Automatic industrial equipment PLC data acquisition monitoring control method and system

Abstract

The invention relates to the technical field of automatic industrial control, and discloses an automatic industrial equipment plc data acquisition monitoring control method and system. The method comprises the following steps: receiving a multi-source PLC data stream, and processing the multi-source PLC data stream through a multi-modal data fusion model to generate an equipment health degree evaluation matrix; constructing a dynamic control strategy generation model, and outputting a multi-target optimization instruction set; simulating the interaction influence of the equipment group by utilizing a process chain collaborative constraint model, and optimizing real-time control parameters of an instruction set; parameters are adjusted in parallel through a distributed edge computing framework, and an equipment cooperative control signal is output to an industrial monitoring center. The system comprises a distributed data acquisition module, a multi-modal fusion calculation module, a dynamic strategy generation module, an edge cooperative control module and an industrial communication gateway. According to the invention, multi-source data can be efficiently processed, equipment health can be accurately evaluated, a control strategy is optimized, equipment collaboration is realized, and industrial production efficiency and energy utilization rate are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automated industrial control, and specifically to a method and system for PLC data acquisition, monitoring, and control of automated industrial equipment. Background Art

[0002] In modern industrial production, the degree of automation is constantly increasing, and the operating conditions of industrial equipment directly affect production efficiency, product quality, and the economic benefits of enterprises. As the core control equipment for industrial automation, PLCs are widely used in various industrial scenarios, and the data collected by them contains key information about the operation of the equipment. However, current technologies for PLC data processing, monitoring, and control in industrial equipment face many challenges.

[0003] From the perspective of data acquisition, the multi-source PLC data generated by industrial equipment is complex and diverse, including equipment operating status signals, energy consumption time-series data, and environmental sensor feedback information. The sampling frequencies, accuracies, and formats of different types of data vary greatly, resulting in high data fusion difficulty. Traditional data acquisition methods are difficult to comprehensively, accurately, and real-timely collect these multi-source heterogeneous data, causing a large amount of valuable information to be omitted or unable to be effectively integrated and utilized, affecting the accurate judgment of the overall operating status of the equipment.

[0004] In terms of data processing and analysis, existing technologies lack effective multi-modal data fusion means. The noise interference in multi-source PLC data is serious, which will reduce the data quality and affect the accuracy of subsequent analysis. Moreover, the feature extraction between data is not precise enough to deeply explore the potential laws of equipment operation and the features related to the health status. Therefore, it is difficult to construct an accurate equipment health assessment system, and it is impossible to detect early equipment failure hidden dangers in time, resulting in sudden equipment failures, production interruptions, increased maintenance costs, and other problems.

[0005] In the formulation of control strategies, most of the previous control methods are based on fixed rules or experience and lack the ability of dynamic adjustment. With the increasing complexity of industrial production processes, the coupling relationship between equipment is close, and the change in the operating status of a single equipment will trigger a chain reaction. Traditional control strategies cannot quickly generate an effective multi-objective optimization instruction set according to the real-time health status of the equipment and the collaborative requirements of the process chain, and it is difficult to achieve precise control of equipment start-stop scheduling and parameter adjustment, reducing production efficiency and resource utilization rate.

[0006] In addition, there are also deficiencies in the simulation and optimization of process chain collaborative constraints. The interaction effects between industrial equipment groups are complex, and existing technologies cannot fully consider factors such as energy efficiency transfer, delay response, and real-time working condition disturbances between equipment. In actual production, there are often situations of poor equipment collaboration and serious energy waste, which cannot guarantee the stable and efficient operation of the entire process chain.

[0007] Meanwhile, the application of edge computing in industrial control is not yet mature. The distributed edge computing framework fails to fully utilize its advantages. When optimizing multi-node collaboration, the goal is single, and it cannot take into account both energy consumption and production efficiency simultaneously. Moreover, the collaboration efficiency between nodes is low, the weight allocation is unreasonable, and it is impossible to quickly converge to the optimal solution, making it difficult to output device collaborative control signals that meet the actual needs of industrial production. Summary of the Invention

[0008] The purpose of the present invention is to provide an automated industrial equipment PLC data acquisition, monitoring, and control method and system to solve the problems raised in the above background technology.

[0009] To achieve the above purpose, the present invention provides the following technical solution: An automated industrial equipment PLC data acquisition, monitoring, and control method and system, the method includes:

[0010] Receiving multi-source PLC data streams of industrial equipment, the data streams including equipment operation status signals, energy consumption time series data, and environmental sensor feedback information;

[0011] Based on a preset multi-modal data fusion model, denoising, aligning, and feature extraction are performed on the multi-source PLC data to generate an equipment health assessment matrix;

[0012] According to the health assessment matrix, a dynamic control strategy generation model is constructed to output a multi-objective optimization instruction set, and the instruction set includes an equipment start-stop scheduling sequence and a parameter adjustment gradient scheme;

[0013] Based on a preset process chain collaborative constraint model, the interactive influence between equipment groups within a preset future time period is simulated, and the real-time control parameters in the instruction set are optimized;

[0014] The real-time control parameters are adjusted in parallel through a distributed edge computing framework, and device collaborative control signals are output to the industrial monitoring center.

[0015] Preferably, the construction steps of the multi-modal data fusion model include:

[0016] Collecting historical equipment operation logs and fault case libraries to construct a multi-dimensional training set including signal fluctuation patterns, equipment degradation characteristics, and environmental interference factors;

[0017] Performing noise suppression and feature enhancement on the multi-dimensional training set through an adversarial generation network to separate steady-state operation characteristics and transient abnormal components;

[0018] Combining equipment physical constraint equations to construct state space constraint conditions for the time series dependence relationship between features;

[0019] Embedding the constraint conditions into a temporal convolutional neural network to generate the multi-modal data fusion model that supports online update.

[0020] Preferably, the dynamic control strategy generation model includes:

[0021] Dynamically divide the equipment performance degradation levels according to the entropy value change of the health assessment matrix;

[0022] Calculate the system cascading failure risk score based on the equipment topological connection relationship and process coupling strength;

[0023] Dynamically weightedly fuse the risk score and the degradation level through a piecewise linear function to generate an equipment-specific control threshold;

[0024] Trigger the execution conditions of the multi-level process adjustment protocol according to the threshold.

[0025] Preferably, the construction steps of the process chain collaborative constraint model include:

[0026] Collect the historical interaction data between equipment groups to construct a process causal graph data set;

[0027] Infer the energy efficiency transfer coefficient and delay response parameter between equipment through a Bayesian network;

[0028] Construct a virtual simulation environment for the process chain in combination with digital twin technology to quantify the energy flow balance constraints between nodes;

[0029] Input the constraints and real-time working condition disturbance factors into a spatio-temporal graph neural network to generate the process chain collaborative constraint model.

[0030] Preferably, the method further includes:

[0031] Identify key energy flow bottleneck nodes according to the simulation results of the process chain collaborative constraint model;

[0032] Configure a dynamic energy efficiency compensation strategy for the nodes in the instruction set;

[0033] Based on the compensation strategy, automatically generate a cross-device collaborative optimization plan, including load balancing distribution instructions and power factor correction parameters.

[0034] Preferably, the calculation of the system cascading failure risk score includes:

[0035] Obtain the real-time equipment communication delay matrix and spare part inventory distribution map, and construct a system resilience assessment tensor;

[0036] Extract the implicit correlation factors between multi-dimensional features through a tensor decomposition algorithm;

[0037] Perform a Hadamard product operation on the correlation factor and the assessment tensor to obtain a comprehensive risk score.

[0038] Preferably, the embedding of the state space constraint conditions includes:

[0039] Perform Lie group symmetry analysis on the feature enhancement result to screen out the time-series evolution patterns that conform to physical laws;

[0040] Generate state transition trajectories that satisfy the device dynamics constraints through Markov chain Monte Carlo sampling;

[0041] Use the trajectory data to sparsely train the convolutional kernels of the time-series convolutional network to ensure that the model output conforms to the process chain conservation law.

[0042] Preferably, the execution of the distributed edge computing framework includes:

[0043] Define a loss function for multi-node collaborative optimization, which includes the dual objectives of minimizing energy consumption and maximizing production efficiency;

[0044] Train the local control strategies of each edge node through the asynchronous parallel gradient descent algorithm;

[0045] In each round of iteration, dynamically adjust the weight allocation ratio between nodes according to the degree of target deviation;

[0046] Output the device collaborative control signal that satisfies the Pareto optimal condition;

[0047] Among them, the loss function is:

[0048]

[0049] In the formula, represents the loss value, represents the total energy consumption score of the system, represents the production efficiency score, is the dynamic adjustment factor.

[0050] Preferably, the method further includes:

[0051] After configuring the dynamic energy efficiency compensation strategy, monitor the energy efficiency fluctuation entropy value of the energy flow bottleneck node in real time;

[0052] If the entropy value exceeds the preset safety boundary, trigger the chaos optimization algorithm to re-plan the load distribution topology of the device group.

[0053] Preferably, the present invention further includes an automated industrial equipment PLC data acquisition and monitoring control system, and the system includes:

[0054] A distributed data acquisition module, deployed at the industrial field edge node, for obtaining multi-source PLC data streams in real time;

[0055] The multimodal fusion computing module is connected to the data acquisition module and incorporates a temporal convolutional neural network and a generative adversarial network to perform data denoising and feature extraction.

[0056] The dynamic policy generation module integrates a process chain collaborative constraint model and a digital twin simulation environment to generate a multi-objective optimization instruction set.

[0057] The edge collaborative control module adjusts the device control parameters through an asynchronous parallel computing framework and outputs collaborative control signals.

[0058] The industrial communication gateway transmits the control signal to the PLC execution terminal and feeds back the real-time working conditions to the monitoring center.

[0059] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0060] In terms of data processing, through a preset multimodal data fusion model, denoising alignment and feature extraction are performed on multi-source PLC data of industrial equipment. A multi-dimensional training set is constructed by collecting historical equipment operation logs and a fault case database. The generative adversarial network is used to suppress noise and enhance features, and state space constraint conditions are constructed in combination with the device physical constraint equation and embedded into the temporal convolutional neural network. This not only effectively removes data noise and ensures data accuracy, but also accurately extracts key features related to equipment health. The generated equipment health assessment matrix can accurately reflect the actual health status of the equipment. Compared with traditional data processing methods, it greatly improves the data quality and reliability of analysis, helps to detect potential equipment failures in advance, reduces the equipment failure rate, and reduces the production interruption time caused by equipment failures, thereby enhancing the continuity and stability of production.

[0061] For control strategy generation, a dynamic control strategy generation model is constructed based on the health assessment matrix. This model divides the equipment performance degradation level according to the change of entropy value, calculates the system cascade failure risk score in combination with the equipment topology connection relationship and process coupling strength, and generates a control threshold through dynamic weighted fusion of piecewise linear functions to trigger a multi-level process adjustment protocol. This dynamic and intelligent control strategy can adjust the equipment operation in a timely manner according to the real-time state of the equipment and the system risk status, avoid excessive degradation of the equipment, and reduce the system failure risk. Compared with the traditional fixed-rule-based control method, it is more flexible and accurate, can effectively improve the service life and operation safety of the equipment, and reduce the maintenance cost and downtime loss.

[0062] In the collaborative optimization of the process chain, the process chain collaborative constraint model plays an important role. Historical interaction data between equipment groups is collected to construct a process causal graph dataset. Bayesian network inference parameters are used, combined with digital twin technology to construct a virtual simulation environment, quantify the energy flow balance constraint, and input it into the spatio-temporal graph neural network generation model. This model simulates the interactive influence between equipment groups, optimizes real-time control parameters, identifies key energy flow bottleneck nodes and configures dynamic energy efficiency compensation strategies, and generates a cross-device collaborative optimization plan. This series of operations greatly improves the collaboration between devices, makes energy utilization more reasonable, reduces energy waste, improves energy utilization efficiency, and reduces production costs. At the same time, it ensures the stable operation of the process chain, improves product quality and production efficiency.

[0063] In the execution of the distributed edge computing framework, a loss function containing the dual objectives of minimizing energy consumption and maximizing production efficiency is defined. The local control strategies of each edge node are trained through the asynchronous parallel gradient descent algorithm. The weight allocation ratio between nodes is dynamically adjusted according to the degree of target deviation, and a device collaborative control signal that meets the Pareto optimal condition is output. This realizes the efficient collaborative optimization of multiple nodes, makes full use of the computing resources of edge nodes, takes into account two key indicators of energy consumption and production efficiency, enables industrial equipment to maintain high production efficiency while saving energy, and enhances the comprehensive competitiveness of industrial production.

[0064] Generally speaking, the methods and systems of the present invention comprehensively improve the intelligent management level of industrial equipment. From data collection, processing and analysis to the generation and execution of control strategies, full-process optimization is realized, bringing significant economic and social benefits to industrial enterprises and promoting the development of industrial automation to a higher level. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is the working principle diagram of the PLC data acquisition, monitoring and control method for the automated industrial equipment described in the present invention;

[0066] Figure 2 It is the flow chart of equipment performance degradation evaluation and process adjustment;

[0067] Figure 3 It is the flow chart of system resilience risk assessment and early warning;

[0068] Figure 4 It is the flow chart of energy efficiency fluctuation entropy value monitoring and load distribution optimization. DETAILED DESCRIPTION OF THE INVENTION

[0069] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0070] Please refer to Figures 1-4 , the present invention provides an automated industrial equipment PLC data acquisition, monitoring and control method, and its specific implementation steps are as follows:

[0071] Receive multi-source PLC data streams: In the industrial field, through sensors and data acquisition devices deployed at key positions, multi-source PLC data streams of industrial equipment are received in real time. These data streams cover equipment operation status signals, such as whether the equipment is operating normally, whether there is abnormal vibration, etc.; energy consumption time series data, that is, the energy consumption of the equipment at different times; and environmental sensor feedback information, including environmental factors data such as environmental temperature, humidity, and dust concentration that may affect the operation of the equipment.

[0072] Process multi-source PLC data and generate an equipment health assessment matrix: Based on a preset multi-modal data fusion model, the above multi-source data is processed. This model first denoises and aligns the data to remove noise interference in the data and ensure the accuracy and consistency of the data; then feature extraction is performed to extract key features related to the equipment health status from the complex data. Through these operations, an equipment health assessment matrix that can accurately reflect the equipment health degree is generated.

[0073] Construct a dynamic control strategy generation model and output an instruction set: According to the generated health assessment matrix, a dynamic control strategy generation model is constructed. This model will comprehensively consider various factors and output a multi-objective optimization instruction set, which includes an equipment start-stop scheduling sequence, that is, according to factors such as the equipment health status and production requirements, reasonably arrange the start and stop times of the equipment; and a parameter adjustment gradient scheme to clarify the adjustment direction and amplitude of the equipment operation parameters.

[0074] Simulate interactive impacts and optimize real-time control parameters: Based on a preset process chain collaborative constraint model, simulate the interactive impacts between equipment groups within a preset future time period. For example, a change in the operation status of one equipment may have a chain reaction on other related equipment, and this impact can be predicted through this model. Then, optimize the real-time control parameters in the instruction set according to the simulation results to make the control strategy more in line with the actual production requirements.

[0075] Adjust the parameters and output the collaborative control signal: Parallelly adjust the optimized real-time control parameters through the distributed edge computing framework. The distributed edge computing framework can make full use of the computing resources of the edge nodes in the industrial field to achieve efficient parallel computing. Finally, convert the adjusted parameters into the device collaborative control signal and output it to the industrial monitoring center, where the monitoring center uniformly monitors and controls the industrial devices.

[0076] The following further illustrates the implementation of the present invention in conjunction with Embodiments 1 to 6.

[0077] Embodiment 1:

[0078] The construction process of the multi-modal data fusion model is described in this embodiment.

[0079] Collect a large number of historical device operation logs and a fault case database. The historical device operation logs contain various types of data during the long-term operation of the device, such as changes in operation parameters and performance under different working conditions; the fault case database records the specific situations when the device fails, including fault phenomena, occurrence times, maintenance measures, etc. By sorting and analyzing these data, a multi-dimensional training set including signal fluctuation patterns, device degradation characteristics, and environmental interference factors is constructed. For example, the signal fluctuation pattern can reflect the change rules of some key signals during the device operation, the device degradation characteristics can reflect the performance decline trend of the device as the usage time increases, and the environmental interference factors record the influence of different environmental factors on the device operation.

[0080] Use the generative adversarial network to process the multi-dimensional training set. The generative adversarial network consists of a generator and a discriminator. The generator is responsible for generating false data, and the discriminator is used to judge whether the data is real or generated. In this process, the generative adversarial network suppresses the noise in the multi-dimensional training set, removes the noise interference in the data, and makes the data purer; at the same time, it enhances the features, highlighting those features that are of great significance for device health assessment. In this way, the steady-state operation characteristics and transient abnormal components are separated, and the characteristics of the device during normal operation and the possible abnormal characteristics are distinguished.

[0081] Combine the device physical constraint equations to construct the state space constraint conditions for the temporal dependence relationship between features. The device physical constraint equations are mathematical models established based on the physical principles and operation mechanisms of the device, which describe the internal relationships between various parameters of the device. For example, for an electric motor device, there are certain physical relationships between its rotational speed, torque, and power, and these relationships can be expressed by the physical constraint equations. According to these equations, determine the temporal dependence relationship between features and construct the state space constraint conditions.

[0082] Embed the above constraints into a temporal convolutional neural network. A temporal convolutional neural network is a neural network specialized for processing temporal data, which can effectively extract temporal features from the data. By embedding the constraints into it, a multi-modal data fusion model that supports online update is trained. During the model training process, a large amount of training data is used to optimize the network, enabling the model to accurately denoise, align, and extract features from multi-source PLC data, providing reliable data support for subsequent equipment health assessment.

[0083] Embodiment 2:

[0084] This embodiment mainly focuses on the dynamic control strategy generation model.

[0085] When constructing a dynamic control strategy generation model based on the health assessment matrix, first dynamically divide the equipment performance degradation levels according to the entropy value change of the health assessment matrix. The entropy value is an index to measure the uncertainty of data. In the health assessment matrix, the change of the entropy value can reflect the stability of the equipment performance. When the entropy value is low, it indicates that the equipment is operating relatively stably and the performance degradation degree is low; when the entropy value increases, it means that the equipment may have some abnormal conditions and the performance degradation degree intensifies. By setting different entropy value thresholds, the equipment performance degradation levels are divided into different levels, such as mild degradation, moderate degradation, and severe degradation.

[0086] Based on the equipment topological connection relationship and process coupling strength, calculate the system cascading failure risk score. Obtain the real-time equipment communication delay matrix and spare parts inventory distribution map, and construct a system resilience assessment tensor. The equipment topological connection relationship describes the physical and logical relationships between industrial equipment, and the process coupling strength reflects the degree of association between different equipment in the production process. The communication delay matrix records the delay situation of data transmission between equipment, and the spare parts inventory distribution map shows the storage locations and quantities of spare parts for each equipment. By integrating this information, a system resilience assessment tensor is constructed, which can comprehensively reflect the anti-interference ability and recovery ability of the system when facing various interferences.

[0087] Extract the implicit correlation factors between multi-dimensional features through the tensor decomposition algorithm. The tensor decomposition algorithm is a technology that can decompose a high-dimensional tensor into the product of multiple low-dimensional tensors. Through this algorithm, the hidden correlation relationships between multi-dimensional features can be extracted from the system resilience assessment tensor to obtain the implicit correlation factors. Perform the Hadamard product operation on the correlation factors and the assessment tensor to obtain the comprehensive risk score. The Hadamard product operation is an operation of multiplying corresponding elements. Through this operation, the implicit correlation factors can be combined with each element in the assessment tensor to obtain a score that comprehensively reflects the system cascading failure risk.

[0088] Dynamically weighted fusion of the risk score and the degradation level is performed through a piecewise linear function to generate a device-specific control threshold. A piecewise linear function is a function with different linear expressions in different intervals. According to the different situations of the risk score and the degradation level, the parameters of the piecewise linear function are reasonably selected to perform dynamic weighted fusion on the two. For example, when the device performance degradation level is high and the risk score is also high, the weight of the risk score in the fusion is appropriately increased to make the generated control threshold more stringent. The execution conditions of the multi-level process adjustment protocol are triggered according to the generated threshold. When the operating parameters of the device exceed the control threshold, the corresponding process adjustment protocol is immediately triggered, such as adjusting the operating speed of the device, changing the working mode of the device, etc., to ensure that the device can continue to operate safely and stably.

[0089] Embodiment 3:

[0090] This embodiment details the construction steps of the process chain collaborative constraint model.

[0091] Collect historical interaction data between device groups to construct a process causal graph data set. In the industrial production process, there are various interaction relationships between devices, and these interaction relationships will affect the production process. By long-term collection and collation of historical interaction data between device groups, including information such as signal transmission, energy transfer, and material flow between devices, a process causal graph data set is constructed. This data set can clearly show the causal relationships between devices and provide a basis for subsequent analysis.

[0092] Infer the energy efficiency transfer coefficient and delay response parameter between devices through a Bayesian network. A Bayesian network is a graphical model based on probabilistic inference, which can use known data and prior knowledge to infer unknown parameters. In this embodiment, the Bayesian network is used to analyze the process causal graph data set to infer the energy efficiency transfer coefficient and delay response parameter between devices. The energy efficiency transfer coefficient reflects the degree of influence of the energy consumption of one device on other devices, and the delay response parameter represents the response delay time between devices.

[0093] Construct a process chain virtual simulation environment in combination with digital twin technology. Digital twin technology is a technology that maps and simulates physical entities through digital means. In the process chain virtual simulation environment, virtual models identical to the actual industrial devices and process chain are created, including the structure, performance parameters, operating status, etc. of the devices. By simulating the operation of the virtual model, the operation conditions of each node in the process chain can be monitored and analyzed in real time, and the energy flow balance constraints between nodes can be quantified. For example, in the virtual simulation environment, the operating conditions of different devices can be simulated, the energy flow between nodes can be calculated, and the energy flow balance constraint conditions can be determined.

[0094] These constraints and real-time operating condition disturbance factors are input into a spatio-temporal graph neural network to generate a process chain collaborative constraint model. A spatio-temporal graph neural network is a neural network that can process spatial and temporal information simultaneously, and it can effectively capture the spatial relationships and time series information between devices. The energy flow balance constraint and real-time operating condition disturbance factors are input into the spatio-temporal graph neural network, and through the learning and training of the network, a process chain collaborative constraint model that can accurately reflect the interactive effects between device groups is generated. This model can provide an important basis for the subsequent optimization of real-time control parameters.

[0095] Embodiment 4:

[0096] In this embodiment, further operations are performed based on the simulation results of the process chain collaborative constraint model.

[0097] According to the simulation results of the process chain collaborative constraint model, key energy flow bottleneck nodes are identified. In the industrial production process, some devices or nodes may become bottlenecks restricting the energy flow of the entire process chain due to insufficient energy supply, low transmission efficiency, etc. By analyzing the simulation results of the process chain collaborative constraint model, those key energy flow bottleneck nodes that have a greater impact on the energy flow are found. For example, by simulating the energy flow in the process chain under different operating conditions and observing which nodes show phenomena such as energy accumulation or insufficient supply, these nodes are determined as key energy flow bottleneck nodes.

[0098] A dynamic energy efficiency compensation strategy is configured for these nodes in the instruction set. The dynamic energy efficiency compensation strategy is a strategy that automatically adjusts energy distribution according to the real-time operating conditions and energy demands of the nodes. For example, for key energy flow bottleneck nodes with insufficient energy supply, their energy supply can be increased; for nodes with low energy transmission efficiency, measures such as optimizing the transmission path and improving the performance of transmission equipment can be taken to improve the energy transmission efficiency. By configuring the dynamic energy efficiency compensation strategy, the energy utilization efficiency of the entire process chain is improved.

[0099] Based on the compensation strategy, a cross-device collaborative optimization scheme is automatically generated, including load balancing distribution instructions and power factor correction parameters. The load balancing distribution instructions are used to reasonably distribute the workloads of each device, avoiding the situation where some devices are overloaded while others are underloaded. The power factor correction parameters are used to improve the power factor of the device and reduce energy waste. For example, according to the actual operating power and power factor of the device, the power factor correction parameters that need to be adjusted are calculated, and by adjusting the circuit parameters of the device, the power factor is improved.

[0100] After configuring the dynamic energy efficiency compensation strategy, the energy efficiency fluctuation entropy value of the energy flow bottleneck node is monitored in real time. The energy efficiency fluctuation entropy value can reflect the stability of the energy utilization efficiency of the energy flow bottleneck node. If the entropy value exceeds the preset safety boundary, the chaotic optimization algorithm is triggered to re-plan the load distribution topology of the device group. The chaotic optimization algorithm is an optimization algorithm based on chaos theory, which has the advantages of strong global search ability and fast convergence speed. When the energy efficiency fluctuation entropy value exceeds the preset safety boundary, it indicates that the energy utilization efficiency of the energy flow bottleneck node has fluctuated greatly, which may affect the stable operation of the entire process chain. At this time, the chaotic optimization algorithm is triggered to re-plan the load distribution topology of the device group, making the load distribution between devices more reasonable and improving the energy utilization efficiency and stability of the entire system.

[0101] Example 5:

[0102] This example details the embedding process of the state space constraint conditions. After constructing the state space constraint conditions in the multi-modal data fusion model, they need to be embedded into the temporal convolutional neural network.

[0103] Perform Lie group symmetry analysis on the feature enhancement results to screen out the temporal evolution patterns that conform to physical laws. A Lie group is a mathematical structure with continuous symmetry and has extensive applications in physics and engineering. By performing Lie group symmetry analysis on the feature enhancement results, the symmetric relationships existing in the data can be found, thereby screening out the temporal evolution patterns that conform to physical laws. For example, during the operation of a device, the changes in certain physical quantities may have a certain symmetry, and these laws can be discovered through Lie group symmetry analysis, excluding the abnormal patterns that do not conform to physical laws.

[0104] Generate state transition trajectories that satisfy the device dynamics constraints through Markov chain Monte Carlo sampling. Markov chain Monte Carlo sampling is a stochastic sampling method based on Markov chains, which can perform efficient sampling in high-dimensional spaces. In this example, the Markov chain Monte Carlo sampling method is used to generate a series of state transition trajectories that meet the requirements according to the device dynamics constraint conditions. The device dynamics constraint conditions are established based on the physical principles and motion equations of the device, which describe the limiting conditions for the device to transfer between different states. By generating these state transition trajectories, more sample data can be provided for subsequent model training.

[0105] Sparsify the convolutional kernels of the temporal convolutional network using trajectory data to ensure that the model output conforms to the law of conservation of the process chain. The convolutional kernels of the temporal convolutional network are the key components in the network for extracting temporal features. By sparsifying the training of the convolutional kernels, the number of model parameters can be reduced, and the computational efficiency and generalization ability of the model can be improved. During the training process, the generated state transition trajectory data is used to optimize the convolutional kernels, so that the results output by the model conform to the law of conservation of the process chain. For example, in terms of energy consumption, the results output by the model should satisfy the law of conservation of energy, that is, the total input energy should be equal to the total output energy plus the energy consumed by the equipment. In this way, it is ensured that the multi-modal data fusion model can accurately reflect the actual operation of the equipment, providing reliable support for equipment health assessment and control strategy generation.

[0106] Example 6:

[0107] When using the distributed edge computing framework to parallelly adjust the real-time control parameters, first define the loss function for multi-node collaborative optimization, which includes the dual objectives of minimizing energy consumption and maximizing production efficiency. The loss function is:

[0108] ,

[0109] where, represents the loss value, which is an indicator to measure the effect of the entire optimization process; represents the total system energy consumption score, which is used to quantify the energy consumption of the system during operation; represents the production efficiency score, which reflects the production efficiency of the system; is a dynamic adjustment factor, which can dynamically adjust the weights of energy consumption and production efficiency in the loss function according to actual production requirements. For example, when the energy cost is high, the value of can be appropriately increased to make the optimization process pay more attention to minimizing energy consumption; when the production task is urgent, the value of can be reduced to pay more attention to maximizing production efficiency.

[0110] Train the local control strategies of each edge node through the asynchronous parallel gradient descent algorithm. The asynchronous parallel gradient descent algorithm is a commonly used optimization algorithm in a distributed computing environment, which allows each edge node to perform calculations and updates independently, improving the computational efficiency. On each edge node, according to the locally collected data and the defined loss function, calculate the gradient and update the local control strategy. For example, for an edge node responsible for controlling a certain device, it will calculate the gradient under the current control strategy based on the real-time operation data of the device and the loss function, and then adjust the parameters of the control strategy according to the gradient information.

[0111] In each iteration, the weight allocation ratio between nodes is dynamically adjusted according to the degree of target deviation. The degree of target deviation can be measured by calculating the gap between the current loss value and the ideal loss value. When the control strategy of a certain edge node makes a greater contribution to reducing the loss value, the proportion of this node in the weight allocation is appropriately increased; otherwise, its proportion is decreased. By this way of dynamically adjusting the weight allocation ratio, the distributed edge computing framework can converge to the optimal solution more efficiently.

[0112] Output the device collaborative control signal that meets the Pareto optimality condition. Pareto optimality refers to a state where a certain objective cannot be further optimized without sacrificing other objectives. In this embodiment, by continuously adjusting the control strategies and weight allocation ratios of each edge node, the system reaches the Pareto optimal state and outputs the device collaborative control signal that meets this condition. These signals are transmitted to the industrial monitoring center for precise control of industrial equipment, achieving the dual goals of minimizing energy consumption and maximizing production efficiency.

[0113] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0114] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An automated industrial equipment PLC data acquisition, monitoring and control method, characterized in that Including: Receiving the multi-source PLC data stream of industrial equipment, where the data stream includes equipment operation status signals, energy consumption time-series data, and environmental sensor feedback information; Based on a preset multi-modal data fusion model, denoising, aligning, and feature extracting the multi-source PLC data to generate an equipment health assessment matrix; According to the health assessment matrix, constructing a dynamic control strategy generation model and outputting a multi-objective optimization instruction set, where the instruction set includes an equipment start-stop scheduling sequence and a parameter adjustment gradient scheme; Based on a preset process chain collaborative constraint model, simulating the interactive influence among equipment groups within a preset future time period and optimizing the real-time control parameters in the instruction set; Parallelly adjusting the real-time control parameters through a distributed edge computing framework and outputting equipment collaborative control signals to the industrial monitoring center.

2. The PLC data acquisition, monitoring and control method for automated industrial equipment according to claim 1, characterized in that, The construction steps of the multi-modal data fusion model include: Collecting historical equipment operation logs and fault case libraries, and constructing a multi-dimensional training set containing signal fluctuation patterns, equipment degradation characteristics, and environmental interference factors; Performing noise suppression and feature enhancement on the multi-dimensional training set through an adversarial generation network to separate steady-state operation characteristics and transient abnormal components; Combining equipment physical constraint equations to construct state space constraint conditions for the time-series dependence relationship between features; Embedding the constraint conditions into a time-series convolutional neural network to generate the multi-modal data fusion model that supports online update.

3. The PLC data acquisition, monitoring and control method for the automated industrial equipment according to claim 1, characterized in that, The dynamic control strategy generation model includes: Dynamically dividing the equipment performance degradation levels according to the entropy value change of the health assessment matrix; Calculating the system cascade failure risk score based on the equipment topological connection relationship and process coupling strength; Dynamically weighted fusing the risk score and the degradation level through a piecewise linear function to generate an equipment-specific control threshold; Triggering the execution conditions of a multi-level process adjustment protocol according to the threshold.

4. The PLC data acquisition, monitoring and control method for automated industrial equipment according to claim 1, characterized in that, The construction steps of the process chain collaborative constraint model include: Collecting historical interaction data among equipment groups and constructing a process causal graph data set; Inferring the energy efficiency transfer coefficient and delay response parameter between equipment through a Bayesian network; Combining digital twin technology to construct a virtual simulation environment for the process chain and quantifying the energy flow balance constraint between nodes; Inputting the constraint and real-time working condition disturbance factors into a spatio-temporal graph neural network to generate the process chain collaborative constraint model.

5. The PLC data acquisition, monitoring and control method for automated industrial equipment according to claim 4, characterized in that It also includes: Identifying key energy flow bottleneck nodes according to the simulation results of the process chain collaborative constraint model; Configuring a dynamic energy efficiency compensation strategy for the nodes in the instruction set; Automatically generating a cross-equipment collaborative tuning scheme based on the compensation strategy, including load balancing distribution instructions and power factor correction parameters.

6. The PLC data acquisition, monitoring and control method for the automated industrial equipment according to claim 3, wherein, The calculation of the system cascade failure risk score includes: Obtaining the real-time equipment communication delay matrix and spare parts inventory distribution map, and constructing a system resilience assessment tensor; Extracting implicit correlation factors among multi-dimensional features through a tensor decomposition algorithm; Performing a Hadamard product operation on the correlation factors and the assessment tensor to obtain a comprehensive risk score.

7. The PLC data acquisition, monitoring and control method for automated industrial equipment according to claim 2, characterized in that, The embedding of the state space constraint conditions includes: Performing Lie group symmetry analysis on the feature enhancement results to screen out time-series evolution patterns that conform to physical laws; Generate state transition trajectories that satisfy the device dynamics constraints through Markov chain Monte Carlo sampling; Use the trajectory data to sparsely train the convolutional kernels of the temporal convolutional network to ensure that the model output conforms to the conservation law of the process chain.

8. The PLC data acquisition, monitoring and control method for automated industrial equipment according to claim 1, characterized in that The execution of the distributed edge computing framework includes: Define a loss function for multi-node collaborative optimization, including the dual objectives of minimizing energy consumption and maximizing production efficiency; Train the local control strategies of each edge node through the asynchronous parallel gradient descent algorithm; In each iteration, dynamically adjust the weight allocation ratio between nodes according to the degree of target deviation; Output the device collaborative control signal that satisfies the Pareto optimal condition; Among them, the loss function is: Wherein, represents the loss value, represents the total system energy consumption score, represents the production efficiency score, is the dynamic adjustment factor.

9. The PLC data acquisition, monitoring and control method for automated industrial equipment according to claim 5, characterized in that It also includes: After configuring the dynamic energy efficiency compensation strategy, monitor the energy efficiency fluctuation entropy value of the energy flow bottleneck node in real time; If the entropy value exceeds the preset safety boundary, trigger the chaotic optimization algorithm to re-plan the load allocation topology of the device group.

10. An automated industrial equipment PLC data acquisition, monitoring and control system, characterized in that, It includes: Distributed data acquisition module, deployed at the edge nodes of the industrial site, used to obtain multi-source PLC data streams in real time; Multi-modal fusion computing module, connected to the data acquisition module, built-in temporal convolutional neural network and generative adversarial network, to perform data denoising and feature extraction; Dynamic policy generation module, integrating the process chain collaborative constraint model and the digital twin simulation environment, to generate a multi-objective optimization instruction set; Edge collaborative control module, adjust the device control parameters through the asynchronous parallel computing framework, and output the collaborative control signal; Industrial communication gateway, transmit the control signal to the PLC execution terminal, and feedback the real-time working conditions to the monitoring center.

Images