Transformer body one-piece flow drying system based on data driving and control method

Abstract

The invention relates to a transformer body one-piece flow drying system based on data driving and a control method, and belongs to the technical field of transformer manufacturing, the system comprises a three-dimensional drying warehouse unit, a material transfer unit, a distributed drying execution unit, a data acquisition and monitoring unit and an integrated control and scheduling unit; each storage location of the three-dimensional drying warehouse is an independent drying tank; the method comprises the steps that an MES creates a drying work order and dynamically assigns an optimal drying station based on a multi-objective optimization algorithm; the stacking machine carries out warehousing; the PLC controls the drying process according to the process curve and carries out intelligent end point judgment based on the insulation resistance change rate and the condensate water amount; after drying is completed, dispatching and warehouse-out are conducted; the whole process data is used for analysis and optimization. The continuous one-piece flow mode is adopted, the problems that a traditional drying mode is long in production takt, high in energy consumption and poor in flexibility are solved through three-dimensional layout and data-driven intelligent scheduling and control, and the production efficiency, the energy utilization rate and the product quality consistency are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of transformer manufacturing technology, specifically relating to a data-driven single-piece flow drying system and control method for transformer bodies. Background Technology

[0002] In the production of oil-immersed transformers, the drying process after the transformer body is assembled is a crucial step. The purpose of drying is to completely remove moisture from the insulation material of the transformer body, improve insulation performance, prevent partial discharge during operation, and avoid oxidation of metal components at high temperatures.

[0003] Currently, there are two main drying methods commonly used in the industry: tunnel furnace drying and batch drying in large vacuum drying tanks.

[0004] Tunnel furnace drying: The drying chambers pass sequentially through a long tunnel with multiple temperature zones via a conveyor chain. While this method offers a degree of continuity, it requires a large footprint, has high thermal inertia, and makes it difficult to precisely and quickly adjust process parameters (such as temperature and vacuum) for the characteristics of individual chambers. Furthermore, the chambers in front and behind influence each other, resulting in poor consistency in drying quality.

[0005] Batch drying in large vacuum drying tanks: Multiple assembled drying chambers are simultaneously hoisted into a large vacuum tank for centralized drying. This method has significant drawbacks: Production cycle bottleneck: Drying can only begin after all batches of equipment are assembled, which takes a long time (usually tens of hours). This causes the next process (such as oil immersion) to wait, severely restricting the overall production cycle and making it impossible to meet the modern production needs of multi-variety, small-batch, and fast delivery.

[0006] Energy waste: Regardless of the number of containers inside the tank, the entire large tank needs to be heated and evacuated, resulting in low energy efficiency.

[0007] Poor flexibility: The drying process curves (temperature, vacuum degree, time) required for different models and specifications of the vessel body may be different, and batch processing cannot achieve personalized processes.

[0008] Process black box: The lack of independent monitoring of the microscopic state of each part of the tank makes the determination of the drying endpoint dependent on experience or fixed time, which can easily lead to insufficient drying or over-drying.

[0009] Therefore, existing technologies suffer from problems such as low production efficiency, high energy consumption, poor flexibility, and crude quality control. There is an urgent need for a new drying system and method that can achieve flexible, refined, and continuous flow production. Summary of the Invention

[0010] Purpose of the Invention: This invention aims to overcome the shortcomings of existing technologies and provide a data-driven transformer body single-piece flow drying system and intelligent control method. This system achieves continuous single-piece flow operation for transformer body drying through a three-dimensional warehouse layout and intelligent scheduling. Utilizing comprehensive data acquisition and intelligent algorithms, it achieves precise control of the drying process, energy consumption optimization, and quality traceability, ultimately improving production efficiency, reducing energy consumption, ensuring quality consistency, and enhancing production flexibility.

[0011] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a data-driven transformer core single-piece flow drying system, comprising: The system includes a three-dimensional drying warehouse unit, a material transfer unit, a distributed drying execution unit, a data acquisition and monitoring unit, and an integrated control and scheduling unit. The three-dimensional drying warehouse unit includes multi-layer and multi-column three-dimensional racks, and each storage position of the three-dimensional racks is configured as an independent modular drying tank, forming a distributed drying workstation; The material transfer unit includes a conveyor chain for the conveyor body and a stacker crane for transferring the conveyor body between the conveyor chain and any of the modular drying tanks. The distributed drying execution unit includes a heating device, a vacuum device, and a temperature control valve installed in each of the modular drying tanks; The data acquisition and monitoring unit includes a sensor network and a central monitoring system installed in each of the modular drying tanks. The sensor network is used to collect process parameters, equipment status and energy consumption data of the corresponding drying tanks in real time and upload them to the central monitoring system. The integrated control and scheduling unit includes a manufacturing execution system, a warehouse control system, and a group of programmable logic controllers. The manufacturing execution system is used to generate drying work orders according to the production plan, and based on the system status information fed back by the data acquisition and monitoring unit, dynamically assign the optimal modular drying tank to each drying work order, and issue drying process instructions. The warehouse control system is configured to receive instructions from the manufacturing execution system and control the storage and retrieval operations of the material transfer unit actuator. The programmable logic controller group is connected to each of the modular drying tanks, and is used to receive and execute drying process instructions from the manufacturing execution system, control the drying process of the corresponding drying tank, and upload process data.

[0012] In a further embodiment, the three-dimensional drying warehouse unit has a layout of at least two rows, multiple columns, two layers or more, and the number of modular drying tanks is dozens to hundreds.

[0013] In a further embodiment, the sensor network includes at least a temperature sensor, a vacuum sensor, an online insulation resistance tester, and a smart meter.

[0014] In a further embodiment, the integrated control and scheduling unit further includes a data intelligence analysis engine, which connects to the databases of the manufacturing execution system and the central monitoring system, for analyzing historical drying data and performing process optimization, energy efficiency analysis, and / or predictive maintenance analysis of equipment.

[0015] Secondly, the present invention provides a transformer body single-piece flow control method applied to the above-mentioned system, comprising the following steps: S1: Task creation and binding: When the assembly of the vessel body is completed and put into operation, its identity information is scanned and entered into the manufacturing execution system. The manufacturing execution system creates a corresponding drying work order according to the vessel body model. S2: Dynamic Assignment of Drying Station: The manufacturing execution system obtains the real-time status of all modular drying tanks and dynamically assigns an optimal modular drying tank to the current drying work order based on a multi-objective optimization algorithm. S3: Automatic warehousing and process start-up: The warehouse control system dispatches the stacker crane to transfer the container body to the assigned modular drying tank. After the tank door is closed, the manufacturing execution system sends the corresponding drying process curve to the programmable logic controller corresponding to the tank to start the drying program. S4: Data-driven drying process control and endpoint determination: During the drying process, the programmable logic controller controls the process according to the drying process curve and monitors the insulation resistance change rate and / or condensate volume of the unit body in real time; based on the real-time monitoring data and preset endpoint determination conditions, it intelligently determines the drying endpoint. S5: Outbound scheduling and transfer: When the drying endpoint is determined to be reached, the manufacturing execution system marks the modular drying tank as ready for outbound, and schedules the warehouse control system to control the stacker crane to take out the dried tank body and transfer it to the next process. S6: Data Aggregation and Analysis: Store the data from the entire drying process along with the equipment's identity information for quality traceability, process optimization, and energy efficiency analysis.

[0016] In a further embodiment, in step S2, the optimization objectives of the multi-objective optimization algorithm include: minimizing the total drying completion time, maximizing the overall comprehensive efficiency of the equipment, and balancing the load of each drying station; its constraints include process compatibility and equipment failure avoidance; the manufacturing execution system uses a genetic algorithm or a particle swarm optimization algorithm to solve for the optimal assignment scheme.

[0017] In a further embodiment, in step S4, the preset endpoint determination condition includes: Condition 1: The actual drying time reaches or exceeds the minimum time preset in the process curve; Condition 2: The real-time monitored insulation resistance value of the transformer body reaches or exceeds the target threshold, and its rate of change is lower than the stable threshold; Condition 3: The amount of condensate per unit time is lower than the condensate threshold; When conditions one, two, and three are met simultaneously, the drying endpoint is determined to have been reached.

[0018] In a further embodiment, in the later stage of the drying process, the programmable logic controller takes the drying data from the previous stage as input, predicts the subsequent trend of insulation resistance change through a drying kinetic model, and adaptively adjusts the holding time in the process curve accordingly.

[0019] In a further embodiment, step S6, the data analysis includes: Process optimization analysis: By comparing the drying process data and final insulation performance of different bodies of the same model, the standard drying process curve was iteratively optimized; Energy efficiency benchmarking analysis: Statistical analysis of unit product energy consumption for different models of modular drying tanks to establish energy efficiency benchmarks; Predictive maintenance analysis: Analyze the operating current, vibration, or noise data of key equipment, establish a health model, and generate maintenance warnings.

[0020] In a further embodiment, in step S5, the outbound scheduling also considers the production cycle time of downstream processes and the waiting time of the outbound equipment to coordinate the overall production process.

[0021] Compared with the prior art, the present invention has the following significant advantages: Revolutionary improvement in production efficiency: It achieves true single-piece continuous drying, eliminates the "waiting" and "congestion" bottlenecks of batch drying, significantly shortens the production cycle, and significantly increases capacity, making it particularly suitable for flexible and fast-paced production.

[0022] Extremely high space utilization: Adopting a three-dimensional warehouse layout, it provides dozens or even hundreds of independent drying stations within a unit area, solving the problem of large footprint of traditional drying equipment.

[0023] Energy-efficient and refined: Each drying tank has independent temperature control and vacuuming, avoiding energy waste caused by "small batches occupying large equipment". Combined with load- and process-based energy consumption optimization scheduling, the overall energy efficiency of the system is more than 30% higher than that of traditional methods.

[0024] Excellent and consistent drying quality: By implementing independent and traceable process curve control for each unit and intelligent endpoint determination based on multiple parameters (insulation resistance, vacuum degree, condensate volume), the drying effect of each unit is ensured to reach the best, resulting in extremely high quality consistency.

[0025] The system offers exceptional production flexibility: it can simultaneously process different models and process requirements of vessel bodies, enabling mixed-flow production. Process formulations can be adjusted online via MES, allowing for rapid response to engineering changes.

[0026] High degree of automation and intelligence: From warehousing, distribution, drying control to outbound delivery, the entire process is unmanned. Through data-driven processes, intelligent scheduling, adaptive process control, and predictive maintenance are achieved.

[0027] Full lifecycle data traceability: A complete "drying digital profile" has been established for each transformer body, providing a solid data foundation for product quality analysis, process improvement and customer problem tracing. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the overall architecture of the single-piece flow drying system described in this invention.

[0029] Figure 2 This is the overall flowchart of the intelligent control method described in this invention.

[0030] Figure 3 This is a detailed logic diagram of the adaptive control and endpoint determination algorithm for the drying process in this invention.

[0031] Figure 4 This is a flowchart of the dynamic scheduling algorithm for drying stations based on multi-objective optimization in this invention.

[0032] Figure 5 This is an example of the algorithm of the present invention (simplified pseudocode). Detailed Implementation

[0033] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0034] A data-driven transformer body single-piece flow drying system, such as Figures 1 to 4 As shown, it includes: a three-dimensional drying warehouse unit, a material transfer unit, a distributed drying execution unit, a data acquisition and monitoring unit, and an integrated control and scheduling unit.

[0035] The automated drying warehouse unit consists of at least two rows, multiple columns, and multiple layers (preferably 2 rows, 25 columns, and 2 layers, totaling 100 storage locations). Each standard storage location is equipped with an independent drying station, which integrates an insulation shell, heating device, vacuum system interface, temperature, humidity and vacuum sensors, automatic door mechanism, etc., forming a modular drying tank.

[0036] Material handling unit: includes inbound conveyor chain, outbound conveyor chain, and high-speed stacker crane. The stacker crane is responsible for automatically storing and retrieving equipment between the conveyor chain and any drying station in the automated drying warehouse.

[0037] Distributed drying execution unit: Composed of actuators arranged in each drying station, including: electric heater / hot air circulation system, vacuum pump group and pipeline valves, cold trap, circulating fan, etc. Each drying tank can be independently controlled for start / stop and process parameters.

[0038] Data acquisition and monitoring unit: Equipment-level data acquisition: Through a sensor network installed in each drying tank, process parameters (such as tank temperature, hot spot temperature, vacuum degree, insulation resistance telemetry value, leakage rate), equipment status (heater status, pump status, door status, fan status, fault codes) and energy consumption data (independent electricity meter data for each tank) are collected in real time.

[0039] Transmission network: The above data is uploaded to the SCADA system in real time using a combination of industrial Ethernet and fieldbus.

[0040] SCADA system: Responsible for centralized monitoring, alarming, historical data storage, and trend analysis of the operating status and process parameters of all drying tanks in the entire system. Provides a visual human-machine interface.

[0041] Integrated control and scheduling unit: Consists of a multi-layered system that communicates with each other. WCS: Receives production instructions from MES and is responsible for directing stacker cranes and conveyor chains to complete specific inbound and outbound handling tasks, and optimizing the stacker crane's travel path.

[0042] MES: Receives the production plan from the upper-level ERP system and generates a unique "drying work order" for each drying tank. The work order includes the tank model, number, and preset drying process curve. Based on the system status (occupancy of each tank, drying progress, equipment health) and optimization algorithms, MES dynamically assigns the optimal drying station and sends the drying instructions (process curve) to the corresponding drying tank's PLC.

[0043] PLC control group: Each or each drying tank is controlled by an independent PLC, which is responsible for receiving process instructions from MES, controlling the heating, vacuuming and other processes of the tank in a closed loop, ensuring that the actual process parameters follow the set curve, and uploading real-time data.

[0044] Based on the above system, the present invention provides a transformer body single-piece flow control method applied to the above system.

[0045] The method includes the following steps: S1: Drying Unit Binding and Task Creation: When the assembled drying unit comes online, its identity information (ID, model) is entered into the MES via RFID or QR code scanning. The MES generates a drying work order with a unique task ID based on the standard drying process library for that model and places the drying unit in the inbound buffer.

[0046] S2: Dynamic scheduling of drying stations based on multi-objective optimization: The scheduling module in the MES obtains the real-time status (idle, working, faulty, under maintenance) of all drying stations and the remaining time for the current task. When there is a drying work order to be processed, the scheduling module executes the following algorithm: Input: Collection of dryer body Drying station collection Status of each workstation, historical OEE data for each workstation, and energy consumption coefficient.

[0047] Optimization objectives: Minimize total drying completion time (MakeSpan), maximize overall equipment OEE, balance the load of each workstation, and consider energy consumption factors.

[0048] Constraints: process compatibility (e.g., large equipment bodies can only be assigned to the lowest level workstations), equipment failure avoidance.

[0049] Algorithm Flow: An improved genetic algorithm or particle swarm optimization algorithm can be used to solve the problem. The evaluation function comprehensively considers the shortest path (stacker access time), completion time estimation (based on process duration), and energy efficiency priority (prioritizing the allocation of workstations with high energy efficiency). The final output is the optimal workstation allocation scheme. .

[0050] S3: Automatic Inventory Management and Process Start-up Based on the assignment results from MES, WCS schedules the stacker crane to move the stacker body. Transported from the receiving buffer to the designated drying tank. After the tank door closes automatically, the MES will display the corresponding drying process curve (temperature-time curve). Vacuum degree-time curve Distribute to designated drying tank The PLC starts the drying program and begins controlling heating and vacuuming according to the curve.

[0051] S4: Data-driven adaptive control and endpoint prediction for the drying process: During the drying process, the PLC not only follows the preset curve, but also incorporates real-time data for fine-tuning.

[0052] Insulation resistance online tracking: Real-time monitoring of insulation resistance value of the transformer body Establish a drying kinetic model, for example, by monitoring... The drying rate is determined by the rate of change in insulation resistance. Continuously below the set threshold After a certain period of time, it indicates that the rate of water separation has slowed down.

[0053] Combined analysis of vacuum level and leakage rate: During the high temperature and high vacuum stage, the system leakage rate is calculated in real time. If the leakage rate remains stable at an extremely low level and the vacuum level meets the standard, it can help determine the drying effect.

[0054] Endpoint intelligent determination algorithm: The drying endpoint is not a fixed time, but is determined by the following conditional logic and judgment: The preset heat preservation and pressure holding time has been met; Insulation resistance value Target value reached ,and ; The amount of condensate in the cold trap per unit time is lower than the threshold W_threshold.

[0055] When the PLC determines that the endpoint conditions are met, it sends a "drying complete" signal to the MES.

[0056] S5: Outbound scheduling based on real-time status and global optimization: After receiving the "drying complete" signal, the MES will move the workstation... Marked as "Pending Outbound". Outbound scheduling considerations: Downstream process requirements: Coordination with the cycle time of the oil immersion process.

[0057] The stacker crane's current task queue.

[0058] Priority outbound principle: Prioritize the outbound of completed equipment with the longest waiting time, or prioritize the outbound of high-priority urgent orders.

[0059] The WCS dispatch stacker crane removes the dried container body and places it on the outbound conveyor chain for transport to the next process.

[0060] S6: Full-process data aggregation and in-depth analysis: All process data (process parameters, equipment status, energy consumption, quality results) are associated with the equipment ID and task ID and stored in the database.

[0061] Process optimization: By comparing the drying curves and final insulation performance of different transformer bodies of the same model, the standard process library was iteratively optimized. , curve.

[0062] Energy efficiency analysis: Statistical analysis of unit energy consumption for each workstation and product model to identify energy efficiency gaps.

[0063] Predictive maintenance: Analyze the operating current, vibration, and temperature trends of key equipment such as vacuum pumps and heaters, establish a health model, and provide early warnings before failures occur.

[0064] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the invention.

[0065] Example 1: This embodiment takes a system with 100 drying stations (2 rows × 25 columns × 2 layers) as an example.

[0066] System Configuration: Automated warehouse: rack height 15 meters, internal dimensions of individual drying tanks designed according to the largest tank size, equipped with double-layer insulation walls, built-in nickel-chromium alloy heating belts, and adjustable power.

[0067] Stacker crane: Double column type, with double deep forks, traveling speed 180m / min, lifting speed 60m / min.

[0068] Data Acquisition: Each tank is equipped with: a type K thermocouple (for measuring multiple temperatures inside the tank and the surface temperature of the heater), a capacitive thin-film vacuum gauge, an online insulation resistance tester (connected to the tank body via leads), and a smart meter. All sensor signals are connected to a local remote I / O station, and then to the central control room via the Profinet ring network.

[0069] Control system: Siemens S7-1500 series PLCs are used as the main control PLCs for each row of drying tanks. The SCADA system uses WinCC OA and is deployed on an industrial server. The MES uses a customized development module and interacts with the ERP (SAP) and WCS (integrated into the stacker crane control system) via WebService / RESTful API.

[0070] Method implementation process: like Figure 5As shown, a transformer body of model SZ11-50000 / 110 has been assembled, RFID tag (record ID: TR-2023-08001) has been attached, and it has been placed at point A of the inbound conveyor chain by the overhead crane.

[0071] A fixed RFID reader reads a tag, and the MES system creates a task, Job_TR-2023-08001. The baseline drying curve for this model in the process library is: preheating at 80℃ for 4 hours, then gradually increasing the temperature to 115℃ and holding for 30 hours, during which the vacuum level is gradually increased to below 5 Pa.

[0072] At the time of scheduling, there are 10 idle workstations in the system, 3 of which are located at the bottom layer and are suitable for large equipment bodies. After calculation by the scheduling algorithm, the task is assigned to workstation B-01-01 (B row, 1 column, 1st layer, bottom layer) because it is closest to the warehouse entrance and has the highest periodic OEE.

[0073] The stacker crane transports the equipment body to workstation B-01-01 and closes the door. The MES system issues the process curve. The PLC controls the start-up: first, the circulating fan and low-temperature heating are started to begin preheating.

[0074] During the drying process, the SCADA screen displays the real-time status of all tanks. The insulation resistance value of tank B-01-01 slowly increases from an initial 100MΩ. Once it enters the 115℃ holding stage, the system calculates... In the 28th hour, The flow rate dropped to below 0.5 MΩ / h and stabilized. Meanwhile, monitoring of the cold trap condensate flow showed no new condensate generation in the past two hours.

[0075] The PLC's built-in endpoint determination model comprehensively judges that the heat preservation time (30h) has been met. and The condensate volume meets the standard. Therefore, the "drying complete" signal is triggered at 30.5 hours (slightly longer than the preset time, as it is waiting for a stable judgment).

[0076] MES received a signal marking B-01-01 as ready for outbound shipment. At this time, the oil immersion process had a spare space, and WCS immediately dispatched the stacker crane to remove the container and send it to point C of the outbound conveyor chain, ready for oil immersion.

[0077] Meanwhile, the data engine recorded a total power consumption of 1850 kWh and an average final insulation resistance of 8500 MΩ for this drying process, and updated the data to the digital profile of the device and the process statistical analysis library for this model.

[0078] Example 2: Building upon Example 1, this example focuses on the application of predictive maintenance. A data intelligence analysis engine continuously monitors the startup current curve and operating noise spectrum of each vacuum pump (using a vibration microphone). It learns the pump's normal operating status through machine learning models (such as Isolation Forest). When the startup current peak of a pump (e.g., serving workstation A-10-02) exceeds the normal range by 10% multiple times consecutively, and the noise spectrum shows an energy increase in a specific frequency band, the system generates a yellow warning, prompting maintenance personnel to check the pump's lubricating oil and bearing condition. The workstation is then marked as "requiring maintenance" in the MES system and temporarily avoided during subsequent scheduling, thus preventing sudden downtime due to production failures.

[0079] In summary, this invention provides a complete intelligent solution from hardware layout to software algorithms. It not only revolutionizes the drying mode of transformer bodies but also achieves optimal quality, efficiency, and cost in the manufacturing process through deep data fusion and intelligent control, possessing extremely high industrial application value and market prospects. Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A data driven based transformer body single piece flow drying system characterized by, The application relates to a three-dimensional drying warehouse unit, a material transfer unit, a distributed drying execution unit, a data acquisition and monitoring unit and an integrated control and scheduling unit. The three-dimensional drying warehouse unit comprises a plurality of layers and columns of three-dimensional shelves, each storage position of the three-dimensional shelves is configured as an independent modular drying tank, and the three-dimensional shelves constitute distributed drying stations. The material transfer unit comprises a conveying chain for conveying a body and a stacker for transferring the body between the conveying chain and any of the modular drying tanks. The distributed drying execution unit comprises a heating device, a vacuum device and a temperature control valve arranged in each of the modular drying tanks. The data acquisition and monitoring unit comprises a sensor network arranged in each of the modular drying tanks and a central monitoring system, the sensor network is used for acquiring process parameters, equipment states and energy consumption data of the corresponding drying tank in real time and uploading the data to the central monitoring system. The integrated control and scheduling unit comprises a manufacturing execution system, a warehouse control system and a programmable logic controller group. The manufacturing execution system is used for generating a drying work order according to a production plan, dynamically assigning an optimal modular drying tank for each drying work order based on system state information fed back by the data acquisition and monitoring unit and issuing drying process instructions. The warehouse control system is configured to receive the assignment instructions of the manufacturing execution system and control the material transfer unit to perform access work of the body. The programmable logic controller group is connected with each of the modular drying tanks and is used for receiving and executing the drying process instructions from the manufacturing execution system, controlling the drying process of the corresponding drying tank and uploading process data. The three-dimensional drying warehouse unit has a layout of at least two rows, multiple columns, two layers or more than two layers, and the number of the modular drying tanks is dozens to hundreds.

2. The data driven transformer body single piece flow drying system of claim 1, wherein, The sensor network at least comprises a temperature sensor, a vacuum degree sensor, an insulation resistance online tester and a smart meter.

3. The data driven single piece stream drying system for transformer cores of claim 1, wherein, The integrated control and scheduling unit further comprises a data intelligent analysis engine connected with databases of the manufacturing execution system and the central monitoring system and used for analyzing historical drying data, performing process optimization, energy efficiency analysis and / or predictive maintenance analysis of equipment.

4. The data driven single piece stream drying system for transformer cores of claim 1, wherein, The application further discloses a method for dynamically assigning a drying station, which comprises the following steps:

5. A method for controlling the flow of a single piece of transformer core applied to the system of any one of claims 1-4, characterized by, S1: task creation and binding: when the body is completed and put into operation, the identity information of the body is scanned and recorded in a manufacturing execution system, and the manufacturing execution system creates a corresponding drying work order according to the body model; S2: dynamic assignment of a drying station: the manufacturing execution system acquires real-time states of all modular drying tanks and dynamically assigns an optimal modular drying tank for a current drying work order based on a multi-objective optimization algorithm; S3: automatic warehousing and process starting: a warehouse control system schedules a stacker to transfer the body to the assigned modular drying tank, after the tank door is closed, the manufacturing execution system issues a corresponding drying process curve to a programmable logic controller of the tank to start a drying program. ​ S4: Data-driven drying process control and end-point determination: During the drying process, the programmable logic controller controls according to the drying process curve, and monitors the insulation resistance change rate and / or condensate of the body in real time; according to the real-time monitoring data and the preset end-point determination condition, intelligent determination of the drying end-point is carried out; S5: Outbound scheduling and transfer: when the drying end-point is determined to arrive, the manufacturing execution system marks the modular drying tank as a state to be out of the warehouse, and schedules the warehouse control system to control the stacker to take out the dried body and transfer it to the next process; S6: Data aggregation and analysis: store the data of the whole drying process associated with the body identity information for quality traceability, process optimization and energy efficiency analysis.

6. The transformer core single piece flow control method of claim 5, wherein, In step S2, the optimization objectives of the multi-objective optimization algorithm include: minimizing the total drying completion time, maximizing the overall comprehensive efficiency of the equipment, and balancing the load of each drying station; the constraint conditions include process compatibility and equipment failure avoidance; the manufacturing execution system uses genetic algorithm or particle swarm algorithm to solve the optimal assignment scheme.

7. The transformer core single piece flow control method of claim 5, wherein, In step S4, the preset end-point determination conditions include: Condition one: the actual drying time reaches or exceeds the minimum time preset in the process curve; Condition two: the real-time monitored insulation resistance value of the body reaches or exceeds the target threshold, and its change rate is lower than the stable threshold; Condition three: the condensate amount per unit time is lower than the condensate threshold; When conditions one, two and three are met at the same time, the drying end-point is determined to arrive.

8. The transformer core single piece flow control method of claim 7, wherein, In the later stage of the drying process, the programmable logic controller takes the previous stage drying data as input, predicts the subsequent insulation resistance change trend through the drying kinetics model, and adjusts the holding time in the process curve accordingly.

9. The transformer core single piece flow control method of claim 5, wherein, In step S6, the data analysis includes: Process optimization analysis: compare the drying process data and final insulation performance of different bodies of the same model to iteratively optimize the standard drying process curve; Energy efficiency benchmarking analysis: statistics of the unit product energy consumption of each modular drying tank processing different models of bodies, establishment of energy efficiency benchmark; Predictive maintenance analysis: analyze the running current, vibration or noise data of the key equipment, establish a health model and generate a maintenance warning.

10. The transformer core single piece flow control method of claim 9, wherein, In step S5, the outbound scheduling also considers the production rhythm of the downstream process and the waiting time of the body to be out of the warehouse, so as to coordinate the overall production process.

Images