data benchmark is set based on the optimal working curve, a state space is formed by combining the acquired potential features, and the Euclidean distance between each training data and the normal operating data benchmark is calculated as the state transition probability and a state transition matrix is ​​constructed;

[0090] A reconstructed sequence is generated using a state transfer matrix, a reconstruction error between the sequence and the original sequence is calculated, an error threshold is preset, the calculated reconstruction error is judged against the error threshold, model parameters are adjusted according to the judgment result, and an abnormal operation detection model that meets expectations is obtained;

[0091] Real-time operation monitoring information is obtained and input into the abnormal operation monitoring model to obtain a reconstruction sequence of real-time operation data and calculate a reconstruction error, and abnormal operation data is determined according to the reconstruction error to obtain abnormal operation detection information.

[0092] It should be noted that after obtaining the optimal production plan, the track light production is carried out. At the same time, the actual operation of the production equipment is monitored for anomaly detection. By constructing an abnormal operation detection model for detection, a variational autoencoder (VAE) is used for construction, and a Markov chain is introduced for optimization. Considering the time series properties of the data, a more complete anomaly detection is performed. The optimal working curve is generated according to the optimal production plan, and the model training is carried out based on the optimal working curve. In the actual production process, the production equipment may not be able to operate according to the preset production operation parameters throughout the process, and there will be fluctuations. Therefore, a normal fluctuation range is set, and random fluctuations are added to the optimal working curve according to this range, thereby expanding the training samples. Specifically, the input training data is mapped to a low-dimensional space to extract the potential features of the training data. Based on the optimal working curve, a benchmark for normal operation data is set. Combined with the potential features extracted from the training data, a state space is constructed, and the Euclidean distance between each training data and the benchmark for normal operation data is calculated. These distances are used as state transition probabilities to construct a state transition matrix. The state transition matrix is ​​used to generate a reconstructed sequence, and the reconstruction error between these reconstructed sequences and the original sequence is calculated. An error threshold is set, and the calculated reconstruction error is compared with a preset threshold. According to the comparison results, the model parameters are adjusted until the desired abnormal operation detection model is obtained. Finally, the acquired real-time operation monitoring information is input into the trained abnormal operation detection model. The model generates a reconstruction sequence of real-time operation data and calculates the reconstruction error. By analyzing the reconstruction error, it can be determined whether there is abnormal operation data, thereby obtaining abnormal operation detection information.

[0093] Furthermore, in a preferred embodiment of the present invention, the real-time status evaluation of abnormally operating equipment and the formulation of a control plan to control the operation of the equipment according to the evaluation results specifically include:

[0094] Acquire abnormal operation detection information and normal operation fluctuation range, calculate the deviation value between abnormal operation data and normal operation fluctuation, build a state assessment model, input the calculated deviation value into the state assessment model for assessment, and obtain real-time state assessment information;

[0095] Extract abnormal operation data according to the abnormal operation detection information, obtain expected operation data and preset input control parameters, perform input parameter regulation analysis, calculate the regulated input parameters based on the abnormal operation data, the expected operation data and the preset control parameters, and obtain the regulated input control parameters;

[0096] Calculate the adjusted input control parameters and the preset input control parameters to obtain compensation parameters, and generate an initial equipment control plan based on the compensation parameters;

[0097] Based on the initial equipment control plan combined with the real-time status evaluation information, it is evaluated whether the abnormal equipment can withstand the input parameters after control, and first evaluation information is obtained. According to the first evaluation information, the final equipment control plan is selected to perform equipment operation control.

[0098] It should be noted that, in the production process, after the abnormal operation is found through abnormal operation detection, because the existing operation data is not the expected operation data, but the existing operation data is obtained based on the preset input control parameters, it may be greater than the expected operation data or less than the expected operation data. At this time, in order to obtain the expected operation data, the input control parameters need to be adjusted to avoid affecting the production efficiency and quality. The specific data and equipment where the abnormality occurs can be obtained through the abnormal operation detection information, and then the expected operation data and the preset input control parameters of the corresponding equipment are obtained, such as voltage, current, etc. Then, based on the abnormal data, the expected operation data and the preset input control parameters, the actual input control parameters that reach the expected operation data are calculated. Then the calculated actual input control parameters are calculated with the preset input control parameters to obtain the difference, which is used as the control parameter for control, and the corresponding input control parameters are increased or decreased to obtain the expected operation data. In addition, after obtaining the control parameters, it is necessary to evaluate whether the abnormal equipment corresponding to the input control parameters after the control can bear it, that is, whether it will increase the equipment operation burden. If so, the control parameters are adjusted, and the equipment bearing capacity is considered during the control, so as to improve production efficiency and avoid equipment failure.

[0099] Figure 2 A flowchart of abnormal operation detection of production equipment provided by an embodiment of the present invention;

[0100] like Figure 2 As shown, the present invention provides a flow chart for detecting abnormal operation of production equipment, including:

[0101] S202, monitor the production equipment in real time to obtain real-time operation monitoring information;

[0102] S204, inputting the real-time operation monitoring information into the abnormal operation detection model to perform abnormality detection;

[0103] S206, generating a reconstructed sequence according to the input real-time operation monitoring information, and calculating a reconstruction error with the original sequence;

[0104] S208, judging the calculated reconstruction error and the error threshold, detecting abnormal data according to the judgment result, and obtaining abnormal operation detection information.

[0105] In addition, in the LED track light production control method based on the Internet of Things provided by the present invention, the following steps are also included:

[0106] Monitor the target production line to obtain production line monitoring information, and perform image preprocessing on the production line monitoring information to obtain preprocessing information;

[0107] Performing personnel target detection based on a target detection algorithm combined with the preprocessing information to obtain target detection information, and performing image segmentation on the detected target according to the target detection information to obtain image segmentation information;

[0108] Extracting posture features of the detection target based on the image segmentation information based on a skeleton extraction algorithm, extracting the human skeleton of the detection target from each frame image, and marking key points as skeleton coordinates;

[0109] Generate a skeleton change sequence of the detection target according to the skeleton coordinates, and dynamically combine to obtain the posture change characteristics of the detection target to obtain posture feature information;

[0110] Building an abnormal behavior analysis model based on CNN, inputting the posture feature information into the abnormal behavior analysis model for analysis, and obtaining abnormal behavior analysis information;

[0111] According to the abnormal behavior analysis information, the danger of the detected person is determined, the duration of the dangerous behavior of the monitored person is calculated, and the result of the dangerous behavior is compared with the preset threshold to obtain the danger determination result information;

[0112] A rescue warning is issued according to the danger determination result information.

[0113] It should be noted that during the production of track lights, there may be abnormal situations such as production personnel falling. If there are no other production personnel near the dangerous personnel, safety accidents will occur. By monitoring the production line and analyzing the posture and behavior of the staff, it is possible to determine whether there is danger, and then timely warning and rescue.

[0114] Figure 3 An embodiment of the present invention provides an LED track light production control system 3 based on the Internet of Things, the system comprising: a memory 31, a processor 32, the memory 31 containing an LED track light production control method program based on the Internet of Things, the LED track light production control method program based on the Internet of Things is executed by the processor 32 to implement the following steps:

[0115] Obtain historical production information and production process information, extract the process characteristics of each production process, conduct optimal process analysis for each production process, and construct a production knowledge graph;

[0116] Obtain production demand information, build a production plan formulation model based on the particle swarm algorithm, formulate the optimal production plan for the current desired production product in combination with the production knowledge graph, perform production control and obtain real-time operation monitoring information;

[0117] Building an abnormal operation detection model based on VAE, and performing abnormal operation detection on the production equipment according to the real-time operation monitoring information to obtain abnormal operation detection information;

[0118] Conduct real-time status assessment of abnormally operating equipment and formulate control plans to regulate equipment operation based on the assessment results.

[0119] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0120] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0121] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0122] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0123] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0124] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A production control method for LED track lights based on the Internet of Things, characterized in that: include: Obtain historical production information and production process information, extract the process characteristics of each production process, conduct optimal process analysis for each production process, and construct a production knowledge graph; Obtain production demand information, build a production plan formulation model based on the particle swarm algorithm, formulate the optimal production plan for the current desired production product in combination with the production knowledge graph, perform production control and obtain real-time operation monitoring information; Building an abnormal operation detection model based on VAE, and performing abnormal operation detection on the production equipment according to the real-time operation monitoring information to obtain abnormal operation detection information; Conduct real-time status assessment of abnormally operating equipment and formulate control plans to control equipment operation based on the assessment results; The obtaining of production demand information, building a production plan formulation model based on a particle swarm algorithm, formulating the optimal production plan for the currently expected production product in combination with the production knowledge graph, performing production control and obtaining real-time operation monitoring information specifically include: Acquire production demand information, extract production specification features according to the production demand information, and divide the production process flow to obtain first analysis information; Calculate the similarity between the first analysis information and the constructed production knowledge graph to obtain a similarity value, judge the calculated similarity value with a preset threshold, obtain historical production information of each production process according to the judgment result, and obtain second analysis information; Extracting the production efficiency characteristics corresponding to each historical production operation parameter based on the second analysis information, sorting each historical production operation parameter, and selecting candidate production operation parameters for each production process according to the sorting result to obtain candidate production operation parameter information; Extracting historical production equipment status information corresponding to each candidate production operation parameter according to the candidate production operation parameter information, and associating it with the corresponding candidate production operation parameter to obtain third analysis information; Constructing a production plan formulation model based on a particle swarm algorithm, obtaining equipment status information, combining the third analysis information with the input into the production plan formulation model, formulating a production plan for the desired product, and obtaining optimal production plan information; Performing production control according to the optimal production plan information, and performing real-time operation monitoring of production equipment to obtain real-time operation monitoring information; The production plan formulation model is constructed based on the particle swarm algorithm to formulate a production plan for the desired product, specifically including: Building a production plan formulation model based on a particle swarm algorithm, introducing a genetic algorithm to optimize the model, and generating an initial particle swarm and an initial population according to the input third analysis information; A preset objective function is used to obtain candidate production operation parameters according to the third analysis information, and the production equipment state corresponding to each candidate production operation parameter is used as an auxiliary feature to generate an initial particle swarm for each production process flow. Calculate the fitness value of each particle in the initial particle swarm of each production process, compare the fitness value of each particle in each initial particle swarm with the preset threshold, and obtain the global optimal solution based on the judgment result; Randomly select several particle individuals from the initial particle group to perform crossover operations to obtain newly generated particle individuals, and perform mutation operations on the newly generated particle individuals to obtain a mutant particle group; The speed and position of the mutant particle swarm are updated according to the global optimal solution, a penalty function is constructed based on the input device state information, the global optimal solution is penalized, and iterative optimization is performed until the maximum number of iterations is reached to obtain the optimal solution; An initial production plan is generated according to the optimal solution, energy consumption of each initial production plan is calculated to obtain the energy consumption of each initial production plan, the initial production plans are sorted in descending order according to the energy consumption, and the optimal production plan is selected based on the sorting result.

2. The method for controlling the production of LED track lights based on the Internet of Things according to claim 1, characterized in that: The acquisition of historical production information and production process information, extraction of process characteristics of each production process, and optimal process analysis of each production process to form a production knowledge graph specifically include: Obtain historical production information, including historical production specification information, historical production benefit information, historical production equipment status information and historical production operation parameter information; Acquire production process information based on the production line of the LED track light, perform feature extraction on the production process information, and obtain production process feature information; De-redundancy and segmentation operations are performed according to the production process characteristic information to form a production process subsequence, and the production process category is set through the production process subsequence to obtain production process category information; Introducing the K-means clustering algorithm, initializing the number of cluster centers based on the production process category information, constructing sample data based on the historical production information, and calculating the Euclidean distance between each sample data and each cluster center; Obtain the Euclidean distance between each sample data and each cluster center, and sort them according to the Euclidean distance to obtain the sorting result of the Euclidean distance between each sample data and each cluster center, and obtain the cluster center corresponding to the minimum Euclidean distance based on the sorting result; The cluster center corresponding to the minimum Euclidean distance is used as the target cluster of the corresponding sample data, all sample data are classified, and a production knowledge graph is constructed in combination with the production process feature information.

3. The method for controlling the production of LED track lights based on the Internet of Things according to claim 1, characterized in that: The abnormal operation detection model is constructed based on VAE, and abnormal operation detection is performed on the production equipment according to the real-time operation monitoring information to obtain abnormal operation detection information, specifically including: Obtaining optimal production plan information, generating an optimal operation working curve of the production equipment according to the optimal production plan information, presetting a normal operation fluctuation range, adding random fluctuations to the optimal operation working curve of the production equipment, performing sample enhancement and forming a training data set; An abnormal operation detection model is constructed based on VAE, the model is trained according to the training data set, a Markov chain is introduced to optimize the model, and the input training data is mapped to a low-dimensional space to obtain potential features of the training data; A normal operating data benchmark is set based on the optimal working curve, a state space is formed by combining the acquired potential features, and the Euclidean distance between each training data and the normal operating data benchmark is calculated as the state transition probability and a state transition matrix is ​​constructed; A reconstructed sequence is generated using a state transfer matrix, a reconstruction error between the sequence and the original sequence is calculated, an error threshold is preset, the calculated reconstruction error is judged against the error threshold, model parameters are adjusted according to the judgment result, and an abnormal operation detection model that meets expectations is obtained; Real-time operation monitoring information is obtained and input into the abnormal operation detection model to obtain a reconstruction sequence of real-time operation data and calculate a reconstruction error, and abnormal operation data is determined according to the reconstruction error to obtain abnormal operation detection information.

4. The method for controlling the production of LED track lights based on the Internet of Things according to claim 1, characterized in that: The real-time status evaluation of abnormally operating equipment and the formulation of a control plan for equipment operation control based on the evaluation results specifically include: Obtain abnormal operation detection information and normal operation fluctuation range, calculate the deviation value between abnormal operation data and normal operation fluctuation, build a state assessment model, input the calculated deviation value into the state assessment model for assessment, and obtain real-time state assessment information; Extract abnormal operation data according to the abnormal operation detection information, obtain expected operation data and preset input control parameters, perform input parameter regulation analysis, calculate the regulated input parameters based on the abnormal operation data, the expected operation data and the preset control parameters, and obtain the regulated input control parameters; Calculate the adjusted input control parameters and the preset input control parameters to obtain compensation parameters, and generate an initial equipment control plan based on the compensation parameters; Based on the initial equipment control plan combined with the real-time status evaluation information, it is evaluated whether the abnormal equipment can withstand the input parameters after control, and first evaluation information is obtained. According to the first evaluation information, the final equipment control plan is selected to perform equipment operation control.

5. An LED track light production control system based on the Internet of Things, characterized in that: The system includes: a memory and a processor, wherein the memory contains an LED track light production control method program based on the Internet of Things, and when the LED track light production control method program based on the Internet of Things is executed by the processor, the following steps are implemented: Obtain historical production information and production process information, extract the process characteristics of each production process, conduct optimal process analysis for each production process, and construct a production knowledge graph; Obtain production demand information, build a production plan formulation model based on the particle swarm algorithm, formulate the optimal production plan for the current desired production product in combination with the production knowledge graph, perform production control and obtain real-time operation monitoring information; Building an abnormal operation detection model based on VAE, and performing abnormal operation detection on the production equipment according to the real-time operation monitoring information to obtain abnormal operation detection information; Conduct real-time status assessment of abnormally operating equipment and formulate control plans to control equipment operation based on the assessment results; The obtaining of production demand information, building a production plan formulation model based on a particle swarm algorithm, formulating the optimal production plan for the currently expected production product in combination with the production knowledge graph, performing production control and obtaining real-time operation monitoring information specifically include: Acquire production demand information, extract production specification features according to the production demand information, and divide the production process flow to obtain first analysis information; Calculate the similarity between the first analysis information and the constructed production knowledge graph to obtain a similarity value, judge the calculated similarity value with a preset threshold, obtain historical production information of each production process according to the judgment result, and obtain second analysis information; Extracting the production efficiency characteristics corresponding to each historical production operation parameter based on the second analysis information, sorting each historical production operation parameter, and selecting candidate production operation parameters for each production process according to the sorting result to obtain candidate production operation parameter information; Extracting historical production equipment status information corresponding to each candidate production operation parameter according to the candidate production operation parameter information, and associating it with the corresponding candidate production operation parameter to obtain third analysis information; Constructing a production plan formulation model based on a particle swarm algorithm, obtaining equipment status information, combining the third analysis information with the input into the production plan formulation model, formulating a production plan for the desired product, and obtaining optimal production plan information; Performing production control according to the optimal production plan information, and performing real-time operation monitoring of production equipment to obtain real-time operation monitoring information; The production plan formulation model is constructed based on the particle swarm algorithm to formulate a production plan for the desired product, specifically including: Building a production plan formulation model based on a particle swarm algorithm, introducing a genetic algorithm to optimize the model, and generating an initial particle swarm and an initial population according to the input third analysis information; A preset objective function is used to obtain candidate production operation parameters according to the third analysis information, and the production equipment state corresponding to each candidate production operation parameter is used as an auxiliary feature to generate an initial particle swarm for each production process flow. Calculate the fitness value of each particle in the initial particle swarm of each production process, compare the fitness value of each particle in each initial particle swarm with the preset threshold, and obtain the global optimal solution based on the judgment result; Randomly select several particle individuals from the initial particle group to perform crossover operations to obtain newly generated particle individuals, and perform mutation operations on the newly generated particle individuals to obtain a mutant particle group; The speed and position of the mutant particle swarm are updated according to the global optimal solution, a penalty function is constructed based on the input device state information, the global optimal solution is penalized, and iterative optimization is performed until the maximum number of iterations is reached to obtain the optimal solution; An initial production plan is generated according to the optimal solution, energy consumption of each initial production plan is calculated to obtain the energy consumption of each initial production plan, the initial production plans are sorted in descending order according to the energy consumption, and the optimal production plan is selected based on the sorting result.

6. The LED track light production control system based on Internet of Things according to claim 5, characterized in that: The acquisition of historical production information and production process information, extraction of process characteristics of each production process, and optimal process analysis of each production process to form a production knowledge graph specifically include: Obtain historical production information, including historical production specification information, historical production benefit information, historical production equipment status information and historical production operation parameter information; Acquire production process information based on the production line of the LED track light, perform feature extraction on the production process information, and obtain production process feature information; De-redundancy and segmentation operations are performed according to the production process characteristic information to form a production process subsequence, and the production process category is set through the production process subsequence to obtain production process category information; Introducing the K-means clustering algorithm, initializing the number of cluster centers based on the production process category information, constructing sample data based on the historical production information, and calculating the Euclidean distance between each sample data and each cluster center; Obtain the Euclidean distance between each sample data and each cluster center, and sort them according to the Euclidean distance to obtain the sorting result of the Euclidean distance between each sample data and each cluster center, and obtain the cluster center corresponding to the minimum Euclidean distance based on the sorting result; The cluster center corresponding to the minimum Euclidean distance is used as the target cluster of the corresponding sample data, all sample data are classified, and a production knowledge graph is constructed in combination with the production process feature information.

Images