Energy saving control method of injection machine

Abstract

An energy-saving control method for an injection molding machine includes: training an energy consumption prediction model based on the historical production parameters of the injection molding machine and the historical production energy consumption corresponding to the historical production parameters; training a weight difference prediction model based on the historical production parameters and the historical product weight difference corresponding to the historical production parameters; executing an optimization algorithm based on an objective function to try multiple sets of production parameters, inputting each set of production parameters into the energy consumption prediction model and the weight difference prediction model to obtain predicted production energy consumption and predicted product weight difference, and finding a set of optimal production parameters for optimizing the objective function.

Description

Technical Field

[0001] This invention relates to an energy-saving control method, and more particularly to an energy-saving control method for an injection molding machine. Prior Technology

[0002] Traditional injection molding machines typically use fixed-frequency motors, which operate at full speed throughout their lifespan. They cannot be properly adjusted according to actual needs or the current operating status of the injection molding machine, resulting in the same energy consumption even when operating under lower loads, thus wasting energy.

[0003] Currently, there are also proposals to control the injection molding machine motor using frequency conversion. However, most of these methods rely on sensor values ​​to control the frequency conversion output. This method has a slow response time and cannot issue control commands in real time, which affects the quality of the injection molded products. In addition, this method generates large instantaneous voltage changes, resulting in high energy consumption, and is prone to impact and wear, thus reducing the equipment's lifespan. Summary of the Invention

[0004] The purpose of this invention is to propose an energy-saving control method for injection molding machines, comprising: training an energy consumption prediction model based on the historical production parameters of the injection molding machine and the historical production energy consumption corresponding to the historical production parameters; training a weight difference prediction model based on the historical production parameters and the historical product weight difference corresponding to the historical production parameters; executing an optimization algorithm based on an objective function, thereby trying multiple sets of production parameters, inputting each set of production parameters into the energy consumption prediction model and the weight difference prediction model to obtain predicted production energy consumption and predicted product weight difference, and finding a set of optimal production parameters for optimizing the objective function.

[0005] In some embodiments, the objective functions are minimizing the predicted production energy consumption and minimizing the predicted product weight difference, and the set of optimal production parameters minimizes the predicted production energy consumption and the predicted product weight difference.

[0006] In some embodiments, the historical product weight difference is the difference between the actual weight of the injected finished product and the preset weight.

[0007] In some embodiments, the historical production parameters include injection speed, injection pressure, holding pressure, holding speed, and inverter frequency.

[0008] In some embodiments, the historical production parameters further include pipe temperature, screw pressure, cooling time, and process stage, where the process stage is mold locking, pressure holding, feeding, cooling, or mold opening.

[0009] In some embodiments, the energy-saving control method for the injection molding machine further includes: controlling the injection molding machine according to the injection speed, injection pressure, holding pressure, holding speed, and process stage included in the set of optimal production parameters, and smoothly adjusting the frequency of the frequency converter in a variable frequency manner according to the process stage and frequency converter frequency included in the set of optimal production parameters.

[0010] In some embodiments, the frequency converter is coupled to the injection molding machine motor, and the injection molding machine motor is coupled to the injection molding machine, wherein the injection molding machine motor is a hydraulic pump.

[0011] In some embodiments, after each set of production parameters is input into the weight difference prediction model, a quality compliance judgment is performed to determine whether the predicted product weight difference is compliant, and the weight difference prediction model outputs the predicted product weight difference only when it is compliant.

[0012] In some embodiments, the optimization algorithm is a gene algorithm or a particle swarm optimization algorithm.

[0013] In some embodiments, each of the energy consumption prediction model and the weight difference prediction model is a support vector machine (SVM), a random forest, or a neural network.

[0014] To make the above features and advantages of the present invention more apparent and understandable, specific embodiments are described below in conjunction with the accompanying drawings for detailed explanation. Simple Explanation of the Diagram

[0015] A better understanding of the invention can be obtained from the following detailed description taken in conjunction with the accompanying drawings. It should be noted that, according to industry standard practice, the features are not drawn to scale. In fact, the dimensions of the features can be arbitrarily increased or decreased for clarity of discussion. [Figure 1] is a block diagram of an energy-saving control system for an injection molding machine according to an embodiment of the present invention. [Figure 2] is a flowchart of an energy-saving control method for an injection molding machine according to an embodiment of the present invention. [Figure 3] is used to illustrate the difference between the change in the injection molding machine motor speed in the embodiment of the present invention and the conventional practice. Implementation

[0016] The embodiments of the present invention will be discussed in detail below. However, it will be understood that the embodiments provide many applicable concepts that can be implemented in a wide variety of specific contexts. The discussed and disclosed embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0017] Figure 1 is a block diagram of an energy-saving control system for an injection molding machine according to an embodiment of the present invention. An injection molding machine motor 110 is coupled to an injection molding machine 120, and a frequency converter 130 is coupled to the injection molding machine motor 110. In some embodiments, the injection molding machine motor 110 is a hydraulic pump used to output hydraulic kinetic energy to drive the injection molding machine 120, while the frequency converter 130 outputs an AC voltage signal to drive the injection molding machine motor 110 and adjusts the speed of the injection molding machine motor 110 by changing its frequency.

[0018] The inverter 130 reduces energy consumption. For example, when the injection molding machine 120 is in a cooling state or under low load, the inverter 130 reduces the speed of the injection molding machine motor 110, unlike a fixed-frequency motor which continues to run at full speed under low load or cooling conditions, resulting in significant energy consumption. Furthermore, the oil temperature is less likely to rise due to the reduced speed of the injection molding machine motor 110, thus extending the life of the hydraulic cylinder oil seals and increasing the stability of the hydraulic pressure. This invention proposes an energy-saving control method for injection molding machines, which can better control the speed of the injection molding machine motor 110 and multiple production parameters of the injection molding machine 120 to achieve optimal energy-saving effects.

[0019] The energy-saving control method for the injection molding machine proposed in this invention is implemented in the frequency converter energy-saving box 140. The frequency converter energy-saving box 140 is an electronic device, which includes a processor and memory. The electronic device can be various forms of control computer. The processor can be a central processing unit, microprocessor, microcontroller, digital signal processor, special application integrated circuit, etc. The memory can be volatile memory or non-volatile memory, which stores multiple instructions. The processor executes these instructions to complete the energy-saving control method for the injection molding machine.

[0020] The frequency converter energy-saving box 140 is communicatively connected to the injection molding machine 120 to obtain production parameters from the injection molding machine 120 or to transmit commands to the injection molding machine 120 to adjust its production parameters. In some embodiments, the frequency converter energy-saving box 140 communicates with the programmable logic controller (PLC) of the injection molding machine 120 through a serial communication interface. The frequency converter energy-saving box 140 is communicatively connected to the frequency converter 130 to obtain the frequency converter frequency from the frequency converter 130 or to transmit commands to the frequency converter 130 to adjust its frequency converter frequency. In some embodiments, the frequency converter energy-saving box 140 communicates with the frequency converter 130 through an RS-485 communication interface. The electricity meter 150 is coupled to the injection molding machine motor 110 to measure production energy consumption (the unit can be kilovolt-amperes (kVA) or kilowatts (kW)), and the frequency converter energy-saving box 140 is communicatively connected to the electricity meter 150 to obtain the production energy consumption from the electricity meter 150.

[0021] Figure 2 is a flowchart of an energy-saving control method for an injection molding machine according to an embodiment of the present invention. In step S1, historical production parameters of the injection molding machine 120 and the corresponding historical production energy consumption and historical product weight difference are obtained. The historical production parameters are relevant production parameters used by the injection molding machine 120 in past production processes. In some embodiments, the historical production parameters include the injection speed, injection pressure, holding pressure and holding speed of the injection molding machine 120, and the inverter frequency of the inverter 130. In some embodiments, the historical production parameters further include the barrel temperature, screw pressure, cooling time, and process stage of the injection molding machine 120, wherein the process stage of the injection molding machine 120 is mold clamping, holding pressure, feeding, cooling, or mold opening. In some embodiments, some preprocessing can also be performed on the historical production parameters, including data cleaning (e.g., excluding unreasonable data), outlier filtering, and normalization, etc., but the present invention does not limit the content of these preprocessing. The historical production energy consumption is obtained by the inverter energy-saving box 140 from the electricity meter 150. Historical product weight difference is the difference between the actual weight of the injected finished product and the preset weight.

[0022] In step S2, an energy consumption prediction model is trained based on historical production parameters and corresponding historical energy consumption. The energy consumption prediction model employs machine learning regression algorithms, such as support vector machines, random forests, or neural networks. The purpose of the energy consumption prediction model is to predict production energy consumption based on production parameters. In other words, step S2 is the training phase of the energy consumption prediction model; historical production parameters serve as the input, and historical production energy consumption serves as the output. In the subsequent prediction phase, step S4, the production parameters are input into the trained energy consumption prediction model, and the output of the model is the production energy consumption.

[0023] In step S3, a weight difference prediction model is trained based on historical production parameters and the corresponding historical product weight differences. The weight difference prediction model employs machine learning regression algorithms, such as support vector machines, random forests, or neural networks. The purpose of the weight difference prediction model is to predict product weight differences based on production parameters. In other words, step S3 is the training phase of the weight difference prediction model; historical production parameters serve as the input, and historical product weight differences are the output. In the subsequent prediction phase, step S4, the production parameters are input into the trained weight difference prediction model, and the output of the model is the product weight difference. In some embodiments, the weight difference prediction model may also first perform a quality compliance check to determine whether the historical product weight differences are compliant (e.g., whether the historical product weight differences are below a threshold to ensure the actual weight of the injected finished product is compliant), and only if compliant will the weight difference prediction model be trained accordingly.

[0024] In step S4, an optimization algorithm is executed based on the objective function to try multiple sets of production parameters. Each set of production parameters is input into the energy consumption prediction model and the weight difference prediction model to obtain the predicted production energy consumption and the predicted product weight difference, thereby finding the optimal set of production parameters for optimizing the objective function. Specifically, the production parameters can be adjusted to input different combinations of production parameters into the trained energy consumption prediction model and the weight difference prediction model to obtain the corresponding predicted production energy consumption and predicted product weight difference.

[0025] Step S4 uses minimizing the difference between predicted production energy consumption and predicted product weight as the objective function. In other words, step S4 searches for the minimum value of this objective function. In some embodiments, the optimization algorithm can employ a bio-inspired algorithm, such as a genetic algorithm or a particle swarm optimization (PSO) algorithm, to search for production parameters, thereby finding the optimal set of production parameters that minimizes the difference between predicted production energy consumption and predicted product weight. In other words, the optimization algorithm searches for the set of production parameters that minimizes the aforementioned objective function. Those skilled in the art will understand how bio-inspired algorithms can be used given an objective function, and will not be elaborated upon here.

[0026] In some embodiments, after each set of production parameters is input into the weight difference prediction model, a quality compliance judgment is performed to determine whether the predicted product weight difference is compliant (e.g., whether the predicted product weight difference is below a threshold), and the weight difference prediction model outputs the predicted product weight difference only when it is compliant.

[0027] In step S5, the injection machine is controlled according to the injection speed, injection pressure, holding pressure, holding speed and process stage included in the set of optimal production parameters, and the frequency of the inverter is smoothly adjusted in a variable frequency manner according to the process stage and inverter frequency included in the set of optimal production parameters.

[0028] Figure 3 illustrates the difference between the variation of the injection molding machine motor speed in an embodiment of the present invention and conventional methods. Line segment L1 in Figure 3 represents a conventional injection molding machine using a fixed-frequency motor, where the motor operates at full speed (100% speed) throughout its operation, resulting in significant energy waste. Line segment L2 in Figure 3 represents a conventional method using a single frequency converter to control the injection molding machine motor. This method employs different motor speeds depending on the different stages of the injection molding process. However, it generates instantaneous voltage changes at the boundary between two process stages, resulting in higher energy consumption and potentially causing shocks and wear, thus reducing equipment lifespan. Furthermore, while a single frequency converter can save energy, improper frequency conversion can lead to poor injection molding quality. In Figure 3, line segment L3 represents the injection molding machine motor controlled by the energy-saving control method of the present invention. The frequency of the inverter is smoothly adjusted in a variable frequency manner. In other words, the present invention controls the change of the injection molding machine motor speed in advance according to the process stage of the injection molding machine, so that the change of the injection molding machine motor speed is smooth and avoids instantaneous voltage changes. This not only achieves the effect of energy saving, but also reduces impact and wear to increase the life of the equipment.

[0029] In summary, this invention proposes an energy-saving control method for injection molding machines. This method can issue control commands in real-time based on optimal production parameters and smoothly adjust the speed of the injection molding machine motor using frequency conversion, thereby controlling the injection molding machine for production in a fast-response and energy-saving manner. Specifically, this invention uses machine learning combined with optimization algorithms to adaptively control the speed of the injection molding machine motor, ensuring injection quality while achieving energy saving and carbon reduction.

[0030] The foregoing has outlined the features of several embodiments, thus enabling those skilled in the art to better understand the nature of the invention. Those skilled in the art should recognize that the invention can be readily used as a basis to design or modify other processes and structures, thereby achieving the same objectives and / or advantages as the embodiments described herein. Those skilled in the art should also understand that these equivalent constructions do not depart from the spirit and scope of the invention, and that various changes, substitutions, and modifications can be made without departing from the spirit and scope of the invention.

[0031] 110: Injection molding machine motor 120: Injection Molding Machine 130: Frequency converter 140: Variable Frequency Energy Saving Box 150: Electricity meter L1, L2, L3: Line segments S1, S2, S3, S4, S5: Steps

Claims

1. An energy-saving control method for an injection molding machine, executed by a control computer, comprising: An energy consumption prediction model is trained based on historical production parameters of an injection molding machine and historical production energy consumption corresponding to those parameters. A historical product weight difference is determined to be compliant based on those parameters. Only when the historical product weight difference is deemed compliant is a weight difference prediction model trained based on those parameters and the historical product weight difference, where the historical product weight difference is the difference between the actual weight of the injected product and a preset weight. An optimization algorithm is executed based on an objective function, thereby trying multiple sets of production parameters. Each of these multiple sets of production parameters is input into the energy consumption prediction model and the weight difference prediction model to obtain a predicted production energy consumption and a predicted product weight difference, and based on this, a set of optimal production parameters for optimizing the objective function is found. The method for determining whether the historical product weight difference is compliant is: whether the historical product weight difference is below a threshold; if the historical product weight difference is below the threshold, the historical product weight difference is deemed compliant.

2. An energy-saving control method for an injection molding machine as described in claim 1, wherein the objective function is to minimize the predicted production energy consumption and the predicted product weight difference, wherein the set of optimal production parameters minimizes the predicted production energy consumption and the predicted product weight difference.

3. An energy-saving control method for an injection molding machine as described in claim 1, wherein the historical production parameters include injection speed, injection pressure, holding pressure, holding speed, and inverter frequency.

4. An energy-saving control method for an injection molding machine as described in claim 3, wherein the historical production parameters further include barrel temperature, screw pressure, cooling time, and process stage, wherein the process stage is mold clamping, pressure holding, feeding, cooling, or mold opening.

5. The energy-saving control method for the injection molding machine as described in claim 1 further includes: The injection machine is controlled based on the injection speed, injection pressure, holding pressure, holding speed, and process stage included in the set of optimal production parameters, and the frequency of a frequency converter is smoothly adjusted in a variable frequency manner based on the process stage and frequency converter frequency included in the set of optimal production parameters.

6. The energy-saving control method for an injection molding machine as described in claim 5, wherein the frequency converter is coupled to an injection molding machine motor, and the injection molding machine motor is coupled to the injection molding machine, wherein the injection molding machine motor is a hydraulic pump.

7. The energy-saving control method for the injection molding machine as described in claim 1, wherein after each of the multiple sets of production parameters is input into the weight difference prediction model, a quality acceptance judgment is performed to determine whether the predicted product weight difference is compliant, and the weight difference prediction model outputs the predicted product weight difference only when it is compliant.

8. An energy-saving control method for an injection machine as described in claim 1, wherein the optimization algorithm is a genetic algorithm or a particle swarm optimization algorithm.

9. An energy-saving control method for an injection machine as described in claim 1, wherein each of the energy consumption prediction model and the weight difference prediction model is a support vector machine (SVM), a random forest, or a neural network.

Images