Networked loom intelligent control method and system based on software of process knowledge

Abstract

This application provides a networked intelligent control method and system for looms based on process knowledge software. The method includes: collecting main motor current data, start-stop time series data, and vibration characteristic data from multiple looms; firstly, calculating the mechanical impact energy value from the vibration characteristic data and converting it into storable electrical energy; then, combining the main motor current data and start-stop time series data to generate spatiotemporal correlation data of energy consumption; analyzing this data using a multi-agent reinforcement learning algorithm to generate start-stop control commands and energy storage scheduling commands; dynamically adjusting the start-stop sequence of each loom and scheduling storable electrical energy according to the commands; and finally achieving overall optimized control of multiple looms through coordinated control signals. This application realizes energy consumption optimization and grid load balancing under multi-loom coordinated control.

Description

Technical Field

[0001] This application relates to the field of intelligent manufacturing technology for textile machinery, and in particular to a networked intelligent control method and system for looms based on process knowledge software. Background Technology

[0002] In the clustered production scenarios of the textile industry, the overlapping of energy consumption peaks caused by multiple looms operating in parallel leads to large fluctuations in grid load and low energy utilization. There is an urgent need for an intelligent control solution that can coordinate the loom's operating status and energy scheduling in real time, achieving load balancing and energy optimization while ensuring production efficiency.

[0003] Currently, solutions employing centralized control systems monitor the total power consumption data of each loom and, based on preset energy consumption threshold rules, uniformly adjust the start-up and shutdown sequence of the looms. This system distributes load based on fixed time windows and utilizes buffer capacitor banks to smooth out power grid fluctuations.

[0004] The centralized control system in this scheme has a coarse-grained perception of the individual loom's operating status, and it is difficult to accurately identify transient characteristics such as mechanical impact energy by relying solely on total power consumption data. The fixed threshold rules lack dynamic adaptability and are prone to scheduling delays when the loom switches operating modes, resulting in frequent charging and discharging of the capacitor bank and reduced energy storage efficiency. Summary of the Invention

[0005] This application provides a networked intelligent control method and system for looms based on process knowledge software, in order to solve the problems of poor energy consumption optimization and unbalanced grid load under multi-loom collaborative control in the prior art.

[0006] Firstly, this application provides a networked intelligent control method for looms based on software-defined process knowledge, including:

[0007] Collect main motor current data, start-stop time series data, and vibration characteristic data of multiple looms;

[0008] The mechanical impact energy value is calculated based on the vibration characteristic data, and the mechanical impact energy value is converted into a storable electrical energy value.

[0009] Based on the main motor current data, the start-stop time sequence data, and the storable electrical energy value, energy consumption spatiotemporal correlation data is generated;

[0010] Based on the aforementioned energy consumption spatiotemporal correlation data, start-stop control commands and energy storage scheduling commands are generated through a multi-agent reinforcement learning algorithm.

[0011] The start-stop time sequence data of each loom is adjusted according to the start-stop control command, and the storable electrical energy value is scheduled according to the energy storage scheduling command.

[0012] Based on the adjusted start-stop time series data and the scheduled storable energy value, a coordinated control signal is generated to control multiple looms.

[0013] Optionally, the step of generating start-stop control commands and energy storage scheduling commands based on the energy consumption spatiotemporal correlation data using a multi-agent reinforcement learning algorithm includes:

[0014] The multiple looms are defined as multiple independent intelligent agents;

[0015] Extract the grid load value corresponding to each independent intelligent agent from the energy consumption spatiotemporal correlation data;

[0016] Based on the grid load value and the storable energy value of each independent intelligent agent, calculate the start-stop time adjustment amount and the energy dispatch amount;

[0017] The agent collaboration module, which uses a multi-agent reinforcement learning algorithm, integrates the start-stop time adjustment and power dispatching quantities of all independent agents to generate start-stop control commands and energy storage dispatching commands.

[0018] Optionally, the process of integrating the start-stop time adjustment amounts and the power dispatch amounts of all independent intelligent agents to generate start-stop control commands and energy storage dispatch commands includes:

[0019] The start-stop time adjustments of all independent agents are aggregated to form a start-stop time adjustment dataset.

[0020] The energy scheduling data of all independent intelligent agents are aggregated to form an energy scheduling dataset.

[0021] Time conflict detection and elimination are performed on the aforementioned start / stop time adjustment dataset;

[0022] The power dispatch dataset is subjected to supply and demand matching and balancing processing.

[0023] Convert the conflict-resolved start / stop time adjustment dataset into start / stop control commands;

[0024] Convert the power dispatch dataset after supply and demand are balanced into energy storage dispatch instructions.

[0025] Optionally, the time conflict detection and elimination process for the start / stop time adjustment dataset includes:

[0026] Extract all planned start and stop time points from the start and stop time adjustment dataset;

[0027] The system detects overlapping time intervals among the planned start and stop times.

[0028] Set the minimum safe time interval required for starting and stopping the loom;

[0029] The planned start and stop times within the detected overlapping time intervals are offset and adjusted according to the minimum safe time interval value.

[0030] The planned start and stop times in the time adjustment dataset are updated using the adjusted start and stop times to obtain the time adjustment dataset after conflict resolution.

[0031] Optionally, calculating the start / stop time adjustment and power dispatch amount based on the grid load value and storable energy value of each independent intelligent agent includes:

[0032] Based on the grid load value and the storable electrical energy value, calculate the state assessment index;

[0033] If the state assessment index indicates that the grid load is higher than the preset grid load threshold, then the start / stop time offset is determined;

[0034] If the state assessment index indicates that the storable energy value is lower than the preset storable energy threshold, then the amount of energy transferred is determined;

[0035] Convert the start / stop time offset into a start / stop time adjustment amount;

[0036] The energy transfer amount is converted into an electrical energy dispatch amount.

[0037] Optionally, the step of calculating the mechanical impact energy value based on the vibration characteristic data and converting the mechanical impact energy value into a storable electrical energy value includes:

[0038] The vibration characteristic data is decomposed into multiple frequency band energy components;

[0039] From the multiple frequency band energy components, the key frequency band energy components that are higher than the preset impact energy threshold are selected as the impact frequency band components;

[0040] The mechanical impact energy value is obtained by performing energy integration calculation on the impact frequency band components.

[0041] The mechanical impact energy value is converted into a storable electrical energy value.

[0042] Optionally, the step of generating a coordinated control signal to control multiple looms based on the adjusted start-stop time series data and the scheduled storable energy value includes:

[0043] Extract time-period features from the adjusted start-stop time-series data;

[0044] Based on the storable energy value after scheduling, real-time status data is generated;

[0045] The time period characteristics are matched with the real-time status data to determine the optimal time period for energy release or storage;

[0046] Based on the optimized time period, a coordinated control signal is generated.

[0047] Secondly, this application provides a networked intelligent control system for looms based on process knowledge software, comprising:

[0048] The data acquisition module is used to collect main motor current data, start-stop time sequence data, and vibration characteristic data of multiple looms;

[0049] The calculation module is used to calculate the mechanical impact energy value based on the vibration characteristic data, and convert the mechanical impact energy value into a storable electrical energy value;

[0050] The first generation module is used to generate spatiotemporal correlation data of energy consumption based on the main motor current data, the start-stop time sequence data and the storable electrical energy value.

[0051] The second generation module is used to generate start-stop control commands and energy storage scheduling commands based on the energy consumption spatiotemporal correlation data and through a multi-agent reinforcement learning algorithm.

[0052] The adjustment module is used to adjust the start-stop time sequence data of each loom according to the start-stop control command, and simultaneously schedule the storable electrical energy value according to the energy storage scheduling command.

[0053] The third generation module is used to generate coordinated control signals to control multiple looms based on the adjusted start-stop time series data and the scheduled storable energy value.

[0054] Thirdly, this application provides a computing device, including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the networked intelligent control method for looms based on process knowledge software as described in any of the first aspects.

[0055] Fourthly, this application provides a computer storage medium storing computer program instructions thereon, which, when executed by a processor, implement the networked intelligent control method for looms based on process knowledge software as described in any one of the first aspects.

[0056] This application provides a networked intelligent control method for looms based on process knowledge software. The method includes: collecting main motor current data, start-stop time sequence data, and vibration characteristic data from multiple looms; calculating mechanical impact energy values ​​based on the vibration characteristic data and converting the mechanical impact energy values ​​into storable electrical energy values; generating energy consumption spatiotemporal correlation data based on the main motor current data, the start-stop time sequence data, and the storable electrical energy values; generating start-stop control commands and energy storage scheduling commands using a multi-agent reinforcement learning algorithm based on the energy consumption spatiotemporal correlation data; adjusting the start-stop time sequence data of each loom according to the start-stop control commands, and simultaneously scheduling the storable electrical energy values ​​according to the energy storage scheduling commands; and generating a coordinated control signal to control multiple looms based on the adjusted start-stop time sequence data and the scheduled storable electrical energy values.

[0057] The technical solution provided in this application has the following beneficial effects:

[0058] This application achieves comprehensive perception of the loom's operating status, providing a multi-dimensional data foundation for energy consumption analysis. It converts traditionally discarded mechanical impact energy into usable electrical energy, improving energy recovery rates. A correlation model between spatiotemporal dimensions and energy forms is established to accurately reflect the system's energy consumption characteristics. The operating strategy for each loom is dynamically optimized to balance production efficiency and energy consumption demands. Equipment start-up and shutdown and power dispatch are synchronously adjusted to reduce instantaneous load pressure on the power grid. Multi-equipment collaborative operation is achieved to maintain the stable and efficient operation of the production system.

[0059] Furthermore, this application also constructs each loom as an independent intelligent agent, extracting its corresponding grid load value and storable energy value from the spatiotemporal correlation data of energy consumption. Each intelligent agent calculates the localized start-stop time adjustment amount and energy dispatch amount. Finally, all decision outputs are integrated through the intelligent agent collaboration module to form globally optimized start-stop control commands and energy storage dispatch commands.

[0060] Furthermore, this application combines distributed decision-making with centralized coordination. While ensuring the operational needs of individual devices, it effectively reduces the overall energy consumption fluctuation of the cluster system through collaborative learning among intelligent agents, while improving the utilization efficiency of recovered energy.

[0061] These or other aspects of this application will become more apparent from the description of the following embodiments. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 A flowchart illustrating a networked intelligent control method for looms based on process knowledge software, provided in this application embodiment;

[0064] Figure 2 A schematic diagram of a networked intelligent control system for a loom based on process knowledge software, provided in an embodiment of this application;

[0065] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0066] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0067] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0068] Existing centralized control systems have key shortcomings in the clustered production scenarios of the textile industry: the system only monitors the total power consumption data of the looms, resulting in insufficient accuracy in identifying transient characteristics such as mechanical impact energy; fixed threshold rules lack dynamic adaptability, causing scheduling delays when the looms switch operating modes, resulting in frequent charging and discharging of the buffer capacitor bank and reducing energy storage efficiency. This contradiction stems from the coarse-grained perception of the individual operating status and energy characteristics of the looms, and the insufficient response of static control strategies to dynamic production scenarios.

[0069] To address the aforementioned issues, this application proposes a networked intelligent control method for looms based on software-defined process knowledge. The core of this method involves collecting main motor current data, start-stop time series data, and vibration characteristic data from multiple looms. Based on the vibration characteristic data, the mechanical impact energy value is calculated and converted into storable electrical energy. Combined with the main motor current data and start-stop time series data, spatiotemporal correlation data of energy consumption is generated. A multi-agent reinforcement learning algorithm is used to analyze this data to generate start-stop control commands and energy storage scheduling commands. Based on these commands, the start-stop time series data of each loom is dynamically adjusted, and storable electrical energy is scheduled. Finally, overall optimized control of multiple looms is achieved through coordinated control signals. This method effectively solves the problems of scheduling delays and low energy storage efficiency caused by incomplete data perception and static control strategies in existing technologies by fine-grainedly perceiving the individual state and energy characteristics of each loom, combined with dynamic learning optimization strategies. It achieves energy consumption optimization and grid load balancing under the coordinated control of multiple looms.

[0070] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0071] Figure 1 A flowchart illustrating a networked intelligent control method for looms based on process knowledge software, as provided in this application embodiment, is shown below. Figure 1 As shown, the method includes:

[0072] Step 101: Collect the main motor current data, start-stop time series data, and vibration characteristic data of multiple looms.

[0073] In step 101, the main motor current data refers to the current signal generated when the main motor of the loom is working, reflecting the real-time energy consumption status of the equipment. The start-stop time series data represents an ordered dataset recording the time points and durations of each loom's start-up and stop operations. The vibration characteristic data refers to the mechanical vibration signals of the loom collected by sensors, including features such as frequency and amplitude.

[0074] In this embodiment, during the operation of the loom cluster, the real-time current signal of the main motor of each loom is first collected by a current sensor, and the start-up and stop operation times of each device are recorded synchronously to form a time series. At the same time, mechanical vibration waveform data is obtained using a vibration sensor. These three types of data are uploaded to the central processing unit after being standardized by an industrial IoT terminal device. The current data reflects the instantaneous load, the time series data provides operating rhythm information, and the vibration data implies the characteristics of mechanical energy loss.

[0075] For example, a textile workshop has a cluster of looms consisting of three machines: A, B, and C. During the morning production period, the system collects data showing that the main motor current of loom A exhibits a stable fluctuating signal, and its start-stop records show that it completes one work cycle every 30 minutes. Vibration sensors detect impact waveforms in a specific frequency range. Loom B shows periodic peak currents, with a start-stop interval of 45 minutes, and its vibration energy is concentrated in the low-frequency range. Loom C has a stable current but frequent start-stops, and its vibration signal exhibits high-frequency characteristics. These data are filtered and denoised to form a standardized dataset.

[0076] Step 102: Calculate the mechanical impact energy value based on the vibration characteristic data, and convert the mechanical impact energy value into a storable electrical energy value.

[0077] In step 102, the mechanical impact energy value represents a quantitative index of kinetic energy loss in the mechanical system calculated from vibration characteristic data. The storable electrical energy value represents the current regenerative energy reserves available for dispatch, which are converted from mechanical energy into electrical energy that can be accepted by the energy storage system through an energy conversion device.

[0078] In this embodiment, the collected vibration characteristic data is decomposed into frequency bands to extract the characteristic frequency band energy components related to mechanical impact. The transient impact energy value is obtained through integration. This energy value is input into a piezoelectric conversion device, which converts it into equivalent electrical energy according to the electromechanical conversion efficiency coefficient. After rectification and voltage stabilization, the energy is input into the energy storage unit. During the conversion process, the state of charge of the energy storage unit is monitored in real time to ensure that the electrical energy can be safely stored.

[0079] For example, for the vibration signal in the characteristic frequency range detected by loom A, the system calculates its waveform envelope area as the mechanical impact energy value. This energy is then converted into DC power by a piezoelectric conversion module deployed in the workshop with a fixed conversion efficiency. The measured storable energy value is then stored in a lithium battery pack. The low-frequency vibration energy of loom B is converted using the same process, but with adjusted conversion parameters due to its different frequency characteristics.

[0080] Step 103: Generate spatiotemporal correlation data of energy consumption based on the main motor current data, the start-stop time sequence data, and the storable electrical energy value.

[0081] In step 103, the energy consumption spatiotemporal correlation data representation is a structured dataset that integrates the three dimensions of time, space, and energy, reflecting the energy consumption distribution pattern of the cluster.

[0082] In this embodiment, a data table structure with time axis and device number axis as dimensions is established. The main motor current data is filled in according to time-device coordinates as the grid load value, and the start-stop time series data of the corresponding devices are synchronously associated. The storable energy value data is superimposed at the corresponding positions. The data gaps are filled by spatiotemporal interpolation algorithm, and finally a complete associated data table reflecting the spatiotemporal distribution characteristics of cluster energy consumption is formed.

[0083] For example, the current production cycle data of three looms, A, B, and C, can be integrated into a three-dimensional table. The horizontal dimension is a minute-level time scale, and the vertical dimension is the equipment number. The table cells record the current load value, start / stop status markers, and storable energy value at each moment. The system automatically completes the missing data caused by communication delays and generates a continuous and complete energy consumption distribution chart.

[0084] Step 104: Based on the aforementioned energy consumption spatiotemporal correlation data, generate start-stop control commands and energy storage scheduling commands through a multi-agent reinforcement learning algorithm.

[0085] In step 104, the multi-agent reinforcement learning algorithm represents a distributed artificial intelligence method that models each loom as an autonomous decision-making unit and optimizes the group's behavior through interactive learning. The start / stop control command represents a scheduling command that adjusts the working sequence of the looms, including specific time adjustment parameter commands for each device, specifying the exact start and stop times for each loom. The energy storage scheduling command represents a control signal that manages the charging and discharging operations of storable electrical energy, specifying the energy flow path and quantity, and defining the charging and discharging sequence and power of the energy storage device.

[0086] In this embodiment, an independent agent model is constructed for each loom. Each agent receives its corresponding grid load value and storable energy value from the spatiotemporal correlation data of energy consumption as state input, and outputs start-stop time adjustment and energy dispatch suggestions through a policy network. The central coordination module integrates the outputs of all agents, uses a game theory equilibrium algorithm to eliminate decision conflicts, and finally generates a globally optimized set of start-stop control commands and energy storage dispatch commands.

[0087] For example, the intelligent agent of loom A suggests delaying the start-up by 5 minutes based on the current high load status, while the intelligent agent of loom B suggests stopping early to release electrical energy due to sufficient energy storage. After calculation by the coordination module, the final instruction set is generated: loom A delays by 3 minutes, loom B stops immediately, and loom C maintains the original plan. At the same time, the electrical energy converted by loom B is arranged to be supplied to loom A first.

[0088] Step 105: Adjust the start-stop time sequence data of each loom according to the start-stop control command, and simultaneously schedule the storable electrical energy value according to the energy storage scheduling command.

[0089] In this embodiment, the time adjustment amount in the start-stop control command is parsed, and the planned operation timetable of the corresponding loom is modified to ensure that the start-stop peaks of the equipment group are staggered after the time adjustment. The energy storage dispatch command is executed synchronously to control the bidirectional converter to achieve directional transfer of electrical energy, distributing a specified amount of storable electrical energy to the target equipment or feeding it back to the grid as needed.

[0090] For example, the system adjusts the original 08:00 start time of loom A to 08:03, and moves the shutdown time of loom B from 08:15 to 08:12. At the same time, the system controls the energy storage system to send the electrical energy converted by loom B to the power supply circuit of loom A. The adjusted timing completely staggers the high-load periods of the three looms.

[0091] Step 106: Based on the adjusted start-stop time series data and the scheduled storable energy value, generate a coordinated control signal to control multiple looms.

[0092] In step 106, the adjusted start-stop time sequence data represents the optimized equipment operating schedule. The scheduled storable energy value represents the energy storage scheme reallocated according to instructions. The coordinated control signal represents a composite instruction set that coordinates the equipment operating sequence and energy scheduling scheme, including timing synchronization and energy allocation information.

[0093] In this embodiment, the adjusted start-stop times of all looms are integrated with the real-time power scheduling status to generate a control instruction package containing elements such as equipment start-stop schedules, power transmission paths, and power allocation ratios. This signal is synchronously sent to each loom controller and energy storage system execution unit via the industrial bus to ensure precise matching of spatiotemporal parameters with energy flow.

[0094] For example, the final generated signal packet specifies: 08:03 start loom A and use the stored energy provided by loom B; 08:10 start loom C and connect to the power grid; 08:12 stop loom B and activate its energy recovery mode. After the three devices operate according to this scheme, the total load curve of the workshop tends to stabilize.

[0095] This method achieves precise perception and collaborative optimization of the operating status of a loom cluster through multi-dimensional data fusion and distributed intelligent decision-making. While ensuring normal production rhythm, it effectively smooths out grid load fluctuations, improves mechanical energy recovery and utilization, and forms a dynamic balance between energy consumption and production capacity. Compared with traditional control methods, it solves technical bottlenecks such as extensive data perception, delayed scheduling response, and singular energy utilization.

[0096] To address the dynamic optimization problem in multi-loom collaborative control, in some embodiments, step 104: generating start / stop control commands and energy storage scheduling commands based on the energy consumption spatiotemporal correlation data using a multi-agent reinforcement learning algorithm includes:

[0097] Step 201: Define the multiple looms as multiple independent intelligent agents.

[0098] In step 201, an independent intelligent agent refers to a virtual control unit with autonomous decision-making capabilities. Each intelligent agent corresponds to a loom and is responsible for optimizing the operation strategy of the equipment. The intelligent agents achieve group behavior coordination through interactive learning.

[0099] In this embodiment, the system creates a dedicated agent model for each loom in the workshop, with each model possessing independent state perception and decision-making capabilities. These agents share the same learning framework but maintain their own policy parameters, forming a distributed decision-making network.

[0100] Step 202: Extract the grid load value corresponding to each independent intelligent agent from the energy consumption spatiotemporal correlation data.

[0101] In step 202, the grid load value reflects the real-time demand intensity of the loom for the power system and is calculated from the main motor current data.

[0102] In this embodiment, each intelligent agent reads its corresponding grid load curve and energy storage status data from a central database. The load value is taken from the calibration signal of the current sensor, and the energy storage data comes from the output metering of the energy conversion device. The two are aligned by timestamps to form a state input vector.

[0103] Step 203: Calculate the start-stop time adjustment amount and power dispatch amount based on the grid load value and the storable energy value of each independent intelligent agent.

[0104] In step 203, the start / stop time adjustment is the suggested modification value for the device's operating time by the agent; a positive value indicates delayed start-up or early stop-up. The energy dispatch quantity refers to the suggested energy transfer value, a control parameter recognizable by the energy storage system, including magnitude and direction attributes.

[0105] In this embodiment, each intelligent agent calculates the optimal adjustment scheme based on the input load and energy storage data through a built-in policy network. When the load is high, a positive adjustment value is output to suggest delaying the start-up; when energy storage is sufficient, a negative scheduling value is output to suggest releasing electrical energy. The calculation results are accompanied by a confidence assessment.

[0106] Step 204: Through the agent collaboration module of the multi-agent reinforcement learning algorithm, integrate the start-stop time adjustment amount and the power dispatch amount of all independent agents to generate start-stop control instructions and energy storage dispatch instructions.

[0107] In step 204, the agent coordination module is the central unit that coordinates multi-agent decision-making and generates executable instructions by resolving policy conflicts.

[0108] In this embodiment, after receiving the output suggestions from all agents, the collaboration module first establishes a policy influence graph to identify conflicting adjustment schemes. A game-theoretic negotiation mechanism is then used for multi-round policy optimization to ultimately generate an optimal instruction set that balances global energy consumption targets and individual needs.

[0109] Here is a specific example:

[0110] During the operation of three looms (A, B, and C) in the textile workshop, the system first constructs an independent intelligent agent decision-making unit for each loom. The intelligent agent for loom A extracts from the energy consumption spatiotemporal correlation data table that the current grid load is 1.3 times the standard load, with 55 units of storable energy. The intelligent agent for loom B obtains a grid load of 0.8 times the standard value, with 34 units of storable energy. The intelligent agent for loom C monitors a grid load of 1.1 times the standard value, with 0 units of storable energy. Each intelligent agent uses the same decision-making algorithm for calculation. The intelligent agent for loom A, based on the degree of load exceeding the standard, calculates according to the formula Δt = K × (L - L0) that a 5-minute delay in startup is needed, where Δt represents the time adjustment amount, K is a proportionality coefficient of 2, L represents the current load value, and L0 represents the standard load value. Simultaneously, since there is sufficient storable energy, no power scheduling suggestion is proposed. Loom B, detecting a low load and sufficient energy storage, suggests releasing 30 units of energy according to the formula E = E0 × η, where E represents the scheduling amount, E0 represents the storable energy value, and η is the release coefficient, taken as 0.9. Loom C, having no power supply and a slightly high load, is advised to maintain its original operating plan. After these decision suggestions are uploaded to the coordination module, the system first verifies that a 5-minute delay for loom A does not conflict with the operating schedule of loom C. However, immediate energy release by loom B would cause short-term grid fluctuations. Therefore, the system adjusts the schedule: loom A is delayed by 3 minutes, and loom B releases energy in two stages: 15 units at 08:00 for loom A's startup, and the remaining 15 units at 08:30 for regular operation. The final start / stop control command specifies that loom A starts at 08:03, loom B stops at 08:15, and loom C starts as planned at 08:10; the energy storage scheduling command arranges for loom B to supply 15 units of power to loom A at 08:00 and 08:30 respectively.

[0111] In this embodiment, the method combines autonomous decision-making by distributed intelligent agents with centralized coordination to achieve dynamic load balancing in a multi-loom system. While ensuring normal production order, it effectively optimizes the overall energy consumption distribution of the equipment group, improves the utilization rate of renewable energy, and forms an intelligent control system with adaptive capabilities.

[0112] To further improve the accuracy of multi-loom collaborative control, in some embodiments, step 204: integrating the start / stop time adjustment amounts and energy dispatch amounts of all independent intelligent agents to generate start / stop control commands and energy storage dispatch commands, includes:

[0113] Step 301: Summarize the start and stop time adjustments of all independent agents to form a start and stop time adjustment dataset.

[0114] In step 301, the start-stop time adjustment dataset is a collection of device operation time modification schemes proposed by each agent, including fields such as device number, suggested adjustment time, and adjustment direction.

[0115] In this embodiment, the system collects all time adjustment suggestions output by intelligent agents and stores them according to device number. Each record contains an agent identifier, the number of minutes to be adjusted, and the adjustment type (delay or advance), forming a structured data table.

[0116] Step 302: Summarize the power scheduling amounts of all independent intelligent agents to form a power scheduling dataset.

[0117] In step 302, the power dispatch dataset refers to the set of energy allocation schemes proposed by each agent, which records information such as source device, target device, and proposed dispatch power value.

[0118] In this embodiment, the power transfer schemes proposed by each intelligent agent are organized in a unified format, including the power source loom number, the receiving loom number, the suggested scheduling amount and time window, and a power flow relationship graph is established.

[0119] Step 303: Perform time conflict detection and elimination processing on the start / stop time adjustment dataset. In this embodiment, the system scans the start / stop time adjustment dataset to detect whether multiple devices are starting up intensively during the same time period. After a conflict is detected, the start time period is reallocated according to device priority and energy consumption characteristics to ensure that high-load devices receive priority scheduling rights.

[0120] Step 304: Perform supply and demand matching and balancing processing on the power dispatch dataset.

[0121] In step 304, the supply and demand matching and balancing process refers to the process of coordinating the relationship between power supply and demand so that the scheduling scheme meets the actual energy storage and usage needs.

[0122] In this embodiment of the application, the supply and demand relationship in the power dispatch data is analyzed. When the total power demand exceeds the available supply in a certain period, the power is redistributed according to the importance and efficiency of the equipment, giving priority to ensuring the power supply stability of key production lines.

[0123] Step 305: Convert the conflict-resolved start-stop time adjustment dataset into start-stop control commands.

[0124] In this embodiment, the optimized time adjustment dataset is converted into an instruction sequence in the device control protocol format, including control parameters such as device address code, execution time point, and action type, and then distributed to each loom controller through the industrial network.

[0125] Step 306: Convert the power dispatch dataset after supply and demand balance into energy storage dispatch instructions.

[0126] In this embodiment, a control command for the energy storage system is generated based on the balanced power dispatch scheme, specifying the discharge target, discharge period, and discharge power level of the energy storage unit, and is simultaneously sent to the energy management system for execution.

[0127] Here is a specific example:

[0128] In the coordinated control of three looms (A, B, and C) in a textile workshop, the system first aggregates the decision suggestions of each agent to form a dataset. This dataset includes start / stop time adjustment data such as loom A's suggestion to delay start by 5 minutes, loom B's suggestion to stop immediately, and loom C's suggestion to maintain the original schedule. The power dispatch dataset records loom B's suggestion to release 30 units of power to loom A. When the system checks the time adjustment dataset, it finds that if loom A's suggestion to delay by 5 minutes is fully adopted, its new start time of 08:05 will be insufficient to meet the minimum 5-minute interval requirement for safe equipment operation, given the original scheduled start time of loom C at 08:10. The adjusted time is calculated using the formula T_new = T_old + Δt × R, where T_new represents the new time point, T_old represents the original time point, Δt represents the adjustment amount, and R is an adjustment coefficient of 0.6. Ultimately, loom A is delayed by 3 minutes to start at 08:03. For the power dispatch dataset, the system analysis shows that the current total grid load is 2.2 times the standard value. However, a single release of 30 units of energy by loom B would cause the load to increase instantaneously by 0.3 times, exceeding the safety threshold. A phased release strategy is adopted, calculating the initial release amount according to the formula E_s = E_t × (1 - L / L_max), where E_s represents the safe release amount, E_t represents the total release amount, L represents the current load, and L_max represents the maximum allowable load. This results in an initial release of 15 units at 08:00, with the remaining 15 units released at 08:30. The final generated start / stop control commands specify that loom A starts at 08:03, loom B stops at 08:15, and loom C starts at 08:10. The energy storage dispatch commands schedule loom B to release 15 units of energy to loom A at both 08:00 and 08:30.

[0129] In this embodiment, the method achieves accurate generation of control commands for multiple looms by systematically integrating and optimizing distributed decision suggestions. It retains the flexibility of local agent decision-making while ensuring the feasibility of the solution through global coordination, enabling the equipment group to operate collaboratively under optimal energy consumption.

[0130] To further improve the accuracy of loom start-stop timing conflict handling, in some embodiments, step 303: the timing conflict detection and elimination processing of the start-stop time adjustment dataset includes:

[0131] Step 401: Extract all planned start and stop time points from the start and stop time adjustment dataset.

[0132] In step 401, the planned start-stop time point is the specific time when each intelligent agent suggests starting or stopping the device operation, which includes device identification and timestamp information.

[0133] In this embodiment of the application, the system parses the suggested start and stop operation time points for each loom from the time adjustment dataset, including the equipment number and the corresponding planned execution time, and arranges them in chronological order to form an operation time table.

[0134] Step 402: Detect overlapping time intervals among the planned start and stop times.

[0135] In step 402, the overlapping time interval refers to the continuous period in which the planned start and stop times of two or more looms overlap on the time axis. Its length is determined by the earliest start time and the latest end time. The method for determining overlap is as follows: arrange the start and stop times of each device in chronological order. If the interval between the operation times of any two devices is less than the minimum safe time interval, or if the start time of one device falls within the running period of another device, then overlap is determined. For example, if loom A runs from 08:00 to 08:30 and loom B starts at 08:25, then 08:25 to 08:30 is the overlapping interval.

[0136] In this embodiment, the system scans all planned start and stop times, detects whether there are multiple devices whose operation times fall within the same time period, marks the overlapping time periods as conflict intervals, and records the device numbers and overlapping durations involved.

[0137] Step 403: Set the minimum safe time interval required for starting and stopping the loom.

[0138] In step 403, the minimum safe time interval is the shortest buffer time required to ensure the stable operation of the loom, which depends on the mechanical characteristics of the equipment and the power grid restoration requirements.

[0139] In this embodiment of the application, a uniform safety interval standard is set according to the loom model parameters. This value ensures that the next device will only start a new operation after the previous device has completed its operation and the power grid status has returned to stability.

[0140] Step 404: Adjust the planned start and stop times within the detected overlapping time intervals by offsetting them according to the minimum safe time interval value.

[0141] In step 404, offset adjustment refers to shifting the start and stop time points within the conflict interval on the time axis, with the shift amount not less than the minimum safe time interval value.

[0142] In this embodiment, the system reorders the time points within the conflict interval according to the device priority, and allocates a new execution time to each time point in turn, ensuring that the adjacent operation intervals meet the safety requirements and prioritizing the original planned time period of critical equipment.

[0143] Step 405: Update the corresponding planned start and stop times in the time adjustment dataset using the adjusted start and stop times to obtain the time adjustment dataset after conflict resolution.

[0144] In step 405, the time adjustment dataset after conflict resolution is an optimized start-stop schedule, in which all device operation times meet the safety interval constraints.

[0145] In this embodiment of the application, the adjusted new time points are written back to the original dataset to replace the conflicting old time points, generating the final executable start and stop time scheme, while retaining the adjustment records for the agent to learn and reference.

[0146] Here is a specific example:

[0147] During the operation control of three looms (A, B, and C) in the textile workshop, the system first extracts all planned operation time points from the start-stop time adjustment dataset, including the suggested start time of 08:05 for loom A, the stop time of 08:15 for loom B, and the start time of 08:10 for loom C. Time series analysis reveals that the 5-minute interval between the start times of loom A (08:05) and loom C (08:10) does not meet the minimum 7-minute interval standard required for safe equipment operation. This standard is the minimum time required for mechanical recovery set according to the loom model parameters. The system uses a time offset algorithm to adjust the conflicting time periods. Specifically, it calculates the new time points according to the formula T_a = T_o + [D - (T_j - T_i)] / 2, where T_a represents the adjusted time, T_o represents the original time, D represents the minimum safe interval of 7 minutes, and T_j and T_i represent adjacent operation time points. The start time of loom A is adjusted to 08:02, and the start time of loom C is adjusted to 08:09, ensuring that the interval between them reaches 7 minutes. Simultaneously, a 6-minute time interval was detected between the stop time of loom B at 08:15 and the start time of loom C at 08:09. The formula was then applied again to adjust the stop time of loom B to 08:14, resulting in a final conflict-free time arrangement. The updated time adjustment dataset records the new scheme with loom A starting at 08:02, loom B stopping at 08:14, and loom C starting at 08:09. All adjacent operation time intervals meet safety requirements.

[0148] In this embodiment of the application, the method ensures a safe interval for the start-up and shutdown of multiple looms through systematic conflict detection and intelligent adjustment. This avoids the power grid impact caused by the simultaneous operation of equipment groups and preserves the original planned timing characteristics of each equipment to the greatest extent, thus achieving a balance between stability and efficiency.

[0149] To further improve the accuracy of agent decision-making, in some embodiments, step 203: calculating the start-stop time adjustment and power dispatch amount based on the grid load value and storable energy value of each independent agent, includes:

[0150] Step 501: Calculate the state assessment index based on the grid load value and the storable electrical energy value.

[0151] In step 501, the state assessment index is a comprehensive parameter reflecting the grid load pressure and energy storage status, which is calculated by weighting the grid load value and the storable electrical energy value.

[0152] In this embodiment, the intelligent agent normalizes the received grid load data and energy storage data, and linearly combines them according to a preset weight ratio to generate an evaluation value within a standard range, which is used to determine the current system status level.

[0153] Step 502: If the status assessment index indicates that the grid load is higher than the preset grid load threshold, then determine the start-stop time offset.

[0154] In step 502, the grid load threshold is the critical load value that triggers start-stop adjustments. It is set comprehensively based on the transformer's rated capacity, historical load peak values, and safety margin. The preset method is as follows: take 80% of the transformer's rated capacity as the base value, and add 20% of the average load fluctuation over the past 30 days as a dynamic buffer. For example, if the rated capacity is 1000kW, then the threshold is set to 800kW + 160kW = 960kW. The start-stop time offset is the suggested adjustment range for equipment operation time. A positive value indicates delayed operation, and a negative value indicates early operation. Its magnitude is related to the degree to which the grid load exceeds the threshold.

[0155] In this embodiment of the application, when the state assessment index shows that the power grid load is too high, the intelligent agent calculates the time offset suggestion based on the proportion of the load exceeding the threshold. The more the load exceeds the threshold, the longer the suggested delay time will be. The characteristics of the current working stage of the equipment are taken into account during the calculation.

[0156] Step 503: If the state assessment index indicates that the storable energy value is lower than the preset storable energy threshold, then determine the amount of energy transferred.

[0157] In step 503, the storable energy threshold is the minimum energy reserve requirement for the energy storage system's scheduling decision. It is set based on the average energy consumption of the loom group and the grid stability requirements. The preset method is: use three times the starting energy consumption of a single loom as the base threshold to ensure that at least one device can be fully started and stopped three times. For example, if a single start requires 5 units of energy, then the threshold is set to 15 units. The energy transfer amount is the suggested energy allocation value, including magnitude and direction attributes; a positive value indicates energy absorption, and a negative value indicates energy release.

[0158] In this embodiment of the application, when the state assessment index shows insufficient energy storage, the intelligent agent calculates the amount of electrical energy that needs to be transferred based on the degree of the gap, and at the same time determines where the electrical energy should be obtained from or supplied to, forming a directional scheduling suggestion.

[0159] Step 504: Convert the start / stop time offset into a start / stop time adjustment amount. In this embodiment, the intelligent agent converts the calculated time offset into a format specified by the device control protocol, including the specific adjustment minutes and adjustment direction, and adds an execution priority identifier.

[0160] Step 505: Convert the energy transfer amount into an electrical energy dispatch amount.

[0161] In this embodiment, the intelligent agent transforms the energy transfer scheme into standard control commands that include elements such as the target device, transmission power, and duration, ensuring that the energy storage system can execute them accurately.

[0162] Here is a specific example:

[0163] In the coordinated control process of three looms (A, B, and C) in the textile workshop, the system first establishes a state evaluation model for the intelligent agent of each loom. The intelligent agent of loom A obtains a grid load value 1.3 times the standard value and a storeable energy value of 55 units. The state evaluation index is calculated as 1.15 using the formula I = α × L + β × E, where I represents the evaluation index, α is a load weight coefficient of 0.7, L represents the load ratio, β is an energy weight coefficient of 0.3, and E represents the ratio of energy value to full storage energy of 0.55. Since this index is higher than the preset grid load threshold of 1.1, the intelligent agent of loom A needs a 4.5-minute delay to start, calculated according to the formula Δt = K × (I - I0), where K is an adjustment coefficient of 3, I0 is the threshold of 1.1, and the delay is rounded to 5 minutes. Loom B's agent calculated a state assessment index of 0.86, and its storeable energy value of 34 units was lower than the preset threshold of 40 units. Based on the shortage ratio, it was calculated that 15 units of energy needed to be transferred. Considering the safety margin, the final energy transfer amount was determined to be 12 units. Loom C's agent had no power supply and its assessment index of 1.05 was close to the threshold, so no adjustment suggestions were made.

[0164] In this embodiment of the application, the method enables each loom to make reasonable adjustment suggestions based on the real-time system status through quantitative state assessment and intelligent decision-making, thereby ensuring the stable operation of the power grid, improving energy utilization efficiency, and achieving dynamic optimization of the production process.

[0165] To further improve the accuracy of mechanical energy recovery, in some embodiments, step 102: calculating the mechanical impact energy value based on the vibration characteristic data and converting the mechanical impact energy value into storable electrical energy includes:

[0166] Step 601: Decompose the vibration characteristic data into multiple frequency band energy components.

[0167] In step 601, the frequency band energy component is the energy distribution data obtained by decomposing the vibration signal into different frequency ranges, reflecting the intensity characteristics of mechanical vibration in each frequency band.

[0168] In this embodiment, the system uses digital signal processing to decompose the collected vibration waveform data into several adjacent frequency intervals, calculates the energy intensity of the vibration waveform in each interval, and forms a complete frequency band energy distribution map.

[0169] Step 602: Select key frequency band energy components that are higher than the preset impact energy threshold from the multiple frequency band energy components as impact frequency band components.

[0170] In step 602, the key frequency band energy component refers to the vibration frequency band data that exceeds the preset energy threshold. These frequency bands usually correspond to mechanical impact events during the operation of the loom.

[0171] In this embodiment, the system compares the energy values ​​of each frequency band with preset thresholds and filters out characteristic frequency bands with energy higher than the average level. These frequency bands are often directly related to the mechanical impact of key actions such as weft insertion and winding of the loom.

[0172] Step 603: Perform energy integration calculation on the impact frequency band components to obtain the mechanical impact energy value.

[0173] In step 603, the energy integral calculation is the quantification of the energy accumulation of the vibration signal in the time domain. The implementation method is as follows: after taking the absolute value of the vibration waveform of the selected impact frequency band, a definite integral operation is performed over the impact duration. The formula is ∫|f(t)|dt, where f(t) is the vibration waveform function, the upper and lower limits of the integral are the start and end times of the impact, and the unit of the result is joules.

[0174] In this embodiment, waveform envelope extraction and area calculation are performed on the selected key frequency band vibration signals, and the energy values ​​of multiple key frequency bands are accumulated to finally obtain the energy value reflecting the overall mechanical impact intensity.

[0175] Step 604: Convert the mechanical impact energy value into a storable electrical energy value.

[0176] In this embodiment, the calculated mechanical impact energy value is input into the piezoelectric conversion system, the energy form is converted according to the device conversion efficiency coefficient, and after rectification and voltage stabilization, a stable electrical energy value that meets the requirements of the energy storage system is output.

[0177] Here is a specific example:

[0178] During the actual operation of three looms (A, B, and C) in the textile workshop, the system first analyzes and processes the vibration signal of loom A at a specific frequency range. The vibration waveform is decomposed into three characteristic frequency bands. The mid-frequency band, with an energy amplitude reaching 2.5 times the preset threshold, is identified as the impact frequency component. The mechanical impact energy value is calculated as 85 units by integrating the waveform over time using the formula S=∫|f(t)|dt, where S represents the impact energy value, f(t) represents the vibration waveform function, and t represents the time variable. This energy value is converted into 55.25 units of storable electrical energy by a piezoelectric converter with a conversion efficiency coefficient of 0.65 and stored in the energy storage system. For the low-frequency vibration signal of loom B, the system detects that its main frequency band energy exceeds the threshold by 1.8 times. The impact energy value is calculated using the same method to be 62 units. Considering the low-frequency characteristics, the conversion efficiency is adjusted to 0.55, resulting in 34.1 units of storable electrical energy. The high-frequency vibration signal of loom C is not converted because its energy dispersion does not meet the threshold requirement.

[0179] In this embodiment of the application, the method achieves efficient recycling of waste mechanical energy in the textile production process by accurately identifying and quantifying mechanical impact energy, which not only reduces equipment vibration loss, but also provides renewable auxiliary energy for the production system.

[0180] To further improve the accuracy of multi-loom coordinated control, in some embodiments, step 106: generating a coordinated control signal to control multiple looms based on the adjusted start-stop time series data and the scheduled storable energy value includes:

[0181] Step 701: Extract time period features from the adjusted start-stop time series data.

[0182] In step 701, the time period characteristics refer to the equipment operation patterns reflected in the adjusted start-stop time series data, including periodic characteristics such as high load periods and idle periods.

[0183] In this embodiment of the application, the system analyzes the adjusted start-stop schedule, identifies the time periods during which the equipment cluster operates, the time periods during which a single device operates independently, and the idle time periods of the entire system, and extracts characteristic parameters such as the start time, duration, and repetition cycle of these time periods.

[0184] Step 702: Generate real-time status data based on the storable energy value after scheduling.

[0185] In step 702, the real-time status data reflects the current available amount of storable electrical energy and its charging and discharging status, including dynamic parameters such as remaining power and charging / discharging capacity.

[0186] In this embodiment, the system monitors the real-time parameters of the energy storage device, including the current total amount of stored electrical energy, maximum charging and discharging power, remaining capacity percentage, and other information, and organizes this data into a status report in a unified format.

[0187] Step 703: Match the time period characteristics with the real-time status data to determine the optimal time period for energy release or storage.

[0188] In step 703, time period matching involves correlating the characteristics of equipment operation periods with power status data to identify the optimal energy allocation timing. The optimized time period is the best energy allocation time window obtained by matching equipment operating demands with power supply status.

[0189] In this embodiment, the system matches the high-load periods of the device group with the discharge capacity of the energy storage system, and matches the idle periods with the charging demand, so as to ensure that the energy release period coincides with the device demand period and the charging period coincides with the system's low-load period.

[0190] Step 704: Generate a coordinated control signal based on the optimized time period.

[0191] In this embodiment, the system first analyzes and optimizes the operating status and power demand of each loom during the time period, precisely aligns the equipment start-up and shutdown times with the available power supply periods, and establishes a time-energy mapping table. Then, based on the mapping table, it generates an instruction sequence containing a specific timestamp, equipment number, operation type, and energy source. The operation type includes start, stop, power call / storage, etc., and the energy source is marked as grid power supply or energy storage power supply. Finally, the instruction sequence is encoded into a standard protocol format that the equipment controller can recognize and synchronously distributed to the execution terminals of each loom and energy storage system through the industrial network to ensure that all equipment operates collaboratively according to the predetermined plan during the specified time period.

[0192] Here is a specific example:

[0193] In the coordinated control of three looms (A, B, and C) in the textile workshop, the system first analyzes the adjusted start-stop time series data and extracts three key time periods: 08:00-08:15 is the high-load period, during which looms A and C run simultaneously; 08:15-08:30 is the transition period, during which only loom C runs; and after 08:30 is the low-load period. Simultaneously, according to the scheduling scheme, the system obtains real-time data on storable electrical energy, showing that the available power supply is 15 units at 08:00 and increases to 30 units at 08:30. Through a time-period matching algorithm, the high-load period 08:00-08:15 is matched with the power release demand, determining 08:00 and 08:30 as the optimal power release periods. The matching degree is calculated using the formula M = S × E / (L × C), where M represents the matching degree, S represents the power supply, E represents the power utilization efficiency coefficient (0.9), L represents the load value, and C represents the time-period capacity coefficient (1.2). The calculated matching degree is 0.82 for the 08:00 time period and 0.91 for the 08:30 time period, both exceeding the matching threshold of 0.8. Based on this, the system generates a coordinated control signal, specifically including the following instructions: start loom A at 08:00 and use 15 units of electrical energy stored in loom B; start loom C at 08:10 and connect it to the power grid; stop loom B at 08:15 and activate its energy recovery function; and at 08:30, use the newly recovered 15 units of electrical energy from loom B to supply loom A.

[0194] In this embodiment of the application, the method achieves coordinated optimization of equipment operation and energy management through refined time period matching and dynamic scheduling, ensuring the stable and efficient operation of the production system while maximizing energy utilization efficiency.

[0195] Figure 2 A schematic diagram of a networked intelligent control system for looms based on process knowledge software, provided as an embodiment of this application, is shown below. Figure 2 As shown, the system includes:

[0196] The acquisition module 21 is used to acquire the main motor current data, start-stop time series data and vibration characteristic data of multiple looms.

[0197] The calculation module 22 is used to calculate the mechanical impact energy value based on the vibration characteristic data and convert the mechanical impact energy value into a storable electrical energy value.

[0198] The first generation module 23 is used to generate spatiotemporal correlation data of energy consumption based on the main motor current data, the start-stop time sequence data and the storable electrical energy value.

[0199] The second generation module 24 is used to generate start-stop control commands and energy storage scheduling commands based on the energy consumption spatiotemporal correlation data and through a multi-agent reinforcement learning algorithm.

[0200] The adjustment module 25 is used to adjust the start-stop time sequence data of each loom according to the start-stop control command, and simultaneously schedule the storable electrical energy value according to the energy storage scheduling command.

[0201] The third generation module 26 is used to generate a coordinated control signal to control multiple looms based on the adjusted start-stop time sequence data and the scheduled storable energy value.

[0202] Figure 2 The aforementioned networked intelligent control system for looms based on process knowledge software can execute... Figure 1 The implementation principle and technical effects of the networked intelligent control method for looms based on process knowledge software, as described in the illustrated embodiment, will not be repeated here. The specific methods by which each module and unit of the networked intelligent control system for looms based on process knowledge software in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0203] In one possible design, Figure 2 The networked intelligent control system for a loom based on process knowledge software, as shown in the embodiment, can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0204] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0205] The processing component 32 is used to perform the above. Figure 1 The embodiment describes a networked intelligent control method for looms based on process knowledge software.

[0206] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0207] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0208] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0209] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0210] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0211] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0212] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown is a networked intelligent control method for looms based on process knowledge software.

[0213] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0214] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0215] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0216] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A networked intelligent control method for looms based on process knowledge software, characterized in that, include: Collect main motor current data, start-stop time series data, and vibration characteristic data of multiple looms; The mechanical impact energy value is calculated based on the vibration characteristic data, and the mechanical impact energy value is converted into a storable electrical energy value. Based on the main motor current data, the start-stop time sequence data, and the storable electrical energy value, energy consumption spatiotemporal correlation data is generated; Based on the aforementioned energy consumption spatiotemporal correlation data, start-stop control commands and energy storage scheduling commands are generated through a multi-agent reinforcement learning algorithm. The start-stop time sequence data of each loom is adjusted according to the start-stop control command, and the storable electrical energy value is scheduled according to the energy storage scheduling command. Based on the adjusted start-stop time series data and the scheduled storable energy value, a coordinated control signal is generated to control multiple looms; The generation of start-stop control commands and energy storage scheduling commands based on the energy consumption spatiotemporal correlation data using a multi-agent reinforcement learning algorithm includes: The multiple looms are defined as multiple independent intelligent agents; Extract the grid load value corresponding to each independent intelligent agent from the energy consumption spatiotemporal correlation data; Based on the grid load value and the storable energy value of each independent intelligent agent, calculate the start-stop time adjustment amount and the energy dispatch amount; The agent collaboration module of the multi-agent reinforcement learning algorithm integrates the start-stop time adjustment amount and the power dispatch amount of all independent agents to generate start-stop control instructions and energy storage dispatch instructions. The process of integrating the start-stop time adjustment amounts and power dispatch amounts of all independent intelligent agents to generate start-stop control commands and energy storage dispatch commands includes: The start-stop time adjustments of all independent agents are aggregated to form a start-stop time adjustment dataset. The energy scheduling data of all independent intelligent agents are aggregated to form an energy scheduling dataset. Time conflict detection and elimination are performed on the aforementioned start / stop time adjustment dataset; The power dispatch dataset is subjected to supply and demand matching and balancing processing. Convert the conflict-resolved start / stop time adjustment dataset into start / stop control commands; Convert the power dispatch dataset after supply and demand balance into energy storage dispatch instructions; The time conflict detection and elimination process for the start / stop time adjustment dataset includes: Extract all planned start and stop time points from the start and stop time adjustment dataset; The system detects overlapping time intervals among the planned start and stop times. Set the minimum safe time interval required for starting and stopping the loom; The planned start and stop times within the detected overlapping time intervals are offset and adjusted according to the minimum safe time interval value. The planned start and stop times in the time adjustment dataset are updated using the adjusted start and stop times to obtain the time adjustment dataset after conflict resolution.

2. The method according to claim 1, characterized in that, The step of calculating start-stop time adjustment and power dispatch based on the grid load value and storable energy value of each independent intelligent agent includes: Based on the grid load value and the storable electrical energy value, calculate the state assessment index; If the state assessment index indicates that the grid load is higher than the preset grid load threshold, then the start / stop time offset is determined; If the state assessment index indicates that the storable energy value is lower than the preset storable energy threshold, then the amount of energy transferred is determined; Convert the start / stop time offset into a start / stop time adjustment amount; The energy transfer amount is converted into an electrical energy dispatch amount.

3. The method according to claim 1, characterized in that, The step of calculating the mechanical impact energy value based on the vibration characteristic data and converting the mechanical impact energy value into a storable electrical energy value includes: The vibration characteristic data is decomposed into multiple frequency band energy components; From the multiple frequency band energy components, the key frequency band energy components that are higher than the preset impact energy threshold are selected as the impact frequency band components; The mechanical impact energy value is obtained by performing energy integration calculation on the impact frequency band components. The mechanical impact energy value is converted into a storable electrical energy value.

4. The method according to claim 1, characterized in that, The process of generating coordinated control signals to control multiple looms based on adjusted start-stop time series data and scheduled storable energy values ​​includes: Extract time-period features from the adjusted start-stop time-series data; Real-time status data is generated based on the storable energy value after scheduling; The time period characteristics are matched with the real-time status data to determine the optimal time period for energy release or storage; Based on the optimized time period, a coordinated control signal is generated.

5. A networked intelligent control system for looms based on process knowledge software, characterized in that, include: The data acquisition module is used to collect main motor current data, start-stop time series data, and vibration characteristic data of multiple looms; The calculation module is used to calculate the mechanical impact energy value based on the vibration characteristic data, and convert the mechanical impact energy value into a storable electrical energy value; The first generation module is used to generate spatiotemporal correlation data of energy consumption based on the main motor current data, the start-stop time sequence data and the storable electrical energy value. The second generation module is used to generate start-stop control commands and energy storage scheduling commands based on the energy consumption spatiotemporal correlation data and through a multi-agent reinforcement learning algorithm. The adjustment module is used to adjust the start-stop time sequence data of each loom according to the start-stop control command, and simultaneously schedule the storable electrical energy value according to the energy storage scheduling command. The third generation module is used to generate coordinated control signals to control multiple looms based on the adjusted start-stop time series data and the scheduled storable energy value. The generation of start-stop control commands and energy storage scheduling commands based on the energy consumption spatiotemporal correlation data using a multi-agent reinforcement learning algorithm includes: The multiple looms are defined as multiple independent intelligent agents; Extract the grid load value corresponding to each independent intelligent agent from the energy consumption spatiotemporal correlation data; Based on the grid load value and the storable energy value of each independent intelligent agent, calculate the start-stop time adjustment amount and the energy dispatch amount; The agent collaboration module of the multi-agent reinforcement learning algorithm integrates the start-stop time adjustment amount and the power dispatch amount of all independent agents to generate start-stop control instructions and energy storage dispatch instructions. The process of integrating the start-stop time adjustment amounts and power dispatch amounts of all independent intelligent agents to generate start-stop control commands and energy storage dispatch commands includes: The start-stop time adjustments of all independent agents are aggregated to form a start-stop time adjustment dataset. The energy scheduling data of all independent intelligent agents are aggregated to form an energy scheduling dataset. Time conflict detection and elimination are performed on the aforementioned start / stop time adjustment dataset; The power dispatch dataset is subjected to supply and demand matching and balancing processing. Convert the conflict-resolved start / stop time adjustment dataset into start / stop control commands; Convert the power dispatch dataset after supply and demand balance into energy storage dispatch instructions; The time conflict detection and elimination process for the start / stop time adjustment dataset includes: Extract all planned start and stop time points from the start and stop time adjustment dataset; The system detects overlapping time intervals among the planned start and stop times. Set the minimum safe time interval required for starting and stopping the loom; The planned start and stop times within the detected overlapping time intervals are offset and adjusted according to the minimum safe time interval value. The planned start and stop times in the time adjustment dataset are updated using the adjusted start and stop times to obtain the time adjustment dataset after conflict resolution.

6. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the networked intelligent control method for looms based on process knowledge software as described in any one of claims 1 to 4.

7. A computer storage medium, characterized in that, The device stores a computer program, which, when executed by a computer, implements a networked intelligent control method for looms based on software-defined process knowledge as described in any one of claims 1 to 4.

Images