Distributed collaborative control method and system for silicone rubber glass fiber sleeve production equipment

Abstract

The application discloses a distributed collaborative control method and system for a silicon rubber glass fiber sleeve production equipment, and relates to the technical field of industrial automation control.The application ensures the non-tamperability and credibility of data through blockchain technology, solves the problems of easy data tampering and lack of trust mechanism in traditional methods, realizes efficient collaboration and task allocation between equipment by combining a multi-agent system, optimizes equipment operating parameters, improves production efficiency and product quality, constructs a virtual model by using digital twin technology, simulates and verifies the optimized parameters, further ensures the achievement of production targets, and dynamically adjusts equipment operating parameters through a global optimization algorithm and a feedback control mechanism, thereby enhancing the flexibility and dynamic response capability of the system.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of industrial automation control, and particularly relates to a distributed collaborative control method and system for a silicone rubber glass fiber sleeve production device. BACKGROUND

[0002] Intelligent control of a silicone rubber glass fiber sleeve production device is an important research field of high-end manufacturing, which is of key significance to improving production efficiency and guaranteeing product quality. At present, the traditional centralized control method has problems such as low data processing efficiency, slow response speed and insufficient system flexibility when facing complex production environments, and cannot meet the dynamic production requirements. In addition, the existing scheme performs poorly in data security and device collaboration, limiting the intelligent level of the production system. Under this background, distributed collaborative control has become a core challenge to be broken through. Firstly, the production device needs to collect and process massive multi-dimensional data in real time, but the traditional data management method cannot guarantee the security and tamper resistance of the data, resulting in a lack of trust mechanism. The unsolved data security problem further restricts the efficient collaboration between devices, because each device needs to rely on trusted data for interaction and decision-making. The lack of collaboration mechanism directly affects the system's ability to quickly respond to dynamic production targets, especially in complex production scenarios, where devices are difficult to adjust parameters to optimize overall performance. SUMMARY

[0003] The application aims to provide a distributed collaborative control method and system for a silicone rubber glass fiber sleeve production device, which solves the shortcomings of traditional control methods through the combination of blockchain, multi-agent system and digital twin technology, and realizes efficient collaboration and dynamic optimization of the production system.

[0004] The application can be achieved by the following technical solutions:

[0005] The application provides a distributed collaborative control method for a silicone rubber glass fiber sleeve production device, including the following steps:

[0006] The real-time multi-dimensional data set is stored by using the blockchain technology, the data is judged whether it is tampered with through the distributed ledger synchronization, and the tamper-proof data record is obtained;

[0007] The device state parameters are obtained from the tamper-proof data record, the multi-agent system is used to distribute the agent role, and the task allocation scheme is obtained;

[0008] The distributed decision algorithm is executed through the multi-agent system, the running parameter adjustment value of each device is calculated, and the optimized device parameter set is obtained;

[0009] An adjustment instruction is acquired from the optimized device parameter set, a virtual model of the production device is constructed by using a digital twin technology, simulation verification is performed on the optimized device parameter set, and a virtual optimization result is obtained;

[0010] A global optimization algorithm is used in combination with a production objective function to calculate an optimal parameter combination, and a globally optimized parameter is obtained;

[0011] A final adjustment instruction is acquired from the globally optimized parameter, and is issued to each device through a distributed control network to update a running state in real time, new multidimensional data streams are collected, and system dynamic responses are judged to obtain an updated production state;

[0012] A feedback control algorithm is used to analyze deviations of newly collected multidimensional data from expected targets, to calculate deviation correction values, and to redistribute tasks through a multi-agent system to obtain a corrected task scheme.

[0013] Further, an unforgeable data record is obtained, and specifically includes:

[0014] Real-time multidimensional data sets are acquired by using a blockchain technology, a hash timestamp record is generated by using a hash function, is stored in a distributed ledger, and an initial data record is obtained, when hash values of data of device nodes in the distributed ledger are consistent, the latest data record is acquired through a data synchronization protocol, and data consistency is determined;

[0015] A consensus algorithm is used to verify hash timestamps in the distributed ledger, if the verification is passed, the latest data record is stored through the blockchain technology, and unforgeable data is obtained;

[0016] Data records in the distributed ledger are acquired through a real-time processing module, a data check value is generated by using an encryption algorithm, it is judged whether the data is tampered with, when the data check value is consistent with the hash timestamp, the check value is synchronized through an inter-device communication protocol, and unforgeable data records are determined;

[0017] According to the synchronized check value, a data consistency protocol is used to update the multidimensional data set, to generate a final data record, to store the final data record through the blockchain technology, to acquire a hash timestamp in the distributed ledger, and to determine that the data is unforgeable.

[0018] Further, a task allocation scheme is obtained, and specifically includes:

[0019] S21, device state parameters are acquired from the unforgeable data record, data integrity is verified by using a blockchain technology, a trusted state parameter set is obtained, agent nodes are initialized through a multi-agent system, the trusted state parameter set is distributed to each agent according to a preset role allocation rule, and an agent role allocation table is determined;

[0020] S22, each intelligent agent performs distributed data analysis on the assigned parameter set according to the cooperation agreement, divides the equipment state category by using the clustering algorithm, obtains the equipment state classification result, when there is an abnormal category in the equipment state classification result, triggers the task allocation mechanism through the cooperation agreement to generate a preliminary task allocation scheme, and when there is no abnormality, maintains the current task allocation;

[0021] S23, according to the preliminary task allocation scheme, the genetic algorithm is used to optimize the cooperative task allocation between devices, and an optimized task allocation scheme is obtained, and then the device cooperation parameters are extracted from the optimized task allocation scheme and distributed to each device through the message passing interface to determine the device cooperation execution instruction;

[0022] S24, obtaining the running log of the device cooperation execution instruction, using time series analysis to verify the task execution effect, and obtaining the verification result of the final task allocation scheme.

[0023] Further, the optimized device parameter set is obtained, specifically including:

[0024] S31, obtaining real-time data stream, collecting each device operating parameter through a multi-agent system, and then processing through a distributed decision algorithm to calculate the adjustment value of each device operating parameter, and obtaining an adjustment value set;

[0025] S32, when the parameter value in the adjustment value set exceeds the preset threshold value, the adjustment value is normalized through a data processing module to obtain a normalized adjustment value set, and then each device operating parameter is updated through a multi-agent system to obtain an updated parameter set;

[0026] S33, obtaining the dynamic indicators of the production target through the real-time data stream, judging whether the updated parameter set meets the production target, obtaining the meeting state, and when the meeting state indicates that the parameters do not completely meet the production target, recalculating the adjustment value through a distributed decision algorithm to obtain a new adjustment value set;

[0027] S34, according to the new adjustment value set, repeatedly updating the parameters and judging the target meeting state to obtain the final optimized parameter set.

[0028] Further, the virtual optimization result is obtained, specifically including:

[0029] Obtaining the optimized device parameter set, generating an adjustment instruction set by analyzing the parameter set, determining the integrity of the instruction set, then collecting real-time running data from the production equipment, constructing a digital twin model, mapping the device running state, and obtaining a virtual running state;

[0030] When the deviation between the virtual running state and the real-time running data exceeds the preset threshold value, the digital twin model parameters are updated through iteration to adjust the model state, and a calibrated virtual running state is obtained;

[0031] With the calibrated virtual running state, load the adjustment instruction set, execute parameter adjustment simulation, and obtain simulation running results;

[0032] By analyzing the simulation running results, extracting key performance indicators, judging whether the adjustment instruction set meets the optimization target, and obtaining verification results, when the verification results do not reach the preset optimization threshold, the adjustment instruction set is optimized through a genetic algorithm to generate a new instruction set, and an optimized adjustment instruction set is obtained;

[0033] According to the optimized adjustment instruction set, update the digital twin model parameters, and execute the final simulation to obtain the virtual optimization results.

[0034] Further, the global optimization parameters are obtained, specifically including:

[0035] Obtain production data, extract feature values from efficiency indicators, quality indicators, and energy consumption indicators, and then use a preset weight distribution to construct a target function and determine the target function expression;

[0036] Through a global optimization algorithm, iteratively search for parameter combinations from the initial data set to obtain a candidate parameter set, and then calculate the target function value. When the function value meets the preset threshold, it is determined as the optimal parameter. When it does not meet, adjust the parameter combination for continuous iteration;

[0037] According to the optimal parameters, generate production control instructions to obtain the optimized production configuration, obtain the optimized production data, calculate the efficiency indicators, quality indicators, and energy consumption indicators, and judge whether the production target is met. By comparing the data before and after optimization, extract performance improvement features, and obtain the final optimization results.

[0038] Further, the updated production state is obtained, specifically including:

[0039] Extract the final adjustment instructions from the global optimization parameters, use the parameter extraction process, process the global optimization parameters through a preset extraction algorithm to obtain the final adjustment instructions, and then transmit the final adjustment instructions through a distributed control network. Using the instruction transmission mechanism, if the network bandwidth meets the preset threshold, the final adjustment instructions are distributed to each device, and the instruction transmission is determined to be complete;

[0040] Real-time update of the running state of each device, obtain the device list of instruction transmission completion, collect the real-time running state of each device through the state update frequency configuration to obtain real-time running state data, and then collect multi-dimensional data streams. For real-time running state data, use the data stream collection mechanism to obtain multi-dimensional data streams through a sensor interface to obtain a multi-dimensional data stream set;

[0041] The target judgment criterion is adopted to determine whether the multi-dimensional data stream set reaches the dynamic response target, and if the key indicators in the multi-dimensional data stream set meet the preset threshold, it is determined that the dynamic response target is reached, and a target achievement state is obtained.

[0042] The production state is updated according to the target achievement state, the target achievement state is obtained through a production state updating mechanism, updated production state data is generated, a feedback optimization algorithm is used to adjust the global optimization parameters, and optimized global optimization parameters are obtained.

[0043] Further, before obtaining the tamper-proof data record, it further includes: obtaining multi-dimensional data stream from production equipment sensors, pre-processing the data using edge computing nodes, filtering noise data, generating standardized data packets, and obtaining real-time multi-dimensional data sets.

[0044] Further, it further includes: obtaining adjusted operating parameters from the corrected task plan, re-simulating and verifying through a digital twin model, updating the virtual model state, determining whether the production efficiency reaches the preset target, and obtaining the final optimized production parameters.

[0045] The present application provides a distributed collaborative control system for a silicone rubber glass fiber sleeve production device, which is used to realize a distributed collaborative control method for a silicone rubber glass fiber sleeve production device, and specifically includes:

[0046] The data acquisition and preprocessing module acquires multi-dimensional data stream from production equipment sensors, pre-processes and generates standardized data sets through edge computing, and at the same time marks and processes abnormal data;

[0047] The blockchain data storage and verification module stores the multi-dimensional data set using blockchain technology, generates a hash timestamp record, and determines that the data is tamper-proof through a consensus algorithm and data verification;

[0048] The multi-agent task allocation module obtains device state parameters from the blockchain, allocates tasks through a multi-agent system, and optimizes the task allocation scheme using clustering and genetic algorithms;

[0049] The distributed decision-making and parameter optimization module executes a distributed decision-making algorithm through a multi-agent system based on the task allocation scheme, dynamically adjusts the device operating parameters, and optimizes the production target;

[0050] The digital twin simulation verification module constructs a digital twin model of the production device, verifies the parameter adjustment effect through simulation, and ensures that the virtual running state is consistent with the actual production target;

[0051] The global optimization module combines the production objective function, calculates the optimal parameter combination using a global optimization algorithm, and improves the production efficiency, quality and energy consumption performance;

[0052] The distributed control and dynamic response module converts the global optimization parameters into adjustment instructions, and sends the adjustment instructions to the equipment through a distributed network, updates the running state in real time, and judges the dynamic response effect;

[0053] The feedback control and task correction module analyzes the deviation and calculates the correction value by using a feedback control algorithm according to the production state, reassigns the tasks through a multi-agent system, and optimizes the production process.

[0054] The model updating and verification module obtains the running parameters from the corrected task scheme, updates the digital twin model, and simulates and verifies again to ensure that the production efficiency reaches the preset target.

[0055] The beneficial effects of the present application are:

[0056] The present application stores and verifies real-time multi-dimensional data sets through blockchain technology, solves the problem of data tampering and lack of trust mechanism in traditional centralized control methods, uses hash timestamp records and consensus algorithms to ensure data tamper resistance and consistency, provides a trusted data foundation for efficient collaboration between devices, and further solves the problem of insufficient collaboration between devices by assigning tasks and optimizing parameters through a multi-agent system, thereby improving the intelligent level and response capability of the production system.

[0057] By combining global optimization algorithms and digital twin technology, through simulation verification and parameter optimization, the present application solves the problem of slow response speed and insufficient flexibility in complex production environments, can adjust device running parameters in real time, ensures that production efficiency and product quality reach the preset target, and further improves the stability and reliability of production efficiency, significantly improving the overall performance of the production system.

[0058] The present application adopts a distributed control network and edge computing technology to solve the problem of insufficient adaptability of traditional centralized control methods in dynamic production environments, the system can collect and process multi-dimensional data streams in real time, quickly respond to dynamic changes in the production process, and through feedback control algorithms and digital twin model updating and verification, the system can flexibly adjust task allocation and running parameters to ensure the efficiency and stability of the production process, meeting the requirements of industrial production system intelligentization and dynamic response. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to better understand and implement, the technical solutions of the present application are described in detail below with reference to the accompanying drawings.

[0060] Figure 1 The flowchart of the distributed collaborative control method of the silicone rubber glass fiber sleeve production equipment provided in Embodiment 1 of the present application is shown in the figure.

[0061] Figure 2A flowchart of a task allocation scheme obtained by the distributed collaborative control method of the silicone rubber glass fiber sleeve production equipment provided in Embodiment 1 of the present application;

[0062] Figure 3 A flowchart of an optimized equipment parameter set obtained by the distributed collaborative control method of the silicone rubber glass fiber sleeve production equipment provided in Embodiment 1 of the present application;

[0063] Figure 4 A structure diagram of the distributed collaborative control system of the silicone rubber glass fiber sleeve production equipment provided in Embodiment 2 of the present application. DETAILED DESCRIPTION

[0064] To further clarify the technical means and effects adopted by the present application to achieve the predetermined object of the application, exemplary embodiments will be described in detail herein, which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Rather, they are merely examples of methods and systems consistent with some aspects of the present application as detailed in the appended claims.

[0065] The terms used in the present application are merely for the purpose of describing specific embodiments and are not intended to limit the present application. The singular forms "a," "an," and "the" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0066] The specific embodiments, features and effects according to the present application are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0067] Embodiment 1

[0068] Please refer to Figures 1-3 The present embodiment provides a distributed collaborative control method of silicone rubber glass fiber sleeve production equipment, comprising the following steps:

[0069] S1, using blockchain technology to store real-time multi-dimensional data sets, generating hash-based timestamp records, synchronizing through distributed ledgers between devices, judging whether the data is tampered with, and obtaining tamper-proof data records;

[0070] Further, obtaining tamper-proof data records specifically includes:

[0071] Real-time multi-dimensional data sets are obtained through blockchain technology, a hash timestamp record is generated using a hash function, is stored in a distributed ledger, and initial data records are obtained, when the data hash values of each device node in the distributed ledger are consistent, the latest data records are obtained through a data synchronization protocol to determine data consistency;

[0072] The hash timestamp in the distributed ledger is verified using a consensus algorithm, and if the verification is passed, the latest data records are stored through blockchain technology to obtain tamper-proof data;

[0073] Data records in the distributed ledger are obtained through a real-time processing module, a data check value is generated using an encryption algorithm, and it is determined whether the data has been tampered with, when the data check value and the hash timestamp are consistent, the check value is synchronized through an inter-device communication protocol to determine tamper-proof data records;

[0074] According to the synchronized check value, the multi-dimensional data set is updated using a data consistency protocol to generate final data records, and the final data records are stored through blockchain technology to obtain the hash timestamp in the distributed ledger to determine that the data is tamper-proof.

[0075] The process of generating a data check value using an encryption algorithm includes: obtaining stored real-time multi-dimensional data records from the distributed ledger, the data records contain key parameters in the production device running process, such as temperature, pressure, speed and material thickness, etc., then select a suitable encryption algorithm (such as SHA-256 hash algorithm), and encrypt these data, the encryption algorithm will generate a unique check value (hash value) according to the data content, this check value has one-way and uniqueness, that is, the same input data will produce the same check value, and any slight data change will cause the check value to change greatly. By comparing the generated check value with the hash timestamp record stored in the distributed ledger, it can be determined whether the data has been tampered with during transmission or storage.

[0076] Specifically, reliable storage and verification of real-time multi-dimensional data sets of production devices are achieved. Hash timestamp records are generated using blockchain technology and are synchronized between devices through a distributed ledger to ensure data tamper resistance and consistency, the integrity and authenticity of the data are further verified through a consensus algorithm and a data check mechanism, and finally tamper-proof data records are generated, which not only improves the security and credibility of the data, but also provides an accurate and reliable data basis for subsequent production process control, thereby enhancing the intelligence and collaboration capabilities of the entire production system.

[0077] S2, obtain device state parameters from tamper-proof data records, assign agent roles using a multi-agent system, each agent analyzes data according to a predetermined collaboration protocol to determine device collaboration task allocation, and obtain a task allocation scheme;

[0078] Further, the task allocation scheme is obtained, specifically including:

[0079] S21, obtaining the device state parameters from the tamper-proof data record, verifying the data integrity by using the blockchain technology, obtaining the trusted state parameter set, initializing the agent node by using the multi-agent system, distributing the trusted state parameter set to each agent according to the preset role allocation rule, and determining the agent role allocation table;

[0080] S22, each agent performs distributed data analysis on the allocated parameter set according to the cooperation protocol, divides the device state category by using the clustering algorithm, obtains the device state classification result, and when there is an abnormal category in the device state classification result, triggers the task allocation mechanism through the cooperation protocol to generate a preliminary task allocation scheme, and when there is no abnormality, maintains the current task allocation;

[0081] S23, according to the preliminary task allocation scheme, using a genetic algorithm to optimize the collaborative task allocation between devices, obtaining an optimized task allocation scheme, extracting device cooperation parameters from the optimized task allocation scheme, and distributing them to each device through a message passing interface to determine the device cooperation execution instruction;

[0082] S24, obtaining the running log of the device cooperation execution instruction, verifying the task execution effect by using time series analysis, and obtaining the verification result of the final task allocation scheme.

[0083] Specifically, the blockchain technology is used to ensure the credibility and integrity of the data, and the distributed data analysis of the multi-agent system and the optimization algorithm are combined to realize the intelligentization and dynamic adjustment of the task allocation, which not only improves the collaboration efficiency of the production equipment, but also enhances the flexibility and response capability of the system, and finally significantly improves the intelligent level and overall production efficiency of the production process.

[0084] S3, according to the task allocation scheme, executing the distributed decision algorithm by using the multi-agent system, calculating the running parameter adjustment value of each device, judging whether the parameter meets the production target based on the real-time data stream, and obtaining the optimized device parameter set;

[0085] Further, the optimized device parameter set is obtained, specifically including:

[0086] S31, obtaining the real-time data stream, performing distributed collection on the running parameters of each device by using the multi-agent system, and then processing by using the distributed decision algorithm, calculating the adjustment value of each device running parameter, and obtaining the adjustment value set;

[0087] S32, when the parameter value in the adjustment value set exceeds the preset threshold value, the adjustment value is normalized by the data processing module to obtain a normalized adjustment value set, and the multi-agent system is used to update the device operation parameters to obtain an updated parameter set;

[0088] S33, the dynamic indicators of the production target are obtained through real-time data flow, and it is judged whether the updated parameter set meets the production target to obtain a satisfaction state, when the satisfaction state indicates that the parameters do not completely meet the production target, the adjustment value is recalculated through a distributed decision algorithm to obtain a new adjustment value set;

[0089] S34, according to the new adjustment value set, the parameters are repeatedly updated and the target satisfaction state is judged to obtain a final optimized parameter set.

[0090] Specifically, through the multi-agent system and the distributed decision algorithm, the device operation parameters are collected and processed in real time, the adjustment value is dynamically calculated and normalized, the parameters are repeatedly optimized according to the dynamic indicators of the production target, and finally the optimized parameter set meeting the production demand is obtained, which significantly improves the accuracy and efficiency of the production process.

[0091] S4, obtain adjustment instructions from the optimized device parameter set, use digital twin technology to build a virtual model of the production device, map the real-time running state, verify the parameter adjustment effect through simulation, and obtain a virtual optimization result;

[0092] Further, the virtual optimization result specifically includes:

[0093] Obtain the optimized device parameter set, generate the adjustment instruction set by analyzing the parameter set, determine the integrity of the instruction set, then collect real-time running data from the production device, build a digital twin model, map the device running state, and obtain a virtual running state;

[0094] When the deviation between the virtual running state and the real-time running data exceeds the preset threshold value, update the parameters of the digital twin model through iteration to adjust the model state, and obtain a calibrated virtual running state;

[0095] Load the adjustment instruction set using the calibrated virtual running state, execute parameter adjustment simulation, and obtain a simulation running result;

[0096] By analyzing the simulation running result, extracting key performance indicators, judging whether the adjustment instruction set meets the optimization target, and obtaining a verification result, when the verification result does not reach the preset optimization threshold value, the adjustment instruction set is optimized through a genetic algorithm to generate a new instruction set, and an optimized adjustment instruction set is obtained;

[0097] According to the optimized adjustment instruction set, update the parameters of the digital twin model, and execute the final simulation to obtain a virtual optimization result.

[0098] Specifically, through digital twinning technology and simulation verification, combined with genetic algorithm optimization adjustment instruction set, the virtual optimization of production equipment operation parameters is realized. This process can quickly and efficiently verify and optimize the device parameter adjustment scheme, reduce the trial and error cost in actual production, improve production efficiency and product quality, and enhance the flexibility and adaptability of the production system.

[0099] S5、According to the virtual optimization result, a global optimization algorithm is used to calculate the optimal parameter combination in combination with the production objective function F(x) = w1efficiency + w2quality + w3energy consumption (w1, w2, w3 are weights, and x is a parameter set), and it is determined whether the preset threshold is met to obtain the global optimization parameter;

[0100] Further, the global optimization parameter is obtained, specifically including:

[0101] The production data is obtained, the characteristic values are extracted from the efficiency index, quality index and energy consumption index, and then the preset weight distribution is used to construct the objective function F(x) = w1efficiency + w2quality + w3energy consumption (w1, w2, w3 are weights, and x is a parameter set) to determine the objective function expression;

[0102] Through the global optimization algorithm, the parameter combination is iteratively searched from the initial data set to obtain a candidate parameter set, and then the objective function value is calculated. When the function value meets the preset threshold, it is determined as the optimal parameter, and when it does not meet, the parameter combination is adjusted for further iteration.

[0103] According to the optimal parameters, production control instructions are generated to obtain the optimized production configuration, and the optimized production data is obtained to calculate the efficiency index, quality index and energy consumption index, and it is determined whether the production target is met. By comparing the data before and after optimization, the performance improvement characteristics are extracted to obtain the final optimization result.

[0104] Among them, according to the optimal parameters obtained by the global optimization algorithm, the system will analyze these parameters and generate specific production control instructions. These instructions are sent to each production equipment through a distributed control network to guide the equipment to adjust the operating state according to the optimized parameters, thereby realizing efficient cooperation and performance improvement of the production process, and finally forming the optimized production configuration to ensure that the production efficiency, quality and energy consumption indicators reach the best balance.

[0105] Specifically, through the global optimization algorithm combined with the production objective function, the efficiency, quality and energy consumption indicators are systematically analyzed, and the parameter combination is dynamically adjusted to find the optimal solution. This process not only ensures the global optimization of production parameters, but also intuitively shows the performance improvement effect by comparing the data before and after optimization, thereby significantly improving the production efficiency, product quality and energy utilization efficiency, and realizing the fine management and optimization control of the production process.

[0106] S6, obtaining the final adjustment instruction from the global optimization parameter, issuing to each device through the distributed control network, updating the running state in real time, collecting new multi-dimensional data flow, judging whether the system reaches the dynamic response target, obtaining the updated production state;

[0107] Further, obtaining the updated production state, specifically comprising:

[0108] Extracting the final adjustment instruction from the global optimization parameter, using the parameter extraction process, processing the global optimization parameter through the preset extraction algorithm (such as linear regression algorithm) to obtain the final adjustment instruction, and then transmitting the final adjustment instruction through the distributed control network, using the instruction transmission mechanism, if the network bandwidth meets the preset threshold, the final adjustment instruction is distributed to each device, and the instruction transmission is completed;

[0109] Updating the running state of each device in real time, obtaining the device list of the completed instruction transmission, collecting the real-time running state of each device through the state update frequency configuration, obtaining the real-time running state data, and then collecting the multi-dimensional data flow, using the data flow collection mechanism for the real-time running state data, obtaining the multi-dimensional data flow set through the sensor interface;

[0110] Judging whether the multi-dimensional data flow set reaches the dynamic response target, using the target judgment standard, if the key indicators in the multi-dimensional data flow set meet the preset threshold, it is confirmed that the dynamic response target is reached, and the target achievement state is obtained;

[0111] Updating the production state according to the target achievement state, obtaining the target achievement state through the production state update mechanism, generating updated production state data, and then adjusting the global optimization parameter using the feedback optimization algorithm (such as gradient descent algorithm) to obtain the optimized global optimization parameter.

[0112] Specifically, the final adjustment instruction is extracted from the global optimization parameter and distributed to each device, the running state is updated in real time and the multi-dimensional data flow is collected, it is judged whether the dynamic response target is reached, and the production state is updated and the global parameter is optimized accordingly, the dynamic adjustment and optimization of the production process are realized, and the self-adaptability and intelligent level of the system are improved.

[0113] S7, according to the updated production state, using the feedback control algorithm to analyze the deviation between the newly collected multi-dimensional data and the expected target, calculating the deviation correction value, redistributing the tasks through the multi-agent system, and obtaining the corrected task scheme.

[0114] Further, obtaining the corrected task scheme, specifically comprising:

[0115] According to the updated production state, a feedback control algorithm is used to analyze the newly collected multi-dimensional data stream, and each parameter (such as temperature, pressure, speed, material thickness, etc.) in the multi-dimensional data stream is compared with the preset production target to calculate the deviation between the actual value and the target value of each parameter.

[0116] Based on the deviation value, a corresponding correction value is calculated to adjust the operating parameters of the equipment, and the correction value is transmitted to the multi-agent system. Each agent re-evaluates its task allocation based on the corrected parameters.

[0117] Through the cooperation and negotiation mechanism between agents, each agent dynamically adjusts its task allocation according to the corrected parameters and task requirements, and obtains a corrected task plan. After adjusting the operating parameters of each device, it can work better in cooperation, and finally forms a corrected task plan to optimize the efficiency and quality of the entire production process.

[0118] Specifically, by analyzing the deviation between the newly collected multi-dimensional data and the production target through the feedback control algorithm, calculating the correction value and transmitting it to the multi-agent system, each agent dynamically adjusts the task allocation based on the correction value to optimize the cooperation between devices. This process realizes real-time adjustment and optimization of production tasks, improves the flexibility and response capability of the production system, and significantly improves the production efficiency and product quality.

[0119] Further, before obtaining the tamper-proof data record, it further includes: obtaining a multi-dimensional data stream from a production equipment sensor, including temperature, pressure, speed, and material thickness; using an edge computing node to preprocess the data, filter noise data, generate a standardized data package, and obtain a real-time multi-dimensional data set.

[0120] Further, obtaining a real-time multi-dimensional data set specifically includes:

[0121] Obtaining a multi-dimensional data stream from a production equipment sensor, including temperature, pressure, speed, and material thickness; using an edge computing node to preprocess the data, filter noise, generate a standardized data package, and obtain a real-time multi-dimensional data set; when any dimension data in the multi-dimensional data stream exceeds the preset threshold, marking the dimension data to generate an abnormal marked data set;

[0122] According to the abnormal marked data set, a K-means clustering algorithm is used to classify the marked data to obtain an abnormal data class set, and then the statistical characteristics of each class are obtained, including mean and variance, to generate an abnormal feature data set. When the variance in the abnormal feature data set is greater than the preset threshold, a sliding window method is used to perform time series analysis on the abnormal feature data set to obtain an abnormal change trend.

[0123] According to the abnormal change trend, a decision tree algorithm is used to classify the trend and determine whether the abnormality is persistent, to obtain an abnormality persistence judgment result, and through the abnormality persistence judgment result, an abnormal data processing instruction is generated and output to the edge computing node to obtain an updated real-time multi-dimensional data set.

[0124] Specifically, by preprocessing, abnormal marking, clustering analysis, feature extraction, trend analysis and abnormal processing of the multi-dimensional data stream collected by the production equipment sensor, noise is effectively filtered and abnormal data is identified, ensuring the quality and reliability of the data, providing accurate data basis for subsequent production control and optimization, and thus improving the stability and intelligent level of the production system.

[0125] Further, it also includes: obtaining the adjusted running parameters from the corrected task scheme, re-simulating and verifying through the digital twin model, updating the virtual model state, judging whether the production efficiency reaches the preset target, and obtaining the final optimized production parameters.

[0126] Further, obtaining the final optimized production parameters specifically includes:

[0127] Obtaining the adjusted running parameters from the task scheme, parsing the parameter configuration, extracting the key running indicators, obtaining the optimized running parameter set, and then loading the optimized running parameter set through the digital twin model to perform simulation verification, generate a virtual production scene, and obtain simulation result data;

[0128] Extracting the virtual model state from the simulation result data, updating the running state parameters of the virtual model, and obtaining the updated model state;

[0129] According to the updated model state, calculating the production efficiency index, if the efficiency index reaches the preset target, determining the current parameters as the optimized parameters, if not, adjusting the parameters and returning to the simulation verification, obtaining the optimized parameter set;

[0130] Extracting the key running parameters from the optimized parameter set, generating a parameter adjustment instruction, updating the task scheme configuration, and obtaining an updated task scheme;

[0131] Through the digital twin model, the updated task scheme is simulated and verified again to generate a production efficiency report, judge whether the efficiency is stable, and obtain the final production parameters, extract the running configuration from the final production parameters, verify through the preset threshold, generate a production deployment instruction, and determine the deployment state of the production parameters.

[0132] Specifically, by obtaining the adjusted operating parameters from the revised task plan and utilizing the digital twin model for simulation verification, the system can dynamically update the virtual model state and assess whether the production efficiency meets the preset target, ensuring not only the optimization of production parameters but also the further improvement of the stability and reliability of production efficiency through multiple simulation verifications and parameter adjustments. Ultimately, the generated production deployment instructions provide verified optimal operating parameters for actual production, significantly improving the overall performance and intelligent level of the production system.

[0133] Embodiment 2

[0134] Please refer to Figure 4 The embodiment provides a distributed collaborative control system for a silicone rubber glass fiber sleeve production equipment, which is used to realize a distributed collaborative control method for a silicone rubber glass fiber sleeve production equipment, and specifically comprises the following steps:

[0135] A data acquisition and preprocessing module acquires multi-dimensional data streams from production equipment sensors, pre-processes and generates standardized data sets through edge computing, and simultaneously marks and processes abnormal data.

[0136] A blockchain data storage and verification module stores multi-dimensional data sets using blockchain technology, generates a hash timestamp record, and determines data immutability through consensus algorithms and data verification.

[0137] A multi-agent task allocation module acquires device state parameters from the blockchain, allocates tasks through a multi-agent system, and optimizes task allocation schemes using clustering and genetic algorithms.

[0138] A distributed decision-making and parameter optimization module executes a distributed decision-making algorithm through a multi-agent system based on the task allocation scheme, dynamically adjusts device operating parameters, and optimizes production targets.

[0139] A digital twin simulation verification module constructs a digital twin model of the production equipment, verifies the effects of parameter adjustments through simulation, and ensures that the virtual operating state is consistent with the actual production target.

[0140] A global optimization module calculates the optimal parameter combination using a global optimization algorithm in combination with the production objective function, improving production efficiency, quality, and energy consumption performance.

[0141] A distributed control and dynamic response module converts the global optimization parameters into adjustment instructions, which are issued to the equipment through a distributed network, updates the operating state in real time, and judges the dynamic response effect.

[0142] A feedback control and task correction module analyzes deviations and calculates correction values using a feedback control algorithm based on production states, reallocates tasks through a multi-agent system, and optimizes the production process.

[0143] The model updating and verifying module obtains the running parameters from the correction task scheme, updates the digital twin model and simulates and verifies again to ensure that the production efficiency reaches the preset target.

[0144] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with a preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, without departing from the technical solution of the present application. Any modification, equivalent change and modification of the above embodiments made according to the technical essence of the present application, without departing from the technical solution of the present application, still belongs to the scope of the technical solution of the present application.

Claims

1. A method of distributed collaborative control of a silicone rubber glass fibre sleeving production plant, characterised in that: Includes the following steps: Blockchain technology is used to store real-time multidimensional datasets. Through distributed ledger synchronization, it is determined whether the data has been tampered with, thus obtaining immutable data records. Device status parameters are obtained from immutable data records, and a task allocation scheme is obtained by assigning agent roles using a multi-agent system. By executing a distributed decision-making algorithm through a multi-agent system, the operating parameter adjustment values ​​of each device are calculated to obtain an optimized set of device parameters. Adjustment instructions are obtained from the optimized equipment parameter set, a virtual model of the production equipment is constructed using digital twin technology, and the optimized equipment parameter set is simulated and verified to obtain the virtual optimization results. By employing a global optimization algorithm and combining it with the production objective function, the optimal parameter combination is calculated to obtain the globally optimized parameters; The final adjustment instructions are obtained from the global optimization parameters, distributed to each device through the distributed control network, the operating status is updated in real time, new multi-dimensional data streams are collected and the dynamic response of the system is judged to obtain the updated production status. A feedback control algorithm is used to analyze the deviation between the newly collected multidimensional data and the expected target, calculate the deviation correction value, and then reallocate tasks through a multi-agent system to obtain a corrected task plan.

2. The distributed collaborative control method of a silicone rubber glass fiber sleeve production apparatus according to claim 1, characterized by: Obtaining immutable data records specifically includes: Real-time multidimensional datasets are obtained through blockchain technology, hash timestamp records are generated using hash functions, and stored in a distributed ledger to obtain initial data records. When the data hash values ​​of each device node in the distributed ledger are consistent, the latest data records are obtained through a data synchronization protocol to determine data consistency. A consensus algorithm is used to verify the hash timestamps in the distributed ledger. If the verification is successful, the latest data record is stored using blockchain technology to obtain tamper-proof data. The real-time processing module obtains data records from the distributed ledger, uses an encryption algorithm to generate data verification values, and determines whether the data has been tampered with. When the data verification value matches the hash timestamp, the verification value is synchronized through the device communication protocol to determine that the data record is tamper-proof. Based on the synchronized verification value, the multidimensional dataset is updated using a data consistency protocol to generate the final data record. The final data record is then stored using blockchain technology, and the hash timestamp in the distributed ledger is obtained to ensure that the data is tamper-proof.

3. The distributed collaborative control method of a silicone rubber glass fiber sleeve production apparatus according to claim 1, characterized by: The task allocation plan includes: Device status parameters are obtained from immutable data records, and data integrity is verified using blockchain technology to obtain a set of trusted status parameters. Then, the intelligent agent nodes are initialized through a multi-agent system, and the set of trusted status parameters is distributed to each intelligent agent according to the preset role allocation rules to determine the intelligent agent role allocation table. Each intelligent agent performs distributed data analysis on the allocated parameter set according to the cooperation protocol, uses a clustering algorithm to divide the device status into categories, and obtains the device status classification results. When there are abnormal categories in the device status classification results, the task allocation mechanism is triggered through the cooperation protocol to generate a preliminary task allocation scheme. When there are no abnormalities, the current task allocation is maintained. Based on the initial task allocation scheme, a genetic algorithm is used to optimize the collaborative task allocation between devices to obtain an optimized task allocation scheme. Then, device collaboration parameters are extracted from the optimized task allocation scheme and distributed to each device through a message passing interface to determine the device collaborative execution instructions. Obtain the operation logs of the device collaborative execution instructions, use time series analysis to verify the task execution effect, and obtain the verification results of the final task allocation scheme.

4. The distributed collaborative control method for silicone rubber glass fiber sleeve production equipment according to claim 1, characterized in that: The optimized set of device parameters is obtained, specifically including: The system acquires real-time data streams, collects operating parameters of each device in a distributed manner through a multi-agent system, processes the data through a distributed decision-making algorithm, calculates the adjustment values ​​of the operating parameters of each device, and obtains a set of adjustment values. When the parameter values ​​in the adjustment value set exceed the preset threshold, the adjustment values ​​are normalized by the data processing module to obtain a normalized adjustment value set. Then, the multi-agent system is used to update the operating parameters of each device to obtain an updated parameter set. By acquiring dynamic indicators of production targets through real-time data streams, it is determined whether the updated parameter set meets the production targets and obtains the satisfaction status. If the satisfaction status indicates that the parameters do not fully meet the production targets, the adjustment values ​​are recalculated through a distributed decision-making algorithm to obtain a new set of adjustment values. Based on the newly adjusted value set, the parameters are repeatedly updated and the target satisfaction status is determined to obtain the final optimized parameter set.

5. The distributed collaborative control method for silicone rubber glass fiber sleeve production equipment according to claim 1, characterized in that: The virtual optimization results obtained include: Obtain the optimized equipment parameter set, generate the adjustment instruction set by parsing the parameter set, determine the integrity of the instruction set, collect real-time operating data from the production equipment, build a digital twin model, map the equipment operating status, and obtain the virtual operating status; When the deviation between the virtual operating state and the real-time operating data exceeds a preset threshold, the parameters of the digital twin model are iteratively updated to adjust the model state and obtain the calibrated virtual operating state. Using the calibrated virtual operating state, the adjustment instruction set is loaded, the parameter adjustment simulation is executed, and the simulation results are obtained. By analyzing the simulation results, key performance indicators are extracted to determine whether the adjusted instruction set meets the optimization objective and obtain the verification results. If the verification results do not reach the preset optimization threshold, the genetic algorithm is used to optimize the adjusted instruction set, generate a new instruction set, and obtain the optimized adjusted instruction set. Based on the optimized adjustment instruction set, the parameters of the digital twin model are updated, the final simulation is performed, and the virtual optimization results are obtained.

6. The distributed collaborative control method for silicone rubber glass fiber sleeve production equipment according to claim 1, characterized in that: The global optimization parameters are obtained, specifically including: Acquire production data, extract feature values ​​from efficiency indicators, quality indicators, and energy consumption indicators, then construct an objective function using a preset weight allocation, and determine the expression of the objective function. The global optimization algorithm iteratively searches for parameter combinations from the initial dataset to obtain a candidate parameter set. Then, it calculates the objective function value. If the function value meets the preset threshold, it is determined to be the optimal parameter. If it does not meet the threshold, the parameter combination is adjusted and the iteration continues. Based on the optimal parameters, production control instructions are generated to obtain the optimized production configuration, the optimized production data is acquired, efficiency indicators, quality indicators and energy consumption indicators are calculated, it is determined whether the production target is met, and by comparing the data before and after optimization, performance improvement features are extracted to obtain the final optimization result.

7. The distributed collaborative control method for silicone rubber glass fiber sleeve production equipment according to claim 1, characterized in that: The updated production status includes: The final adjustment command is extracted from the global optimization parameters. The parameter extraction process is adopted, and the global optimization parameters are processed by a preset extraction algorithm to obtain the final adjustment command. The final adjustment command is then transmitted through a distributed control network. The command transmission mechanism is adopted. If the network bandwidth meets the preset threshold, the final adjustment command is distributed to each device to determine that the command transmission is complete. The system updates the operating status of each device in real time, obtains a list of devices that have completed instruction transmission, collects the real-time operating status of each device by configuring the status update frequency, obtains real-time operating status data, and then collects multi-dimensional data streams. For the real-time operating status data, a data stream acquisition mechanism is adopted to obtain multi-dimensional data streams through sensor interfaces, resulting in a multi-dimensional data stream set. To determine whether the multidimensional data stream set has reached the dynamic response target, a target judgment standard is adopted. If the key indicators in the multidimensional data stream set meet the preset threshold, it is confirmed that the dynamic response target has been reached, and the target achievement status is obtained. The production status is updated based on the target achievement status. The target achievement status is obtained through the production status update mechanism, the updated production status data is generated, and then the global optimization parameters are adjusted using a feedback optimization algorithm to obtain the optimized global optimization parameters.

8. The distributed collaborative control method for silicone rubber glass fiber sleeve production equipment according to claim 1, characterized in that: Before obtaining tamper-proof data records, the process also includes: acquiring multidimensional data streams from production equipment sensors, preprocessing the data using edge computing nodes, filtering noisy data, generating standardized data packets, and obtaining real-time multidimensional datasets.

9. The distributed collaborative control method for silicone rubber glass fiber sleeve production equipment according to claim 1, characterized in that: Also includes: The adjusted operating parameters are obtained from the revised task plan, and then simulated and verified again using a digital twin model. The virtual model status is updated, and it is determined whether the production efficiency has reached the preset target, thus obtaining the final optimized production parameters.

10. A distributed collaborative control system for a silicone rubber glass fiber duct production equipment, used to implement the distributed collaborative control method for a silicone rubber glass fiber duct production equipment as described in any one of claims 1-9, characterized in that: include: The data acquisition and preprocessing module collects multidimensional data streams from sensors on production equipment, preprocesses them through edge computing, generates standardized datasets, and simultaneously marks and processes abnormal data. The blockchain data storage and verification module uses blockchain technology to store multidimensional datasets, generates hash timestamp records, and uses consensus algorithms and data verification to ensure that the data is tamper-proof. The multi-agent task allocation module obtains device status parameters from the blockchain, allocates tasks through the multi-agent system, and optimizes the task allocation scheme using clustering and genetic algorithms. The distributed decision-making and parameter optimization module, based on the task allocation scheme, executes a distributed decision-making algorithm through a multi-agent system to dynamically adjust equipment operating parameters and optimize production targets. The digital twin simulation verification module constructs a digital twin model of the production equipment and verifies the effect of parameter adjustment through simulation to ensure that the virtual operating state is consistent with the actual production target. The global optimization module, combined with the production objective function, uses a global optimization algorithm to calculate the optimal parameter combination, thereby improving production efficiency, quality, and energy consumption performance. The distributed control and dynamic response module converts global optimization parameters into adjustment commands, which are then sent to the device via a distributed network to update the operating status and determine the dynamic response effect in real time. The feedback control and task correction module analyzes deviations and calculates correction values ​​based on production status using a feedback control algorithm. It then reallocates tasks through a multi-agent system to optimize the production process. The model update and verification module obtains operating parameters from the revised task plan, updates the digital twin model, and performs simulation verification again to ensure that production efficiency reaches the preset target.

Images