Intelligent control method and system for double-table-board laser cutting machine

Abstract

The invention relates to an intelligent control method and system for a double-table-board laser cutting machine in the field of new-generation information technology, and the method comprises the steps: collecting complex workpiece parameters and processing duration data of each task in the double-table-board laser cutting machine in real time, and carrying out the preliminary sorting of an initial task sequence through a genetic algorithm, thereby obtaining a preliminarily optimized task distribution sequence; for a low-power operation scheme, a cooperative operation time table is generated in combination with the balanced task allocation sequence, potential energy waste points are extracted from the time table, and a refined energy management sequence is obtained by fusing task duration data through an information processing link; monitoring a real-time equipment state, if an idle state is detected, switching to low-power operation, and determining a final task coordination and energy consumption control model by continuously updating sequence parameters; and for the final task cooperation and energy consumption control model, historical operation data is obtained for verification, and if a verification result shows that the production efficiency is improved, an optimized double-table-board laser cutting machine operation protocol is output.

Description

Technical Field

[0001] This invention relates to the field of next-generation information technology, and in particular to an intelligent control method and system for a dual-table laser cutting machine. Background Technology

[0002] In modern manufacturing, laser cutting technology has become an indispensable core process due to its high precision and efficiency, especially in metal processing and automobile manufacturing. The operational efficiency and energy utilization of this equipment directly impact a company's production costs and environmental responsibility. With the promotion of green manufacturing concepts, how to improve equipment efficiency while reducing energy consumption has become a critical issue that the industry urgently needs to address. Dual-table laser cutting machines, as advanced equipment, can theoretically significantly improve production efficiency through the alternating operation of two worktables; however, they face numerous challenges in practical applications.

[0003] Current technological solutions have significant shortcomings in task allocation and energy management. Many devices rely on fixed operating modes and cannot be flexibly adjusted according to actual production needs, resulting in uneven resource utilization and limited efficiency improvements. A deeper problem lies in the lack of precise control over energy consumption during equipment operation, often leading to unnecessary energy waste, especially during equipment downtime or non-full-load operation, where energy utilization efficiency is low. This situation not only increases operating costs but also contradicts the goals of energy conservation and emission reduction. Focusing on specific technical challenges, the core issues of dual-table laser cutting machines lie in the imbalance of task allocation and the crude management of energy consumption. The imbalance in task allocation manifests in the difficulty of synchronizing the working times of the two worktables. When one table takes a long time to process complex workpieces, the other table is often idle, causing a decrease in overall equipment efficiency. This imbalance further exacerbates the difficulty of energy management, because when the equipment is waiting or idling, core components such as the laser and dust extraction system continue to operate at high power, generating a large amount of unnecessary energy consumption. For example, in actual production, one workbench may take half an hour to process heavy materials, while another workbench may only take a few minutes to complete due to its simple task. Afterward, the workbench will be idle for a long time, but the laser and dust extraction equipment will still be running at full power, resulting in significant energy waste.

[0004] Therefore, achieving efficient collaboration between the two worktables in dynamically changing production tasks, while simultaneously implementing precise energy consumption control based on equipment operating status, has become a key issue in improving the overall performance of dual-table laser cutting machines. This issue not only concerns improving production efficiency but also directly impacts a company's competitiveness in energy conservation, environmental protection, and cost control. Summary of the Invention

[0005] This invention provides an intelligent control method for a dual-table laser cutting machine, which mainly includes the following steps:

[0006] By real-time acquisition of complex workpiece parameters and processing time data for each task in a dual-table laser cutting machine, a genetic algorithm is used to initially sort the initial task sequence, resulting in a preliminary optimized task allocation sequence. Based on this preliminary optimized sequence, the expected completion time difference between the two worktables is obtained. If the time difference exceeds a preset threshold, an imbalance is identified, and the balanced task allocation sequence is determined by adjusting the position of the complex workpieces within the sequence. For the balanced task allocation sequence, the operating status data of core components is monitored, extracting high-power operating periods during idle periods. A neural network model is used to predict energy consumption during these periods, resulting in a predicted energy consumption distribution. Based on the predicted energy consumption distribution, the overall production efficiency index of the equipment is obtained. If the predicted energy consumption distribution shows that the proportion of idle periods is higher than a preset threshold, optimization is required, and a low-power operation scheme is determined by dynamically adjusting the power level of core components. For the low-power operation scheme, a collaborative operation schedule is generated based on the balanced task allocation sequence. Potential energy waste points are extracted from the schedule, and a refined energy management sequence is obtained by integrating task duration data through information processing. Based on the refined energy management sequence, the real-time equipment status is monitored. If an idle state is detected, the system switches to low-power operation. The final task coordination and energy consumption control model is determined by continuously updating the sequence parameters. Historical operating data is obtained to verify the final task coordination and energy consumption control model. If the verification results show an improvement in production efficiency, the optimized dual-table laser cutting machine operation protocol is output.

[0007] This invention provides an intelligent control system for a dual-table laser cutting machine, mainly comprising: a task sequence preliminary optimization module, used to collect complex workpiece parameters and processing time data of each task in the dual-table laser cutting machine in real time, and use a genetic algorithm to perform preliminary sorting of the initial task sequence to obtain a preliminary optimized task allocation sequence; a task allocation balance adjustment module, used to obtain the expected completion time difference between the two worktables based on the preliminary optimized task allocation sequence, and if the time difference exceeds a preset threshold, it is judged that an imbalance exists, and the balanced task allocation sequence is determined by adjusting the position of the complex workpiece in the sequence; an energy consumption prediction module, used to monitor the operating status data of the core components for the balanced task allocation sequence, extract the high-power running periods in the idle state, and use a neural network model to predict the energy consumption of these periods to obtain the predicted energy consumption distribution; and a power optimization decision module, used to obtain the overall production efficiency index of the equipment based on the predicted energy consumption distribution, and if the predicted energy consumption distribution shows that the proportion of the idle state is higher than a preset threshold, it is judged that optimization is needed, and a low-power operation scheme is determined by dynamically adjusting the power level of the core components. A collaborative operation generation module is used to generate a collaborative operation schedule based on a balanced task allocation sequence for low-power operation schemes. Potential energy waste points are extracted from the schedule, and a refined energy management sequence is obtained by integrating task duration data through information processing. A real-time energy consumption control module monitors the real-time equipment status based on the refined energy management sequence. If an idle state is detected, it switches to low-power operation. The final task collaboration and energy consumption control model is determined by continuously updating sequence parameters. A model verification output module verifies the final task collaboration and energy consumption control model using historical operating data. If the verification results show improved production efficiency, an optimized dual-table laser cutting machine operation protocol is output. The technical solution provided by this invention can include the following beneficial effects:

[0008] This invention discloses an intelligent control method for task coordination and energy consumption optimization in dual-table laser cutting machines, aiming to solve interconnected business scenario problems such as unbalanced task allocation, energy waste, and low production efficiency. By real-time acquisition of complex workpiece parameters and processing times, this invention employs a genetic algorithm to initially optimize the task sequence and adjusts the workpiece position based on the expected completion time difference between the two tables to achieve load balancing. Simultaneously, it utilizes a neural network to predict energy consumption distribution during idle periods, dynamically adjusting the power levels of core components to generate a low-power operation scheme. Combining the collaborative operation schedule and task duration data, the energy management sequence is refined, ultimately forming a task coordination and energy consumption control model, which is then validated using historical data to output an optimized operation protocol. This invention, through the deep integration of task allocation, energy management, and power adjustment, significantly improves equipment production efficiency, reduces energy waste, and provides an innovative solution for the efficient operation of dual-table laser cutting machines. Attached Figure Description

[0009] Figure 1 This is a flowchart of the intelligent control method for the dual-table laser cutting machine of the present invention.

[0010] Figure 2 This is a schematic diagram of the intelligent control method for the dual-table laser cutting machine of the present invention.

[0011] Figure 3 This is another schematic diagram of the intelligent control method for the dual-table laser cutting machine of the present invention.

[0012] Figure 4 This is a schematic diagram of the framework of the intelligent control system of the dual-table laser cutting machine of the present invention. Detailed Implementation

[0013] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] like Figure 1-4 The intelligent control method for the dual-table laser cutting machine in this embodiment may specifically include:

[0015] Step S101: By collecting the complex workpiece parameters and processing time data of each task in the dual-table laser cutting machine in real time, a genetic algorithm is used to perform preliminary sorting of the initial task sequence to obtain a preliminary optimized task allocation sequence.

[0016] Real-time complex workpiece parameters and processing time data are collected from the dual-table laser cutting machine via an interface to determine a complete data set. The workpiece complexity index is quantified based on this complete data set, and a quantified value of complexity is obtained by weighted summation of the parameters and processing time. An initial task sequence is generated using this complexity quantification value to obtain an initial version of the sequence. A genetic algorithm is used to select and mutate the initial sequence version, where the input is the task priority in the initial sequence version and the output is the adjusted sequence, determining the optimal sequence path. A table load balancing mechanism is incorporated into the optimized sequence path, distributing tasks from the path across both tables to balance processing time, resulting in a preliminarily optimized task allocation sequence.

[0017] Specifically, in one implementation, a dual-table laser cutting machine is used to process various metal workpieces, such as stainless steel sheets or aluminum alloy parts, which have different complex parameters, including contour curvature, cutting path length, and material thickness. The process of acquiring these parameters in real time is achieved through sensors mounted on the cutting table.

[0018] For example, a laser scanner captures workpiece contour data while a timer records the processing time for each task. This acquisition ensures real-time data updates, providing a basis for subsequent optimization.

[0019] Specifically, during the cutting process, the system automatically records the complex workpiece parameters and cumulative processing time of the current task at fixed intervals, such as every 5 seconds, thus forming a dataset of the initial task sequence. Further, a genetic algorithm is used to initially sort the initial task sequence. The genetic algorithm is an optimization method based on the principle of natural selection, which simulates the biological evolution process to search for the optimal solution. In this embodiment, the initial task sequence is first encoded as chromosomes, with each chromosome representing a task arrangement.

[0020] For example, tasks A, B, and C are encoded as sequences [1,2,3]. Then, the quality of each sequence is evaluated using a fitness function, calculated based on the total processing time and the complexity parameters of the task, such as incorporating complexity parameter weights into the formula for minimizing the total processing time. In this way, the algorithm starts optimization from a randomly generated initial population. Preferably, the detailed process of the genetic algorithm includes selection, crossover, and mutation steps. In the selection step, the system selects chromosomes with high fitness based on their fitness probabilities.

[0021] For example, roulette wheel selection is used to retain sequences with shorter processing times. The crossover step involves exchanging parts of genes between two chromosomes, such as crossing sequences [1,2,3] and [3,1,2] to generate new sequences [1,1,2] and [3,2,3], to explore new permutations. The mutation step randomly changes gene positions, such as mutating [1,2,3] to [1,3,2], introducing diversity to avoid local optima. After multiple iterations, typically 50 generations, the algorithm gradually converges to the optimal sequence. This method is particularly suitable for dual-table laser cutting machines because it can balance the load on both tables, reducing waiting time. For example, in one possible implementation, for a batch of sequences containing 10 tasks, each task has complex workpiece parameters such as cutting path lengths ranging from 50 cm to 200 cm and processing times ranging from 2 minutes to 10 minutes. The genetic algorithm first initializes the population size to 100, then iteratively optimizes through the above steps, finally outputting a preliminarily optimized task allocation sequence, such as allocating highly complex tasks to idle tables for parallel processing. It should be noted that the preliminary optimized task allocation sequence is the sorting result obtained through a genetic algorithm, which takes into account the parallel characteristics of the two platforms.

[0022] For example, the sequence divides tasks into two groups, assigning one group to the first worktable and the other to the second, ensuring that the total processing time is minimized. This sequence can be directly imported into the control system for execution in actual production. In another implementation, for different workpiece types, such as the cutting of precision electronic components, the system adjusts the parameters of the genetic algorithm, for example, increasing the mutation probability to 0.1, to adapt to more complex parameter variations. In this way, the optimized sequence can better handle sudden task insertions, demonstrating the algorithm's flexibility. Specifically, the innovation of the genetic algorithm lies in its integration of complex workpiece parameters.

[0023] For example, parameters can be converted into weighting factors and incorporated into fitness calculations. This allows the system to dynamically respond to production demands, achieving efficient scheduling in the laser cutting field. Furthermore, through this initial sorting, the system can reduce the overall production cycle.

[0024] For example, when processing a batch of automotive parts, the optimized sequence places long-duration tasks on lower-load tables, thereby improving throughput. Similarly, when processing building materials in a dual-table laser cutting machine, data including workpiece thickness parameters is collected, and a genetic algorithm sorts the data accordingly to ensure that tasks involving thick materials do not overlap on the same table. Understandably, this method is applicable to various laser cutting scenarios, such as cutting wood composite boards in furniture manufacturing; the algorithm similarly generates optimized sequences and supports multi-task parallelism.

[0025] Step S102: Based on the preliminary optimized task allocation sequence, obtain the expected completion time difference between the two workbenches. If the time difference exceeds a preset threshold, it is determined that there is an imbalance. The balanced task allocation sequence is determined by adjusting the position of the complex workpiece in the sequence.

[0026] The overall weight of this technical topic indicates a medium level of difficulty, involving complex techniques such as sequence optimization and adjustment, unclear business processes such as the initial sequence acquisition method, and technical business attribute details such as the definition of complex workpieces that require further explanation. The expected completion time difference between two workstations is obtained from the task allocation sequence. If the time difference exceeds a preset threshold, an imbalance is identified. The balanced task allocation sequence is determined by adjusting the position of complex workpieces in the sequence. Based on the initially optimized task allocation sequence, the total processing time of each workstation is obtained, yielding the expected completion time difference. This expected completion time difference is compared with a preset threshold; if it exceeds the threshold, an imbalance is identified. To address this imbalance, the positions of complex workpieces are extracted from the task allocation sequence. A genetic algorithm is used to adjust these positions by initializing the population and performing selection, crossover, and mutation operations, resulting in an adjusted sequence. The positions of complex workpieces are determined based on workpieces whose processing time exceeds the average. The updated expected completion time difference is obtained from the adjusted sequence to determine the balanced task allocation sequence.

[0027] Specifically, in one implementation, the system first obtains a preliminary optimized task allocation sequence, which is arranged based on the workpiece processing requirements of the production line. For example, on an automotive parts assembly line, parts of different complexities are assigned to two parallel worktables. The expected completion time is calculated by summing the processing time of each workpiece on its corresponding worktable.

[0028] Specifically, for each workstation, the system iterates through the sequence of workpieces assigned to it, summing the standard processing times for each workstation. These times are derived from historical data or simulation estimates to obtain the total expected completion time for each workstation. Further, the difference between the expected completion times of two workstations is calculated, i.e., the larger completion time minus the smaller completion time. If this difference exceeds a preset threshold, for example, 10% of the total processing time, an imbalance is identified. This mechanism helps identify load deviations and ensures the necessity of subsequent adjustments.

[0029] It should be noted that the threshold can be dynamically set according to the scale of the production line. In small assembly lines, the threshold is tighter to pursue a fine balance, while in large production lines, the threshold is slightly wider to tolerate slight deviations.

[0030] For example, after determining an imbalance, the system determines a balanced task allocation sequence by adjusting the positions of complex workpieces in the sequence. Complex workpieces are defined as those whose processing time exceeds the average; for example, on an electronics assembly line, circuit boards involving multi-step soldering are considered complex workpieces. The adjustment process involves identifying the current positions of these complex workpieces in the sequence and then attempting to move them from heavier workbenches to lighter workbenches, or swap the positions of adjacent workpieces, to minimize the time difference.

[0031] Specifically, the adjustment logic employs an iterative approach: first, complex workpieces are sorted in descending order of time; then, the new time difference after moving each complex workpiece to another workbench is tested one by one. If the new difference is less than the original difference and does not exceed a threshold, the adjustment is adopted; otherwise, the next position is tested. For example, in a furniture manufacturing production line, if a workbench sequence contains multiple large wooden processing tasks that result in excessive time, the system can move one of the complex wooden parts to the middle position of another sequence, thereby balancing the total time. This method ensures sequence balance through multiple iterations and typically converges after several rounds of adjustments. Preferably...

[0032] In one possible implementation, the system can introduce auxiliary constraints, such as keeping the workpiece dependency order unchanged, to avoid adjustments disrupting the production logic.

[0033] For example, on a textile processing line, dyeing of workpieces must be done after cutting, and adjustments are limited to positional exchanges that do not violate this sequence. This enhances the practicality of the solution.

[0034] Understandably, the technical effect of this balancing process is to reduce waiting time and increase the overall throughput of the production line. In practical applications, such as in the assembly of mechanical parts, the adjusted sequence can reduce the time difference between two workstations to within a threshold, thereby achieving more efficient resource utilization. In another embodiment, for continuous production environments, such as food packaging lines, the initial sequence may be generated based on order priority, and the time difference calculation takes into account machine changeover time. When adjusting the position of complex workpieces, such as large packaging tasks, the system prioritizes moving the end of the sequence to minimize the impact on earlier tasks. Furthermore, this adjustment mechanism supports multiple parameter configurations; for example, the threshold can be set to a fixed or adaptive value, adjusted according to real-time load feedback, thereby adapting to dynamic production changes.

[0035] Step S103: For the balanced task allocation sequence, monitor the operating status data of the core components, extract the high-power operating segments in the idle state, and use a neural network model to predict the energy consumption of these segments to obtain the predicted energy consumption distribution.

[0036] For the balanced task allocation sequence, the operating status data of core components is monitored, and high-power operating segments in idle states are extracted. A neural network model is used with these high-power operating segments as input, and the data is processed through preset hidden layer nodes and activation functions to predict the energy consumption of these segments, resulting in a predicted energy consumption distribution. Based on the predicted energy consumption distribution, relevant data on component load balancing is obtained. By calculating the maximum difference in power values ​​in the distribution, the range of power peak variation for each period is determined, resulting in a peak adjustment sequence. Using the peak adjustment sequence, during the sequence balancing adjustment process, if the peak exceeds a preset threshold, a convolutional neural network is used with the peak adjustment sequence as input. Spatial features are extracted through convolutional layers, and classification is performed through fully connected layers to extract features for consumption distribution optimization, resulting in an optimized distribution feature. Using the optimized distribution feature, additional backup periods are extracted from component load balancing. By calculating the similarity between the distribution feature and the backup periods, the matching degree between the backup periods and high-power periods is determined, resulting in a set of matched periods. Based on the set of matched periods, for the backup allocation of energy consumption, the consumption values ​​of the matched periods in the set are added to the original distribution to obtain an extended consumption distribution, resulting in an extended energy consumption distribution.

[0037] Specifically, in one implementation, for the balanced task allocation sequence, the operating status data of the core components are first monitored.

[0038] Specifically, core components can be processors or memory modules in data center servers. These components generate real-time operational data after task allocation, including load levels, power consumption, and status indicators. This data is collected in real time by deploying sensors or software monitoring tools, such as recording processor utilization and power values ​​once per second. This monitoring ensures the continuity of data center energy management and provides a foundation for subsequent extraction. Further, high-power operating segments in idle states are extracted from the monitored operational status data. Idle state refers to a situation where the load of core components is below a preset threshold, such as utilization less than 10%, but power consumption is still higher than normal idle levels. The extraction process for high-power operating segments involves data filtering and time period segmentation: first, the data sequence is scanned to identify continuous time segments with low load but high power; then, a threshold comparison method is used, for example, setting a power threshold of 1.5 times the average idle power, to mark the time periods that meet the criteria. This extraction helps identify potential energy waste points, which often occur in data center scenarios, such as server idle periods after virtual machine migration. In this way, the extracted time period list can be used for further analysis.

[0039] Preferably, a neural network model is used to predict energy consumption during these periods. The neural network model is a multilayer perceptron-based structure, including an input layer, hidden layers, and an output layer, used to learn patterns from historical data.

[0040] Specifically, the model takes extracted time-period features as input, such as time period length, average power, and ambient temperature. Through forward propagation, the hidden layer applies activation functions like ReLU to handle non-linear relationships, and the output layer generates predicted energy consumption values. The model training process uses historical datasets, such as server operation logs from the past week, and adjusts the weights using a backpropagation algorithm to minimize the mean squared error. This prediction, in data center energy optimization, can assess consumption during idle high-power periods in advance, helping the scheduling system adjust task allocation.

[0041] In one possible implementation, the construction of the neural network model includes data preprocessing and parameter optimization.

[0042] For example, the extracted time-period data is first normalized, scaling the power values ​​to the range of 0 to 1; then, a training set and a test set are partitioned, with 80% used for training. The model architecture can be a three-layer fully connected network with 64 hidden layer nodes and 100 training iterations. This setup ensures the model's applicability to different load scenarios in data centers, such as energy forecasting during peak and off-peak periods.

[0043] It should be noted that the predicted energy consumption distribution is obtained by summarizing the output of the model.

[0044] Specifically, the predicted values ​​from multiple extraction periods are statistically analyzed to generate distribution maps, such as histograms showing the probability density of consumption. This distribution reflects the energy patterns during idle high-power periods, and in data center management, it can guide cooling system adjustments or task reallocation to achieve more efficient resource utilization.

[0045] For example, in a data center implementation scenario, after monitoring processor operating status data for the task allocation sequence of a cloud server cluster, several high-power periods during the nighttime idle period are extracted, such as each period lasting 5 to 10 minutes. Using the aforementioned neural network model for prediction, the results show that the average energy consumption during these periods is 1.2 times the normal level, and the distribution follows a normal pattern. This prediction helps administrators optimize power management strategies. Furthermore, in another implementation, a combination of different core components is considered, such as simultaneously monitoring processors and storage devices. When extracting periods, a joint threshold is introduced; for example, high-power idle periods are marked when processor power exceeds a threshold and storage access rate is low. The neural network model input is expanded into a multi-dimensional feature vector, and the predicted distribution is enhanced robustness through Monte Carlo simulation. This variant is suitable for large-scale data centers, covering scenarios from small servers to large clusters.

[0046] Understandably, the technical advantage of this method lies in improving data center energy efficiency through accurate prediction.

[0047] For example, in practical applications, the prediction distribution can be used to dynamically adjust task sequences, reduce the occurrence of idle high-power periods, and thus reduce overall energy consumption.

[0048] In one embodiment, for task allocation in edge computing data centers, the operating status of core components such as GPUs is monitored. After extracting idle high-power periods from the data, a variant of a convolutional neural network is used for prediction. This model takes time-series data as input, and convolutional layers extract local patterns to output energy consumption distribution. This implementation demonstrates the versatility of the method in high-performance computing scenarios.

[0049] Step S104: Based on the predicted energy consumption distribution, obtain the overall production efficiency index of the equipment. If the predicted energy consumption distribution shows that the idle state ratio is higher than the preset threshold, it is determined that optimization is needed. A low-power operation scheme is determined by dynamically adjusting the power level of the core components.

[0050] By analyzing the energy consumption data recorded in historical operation logs, a production efficiency index is obtained by analyzing the energy consumption distribution. Based on the proportion of idle periods in the production efficiency index, an optimization requirement is determined when the idle period percentage exceeds a preset threshold. To address this optimization requirement, a low-power solution is obtained by dynamically adjusting the power level based on the current load data of core components. From the time-series power changes in the low-power solution, the overall standby scheduling distribution of the equipment is obtained to determine the allocation of backup energy.

[0051] Specifically, in one implementation, predicting energy consumption distribution is achieved by collecting historical energy data during equipment operation.

[0052] Specifically, the process begins with real-time monitoring of equipment such as robotic arms or conveyor systems on a production line, recording energy consumption for each time period, such as hourly electricity usage. Then, statistical methods are used to construct a distribution model, categorizing energy consumption into different states, such as operating, idle, and peak states. This distribution can be represented as a probability density function, helping to identify patterns in energy use.

[0053] It should be noted that the forecasting process is based on trend analysis of data from the past week to ensure that the distribution reflects the actual operating patterns of the equipment, thus providing a foundation for subsequent efficiency assessment. Furthermore, obtaining the overall production efficiency index of the equipment relies on the aforementioned analysis of the predicted energy consumption distribution. In this step, the total energy consumption in the distribution is correlated with the equipment's output output.

[0054] For example, production efficiency can be defined as the quantity of product produced per unit of energy consumption, quantified using formulas such as Efficiency = Total Output / Total Energy Consumption. If the distribution shows peak consumption concentrated during high-load periods, the efficiency indicator will be adjusted accordingly to reflect overall performance. This method ensures the objectivity of the indicator and is applicable in industrial equipment fields such as manufacturing workshops, covering different production line scenarios.

[0055] For example, if the predicted energy consumption distribution shows that the proportion of idle time is higher than a preset threshold, then optimization is deemed necessary. Here, the proportion of idle time refers to the percentage of time in the distribution without production activity, for example, calculated using an integral distribution curve. The preset threshold can be set to 20%, based on the equipment type, such as an injection molding machine or an assembly line. If the proportion exceeds the threshold, the system will trigger an optimization signal. This judgment logic is applicable to various scenarios in the same field, such as monitoring intermittent idle time in a continuous production environment, or evaluating standby time in intermittent operations to identify energy waste points.

[0056] In one possible implementation, a low-power operating scheme is determined by dynamically adjusting the power levels of core components. Specifically, core components such as motors or heating elements are identified as primary energy consumers. The adjustment process involves real-time monitoring of the component load and gradually reducing power, for example, from 100% full load to 60%, while ensuring uninterrupted production. When the scheme is determined, multiple power combinations are simulated, and the configuration that maintains the minimum output requirement is selected. This dynamic mechanism demonstrates versatility in equipment optimization, such as in textile machinery or food processing equipment, to ensure improved energy efficiency.

[0057] Preferably, in the above scheme, a feedback loop can be introduced to improve and adjust.

[0058] For example, energy distribution is re-predicted after adjustments, and the new efficiency metrics are compared with the original values. If the idle rate decreases after optimization, the solution is considered effective. This approach enhances the flexibility of the technology and is applicable to different implementations within the same industrial context, such as small workshops or large factories.

[0059] Specifically, the detailed process of predicting energy consumption distribution includes data cleaning and model fitting. First, outliers such as sudden spikes are cleaned, and then regression analysis is used to fit the distribution curve.

[0060] For example, in manufacturing equipment, the distribution might show weekday peaks and weekend slumps. This interpretation allows us to understand how the distribution captures the antecedents of energy patterns, such as overall inefficiency caused by high idle time, thus providing a basis for optimization. Furthermore, the threshold for determining what needs optimization can be dynamically updated based on historical experience.

[0061] In one embodiment, if the equipment operating environment changes, such as seasonal production fluctuations, the threshold is adjusted from 15% to 25%. This flexibility ensures accurate judgment and, when applied in fields such as electronic assembly lines, results in the technical effect of reducing energy waste.

[0062] For example, when determining a low-power operation plan, the dependencies between components can be considered. Dynamic adjustments not only reduce the power of individual components but also coordinate the overall system, such as synchronously reducing the power levels of conveyor belts and pumps. This approach can achieve energy savings in practice while maintaining production continuity.

[0063] Understandably, the logical chain from prediction to optimization ensures consistency throughout the entire process. Through these steps, equipment achieves efficient operation in industrial settings.

[0064] Step S105: For the low-power operation scheme, a collaborative operation schedule is generated by combining the balanced task allocation sequence. Potential energy waste points are extracted from the schedule, and a refined energy management sequence is obtained by integrating task duration data through information processing.

[0065] A low-power scheme is used to obtain balanced allocation results from the task allocation sequence and determine a collaborative schedule. For this collaborative schedule, operational time planning is employed to identify idle periods within the time intervals and extract energy waste points, resulting in potential waste extraction. If the potential waste extraction exceeds a preset threshold, task duration data is fused through an information processing stage to obtain a data fusion process. Based on this data fusion process, an energy optimization sequence is integrated for the refining management sequence to determine load balancing adjustments. Real-time monitoring data is obtained from the load balancing adjustments to obtain the refining energy management sequence.

[0066] Specifically, in one implementation, regarding low-power operation schemes, it is first necessary to understand that low-power operation schemes refer to energy-saving strategies designed to extend battery life in IoT device networks, such as in smart home sensor systems, where unnecessary power consumption is reduced by adjusting device operating modes.

[0067] Specifically, this approach involves monitoring the status of devices and dynamically allocating tasks to ensure that the overall network maintains functionality in a low-power state.

[0068] For example, in a smart lighting system, a low-power operation scheme might involve distributing sensor tasks to devices with more standby modes, thereby avoiding high power consumption peaks. This achieves energy balance among devices, preventing rapid power depletion caused by overloading a single device. Furthermore, the process of generating a collaborative operation schedule based on the balanced task allocation sequence requires explaining what the balanced task allocation sequence is. This refers to a sequence of tasks ordered using a load balancing algorithm, such as a greedy algorithm or genetic algorithm, to distribute computational tasks, ensuring that each IoT device bears approximately an equal load.

[0069] In one possible implementation, assuming a network composed of multiple sensor nodes, the task requirements data of each node are first collected, such as data acquisition and transmission tasks. Then, the estimated power consumption and duration of each task are calculated; for example, task A requires 2 seconds to execute and consumes 0.5 watts, while task B requires 3 seconds and consumes 0.4 watts. Next, a balancing algorithm is applied to distribute tasks across nodes, ensuring a uniform distribution of total power consumption. The generation of a coordinated operation schedule is based on this sequence, creating a timeline table, for example, dividing time into slots, with each slot assigned a specific task. For instance, node 1 executes task A from 0-2 seconds, and node 2 executes task B from 1-4 seconds, thus ensuring coordination between tasks and avoiding energy conflicts caused by overlap. This schedule generation can coordinate the operation of multiple devices, achieving low-power operation of the network as a whole. The step of extracting potential energy waste points from the schedule is the core innovation and requires a detailed explanation of its business process. Potential energy waste points refer to those periods or tasks in the schedule that lead to unnecessary power consumption, such as excessively long idle periods or peak power consumption caused by task overlap. First, the structure of the schedule is analyzed; for example, a schedule records the device status and power consumption values ​​for each time slot. The extraction process can be achieved using a scanning algorithm.

[0070] Specifically, the timeline is traversed to calculate the difference between the actual power consumption of each slot and the ideal low power threshold. If the difference exceeds a preset threshold, such as 0.2 watts, it is marked as a waste point.

[0071] For example, in a smart home network, suppose a schedule shows that node 1 is idle for 5-7 seconds but remains awake, consuming 0.1 watts, while ideally it should be in sleep mode (0 watts). This is identified as a waste point. Further, the process involves comparing adjacent slots; if a node is idle for more than 3 consecutive seconds, it is considered a potential waste, and its start time and duration are recorded. This extraction identifies inefficient energy use caused by task delays or inconsistencies, providing a basis for subsequent optimization.

[0072] In one embodiment, this step can have the technical effect of helping the system adjust its schedule by identifying waste points early, thereby reducing overall energy consumption by more than 15%.

[0073] Preferably, after extracting waste points, an information processing stage is used to integrate task duration data to obtain a refined energy management sequence. Here, it's necessary to explain what the information processing stage is; it refers to a data integration and optimization phase, including filtering, aggregating, and analyzing task-related data. Specifically, the process first collects task duration data, for example, obtaining the actual execution time of each task from a schedule, such as task A actually taking 2.5 seconds instead of the expected 2 seconds. Then, in the integration stage, this data is integrated using weighted averaging or machine learning methods, such as calculating the average duration of all tasks and combining it with the waste point data to generate a refined sequence.

[0074] For example, in an IoT sensor network, the fused sequence might be an optimized list of tasks, such as adjusting the duration of task A to 2.2 seconds and removing idle tasks associated with waste points, thus forming a more compact energy management sequence. This sequence ensures the continuity of low-power operation. In another implementation, for a smart environmental monitoring system within the same IoT domain, the low-power operation scheme can be extended to multi-sensor collaboration.

[0075] For example, when balancing task allocation sequences, environmental factors such as temperature affect power consumption. After generating the schedule, waste points are extracted and the focus is on the additional cooling power consumption during high-temperature periods. By integrating duration data through information processing, the refined sequence can be used to dynamically adjust monitoring frequency, achieving more efficient energy management.

[0076] It should be noted that the logical connection of the above process is that the generation of the timetable directly provides input for waste extraction, and the extraction results are then input into the fusion process, forming a closed optimization chain.

[0077] For example, in a specific scenario, a waste point scan is performed immediately after the collaborative operation schedule is generated to ensure real-time performance. Furthermore...

[0078] In one possible implementation, the detailed process for identifying potential energy waste points may also include a threshold setting step.

[0079] Specifically, the first step is to define quantitative indicators of energy waste, such as power consumption deviation rate. This is calculated by comparing the expected power consumption of each task in the schedule with the actual recorded value. For example, if the expected power consumption is 0.4 watts and the actual power consumption is 0.6 watts, the deviation rate is 50%. If the deviation rate exceeds 20%, it is identified as a waste point. When applied to smart home systems, this method can accurately pinpoint waste such as repeated wake-ups of light sensors due to signal interference, thereby guiding the fusion process to adjust task duration data and generate refined sequences.

[0080] For example, the output of a refined energy management sequence can be an ordered list including task ID, adjusted duration, and power consumption estimate, such as Task 1: 1.8 seconds, 0.3 watts; Task 2: 2.5 seconds, 0.4 watts. This sequence can be directly applied to equipment scheduling in low-power operation scenarios to improve overall efficiency.

[0081] In one embodiment, the entire scheme is applied to wireless sensor networks. The schedule generated after balancing task allocation is synchronized through the cloud. Historical data is fused when waste points are extracted, and the sequence is further refined to adapt to changing network loads.

[0082] Understandably, these implementations demonstrate the versatility of the technology in the field of IoT energy management, ensuring flexible application of low-power operation through different scenarios such as home and environmental monitoring.

[0083] Step S106: Based on the refined energy management sequence, monitor the real-time equipment status. If an idle state is detected, switch to low-power operation. Determine the final task coordination and energy consumption control model by continuously updating the sequence parameters.

[0084] By monitoring real-time device status, idle state detection results are obtained from the energy management sequence. If an idle state is detected, the system switches to low-power operation, providing a basis for sequence parameter updates. Based on this update basis, sequence parameters are continuously updated, and parameters are integrated to determine the collaborative task allocation scheme. The correlation between device status monitoring data and the power switching mechanism is obtained through this collaborative task allocation scheme. If the correlation meets a preset threshold, energy consumption control model elements are integrated to obtain the final model framework. Based on this final model framework, the device load balancing allocation service is integrated to determine the final task collaboration and energy consumption control model.

[0085] Specifically, in one implementation, the energy management sequence refers to a set of predefined equipment operation instruction sequences, refined through data analysis, used to guide the energy allocation of equipment in different states.

[0086] Specifically, the sequence first extracts key parameters from historical operating data, such as equipment load rate and energy consumption curves, and then generates a refined version by filtering noise and optimizing paths. This refining process ensures that the sequence can adapt to real-time changes; for example, in industrial equipment systems, the sequence can schedule energy for machines on a production line to avoid unnecessary consumption. Furthermore, monitoring real-time equipment status involves deploying sensor networks to collect data such as temperature, power utilization, and runtime.

[0087] For example, if the device has no task input within a specified time, i.e., an idle state is detected, the system automatically switches to a low-power operating mode. This switching is achieved by adjusting voltage and frequency, such as reducing the device from full-speed mode to standby mode, thereby reducing energy consumption.

[0088] It should be noted that the detection of idle state is based on threshold judgment, such as power utilization rate being less than 20%, which helps to achieve energy saving without affecting overall performance.

[0089] Preferably, the continuous updating of sequence parameters is accomplished through a feedback loop mechanism. The specific process includes collecting real-time monitoring data, comparing it with the original sequence parameters, and if the deviation exceeds a set range, applying an incremental algorithm to adjust the parameters.

[0090] For example, in data center energy management scenarios, sequence parameters such as task allocation weights can be dynamically updated based on server load to ensure that the parameters reflect current environmental changes. The principle behind this update mechanism is iterative optimization. Initial parameters are inherited from historical models, and then refined step by step through error minimization to form a more accurate energy management sequence.

[0091] In one possible implementation, task coordination refers to the coordination of work among multiple devices, such as assigning high-priority tasks to efficient devices while moving low-priority tasks to backup units. This coordination is achieved through a shared communication protocol, aiming to balance loads and minimize total energy consumption. For example, in a smart building energy system, lighting and air conditioning tasks can be coordinated and energy can be allocated in real time based on occupancy to avoid resource waste.

[0092] Understandably, the determination of the energy consumption control model depends on the updated sequence parameters, and the final model is generated by integrating all the data.

[0093] Specifically, the model adopts a hierarchical structure. First, a base layer is built to represent the state of individual devices. Then, task coordination information is integrated in the upper layer, and finally, a control strategy is output.

[0094] For example, the model can calculate overall energy consumption forecasts and generate optimization schemes accordingly, such as limiting the operation of non-essential equipment during peak periods. This model building process ensures comprehensive energy management. Furthermore, in another embodiment, for energy management in a factory workshop, the system can monitor machine tool status and, if idleness is detected, switch to low power while updating sequence parameters to coordinate multi-machine tasks. In this scenario, the model determination process involves parameter fusion, such as using fan speed and motor power as inputs to generate an energy consumption control model, thereby enabling coordinated operation between equipment.

[0095] In one embodiment, the application of the above technical solution can realize the dynamic adjustment of equipment energy utilization. For example, during continuous operation, ineffective energy consumption is reduced by updating parameters, ensuring that the task is completed efficiently.

[0096] Step S107: For the final task coordination and energy consumption control model, obtain historical operating data for verification. If the verification results show an improvement in production efficiency, output the optimized dual-table laser cutting machine operation protocol.

[0097] Historical operational data is acquired, and energy consumption records and task execution logs for at least one cycle are extracted from the operational data storage of the dual-table laser cutting machine. A task density matrix is ​​constructed by mapping the task distribution in the records to the execution duration in the logs. Using this task density matrix, a preset threshold is applied to analyze the verification results. This threshold is based on historical average efficiency to determine whether production efficiency indicators have improved and to identify optimization trigger conditions. If the optimization trigger conditions are met, the model parameters for the energy consumption monitoring system are adjusted. These parameters include power allocation thresholds and cooling interval durations, resulting in an energy consumption optimization parameter set. Based on this energy consumption optimization parameter set, and combining it with material path planning—a new and unique business beyond the original theme—a dual-table coordination sequence is generated by mapping the cutting trajectory in the path planning to the parameter set, resulting in a draft operation protocol. Using this draft operation protocol, real-time fault detection—a new and unique business beyond the original theme—is integrated. Verified efficiency improvement data is extracted from the laser cutting process. This data is filtered by fault detection and then used to output an optimized dual-table laser cutting machine operation protocol.

[0098] Specifically, in one implementation, the task collaboration model of the dual-table laser cutting machine achieves task allocation by monitoring the cutting status of the two tables in real time.

[0099] Specifically, the model first collects the current load information of each table, such as the type and thickness of the material being cut, and then assigns new tasks to tables with lower loads according to preset priority rules. This allocation mechanism ensures the continuity of the cutting process and avoids downtime caused by overloading a single table.

[0100] It should be noted that the core of the task collaboration model lies in a scheduling algorithm that takes into account the urgency of the task and the expected completion time, and determines the allocation path by comparing the remaining capacity of the two platforms.

[0101] For example, when processing sheet metal cutting, if one face is cutting a thick plate, the model will move the thin plate task to another face, thus balancing the overall process. This model helps complete more cutting tasks in the same amount of time and is suitable for various laser cutting scenarios, such as precision parts machining or mass production. Furthermore, the energy consumption control model is a key component of this technical solution; it optimizes energy use by dynamically adjusting laser power and cutting speed.

[0102] In one possible implementation, the model first establishes an energy consumption function that calculates real-time energy consumption based on the laser's power output, the material's thermal conductivity, and the cutting path length.

[0103] Specifically, the calculation of the energy consumption function involves multiplying the power output by the cut-off time, and adding the energy consumption of auxiliary equipment such as the cooling system. For example, if the power is P and the time is T, then the total energy consumption E is approximately equal to P*T plus a constant correction value. This calculation ensures that the model can accurately predict and control peak energy consumption.

[0104] For example, when cutting aluminum alloys, the model monitors the heat loss of the laser beam and reduces overall energy consumption by decreasing unnecessary power. The detailed process includes initial parameter settings, real-time data feedback, and adjustment cycles. For instance, if energy consumption exceeds a threshold, the model automatically slows down the cutting to balance efficiency and energy saving. This approach is particularly effective in dual-table systems because it coordinates the energy distribution between the two tables, avoiding overall waste caused by high energy consumption on a single table. Through this control, the system can reduce energy costs while maintaining cutting quality, making it suitable for laser cutting environments with long-term continuous operation.

[0105] Preferably, the step of acquiring historical operation data is performed through the device's built-in log system.

[0106] Specifically, this step retrieves past operation records from the laser cutting machine's database, including the start time, end time, energy consumption readings, and production output for each task. This data acquisition process ensures the accuracy of the verification without introducing external variables.

[0107] In one embodiment, the verification process uses acquired historical data to evaluate the effectiveness of the task coordination and energy consumption control model.

[0108] Specifically, the verification process begins by inputting historical data into the model to simulate the operating scenario, and then comparing the simulation results with actual historical efficiency indicators, such as the number of pieces cut per unit time.

[0109] It should be noted that the validation logic is based on statistical comparison methods, such as calculating the percentage improvement in efficiency. If the simulated efficiency is more than 10% higher than the historical average, it is considered valid. This validation helps confirm the reliability of the model. In a scenario of batch cutting stainless steel sheets, validation shows that the model can shorten production time, thus proving its value in practical applications. Furthermore, if the validation results show improved production efficiency, an optimized operating protocol for the dual-table laser cutter is output. This protocol includes detailed operating guidelines, such as task allocation rules and energy consumption threshold settings.

[0110] Specifically, the protocol outlines the steps, such as first initiating a task collaboration model to allocate work, and then activating energy consumption control for real-time adjustments. This output ensures that users can directly apply the optimization solutions.

[0111] For example, in another implementation, the task collaboration model can be extended to handle cutting scenarios involving different material types.

[0112] For example, when cutting a mixture of plastic and metal, the model adjusts the allocation priority based on material properties, ensuring that heat-sensitive materials are processed first on the better-cooled platform. This extension demonstrates the model's versatility without changing the core logic.

[0113] Understandably, the calculation process of the energy consumption control model further refines the concept of wind load coefficient. Although this scheme does not directly involve wind, it draws on the principle of load calculation.

[0114] Specifically, if the cutting load is likened to air pressure, the model calculates the load-bearing capacity of a conductor analogous to air pressure. For example, it estimates the energy distribution along the laser path using formulas, but avoids complex numerical values, emphasizing only the process: first determining coefficients based on material density, then multiplying by a velocity factor to obtain the total value. This analogy helps in understanding the dynamic balance of energy consumption. In the field of laser cutting, this process ensures efficient energy management.

[0115] In one embodiment, the verification step may include multiple iterations, such as testing with a small dataset first and then expanding to the entire historical data. This approach enhances the robustness of the verification and is suitable for high-precision cutting, such as in the fabrication of electronic components.

[0116] Preferably, the output format of the optimized operation protocol is a structured document, which includes security notes and a parameter table.

[0117] For example, the protocol specifies a laser power limit of 2000W to match the control range of the energy consumption model. This design facilitates practical deployment. Furthermore, the overall effect of the solution is to improve the overall performance of the dual-table system through model integration, achieving higher throughput and lower energy consumption without increasing hardware costs.

[0118] This invention provides an intelligent control system for a dual-table laser cutting machine, mainly comprising: a task sequence preliminary optimization module, used to collect complex workpiece parameters and processing time data of each task in the dual-table laser cutting machine in real time, and use a genetic algorithm to perform preliminary sorting of the initial task sequence to obtain a preliminary optimized task allocation sequence; a task allocation balance adjustment module, used to obtain the expected completion time difference between the two worktables based on the preliminary optimized task allocation sequence, and if the time difference exceeds a preset threshold, it is judged that an imbalance exists, and the balanced task allocation sequence is determined by adjusting the position of the complex workpiece in the sequence; an energy consumption prediction module, used to monitor the operating status data of the core components for the balanced task allocation sequence, extract the high-power running periods in the idle state, and use a neural network model to predict the energy consumption of these periods to obtain the predicted energy consumption distribution; and a power optimization decision module, used to obtain the overall production efficiency index of the equipment based on the predicted energy consumption distribution, and if the predicted energy consumption distribution shows that the proportion of the idle state is higher than a preset threshold, it is judged that optimization is needed, and a low-power operation scheme is determined by dynamically adjusting the power level of the core components. The collaborative operation generation module generates a collaborative operation schedule based on a balanced task allocation sequence for low-power operation schemes. It extracts potential energy waste points from the schedule and integrates task duration data through information processing to obtain a refined energy management sequence. The real-time energy consumption control module monitors the real-time equipment status based on the refined energy management sequence. If an idle state is detected, it switches to low-power operation. It continuously updates sequence parameters to determine the final task collaboration and energy consumption control model. The model verification output module verifies the final task collaboration and energy consumption control model using historical operating data. If the verification results show improved production efficiency, it outputs an optimized dual-table laser cutting machine operation protocol. The above description is merely a specific implementation of this specification. Those skilled in the art will understand that, for convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the scope of protection of this specification is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this specification, and these modifications or substitutions should all be covered within the scope of protection of this specification.

Claims

1. A method of intelligent control of a twin table laser cutting machine, characterized in that, The method comprises the following steps: Step S101, by collecting the complex workpiece parameters and processing time data of each task in the double-table laser cutting machine in real time, using genetic algorithm to preliminarily sort the initial task sequence, and obtaining the preliminarily optimized task allocation sequence; Step S102, according to the preliminarily optimized task allocation sequence, obtaining the expected time difference of the two worktables, if the time difference exceeds the preset threshold, judging that the imbalance exists, and determining the balanced task allocation sequence by adjusting the position of the complex workpiece in the sequence; Step S103, for the balanced task allocation sequence, monitoring the running state data of the core component, extracting the high-power running period in the idle state from the data, using a neural network model to predict the energy consumption of these periods, and obtaining the predicted energy consumption distribution; The step S103 comprises: For the balanced task allocation sequence, monitor the running state data of the core component, extract the high-power running period in the idle state from the data, use a neural network model to take the high-power running period as input, process the data through a preset hidden layer node and an activation function, predict the energy consumption of these periods, and obtain the predicted energy consumption distribution; According to the predicted energy consumption distribution, obtain the related data of component load balancing, determine the change range of the peak value of the period power by calculating the maximum difference of the power value in the distribution, and obtain the peak value adjustment sequence; Through the peak value adjustment sequence, for the sequence balance adjustment process, if the peak value exceeds the preset threshold, use a convolutional neural network to take the peak value adjustment sequence as input, extract spatial features through a convolutional layer and classify through a fully connected layer, extract the features of the consumption distribution optimization, and obtain the optimized distribution features; Obtain the optimized distribution features, extract the additional standby period from the component load balancing, determine the matching degree of the standby period and the high-power period by calculating the similarity of the distribution features and the standby period, and obtain the matching period set; According to the matching time period set, the standby allocation of the energy consumption is obtained by accumulating the consumption value of the matching time period in the set to the original distribution to obtain an extended consumption distribution, and an extended energy consumption distribution is obtained; in step S104, the device overall production efficiency index is obtained according to the predicted energy consumption distribution, if the predicted energy consumption distribution shows that the idle state occupies more than a preset threshold, it is judged that optimization is needed, and a low-power operation scheme is determined by dynamically adjusting the power level of the core component; in step S105, for the low-power operation scheme, a cooperative operation timetable is generated in combination with the balanced task allocation sequence, potential energy waste points are extracted from the timetable, and a refined energy management sequence is obtained by fusing task length data in the information processing link; in step S106, according to the refined energy management sequence, if the idle state is detected, the low-power operation is switched to, and the final task cooperation and energy consumption control model is determined by continuously updating the sequence parameters; in step S107, for the final task cooperation and energy consumption control model, the historical operation data is obtained for verification, and if the verification result shows that the production efficiency is improved, the optimized double-table laser cutting machine operation protocol is output.

2. The intelligent control method of twin table laser cutting machine as claimed in claim 1 wherein, The step S101 comprises: Collecting real-time complex workpiece parameters and processing time length data from the double-table laser cutting machine through an interface to determine a complete data set; Quantifying the workpiece complexity index according to the complete data set, and obtaining a complexity quantization value by weighted sum of the parameters and the time length; Generating an initial task sequence using the complexity quantization value to obtain an initial version of the sequence; Selecting and mutating the initial version of the sequence by genetic algorithm, wherein the input is the task priority in the initial version of the sequence, the output is the adjusted sequence, and the sequence optimization path is determined; For the sequence optimization path, a table load balancing mechanism is integrated, the tasks in the path are allocated to the double tables to balance the processing time, and a preliminarily optimized task allocation sequence is obtained.

3. The intelligent control method of twin table laser cutting machine as claimed in claim 1 wherein, The step S102 comprises: The overall difficulty value of the technical subject embodies medium difficulty, there are complex technologies such as sequence optimization adjustment, not clear business processes such as preliminary sequence acquisition method, and technical business attribute details such as complex workpiece definition still need to be explained; Through the task allocation sequence, the expected completion time difference of the two worktables is obtained, and a time difference value is obtained; If the time difference value exceeds a preset threshold, it is judged that imbalance exists; By adjusting the position of the complex workpiece in the sequence, a balanced task allocation sequence is determined; According to the preliminarily optimized task allocation sequence, the total processing time of each worktable is obtained, and the expected completion time difference is obtained; By comparing the expected completion time difference with the preset threshold, if the expected completion time difference exceeds the preset threshold, it is judged that imbalance exists; For the imbalance, the position of the complex workpiece is extracted from the task allocation sequence, the genetic algorithm is used to initialize the population and perform selection, crossover and mutation operations to adjust the position of the complex workpiece, and an adjusted sequence is obtained, wherein the position of the complex workpiece is determined based on the workpieces whose processing time exceeds the average value; By the adjusted sequence, an updated expected completion time difference value is obtained, and a balanced task allocation sequence is determined.

4. The intelligent control method of twin table laser cutting machine as claimed in claim 1 wherein, The step S104 comprises: By the period energy consumption data recorded in the historical operation log, the production efficiency index is obtained from the energy consumption distribution; According to the idle time proportion in the production efficiency index, it is determined that optimization is needed when the idle state proportion is higher than a preset threshold; For the optimization need, a low-power scheme is obtained by dynamically adjusting the power level from the current load data of the core component; From the timing power change in the low-power scheme, the standby energy distribution is obtained by obtaining the overall standby scheduling distribution of the device.

5. The intelligent control method of twin table laser cutting machine as claimed in claim 1 wherein, The step S105 comprises: By the low-power scheme, the balanced allocation result is obtained from the task allocation sequence, and a cooperative timetable is determined; For the cooperative timetable, the energy waste points are obtained by extracting the potential waste extraction from the idle period identified in the time interval using the operation time planning; If the potential waste extraction exceeds a preset threshold, a data fusion process is obtained by fusing task duration data through an information processing link; According to the data fusion process, the load balancing adjustment is determined by integrating the energy optimization sequence for the refined management sequence. The step S106 comprises:

6. The intelligent control method of twin table laser cutting machine as claimed in claim 1 wherein, By monitoring the real-time device state, the idle state detection result is obtained from the energy management sequence, and if the idle state is detected, the low-power operation is switched to, and the sequence parameter update basis is obtained; According to the sequence parameter update basis, the parameter continuous update is fused for the task coordination determination process by continuously updating the sequence parameters, and a cooperative task allocation scheme is determined; By the cooperative task allocation scheme, the association of device state monitoring data and power switching mechanism is obtained, and if the association meets a preset threshold, the energy consumption control model elements are integrated, and a model final determination framework is obtained; According to the model final determination framework, the final task coordination and energy consumption control model are determined by fusing the device load balancing distribution business. The step S107 comprises:

7. The intelligent control method of twin table laser cutting machine as claimed in claim 1 wherein, Obtain historical operation data, extract at least one period of energy consumption record and task execution log from the running data storage of the double-table laser cutting machine, construct a task density matrix by the corresponding relationship between the task distribution in the record and the execution duration in the log, and obtain the task density matrix; By the task density matrix, a preset threshold is used for verification result analysis, the threshold is set based on historical average efficiency, whether the production efficiency index is improved is judged, and an optimization trigger condition is determined; If the optimization trigger condition is established, the model parameters of the energy consumption monitoring system are adjusted, the parameters include power allocation threshold and cooling interval length, and an energy consumption optimization parameter set is obtained; According to the energy consumption optimization parameter set, combined with the material path planning which is a new unique business beyond the original theme, a double-table coordination sequence is generated by corresponding the cutting trajectory in the path planning and the parameter set, and an operation protocol draft is obtained; ​ Through the operation protocol draft, the integrated real-time fault detection, a new unique business beyond the original theme, extracts verified efficiency improvement data from the laser cutting process, and outputs an optimized double table laser cutting machine operation protocol after filtering by fault detection.

8. An intelligent control system for a twin table laser cutting machine, characterized in that, The system is used to implement the intelligent control method of the double table laser cutting machine according to any one of claims 1-7, and the system comprises: a task sequence preliminary optimization module, configured to collect complex workpiece parameters and processing time data of each task in the double table laser cutting machine in real time, and preliminarily sort the initial task sequence by using a genetic algorithm to obtain a preliminarily optimized task allocation sequence; a task allocation balance adjustment module, configured to obtain an expected time difference of the two worktables according to the preliminarily optimized task allocation sequence, and determine that imbalance exists if the time difference exceeds a preset threshold, and determine a balanced task allocation sequence by adjusting the position of the complex workpiece in the sequence; an energy consumption prediction module, configured to monitor the running state data of the core components for the balanced task allocation sequence, extract high-power running periods in the idle state from the running state data, predict the energy consumption of these periods by using a neural network model, and obtain a predicted energy consumption distribution; a power optimization decision module, configured to obtain an overall production efficiency index of the equipment according to the predicted energy consumption distribution, and determine that optimization is needed if the predicted energy consumption distribution shows that the idle state accounts for more than a preset threshold, and determine a low-power running scheme by dynamically adjusting the power level of the core components; a collaborative operation generation module, configured to generate a collaborative operation timetable for the low-power running scheme in combination with the balanced task allocation sequence, extract potential energy waste points from the timetable, and obtain a refined energy management sequence by using an information processing link to fuse the task duration data; a real-time energy consumption control module, configured to monitor the real-time equipment state for the refined energy management sequence, and switch to low-power running if the idle state is detected, and determine a final task coordination and energy consumption control model by continuously updating sequence parameters; and a model verification output module, configured to obtain historical running data for verification for the final task coordination and energy consumption control model, and output an optimized double table laser cutting machine operation protocol if the verification result shows that the production efficiency is improved.

Images