Directional energy deposition cooling rate active regulation and control system and method and storage medium

Abstract

The invention discloses a directional energy deposition cooling rate active regulation and control system and method, and a storage medium, and the system comprises a temperature monitoring module which is used for collecting the temperature information of different positions; the LSTM model prediction module is used for predicting future temperature change according to the real-time temperature data and outputting a prediction result; the MPC control module is used for optimally calculating a cooling rate control strategy; wherein an optimal control parameter is calculated according to a prediction result of the LSTM model prediction module, and a cooling rate control signal is output; the cooling medium control module is used for adjusting the cooling rate according to the cooling rate control signal; and the feedback adjusting module is used for monitoring the actual temperature of the workpiece in real time, evaluating the cooling effect and transmitting feedback data to the MPC optimization control module to complete dynamic adjustment. By the adoption of the technical scheme, the problems that in the existing directional energy deposition process, the cooling rate is not accurately regulated and controlled, and changing working conditions cannot be adapted are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of directed energy deposition, and in particular relates to a directed energy deposition cooling rate active control system and method, and a storage medium. Background Art

[0002] During the directed energy deposition (DED) process, the cooling rate has a crucial influence on the microstructure and mechanical properties of the workpiece. The non-uniformity during the cooling process often leads to deformation of the workpiece, which not only affects the appearance quality of the workpiece, but also may reduce its mechanical properties and service life. Especially in the manufacture of workpieces with complex geometries, the nonlinear and time-varying characteristics of temperature distribution are more prominent, which makes it difficult for traditional cooling control methods, such as passive regulation based on ambient temperature, to cope with the dynamic changes and complexity of temperature distribution. Therefore, a method for active deformation control that can adjust the cooling rate in real time, accurately and dynamically is needed to ensure the quality and performance of the workpiece during additive manufacturing. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a directed energy deposition cooling rate active control system and method, and a storage medium.

[0004] To achieve the above object, the present invention adopts the following technical solution:

[0005] A directed energy deposition cooling rate active control system, comprising:

[0006] Temperature monitoring module, used to collect temperature information at different locations;

[0007] The LSTM model prediction module is used to predict future temperature changes based on real-time temperature data.

[0008] And output the prediction results;

[0009] The MPC control module is used to optimize and calculate the cooling rate control strategy; wherein, according to the prediction results of the LSTM model prediction module, the optimal control parameters are calculated and the cooling rate control signal is output;

[0010] The cooling medium control module is used to adjust the cooling rate according to the cooling rate control signal.

[0011] Preferably, it also includes: a feedback adjustment module, which is used to monitor the actual temperature of the workpiece in real time, evaluate the cooling effect, and transmit the feedback data to the MPC optimization control module to complete dynamic adjustment.

[0012] Preferably, the temperature monitoring module is a temperature sensor or an infrared thermal imager.

[0013] The present invention also provides a method for actively controlling the cooling rate of directed energy deposition, comprising:

[0014] Step S1, collecting temperature information at different locations through a temperature monitoring module;

[0015] Step S2: predict future temperature changes based on real-time temperature data using the LSTM model prediction module, and output the prediction results;

[0016] Step S3, optimizing and calculating the cooling rate control strategy through the MPC control module; wherein, according to the prediction results of the LSTM model prediction module, the optimal control parameters are calculated, and a cooling rate control signal is output;

[0017] Step S4: adjusting the cooling rate according to the cooling rate control signal through the cooling medium control module.

[0018] Preferably, the method further includes: step S5, monitoring the actual temperature of the workpiece in real time through the feedback adjustment module, evaluating the cooling effect, and transmitting the feedback data to the MPC optimization control module to complete dynamic adjustment.

[0019] Preferably, the temperature monitoring module is a temperature sensor or an infrared thermal imager.

[0020] The present invention also provides a storage medium, on which a computer program is stored, and when the computer program is run, the method for actively regulating the cooling rate of directed energy deposition is executed.

[0021] The present invention actively controls the cooling rate of directed energy deposition through an LSTM model prediction module and an MPC optimization control module, predicts the temperature change of the workpiece in real time, and optimizes the cooling rate control strategy based on the prediction result, thereby solving the problem of inaccurate cooling rate control and inability to adapt to changing working conditions in the existing directed energy deposition (DED) process.

[0022] The present invention has the following technical effects:

[0023] 1. Compared with traditional methods, the present invention can accurately control the cooling rate and reduce deformation by using the LSTM model to predict the temperature in time series and combining it with MPC optimization control.

[0024] 2. The prediction of the LSTM model and the real-time optimization of MPC in the present invention reduce unnecessary delays in the cooling process, thereby shortening the production cycle and reducing the scrap rate; improving production efficiency, especially in the manufacturing of high-precision workpieces.

[0025] 3. In the present invention, MPC controls the flow rate of the cooling medium, avoiding overcooling and waste of coolant, thereby saving energy and coolant usage and reducing production costs.

[0026] 4. The LSTM training process in the present invention is based on simulation to train a small data set, adopts reinforcement learning technology, and has low training cost; MPC can adjust the cooling strategy in real time, adapt to environmental changes, and ensure that the cooling process can operate stably under various working conditions.

[0027] 5. The present invention reduces manual intervention, simplifies the operation process, and improves the intelligence and automation level of the system by automatically adjusting the cooling strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0029] Figure 1 This is a schematic diagram of the composition of a directed energy deposition cooling rate control system according to an embodiment of the present invention;

[0030] Figure 2 This is the LSTM network structure diagram;

[0031] Figure 3 This is a data flow chart of the directed energy deposition cooling rate control system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] Embodiment 1:

[0035] like Figure 1 As shown, an embodiment of the present invention provides a directed energy deposition cooling rate active control system, which combines the LSTM model with the directed energy deposition cooling rate active control controlled by MPC, and predicts the changing trend of the workpiece surface temperature in real time and dynamically optimizes the cooling rate, thereby improving the accuracy of the cooling process and reducing deformation.

[0036] The directed energy deposition cooling rate active control system of the present invention includes: a temperature monitoring module, an LSTM model prediction module, an MPC optimization control module, a cooling medium control module and a feedback adjustment module; wherein the temperature monitoring module is a temperature sensor or an infrared thermal imager; the cooling medium control module includes: a coolant pump, a hollow substrate, and a PVC hose; the feedback adjustment module includes: a temperature sensor, a feedback signal transmission channel, and a graphics workstation.

[0037] Temperature monitoring module, used to collect temperature information at different locations;

[0038] The LSTM model prediction module is used to predict future temperature changes based on real-time temperature data.

[0039] And output the prediction results;

[0040] The MPC control module is used to calculate the optimal control parameters according to the prediction results of the LSTM model prediction module and output a cooling rate control signal;

[0041] The cooling medium control module adjusts the flow rate of the coolant according to the control signal of the MPC optimization control module. The coolant pump controls the flow rate of the coolant by changing the speed, and the flow channel inside the hollow base plate evenly distributes the coolant to the bottom surface of the workpiece to achieve uniform cooling. The flow change of the coolant pump will adjust the distribution of the coolant through the hollow base plate to ensure the cooling effect. The PVC hose connects the coolant pump and the hollow base plate to transmit the coolant. By adjusting the size and length of the hose, the flow rate of the coolant can be further adjusted, and the precise control of the cooling rate can be achieved in collaboration;

[0042] The feedback adjustment module measures the temperature of the workpiece surface in real time through the temperature sensor and converts the temperature information into a digital signal. The converted digital signal is transmitted to the graphics workstation through the wired signal transmission channel. The workstation receives and processes the digital signal, and evaluates the cooling effect by visually displaying the temperature data. If the temperature is found to be too high or too low, the workstation will generate a feedback signal and pass it to the MPC optimization control module. The MPC optimization control module adjusts the flow rate or other control parameters of the coolant pump according to the received digital feedback signal to achieve dynamic cooling rate adjustment.

[0043] As an implementation of an embodiment of the present invention, the LSTM model prediction module uses a long short-term memory network (LSTM) to perform time series prediction on the surface temperature of the workpiece during the additive manufacturing process. Due to the time series characteristics of the temperature field during the additive manufacturing process, the temperature change is affected by the temperature at the previous moment and a variety of process parameters. Therefore, the LSTM model predicts the temperature change trend in the future by inputting historical temperature data and related process parameters. Specifically, it includes:

[0044] 1) Data input and processing

[0045] The input data of the LSTM model prediction module includes real-time temperature data in the additive manufacturing process and related process parameter data, such as arc current, cooling medium flow, workpiece deposition speed, etc. These data are collected in real time through temperature sensors and other monitoring equipment, and are used as input to the LSTM network after preprocessing (such as standardization or normalization).

[0046] 2) LSTM network structure

[0047] The LSTM network has multiple layers, each layer includes multiple neuron units. The network structure design usually includes an input layer, several LSTM hidden layers and an output layer. The network structure is as follows: Figure 2 shown.

[0048] Input layer: The input layer receives preprocessed data, including historical temperature data and process parameters.

[0049] LSTM hidden layer: The role of the hidden layer is to learn the temporal dependency of temperature data and capture the temperature change pattern at each moment. The LSTM layer controls the flow of information through a series of gating mechanisms (input gate, forget gate, and output gate), thereby avoiding the gradient vanishing problem in traditional recurrent neural networks and effectively learning long-term dependencies.

[0050] Output layer: The output layer generates temperature prediction data for future time steps based on the prediction results of the LSTM model. These prediction results provide a reference for future temperatures for the MPC optimization module to help it formulate cooling strategies.

[0051] 3) LSTM training and optimization

[0052] The training of the LSTM model is based on historical temperature data and process parameter data. The training process includes the following steps:

[0053] Data collection: By installing temperature sensors and other measuring devices during the additive manufacturing process, temperature variation data under different process parameters are collected.

[0054] Data preprocessing: The collected raw data is standardized or normalized to ensure that each data feature is within the same scale range to facilitate the training of the LSTM network.

[0055] Model training: Use the training dataset to train the LSTM model and optimize the model parameters (such as weights and biases) to minimize the prediction error.

[0056] Model verification and testing: After training is completed, use the verification set to verify the prediction accuracy of the model. If the prediction accuracy is not high, it is necessary to further adjust the structure of the LSTM network or increase the amount of training data.

[0057] As an implementation method of an embodiment of the present invention, the MPC optimization control module is based on the feedback control method of the system model, which can predict the state of the system in the future period of time, and optimize the control input on this basis to achieve the control target. The MPC optimization control module calculates and optimizes the cooling strategy in the additive manufacturing process in real time based on the future temperature prediction information provided by the LSTM model prediction module. The specific working principle is as follows:

[0058] State prediction: MPC predicts the future temperature change trend of the workpiece through the temperature data provided by the LSTM model prediction module. Based on process parameters (such as deposition rate, heat input, etc.), MPC predicts the temperature change of the workpiece in multiple future time steps.

[0059] Optimization goals: MPC defines optimization goals, such as: Keep the temperature of the workpiece within a predetermined safety range. Reduce local thermal stress and cracks by precisely adjusting the cooling rate. Minimize fluctuations in coolant flow to avoid overcooling or overheating.

[0060] Constraints: MPC sets constraints based on the actual limitations of the system and process requirements, including upper and lower limits on coolant flow.

[0061] Constraints on temperature variation range (to prevent local overheating or overcooling).

[0062] Optimization solution: MPC uses numerical optimization algorithms to solve control problems and calculate the optimal cooling medium flow input in the future time step. During the solution process, MPC will continuously adjust the control strategy to achieve the predetermined control target.

[0063] Real-time feedback: The MPC optimization control module not only relies on the prediction results of LSTM, but also receives real-time feedback information on the surface temperature of the workpiece. When there is a large deviation between the actual temperature and the predicted temperature, MPC can adjust the control strategy to ensure that the system always remains within the predetermined temperature range.

[0064] As an implementation method of an embodiment of the present invention, the cooling medium control module is used to accurately adjust the flow rate of the cooling medium according to the control instructions output by the MPC optimization control module, so as to ensure that the temperature distribution of the workpiece during the additive manufacturing process meets the predetermined requirements and avoid quality problems caused by uneven cooling, such as cracks, deformation, etc. The workflow of the cooling medium control module is closely coordinated with the LSTM model prediction module and the MPC optimization control module.

[0065] The workflow is as follows:

[0066] Data collection and preprocessing: During the additive manufacturing process, the temperature sensor monitors the temperature of the workpiece surface in real time and transmits the data to the LSTM model prediction module. LSTM makes predictions based on historical data and generates temperature change trends for a period of time in the future.

[0067] MPC Optimization Control: Based on the LSTM prediction data, the MPC optimization control module calculates the future coolant flow control parameters and sends these instructions to the cooling medium control module.

[0068] Cooling medium flow regulation: The flow control valve of the coolant supply system adjusts the coolant flow according to the instructions of the MPC module. For example, in high temperature areas, MPC will instruct to increase the flow to accelerate cooling; in low temperature areas, reduce the coolant flow to avoid overcooling.

[0069] Temperature feedback adjustment: The temperature change during the coolant injection process is fed back to the MPC optimization control module in real time, and the MPC further adjusts the cooling strategy based on the feedback information. If the temperature exceeds the set range, the MPC can immediately adjust the coolant flow rate to avoid local overheating.

[0070] Precise control and optimization: The cooling medium control module continuously executes the control instructions of MPC throughout the additive manufacturing process, accurately adjusts the flow of coolant, and ensures that the temperature change during each layer deposition process is within the target range. By dynamically adjusting the cooling strategy, the temperature gradient and thermal stress are reduced to avoid workpiece defects caused by overheating or overcooling.

[0071] The active control of cooling rate of directed energy deposition in the embodiment of the present invention can be applied to wire arc additive manufacturing (WAAM) equipment based on cold metal transfer (CMT) technology; the equipment needs to be equipped with the following functional modules:

[0072] (1) Infrared thermal imager: used to monitor the surface temperature of the workpiece in real time and transmit the data to the LSTM model prediction module.

[0073] (2) Cooling system: used to accurately control the flow rate of coolant to ensure that the cooling effect meets the optimized control strategy.

[0074] (3) Control computing platform: Integrates the LSTM model and the MPC control algorithm to calculate and adjust the cooling process in real time.

[0075] like Figure 3 As shown, the process of preparing additive materials using the directed energy deposition cooling rate control system according to the embodiment of the present invention includes:

[0076] (1) Workpiece preparation: Select materials such as aluminum alloy or magnesium alloy suitable for CMT additive manufacturing and load them onto the additive manufacturing platform.

[0077] (2) LSTM model training: Collect historical temperature data from the AM process and use this data to train the LSTM model so that it can predict temperature changes over a period of time in the future.

[0078] (3) MPC control: Based on the prediction results of the LSTM model, MPC optimizes the cooling strategy and adjusts the flow rate, temperature, and injection pattern of the cooling medium in real time to ensure the control of the cooling rate and temperature distribution.

[0079] (4) Additive Manufacturing: When the WAAM process begins, the LSTM and MPC control systems work together in real time to dynamically adjust the cooling rate to ensure that the temperature field of the workpiece is uniform during the additive manufacturing process and to avoid thermal stress, cracks, and deformation.

[0080] The directed energy deposition cooling rate control system of the present invention is suitable for various additive manufacturing applications, especially WAAM based on CMT technology. It is particularly suitable for the manufacture of light metal materials such as aluminum alloys and magnesium alloys. Due to its low heat input characteristics, this technology can effectively control the cooling rate and reduce the deformation defects of light metal materials caused by rapid cooling.

[0081] LSTM model optimization of the embodiment of the present invention According to different additive manufacturing tasks, the architecture and parameters of the LSTM model can be adjusted, such as changing the number of layers and nodes of the LSTM network to improve the accuracy of temperature prediction. MPC control algorithm optimization of the embodiment of the present invention: The optimization goals and constraints of MPC can be adjusted according to actual needs. For example, if the workpiece requires higher accuracy in temperature control, constraints can be added to further improve control accuracy. In the CMT additive manufacturing process, the present invention focuses on regulating the flow parameters of the cooling medium, and the specific process parameters can be optimized according to different materials (such as aluminum alloys, magnesium alloys, etc.) and the required workpiece quality.

[0082] LSTM is a deep learning model widely used in time series prediction, and is particularly suitable for processing and predicting complex data with time series dependencies such as temperature changes. In the additive manufacturing process, LSTM can capture the long-term dependence of the surface temperature of the workpiece over time, and predict the temperature trend in the future based on the historical temperature data. Through this process, the LSTM model can provide accurate temperature prediction data for the cooling strategy, helping to formulate a more scientific and reasonable cooling rate adjustment plan. At the same time, the model predictive control (MPC) technology can combine the temperature prediction results of LSTM to optimize the future cooling process. MPC is a prediction-based optimization control method that calculates the future control strategy (such as coolant flow, cooling time, etc.) in real time and continuously adjusts the control input to minimize the system deviation and ensure the achievement of the goal. Combined with the temperature prediction of LSTM, MPC can dynamically adjust the cooling rate based on real-time feedback data and expected temperature changes, thereby maintaining high robustness and control accuracy under different manufacturing environments and process parameters.

[0083] The present invention uses a combination of LSTM and MPC to achieve active control, and can accurately adjust the cooling rate according to the dynamic changes in the workpiece temperature, thereby effectively avoiding workpiece defects caused by uneven cooling. In addition, since this method can respond to temperature changes and adjust cooling strategies in real time, it can meet manufacturing needs under different materials, processes and environmental conditions, significantly improve the workpiece quality and mechanical properties in the additive manufacturing process, reduce thermal stress and deformation risks, and provide a new solution for the manufacture of high-precision and high-performance workpieces.

[0084] Embodiment 2:

[0085] The embodiment of the present invention also provides a method for actively controlling the cooling rate of directed energy deposition, comprising:

[0086] Step S1, collecting temperature information at different locations through a temperature monitoring module;

[0087] Step S2: predict future temperature changes based on real-time temperature data using the LSTM model prediction module, and output the prediction results;

[0088] Step S3, optimizing and calculating the cooling rate control strategy through the MPC control module; wherein, according to the prediction results of the LSTM model prediction module, the optimal control parameters are calculated, and a cooling rate control signal is output;

[0089] Step S4: adjusting the cooling rate according to the cooling rate control signal through the cooling medium control module.

[0090] As an implementation of the embodiment of the present invention, it also includes: step S5, monitoring the actual temperature of the workpiece in real time through the feedback adjustment module, evaluating the cooling effect, and transmitting the feedback data to the MPC optimization control module to complete dynamic adjustment.

[0091] As an implementation manner of the embodiment of the present invention, the temperature monitoring module is a temperature sensor or an infrared thermal imager.

[0092] Embodiment 3:

[0093] An embodiment of the present invention further provides a storage medium having a computer program stored thereon, and the computer program executes a method for actively controlling the cooling rate of directed energy deposition when the computer program is run.

[0094] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A directed energy deposition cooling rate active control system, characterized in that: include: Temperature monitoring module, used to collect temperature information at different locations; LSTM model prediction module, used to predict future temperature changes based on real-time temperature data and output prediction results; The MPC control module is used to optimize and calculate the cooling rate control strategy; wherein, according to the prediction results of the LSTM model prediction module, the optimal control parameters are calculated and the cooling rate control signal is output; The cooling medium control module is used to adjust the cooling rate according to the cooling rate control signal.

2. The directed energy deposition cooling rate active control system according to claim 1, characterized in that: Also includes: The feedback adjustment module is used to monitor the actual temperature of the workpiece in real time, evaluate the cooling effect, and pass the feedback data to the MPC optimization control module to complete dynamic adjustment.

3. The method for actively controlling the cooling rate of directed energy deposition according to claim 2, characterized in that: The temperature monitoring module is a temperature sensor or an infrared thermal imager.

4. A method for actively controlling the cooling rate of directed energy deposition, characterized in that: include: Step S1, collecting temperature information at different locations through a temperature monitoring module; Step S2: predict future temperature changes based on real-time temperature data using the LSTM model prediction module, and output the prediction results; Step S3, optimizing and calculating the cooling rate control strategy through the MPC control module; wherein, according to the prediction results of the LSTM model prediction module, the optimal control parameters are calculated, and a cooling rate control signal is output; Step S4: adjusting the cooling rate according to the cooling rate control signal through the cooling medium control module.

5. The method for actively controlling the cooling rate of directed energy deposition according to claim 4, characterized in that: The method further includes: step S5, monitoring the actual temperature of the workpiece in real time through the feedback adjustment module, evaluating the cooling effect, and transmitting the feedback data to the MPC optimization control module to complete dynamic adjustment.

6. The method for actively controlling the cooling rate of directed energy deposition according to claim 5, characterized in that: The temperature monitoring module is a temperature sensor or an infrared thermal imager.

7. A storage medium, characterized in that: The storage medium stores a computer program, which, when running, executes the method for actively controlling the cooling rate of directed energy deposition as described in any one of claims 4 to 6.

Images