Control method, system and equipment for injection molding process and storage medium
By building a state space model and real-time update of the control sequence, the Lm optimization algorithm is used to realize dynamic optimization control of the injection molding and filling process, solving the problem of difficult to obtain state changes in the existing technology, improving control accuracy and stability, optimizing production efficiency and Product quality.
Patent Information
- Application Number
- CN202411353510.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-26
- Publication Date
- 2024-11-05
AI Technical Summary
The existing control methods for injection molding filling process cannot effectively consider changes in the state of the filling process, making it difficult to achieve stable and dynamic optimal feedback control, and the calculation complexity is high, and it does not have the simplicity of engineering applications.
By obtaining the state data of injection molding fill, numerical modeling is used to construct a state space model, the initial control sequence is updated in real time based on real-time state data, the optimal control sequence is obtained using the Lm optimization algorithm, and the injection molding fill is controlled in real time.
It improves the dynamic control accuracy and stability of the injection molding filling process, optimizes production efficiency, reduces material losses, and improves product quality.
Smart Images

Figure CN118893798A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of injection molding, and in particular to a control method, system, equipment and storage medium for an injection molding process. Background Art
[0002] The first is the filling process control with feedback correction of preset control curve. According to material properties such as viscoelasticity, a curve in the filling process is pre-set, such as the relationship curve between melt speed and time, the relationship curve between injection pressure and time, etc., to realize the control of the filling process. This type of method realizes the control of the filling process through the control curve, and the control accuracy is limited. At the same time, when the parameter design of the control curve is unreasonable, it is easy to cause fluctuations in the filling process, resulting in defects in the filling process or too long filling time. In addition, this type of control method cannot realize information feedback of the control process and is not adaptive. Feedback control of the injection molding filling process, such as using BP neural network, RBF neural network, least squares method, genetic algorithm, particle swarm algorithm, ant colony computing and other methods to realize the state observer of the filling process, optimize the filling process according to the state feedback of the sensor through the optimization method, and realize effective filling process control. This type of control method realizes information feedback of the filling process, and can greatly improve the control accuracy and the effect of filling process control through feedback control. However, this type of method ignores its solution complexity, requires a large number of matrix operations and iterative operations, and does not have the simplicity of engineering application. Moreover, this type of control method only predicts the terminal state of the filling process, i.e., the filling time, the filling melt quality, etc., and implements feedback control based on the difference between the predicted value and the set value. The control method does not take into account the state changes of the filling process, and the state transition mechanism of the filling process is difficult to obtain, which makes it difficult to achieve stable and dynamically optimal feedback control of the filling process, does not have the effect of dynamic process optimization, and cannot further reduce the uncertainty and material loss of the filling process.
[0003] In the prior art, the state change of the filling process is not taken into consideration, and the state transition mechanism of the filling process is difficult to obtain, which makes it difficult to achieve stable and dynamically optimal feedback control of the filling process. Summary of the invention
[0004] The present invention provides a control method for an injection molding process, so as to solve the problem in the prior art that the state change of the filling process is not taken into consideration, the state transfer mechanism of the filling process is difficult to obtain, and it is difficult to achieve stable and dynamically optimal feedback control of the filling process.
[0005] In a first aspect, in order to solve the above technical problems, the present invention provides a method for controlling an injection molding process, comprising: Get the injection filling status data; Wherein, the status data includes historical status data and real-time status data; Based on the historical state data, a state space model is constructed by using numerical modeling, and an initial control sequence is obtained according to the state space model; According to the real-time state data and the initial control sequence, the initial control sequence is updated in real time using the Lm optimization algorithm to obtain an optimal control sequence, and the injection molding filling is controlled in real time according to the optimal control sequence.
[0006] Preferably, the step of obtaining the injection filling status data includes: Get the injection temperature, injection speed, injection pressure, mold temperature, holding time and cooling time at different times during the injection filling process.
[0007] Preferably, the constructing of a state space model by numerical modeling based on the historical state data, and obtaining an initial control sequence according to the state space model, includes: Initializing the historical state data, and defining a state vector according to the historical state data; According to the state vector, a state space model is established by using a least square method; Based on a preset objective function, the historical state data is processed using the state space model to obtain an initial control sequence.
[0008] Preferably, before obtaining the initial control sequence by processing using the state space equation based on the preset output index, the method further comprises: The preset objective function is constructed according to the following formula: in is the performance indicator value, is the state vector, Control input vector, is the state weight matrix, is the control input weight matrix, For the optimized time range, is the transpose of the state vector, is the transpose of the control input vector.
[0009] Preferably, the method of updating the initial control sequence in real time using an Lm optimization algorithm according to the real-time state data and the initial control sequence to obtain an optimal control sequence, and controlling injection molding filling in real time according to the optimal control sequence comprises: Initializing the real-time status data; Based on the real-time status data, a preset error function is used to make a judgment: If the error value is greater than a preset error threshold, the initial control sequence is updated using the Lm optimization algorithm; If the error value is less than a preset error threshold, the updating is stopped to obtain the optimal control sequence, and the injection molding filling is controlled in real time according to the optimal control sequence.
[0010] Preferably, the method further comprises: After the optimal control sequence is obtained, the initial control sequence is updated to the optimal control sequence as the initial control sequence for the next same injection molding process.
[0011] Preferably, if the error value is greater than a preset error threshold, the initial control sequence is updated using an Lm optimization algorithm, including: The initial control sequence is updated according to the following formula: in, is the Jacobian matrix, is the control vector at the current k moment, is the updated control vector, is the error vector, To adjust the parameters, is the identity matrix, which is used to prevent the Jacobian matrix from being of insufficient rank.
[0012] Preferably, the method further comprises: After the injection molding filling process is completed, the optimal control sequence is updated to the initial control sequence for the next same injection molding process.
[0013] In a second aspect, the present invention provides a control system for an injection molding process, comprising: A data acquisition module, for acquiring injection filling status data; wherein the status data includes historical status data and real-time status data; A model building module, based on the historical state data, uses numerical modeling to build a state space model, and obtains an initial control sequence according to the state space model; A sequence updating module, which updates the initial control sequence in real time using an Lm optimization algorithm according to the real-time state data and the initial control sequence to obtain an optimal control sequence; The injection molding control module controls the injection molding filling in real time according to the optimal control sequence.
[0014] In a third aspect, the present invention further provides a terminal device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the control method of the injection molding process described in any one of the above is implemented.
[0015] In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for controlling the injection molding process.
[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a control method for an injection molding process, comprising: obtaining state data of injection molding filling; wherein the state data comprises historical state data and real-time state data; based on the historical state data, constructing a state space model by using numerical modeling, and obtaining an initial control sequence according to the state space model; according to the real-time state data and the initial control sequence, updating the initial control sequence in real time by using an Lm optimization algorithm to obtain an optimal control sequence, and controlling the injection molding filling in real time according to the optimal control sequence. In the control process, by constructing a state space model according to the historical state data, describing the dynamic evolution of the system and obtaining an initial control sequence, and updating the initial control sequence in real time by using an Lm optimization algorithm, the dynamic control accuracy and stability of the injection molding filling process are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 is a schematic flow chart of a control method for an injection molding process provided by a first embodiment of the present invention; Figure 2 It is a schematic diagram of the control system structure of the injection molding process provided by the second embodiment of the present invention. DETAILED DESCRIPTION
[0018] 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.
[0019] Reference Figure 1 The first embodiment of the present invention provides a method for controlling an injection molding process, comprising the following steps: S11, obtaining injection filling status data; wherein the status data includes historical status data and real-time status data.
[0020] S12, based on the historical state data, construct a state space model by using numerical modeling, and obtain an initial control sequence according to the state space model.
[0021] S13, according to the real-time state data and the initial control sequence, using the Lm optimization algorithm to update the initial control sequence in real time to obtain an optimal control sequence, and controlling the injection molding filling in real time according to the optimal control sequence.
[0022] The injection molding process is a processing method widely used in the production of plastic products. It injects heated and molten plastic material into the mold cavity, cools and solidifies it, and finally forms a product of the desired shape. This process is widely used in the fields of automobiles, electronics, home appliances, and medical treatment, and has the advantages of high production efficiency, good product consistency, and mass production. However, with the increasing complexity of industrial products and the improvement of product quality requirements, the traditional injection molding process faces many challenges.
[0023] In the field of injection molding technology, the control of the injection molding filling process is a key link to ensure product quality and production efficiency. Traditional injection molding filling control methods mainly rely on preset control curves, such as the relationship between melt speed and time or the relationship between injection pressure and time. These methods use pre-set parameters to control the filling process, but the control accuracy is limited, and when the parameter design is unreasonable, it is easy to cause fluctuations in the filling process, resulting in defects or prolonged filling time. In addition, these traditional methods lack information feedback mechanisms, are not adaptive, and cannot adjust control strategies according to real-time changes.
[0024] With the increasing complexity of industrial products and the requirements for product quality, traditional injection molding processes face many challenges. In order to improve control accuracy and filling effect, researchers have begun to explore feedback control methods for injection molding filling processes. These methods include the use of optimization techniques such as BP neural network, RBF neural network, least squares method, genetic algorithm, particle swarm algorithm, ant colony computing, etc., to optimize the filling process through state observers. Although these methods realize information feedback of the filling process and significantly improve the control accuracy, they usually require a large number of matrix operations and iterative calculations, with high computational complexity and inconvenient for engineering applications. In addition, these control methods mainly focus on predicting the terminal state of the filling process, such as filling time and filling melt quality, while ignoring the dynamics of state changes during the filling process. Therefore, the existing control methods are difficult to achieve stable and dynamically optimal feedback control of the filling process, and cannot effectively reduce the uncertainty and material loss of the filling process.
[0025] The present invention aims to solve the problems existing in the prior art, and realizes dynamic optimization control of the injection molding filling process by constructing a state space model and applying the Lm optimization algorithm. This method can not only predict the filling process based on historical state data, but also dynamically adjust the control strategy based on real-time state data, thereby improving the dynamic control accuracy and stability of the injection molding filling process. Through this method, precise control of the injection molding filling process can be achieved, production efficiency can be optimized, material loss can be reduced, and the quality of the final product can be improved.
[0026] In order to facilitate the understanding of the present invention, some preferred embodiments of the present invention are further described below.
[0027] In step S11, the state data of injection filling is obtained; The state data includes historical state data and real-time state data. Historical state data refers to a series of parameter records related to injection molding filling collected and stored in previous injection molding processes, which include key variables such as injection temperature, injection speed, injection pressure, mold temperature, holding time and cooling time at different times. These data are usually monitored and automatically recorded in real time by sensors equipped with injection molding machines, or manually recorded by operators at specific time intervals during the production process. The accumulation of historical state data provides valuable information resources for the analysis and optimization of the injection molding process. They are used to establish and train numerical models of the injection molding process. By analyzing patterns and trends in historical data, it can help predict filling behavior, identify potential quality problems, and adjust and optimize current and future injection molding process parameters accordingly.
[0028] Real-time status data refers to data collected synchronously during the current injection molding filling process. They include the real-time parameters of the injection molding machine during operation, such as the current injection temperature, injection speed, injection pressure, mold temperature, holding time and cooling time. These data are obtained through real-time monitoring by sensors installed on the injection molding machine. The sensors convert the monitored physical quantities into electrical signals, which are then processed and recorded by the data acquisition system. The acquisition of real-time status data is crucial for the dynamic control of the injection molding process. They are used to update and adjust the control strategy to ensure the stability of the injection molding process and the quality of the product.
[0029] It should also be noted that the acquisition of injection filling status data includes: The state data include injection temperature, injection speed, injection pressure, mold temperature, holding time, and cooling time at different times during the injection filling process. Among them, the injection temperature refers to the temperature reached by the plastic material during the injection stage during the injection molding process. This temperature is a key parameter to ensure that the plastic material can be smoothly injected into the mold in a molten state. It affects the fluidity of the material, the filling effect, and the quality and performance of the final product. The injection temperature is usually controlled by the heating system of the injection molding machine, which includes a barrel heater and a nozzle heater, which heat the plastic material to achieve the desired molten state. During the injection molding process, the monitoring of the injection temperature is achieved through thermocouples or thermistors installed on the barrel and nozzle. These sensors can measure and feedback the temperature of the material in real time so that the operator or automatic control system can make adjustments.
[0030] Injection speed refers to the rate at which the plastic melt is injected into the mold cavity during the injection molding filling process. It is an important parameter in the injection molding process, which directly affects the filling quality, appearance and internal structure of the product. The injection speed can be adjusted by the control system of the injection molding machine, and is usually pre-set by the operator based on factors such as material properties, product design and mold structure. In the actual injection molding process, the injection speed is achieved by the servo motor of the injection molding machine driving the injection screw or piston. The speed and torque of the motor determine the injection speed of the melt. Injection molding machines are usually equipped with speed sensors that can monitor and feedback the injection speed in real time to ensure its precise control.
[0031] Injection pressure refers to the pressure applied when the plastic melt is injected into the mold cavity during the injection molding filling process. This parameter is crucial to ensure that the melt can smoothly fill the entire mold and avoid defects such as bubbles and short shots. The injection pressure is usually controlled by the hydraulic system of the injection molding machine by adjusting the output pressure of the hydraulic pump. When setting up the injection molding process, the operator will pre-set the injection pressure based on material properties, mold design and product requirements. During the actual injection molding process, the injection pressure is monitored in real time by the injection molding machine's sensor to ensure that it is within the set range. This data is not only used for real-time feedback control, but also for numerical modeling to analyze the impact of different injection pressures on the filling effect.
[0032] Mold temperature refers to the temperature of the mold cavity surface during the injection molding filling process. It has a significant impact on the cooling rate of the plastic melt, the filling behavior, and the surface quality and dimensional stability of the final product. The mold temperature is usually controlled by a cooling or heating system designed inside the mold, which can be cooling water channels, heating rods or other heat exchange devices. Before injection molding, the operator sets the target mold temperature based on the characteristics of the plastic material and product requirements, and preheats or cools it through the temperature control system. During the injection molding process, the mold temperature is monitored in real time by temperature sensors installed in the mold. These sensors transmit temperature data to the injection molding machine control system for necessary adjustments. The control of mold temperature is crucial to ensure product quality. Appropriate mold temperature can speed up the production cycle, reduce material internal stress, and improve the dimensional accuracy and surface finish of the product.
[0033] The holding time refers to the period of time during which a certain pressure is applied to the melt after the injection molding filling process is completed in order to ensure that the plastic melt fully fills the mold and compensates for the shrinkage of the material during cooling. This parameter is crucial to controlling the dimensional stability of the product and reducing internal defects such as shrinkage and warpage. The setting of the holding time is usually based on the thermal properties of the material, the cooling rate of the mold, and the thickness and structural complexity of the product. On the injection molding machine, the holding time can be preset by the control system and implemented by the holding valve or holding piston of the injection unit. During the injection molding process, the setting of the holding time needs to take into account the cooling curve of the material and the thermal response of the mold. Through experimental and empirical data, the operator can determine the appropriate holding time to ensure that the plastic melt is fully solidified in the mold cavity while avoiding overfilling of the material or mold damage caused by excessive holding pressure. The holding time data is used to control the duration of the holding stage during the injection molding process and adjust the holding strategy by real-time monitoring of pressure and temperature changes in the mold.
[0034] Cooling time is the time required for the plastic melt to cool from a molten state to a solid state in the mold cavity after the injection molding filling process. This period of time is critical to ensure that the plastic product can stably maintain its shape, size and obtain the necessary mechanical properties. The determination of cooling time is usually based on factors such as the thermal properties of the material, the design of the mold's cooling system, the thickness and complexity of the product. On the injection molding machine, the cooling time can be preset by the control system, and the mold temperature can be adjusted by the flow of cooling water channels or other cooling media in the mold. During the injection molding process, the setting of the cooling time needs to be precisely controlled to ensure that the product is completely solidified before the mold is opened to prevent deformation or damage to the product caused by premature mold opening. By monitoring the temperature changes in the mold in real time, the cooling time can be adjusted to suit different production requirements and material properties. The cooling time data is used to control the duration of the cooling stage during the injection molding process, and to optimize the entire injection molding cycle by synergizing with other parameters such as injection speed, injection pressure and holding time.
[0035] In step S12, a state space model is constructed based on the historical state data by using numerical modeling, and an initial control sequence is obtained according to the state space model.
[0036] It should be noted that the state space model is constructed based on the historical state data by using numerical modeling, and the initial control sequence is obtained according to the state space model, including: Initializing the historical state data, and defining a state vector according to the historical state data; According to the state vector, a state space model is established by using a least square method; Based on a preset objective function, the historical state data is processed using the state space model to obtain an initial control sequence.
[0037] Specifically, the state vector is defined according to the historical state data: in, is the injection temperature, is the mold temperature, is the injection speed, is the injection pressure, is the holding time, These variables are used to describe the state of the system and constitute the state vector , represents the state of the system at k time steps.
[0038] Further, based on the state vector, a state space model is established using the least square method; Among them, the state space model is a mathematical model used to describe the time evolution of a dynamic system. In injection molding process control, the state space model consists of two parts: the state equation and the output equation. The state equation defines the law of change of the system state over time, while the output equation links these internal states with observable system outputs (such as product size, surface quality, etc.).
[0039] The state equation is used to describe the evolution of the state of the system between time steps: in, is the state vector of the system at k time steps, which contains all important state vectors in the injection molding process, including injection temperature, injection pressure, etc. is the state vector of the system at k+1 time steps. It is the control input vector, which includes all adjustable control parameters, such as injection temperature, injection speed, injection pressure, etc. These inputs are the means used by the controller to influence the behavior of the system.
[0040] It is the state transfer matrix, which defines how the state vectors affect each other over time. It is obtained by fitting the least squares method and reflects the inherent laws of the system's internal dynamics. In the process of fitting the A matrix, the state vector is mainly found through historical data. and First, obtain the state vector of the system at different time steps from the historical data , and the corresponding state vector at the next moment , and ignore the control input term B and noise , can be simplified to: Arrange the historical data into input-output pairs of states, and the input data is , the output data is , minimize the following sum of squared errors according to the following formula: The double vertical lines in the formula are the modulus of the vector obtained by finding the internal expression. The modulus of a vector is also called the length of the vector or the magnitude of the vector, which refers to the length of the vector in space. The modulus of a vector can be expressed in absolute value notation.
[0041] Then solve the expression of A according to the following formula to obtain the state transfer matrix A: in, is the state vector matrix at time step k+1, is the state vector matrix at time step k.
[0042] is the control input matrix, which describes how the control input affects the change of the state vector. The control input matrix is also obtained through data fitting, which links the control input with the state change; when fitting B, since the fitted A matrix is already available, the influence of the control input on the state is mainly fitted through historical data. Similarly, the control input is obtained from historical data , the corresponding state vector and the state at the next moment , the state change part is composed of Decide: Linearize the relationship between control input and state change and use the least squares method to fit B: The double vertical lines in the formula are the modulus of the vector obtained by finding the internal expression.
[0043] Solve the expression for B as follows: in Indicates the state change part, is the control input vector matrix.
[0044] Represents random disturbances or external interference that cannot be predicted by the model. It is the system noise in the state equation, indicating the error caused by the unmodeled dynamics within the system, external interference, or random factors in the system. These noises come from the complexity of the system, nonlinear effects, modeling simplification, and random interference from the external environment. It is usually assumed to be zero-mean Gaussian white noise, that is, a zero mean and a variance of The value of is obtained by calculating the residual between the system output and the true measurement value and analyzing the residual distribution. Usually, it cannot be directly observed, but its statistical characteristics can be estimated by the errors in historical data.
[0045] The output equation is used to describe the relationship between the output and state of the system: in, It is the output vector, which represents the output of the system, including the dimensional deviation and surface defect rate of the injection molded product and other parameter indicators, which can be directly obtained from the historical status data.
[0046] Dimensional deviation refers to the difference between the actual size of the product after injection molding and the designed size. It is one of the key indicators for quality control of injection molded products, because dimensional accuracy directly affects the fit and functionality of the product. Dimensional deviation is usually caused by a variety of factors, including the thermal expansion coefficient of the material, mold temperature, injection pressure, cooling rate, and holding time. These factors interact during the injection molding process, causing the plastic melt to shrink when it cools and solidifies in the mold, thus affecting the size of the final product.
[0047] The surface defect rate refers to the proportion of surface defects of products produced during the injection molding process. It is one of the important indicators to measure the quality of injection molded products. These defects may include but are not limited to shrinkage marks, bubbles, cracks, unevenness or other surface irregularities, which are usually caused by factors such as uneven material flow, inappropriate cooling rate, improper mold design or unreasonable processing parameter settings. In the injection molding process, the determination of the surface defect rate is usually based on the inspection and quality control standards of the finished product. Surface defects can be identified and recorded by visual inspection, dimensional measurement or other non-destructive testing of the injection molded product. These data are then used to calculate the defect rate, which is the ratio of the number of defective products to the total production quantity.
[0048] is the output matrix, which defines how the state vector is transformed into the observable system output, also obtained by data fitting; It is the direct effect of control input on output, which allows the model to consider the direct effect of control input on system output; the determination process of C and D matrices is similar to that of A and B, and is also obtained by fitting historical data. Usually, by analyzing the control input Output In the state-space model, if the control input significantly affects the output without propagation of the state vector, the value of the D matrix will not be zero. Otherwise, if the control input has no direct effect on the output, the D matrix can be simplified to a zero matrix.
[0049] is the observation noise, reflecting the uncertainty in the measurement, It is generally assumed to be zero-mean Gaussian white noise, Similarly, its specific value depends on the dynamic characteristics of the system and the complexity of the external disturbance. Usually, the difference between the fitted value and the true state in the observed system state equation is analyzed, and the distribution characteristics of these differences are estimated. Statistical properties.
[0050] It should also be noted that, before the state space equation is used for processing based on the preset output index and the preset objective function to obtain the initial control sequence, the method further includes: The preset objective function is constructed according to the following formula: in, is the performance indicator value, is the state vector, is the control input vector, is the state weight matrix, which indicates the degree of attention paid to the system state. Different state vectors are given different weights to control the cost of their deviation from the target state. is the control input weight matrix, which represents the cost or limit of the control input and is used to balance the use of control input to prevent overuse of control resources. is the optimized time range, which represents the time step of system prediction, is the transpose of the state vector, is the transpose of the control input vector.
[0051] The state weight matrix Q determines the weight of the influence of different state vectors (such as temperature, pressure, speed, etc.) on the control target. It measures the cost of the system state deviating from the ideal value. Each element in the matrix represents the degree of attention paid to different state deviations; its determination process is to set the weights in the matrix by analyzing the influence of different state vectors on the quality of injection molded products. For example, if the injection molding process is sensitive to temperature, the temperature-related weights will be set higher, while the weights for insensitive state vectors such as holding time will be set lower. In general, the Q matrix is a diagonal matrix, and the non-zero elements are the weights of the state vectors. By adjusting the value of Q, the control system can be controlled to pay more attention to the accuracy of certain states.
[0052] The control input weight matrix R is used to measure the control input vector It reflects the importance of reducing the use of control inputs and preventing the control system from over-relying on certain control inputs or over-consuming control resources. Its determination process is usually set by staff based on the physical constraints of the system, the control cost, and the focus on energy consumption. For example, in the injection molding process, frequent adjustments to temperature or pressure may consume too much energy or accelerate equipment wear, so higher weights are applied to these control inputs to reduce their frequency of use. The value of the matrix can be determined through experiments or debugging based on the actual equipment's operating costs, energy consumption requirements, and stability requirements. Similar to the state weight matrix, the control input weight matrix R is usually also a diagonal matrix, and the non-zero elements reflect the cost of each control input.
[0053] Finally, based on the preset objective function, the state space model is used to process the historical state data to obtain the initial control sequence. First, the state equation is substituted into the preset objective function to obtain: The objective function becomes the control input vector Since the state equation is linear, a linear quadratic regulator (LQR) is used to calculate an optimal feedback control law to minimize the objective function J.
[0054] The core of LQR control is to obtain the feedback gain matrix K by solving the Riccati equation, and then use the feedback control law to calculate the optimal control input, as follows: Where K is the feedback gain matrix, which represents the current state vector The degree of influence on the control input; Solving the Riccati equation is as follows: Among them, P is a symmetric positive definite matrix, which represents the feedback effect of future cost on the current state. A and B are the state transfer matrix and control input matrix in the state space model respectively; Q is the state weight matrix, and R is the control input weight matrix. By solving this equation, the matrix P can be obtained, and then the feedback gain matrix K is calculated by the following formula: After obtaining the feedback gain matrix K, first, according to the given initial state , calculate the initial control input Then, using the equation of state To predict the state at the next moment Next, based on the new state , continue to calculate the control input , and repeat this process to predict the state of each subsequent moment until the predetermined time step N is reached. Finally, the initial control sequence is obtained .
[0055] In step S13, the initial control sequence is updated in real time using the Lm optimization algorithm according to the real-time state data and the initial control sequence to obtain an optimal control sequence, and injection molding filling is controlled in real time according to the optimal control sequence.
[0056] It should be noted that the method of updating the initial control sequence in real time using the Lm optimization algorithm according to the real-time state data and the initial control sequence to obtain an optimal control sequence, and controlling the injection molding filling in real time according to the optimal control sequence, includes: Initializing the real-time status data; Based on the real-time status data, a preset error function is used to make a judgment: If the error value is greater than a preset error threshold, the initial control sequence is updated using the Lm optimization algorithm; If the error value is less than a preset error threshold, the updating is stopped to obtain the optimal control sequence, and the injection molding filling is controlled in real time according to the optimal control sequence.
[0057] Among them, the preset error threshold is a performance indicator set in the injection molding process control, which defines the acceptable output error range. This threshold is used to evaluate whether the deviation between the injection molding product and the design specifications in terms of size, appearance quality, etc. is within the allowable error range. It is usually determined based on product design requirements, functional requirements, and industry standards, and optimized through historical production data, quality control experience, and customer feedback. During the injection molding process, the real-time monitoring system collects data such as product dimensional deviation, surface defect rate, and residual stress, and compares these actual outputs with the preset target output. If the error between the actual output and the target output exceeds the preset error threshold, this indicates that the current injection molding process needs to be adjusted to improve product quality.
[0058] Specifically, the real-time status data is obtained by sensors in real time during the injection molding process, including the current injection temperature, injection speed, injection pressure, mold temperature, holding time and cooling time, etc. is the current control parameter vector. In an optional implementation, It can be defined as follows: In order to evaluate the control effect, a preset error function is used to measure the difference between the real-time state data and the predicted value of the control model. The preset error function is defined as follows: in is an error vector, which contains error terms of multiple performance indicators. is the actual output vector at the kth time step, obtained from real-time sensor data. is the kth time step, based on the current control parameter vector The calculated model predicted output vector. This predicted value is calculated using the state space model and the initial control sequence, and represents the output that the system should produce under the current control parameters.
[0059] In an optional implementation, since different output indicators may have different impacts on system performance, the error vector Weighted to reflect the importance of different outputs: in, is the weighted error vector, W is the diagonal weight matrix, and the weight values can be set empirically according to the importance of different output variables.
[0060] Furthermore, according to the formula The error value is calculated. If the error value is greater than the preset error threshold, the initial control sequence is updated using the Lm optimization algorithm according to the following formula based on the error vector Compute the Jacobian matrix : Among them, the Jacobian matrix reflects the sensitivity of the current control parameter changes to the system output and is the key to the algorithm iterative update of the control parameters.
[0061] It should also be noted that if the error value is greater than the preset error threshold, the initial control sequence is updated using the Lm optimization algorithm, including: The initial control sequence is updated according to the following formula: in, is the Jacobian matrix, is the transpose of the Jacobian matrix, is the control vector at the current k moment, is the updated control vector, is the error vector, To adjust the parameters, is the identity matrix, which is used to prevent the Jacobian matrix from being of insufficient rank.
[0062] As a tuning parameter, it is usually selected based on experience and experimental data, with the aim of finding a balance between the stability and convergence speed of the algorithm. Specifically, The value of affects the behavior of the algorithm between the Gauss-Newton method (converges quickly but may be unstable) and the gradient descent method (stable but converges slowly). The value of makes the algorithm closer to the Gauss-Newton method, while larger The value will make the algorithm closer to the gradient descent method. In practical applications, adjusting the parameters The initial value of is usually determined based on previous optimization experience of similar systems or through a series of experiments. During the optimization process, according to the change in error after each iteration, It will be adjusted dynamically. If the error decreases, it means that the current The value is valid, reduce it appropriately to speed up the convergence; if the error increases, increase To increase stability.
[0063] Finally, if the error value is less than the preset error threshold, the update is stopped to obtain the optimal control sequence, and the injection molding filling is controlled in real time according to the optimal control sequence. As the injection molding process proceeds, the system will continuously obtain new real-time status data and update the control sequence based on this data to ensure that the process parameters of each stage can be maintained in the optimal state.
[0064] In an optional embodiment, the method further includes: After the injection molding filling process is completed, the optimal control sequence is updated to the initial control sequence for the next same injection molding process.
[0065] Once the injection molding process is completed and the optimal control sequence is obtained, it means that this set of control parameters has been verified as a solution that can achieve the desired effect under the current operating environment and material properties. The optimal control sequence for each injection molding process has been adjusted by the Lm optimization algorithm for many iterations, so it is very close to the optimal state. Using it as the initial control parameters for the next process avoids the optimization process from scratch, which can significantly reduce the number of control sequence updates for future injection molding. In continuous production, the use of optimized control sequences can ensure the consistency of product quality. For the same process, the optimal control sequence can be applied directly, reducing the risk of making drastic adjustments each time, thereby maintaining the stability and accuracy of the product. In addition, if the production conditions change slightly (such as fluctuations in ambient temperature), it is much faster to make fine adjustments based on the optimal control sequence than to re-optimize the entire process. This allows the system to adapt to small changes more quickly. After each injection molding, the update of the optimal control sequence is not just a "reuse" strategy. With more production batches, the system can continuously improve its adaptability to actual production conditions through accumulated optimization experience.
[0066] The optimal control sequence each time is not only used for the same injection molding process, but also provides a reference for different but similar injection molding processes in the future. Therefore, as production continues, the control system will be able to more accurately predict future process changes, thereby improving the adaptability and intelligence level of the entire production system.
[0067] In summary, the present invention provides a control method for an injection molding process, comprising: obtaining state data of injection molding filling; wherein the state data comprises historical state data and real-time state data; based on the historical state data, constructing a state space model by using numerical modeling, and obtaining an initial control sequence according to the state space model; according to the real-time state data and the initial control sequence, updating the initial control sequence in real time by using the Lm optimization algorithm to obtain an optimal control sequence, and controlling the injection molding filling in real time according to the optimal control sequence. In the control process, by constructing a state space model according to the historical state data, describing the dynamic evolution of the system and obtaining an initial control sequence, and updating the initial control sequence in real time by using the Lm optimization algorithm, the dynamic control accuracy and stability of the injection molding filling process are improved.
[0068] Reference Figure 2 The second embodiment of the present invention provides a control system for an injection molding process, comprising: A data acquisition module, for acquiring injection filling status data; wherein the status data includes historical status data and real-time status data; A model building module, based on the historical state data, uses numerical modeling to build a state space model, and obtains an initial control sequence according to the state space model; A sequence updating module, which updates the initial control sequence in real time using an Lm optimization algorithm according to the real-time state data and the initial control sequence to obtain an optimal control sequence; The injection molding control module controls the injection molding filling in real time according to the optimal control sequence.
[0069] It should be noted that the control system for an injection molding process provided in an embodiment of the present invention is used to execute all process steps of a control method for an injection molding process in the above embodiment, and the working principles and beneficial effects of the two correspond one to one, so they will not be repeated.
[0070] The embodiment of the present invention further provides a terminal device. The terminal device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a control program for an injection molding process. When the processor executes the computer program, the steps in the above-mentioned control method embodiments of the injection molding process are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned system embodiments are realized, such as the model building module.
[0071] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the terminal device.
[0072] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of terminal devices and do not constitute a limitation on the terminal device. The terminal device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.
[0073] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.
[0074] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the terminal device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0075] Wherein, if the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or system capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.
[0076] It should be noted that the system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the system embodiment provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.
[0077] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. A method for controlling an injection molding process, characterized in that: include: Get the injection filling status data; Wherein, the status data includes historical status data and real-time status data; Based on the historical state data, a state space model is constructed by using numerical modeling, and an initial control sequence is obtained according to the state space model; According to the real-time state data and the initial control sequence, the initial control sequence is updated in real time using the Lm optimization algorithm to obtain an optimal control sequence, and the injection molding filling is controlled in real time according to the optimal control sequence.
2. The method for controlling the injection molding process according to claim 1, characterized in that: The step of obtaining injection molding filling status data includes: Get the injection temperature, injection speed, injection pressure, mold temperature, holding time and cooling time at different times during the injection filling process.
3. The control method of the injection molding process according to claim 1, characterized in that: The method of constructing a state space model based on the historical state data by using numerical modeling, and obtaining an initial control sequence according to the state space model, includes: Initializing the historical state data, and defining a state vector according to the historical state data; According to the state vector, a state space model is established by using a least square method; Based on a preset objective function, the historical state data is processed using the state space model to obtain an initial control sequence.
4. The method for controlling the injection molding process according to claim 3, characterized in that: Before obtaining the initial control sequence by processing using the state space equation based on the preset output index and the preset objective function, the method further includes: The preset objective function is constructed according to the following formula: in, is the performance indicator value, is the state vector, Control input vector, is the state weight matrix, is the control input weight matrix, For the optimized time range, is the transpose of the state vector, is the transpose of the control input vector.
5. The method for controlling the injection molding process according to claim 3, characterized in that: The method of updating the initial control sequence in real time using the Lm optimization algorithm according to the real-time state data and the initial control sequence to obtain an optimal control sequence, and controlling injection molding filling in real time according to the optimal control sequence, comprises: Initializing the real-time status data; Based on the real-time status data, a preset error function is used to make a judgment: If the error value is greater than a preset error threshold, the initial control sequence is updated using the Lm optimization algorithm; If the error value is less than a preset error threshold, the updating is stopped to obtain the optimal control sequence, and the injection molding filling is controlled in real time according to the optimal control sequence.
6. The method for controlling the injection molding process according to claim 5, characterized in that: If the error value is greater than a preset error threshold, the initial control sequence is updated using the Lm optimization algorithm, including: The initial control sequence is updated according to the following formula: in, is the Jacobian matrix, is the control vector at the current k moment, is the updated control vector, is the error vector, To adjust the parameters, is the identity matrix, which is used to prevent the Jacobian matrix from being of insufficient rank.
7. The method for controlling the injection molding process according to claim 5, characterized in that: The method further comprises: After the injection molding filling process is completed, the optimal control sequence is updated to the initial control sequence for the next same injection molding process.
8. A control system for an injection molding process, characterized in that: include: A data acquisition module, for acquiring injection filling status data; wherein the status data includes historical status data and real-time status data; A model building module, based on the historical state data, uses numerical modeling to build a state space model, and obtains an initial control sequence according to the state space model; A sequence updating module, which updates the initial control sequence in real time using an Lm optimization algorithm according to the real-time state data and the initial control sequence to obtain an optimal control sequence; The injection molding control module controls the injection molding filling in real time according to the optimal control sequence.
9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the method for controlling the injection molding process according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for controlling the injection molding process according to any one of claims 1 to 7.
Citation Information
Cited By
Injection molding control method and system of injection molding machine, storage medium and injection molding machine
CN119458826A
Intelligent injection method based on lightweight gradient elevator and MPC
CN120921653A
Filling control method and system for integrated vacuum hole plugging machine
CN121057114A
An integrated vacuum plug hole machine filling control method and system
CN121057114B
Real-time monitoring method and system for plastic part production line
CN121290728A
