Device and method for online monitoring of wear status of check valve of injection molding machine

Abstract

The present invention discloses a device and method for online monitoring the wear state of a check valve of an injection molding machine. The device comprises a linear displacement sensor, a load cell pressure sensor, an injection molding machine working condition acquisition device, a microprocessor, and a terminal device. The microprocessor is respectively connected to the linear displacement sensor, the load cell pressure sensor, the injection molding machine working condition acquisition device, and the terminal device. The injection molding machine working condition acquisition device is also connected to the injection molding machine. The linear displacement sensor is used to acquire screw displacement information, the load cell pressure sensor is used to acquire screw force information, and the injection molding machine working condition acquisition device is used to acquire injection molding working condition information. The microprocessor is used to determine the wear state of the check valve according to the screw displacement information, the screw force information, and the injection molding working condition information. The terminal device is used to display the wear state of the check valve. The present invention realizes predictive maintenance of the check valve according to the wear state of the check valve, thereby reducing the maintenance cost.

Description

Technical Field

[0001] The invention relates to the technical field of injection molding machine monitoring, and in particular to a device and method for online monitoring the wear state of a check valve of an injection molding machine. Background Art

[0002] At present, most factories use preventive maintenance to maintain injection molding machines, which wastes a lot of time and maintenance costs. With the development of industrial technology, predictive maintenance technology has gradually developed. It can repair and replace parts in time before a failure occurs, preventing the failure from causing equipment downtime. Predictive maintenance technology can gradually develop injection molding machines into fully automatic injection molding machines that can automatically monitor the use status of all key components and predict their remaining service life.

[0003] The check valve is a consumable part in the injection molding process and a key component that determines the quality of injection molded products. Therefore, predictive maintenance of the check valve is required. The screw head of the existing injection molding machine is composed of a dispensing nozzle, a plug rubber ring, and a glue injection meson. The check valve consists of a plug rubber ring and a glue injection meson. It plays the role of sealing and plasticizing backflow in the injection molding process. The quality problems of many plastic products are closely related to the check valve of the injection molding machine. During the injection molding process, the polymer material contains glass fiber, metal powder, silicate, etc., which causes wear and failure of the check valve, resulting in an increase in the scrap rate. The wear parts of the check valve generally occur on the inner and outer diameters of the plug rubber ring and the mating surface with the glue injection meson. Their wear will make the sealing effect worse, and even directly cause the material to flow back during injection, resulting in glue defects in the product. The check valve of the injection molding machine is a consumable. Many companies have not established relevant inventory. Once the check valve fails and cannot be used, it will cause shutdown and production, affecting the production plan of the entire factory later. Therefore, it is necessary to monitor the wear status of the check valve online to achieve predictive maintenance of the check valve. Summary of the invention

[0004] In order to overcome the defects and shortcomings of the prior art, the present invention provides a device for online monitoring the wear state of a check valve of an injection molding machine. The device adopts a load cell pressure sensor and a linear displacement sensor to respectively collect the force process and displacement process of the screw during injection molding. By combining the current process conditions and raw material performance indicators and using machine learning for data mining, a monitoring model for online evaluation of the wear state of the check valve is established, thereby realizing predictive maintenance of the check valve, reducing the maintenance cost of the injection molding machine for the enterprise, and improving the production efficiency of injection molding.

[0005] A second object of the present invention is to provide a method for online monitoring of the wear status of a check valve of an injection molding machine.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A device for online monitoring the wear state of a non-return valve of an injection molding machine, comprising a microprocessor, a linear displacement sensor, a load cell pressure sensor, an injection molding machine working condition collection device, and a terminal device, wherein the microprocessor is respectively connected to the linear displacement sensor, the load cell pressure sensor, the injection molding machine working condition collection device, and the terminal device, and the injection molding machine working condition collection device is also connected to the injection molding machine;

[0008] The linear displacement sensor, the load cell pressure sensor, and the injection molding machine working condition acquisition device are all arranged on the injection molding machine, the linear displacement sensor is used to acquire screw displacement information, the load cell pressure sensor is used to acquire screw force information, and the injection molding machine working condition acquisition device is used to acquire injection molding working condition information;

[0009] The microprocessor is used to form injection molding input data according to screw displacement information, screw force information, and injection molding working condition information, output the degree of wear of the check valve, set different levels of wear state thresholds, and determine the wear state of the check valve according to the degree of wear of the check valve and the different levels of wear state thresholds;

[0010] The terminal device is used to display the wear status of the check valve.

[0011] As a preferred technical solution, the linear displacement sensor is arranged on one side of the end of the screw.

[0012] As a preferred technical solution, the load cell pressure sensor is arranged at the end edge portion of the screw.

[0013] As a preferred technical solution, the injection molding machine working condition acquisition device is arbitrarily arranged on the injection molding machine.

[0014] As a preferred technical solution, the injection molding working condition information specifically includes injection speed, minimum padding, V / P conversion point and raw material process conditions.

[0015] As a preferred technical solution, the raw material process conditions include melt index performance indicators.

[0016] In order to achieve the above second purpose, the present invention adopts the following technical solutions:

[0017] A method for online monitoring of the wear state of a check valve of an injection molding machine comprises the following steps:

[0018] Injection molding data collection steps: collect screw displacement information, screw force information and injection molding working condition information;

[0019] Injection molding data collection and processing step: converting the screw displacement information and the screw force information from analog signals into digital signals, performing fitting processing based on the screw displacement information and the screw force information to form injection molding process data, wherein the injection molding process data includes a load cell pressure curve and a screw displacement curve;

[0020] Injection molding feature set generation step: extracting data features based on injection molding process data and quantizing features in combination with injection molding working condition information to obtain an injection molding feature set, wherein the data features include maximum, mean, covariance, skewness and power consumption of a single injection;

[0021] Check valve wear state prediction step: prediction is performed based on the injection molding feature set combined with the check valve wear state evaluation model trained by machine learning to determine the wear state of the check valve.

[0022] As a preferred technical solution, the training data collection step specifically sets the injection molding machine to collect data under the same temperature, injection speed, and the same material.

[0023] As a preferred technical solution, the step of generating the injection molding feature set specifically includes the following steps:

[0024] Extracting injection molding process data, extracting first data features, wherein the first data features include maximum, mean, covariance, and skewness, forming a first injection molding data feature matrix based on the first data features, and performing feature vectorization on the first injection molding data feature matrix to obtain a first feature vector of a single injection molding sample;

[0025] Combined with the load cell pressure curve and the screw displacement curve, the injection power consumption formula is used to calculate the power consumption of a single injection. The injection power consumption formula is specifically:

[0026] w=∫f loadcell (t)·f' position (t)dt

[0027] Among them, f loadcell (t) represents the load cell pressure curve function, f′ position (t) represents the screw speed, w represents the screw work of a single injection, and t represents the time variable;

[0028] The screw work w of a single injection is added to the first characteristic vector of the injection molding sample to obtain the second characteristic vector of the injection molding sample, and the second characteristic vector of the injection molding sample and the injection molding working condition information are combined to perform characteristic vectorization to form a third characteristic vector of the injection molding sample;

[0029] The third feature vectors of all injection molding samples are integrated into an injection molding feature set.

[0030] As a preferred technical solution, in the check valve wear state prediction step, the check valve wear state assessment model trained by machine learning is established by the following steps:

[0031] Wear state simulation steps: manufacturing check valves with different degrees of wear defects to simulate check valves with different degrees of wear states, and forming notches of different depths between the plug rubber ring and the injection meson to simulate check valves with different wear states;

[0032] Training data collection steps: Collect screw displacement information, screw force information and injection molding working condition information under different wear conditions of the check valve;

[0033] Training data collection and processing steps: converting the screw displacement information and the screw force information from analog signals into digital signals, performing fitting processing based on the screw displacement information and the screw force information to form injection molding process data, wherein the injection molding process data includes a load cell pressure curve and a screw displacement curve;

[0034] Training set generation steps: extract data features based on injection molding process data and perform feature vectorization based on injection molding working condition information to obtain an injection molding feature set, and use the wear degree of the check valve as a training label to form an injection molding training set;

[0035] Training steps: Use machine learning algorithm to establish a check valve wear state assessment model. When the preset training times are reached and the test accuracy value of the check valve wear degree reaches the preset test accuracy threshold, the training ends, and the trained check valve wear state assessment model is obtained.

[0036] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0037] (1) The present invention installs a load cell pressure sensor and a linear displacement sensor at the rear of the screw and the moving end of the screw, respectively, to realize online monitoring of the wear status of the check valve without damaging the structure of the injection molding machine; extracts data features of process data and combines them with working condition data, and establishes a wear status evaluation model of the check valve based on a machine learning algorithm. The evaluation model can well adapt to the injection molding industry with various working condition characteristics and has a certain generalization effect. The wear status of the check valve can be obtained by monitoring, which helps factories and enterprises avoid unnecessary regular inspection and maintenance and reduce a lot of maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 This is a connection diagram of a device for online monitoring of the wear state of a check valve of an injection molding machine in Example 1 of the present invention;

[0039] Figure 2 Schematic diagram of the structure of the device for online monitoring the wear state of the check valve of the injection molding machine in Example 1 of the present invention;

[0040] Figure 3 This is a flow chart of the method for online monitoring of the wear status of the check valve of an injection molding machine in Example 2 of the present invention.

[0041] Among them, 1-injection molding machine, 2-linear displacement sensor, 3-load cell pressure sensor, 4-microprocessor, 5-terminal equipment. DETAILED DESCRIPTION

[0042] In the description of the present disclosure, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present disclosure and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present disclosure.

[0043] In addition, the terms "first", "second", and "third" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance. Similarly, words such as "one", "an", or "the" do not indicate a quantitative limitation, but rather indicate the presence of at least one. Words such as "include" or "comprise" and the like mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connected" or "connected" and the like are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0044] In the description of the present disclosure, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be internal communication between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present disclosure can be understood according to specific circumstances. In addition, the technical features involved in the different embodiments of the present disclosure described below can be combined with each other as long as they do not conflict with each other.

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0046] Example

[0047] Example 1

[0048] like Figure 1 and Figure 2As shown, this embodiment provides a device for online monitoring the wear state of the check valve of an injection molding machine, the device includes a linear displacement sensor 2, a load cell pressure sensor 3, an injection molding machine working condition collection device, a microprocessor 4, and a terminal device 5. The microprocessor 4 is respectively connected to the linear displacement sensor 2, the load cell pressure sensor 3, the injection molding machine working condition collection device, and the terminal device 5, and the injection molding machine working condition collection device is also connected to the injection molding machine 1. The injection molding machine 1 is provided with a screw.

[0049] Combination Figure 1 As shown, the injection molding machine working condition acquisition device serves as a lower computer of the injection molding machine 1 , and the linear displacement sensor 2 , the load cell pressure sensor 3 , and the injection molding machine working condition acquisition device are all arranged in the injection molding machine 1 .

[0050] Combination Figure 1 As shown, the linear displacement sensor 2 is arranged on one side of the end of the screw. Specifically, one end of the linear displacement sensor 2 is installed on the fixed bracket of the injection molding machine shooting platform, and the other end is installed on the movable plasticizing screw support frame. The linear displacement sensor 2 is used to collect the displacement information of the screw. The load cell pressure sensor 3 is arranged on the end edge of the screw, and the load cell pressure sensor 3 is used to collect the force information of the screw. The injection molding machine working condition collection device is used to collect the injection molding working condition information. The injection molding machine working condition collection device can be arbitrarily set on the injection molding machine, wherein the injection molding working condition information specifically includes the injection speed, the minimum padding, the V / P conversion point and the raw material process conditions, and the raw material process conditions include the melt index and other performance indicators.

[0051] In this embodiment, the microprocessor 4 is used to convert the screw displacement information and the screw force information from analog signals into digital signals, fit the screw displacement information and the screw force information to form injection molding process data, combine the injection molding process data and the injection molding working condition information to form injection molding input data, output the degree of wear of the check valve, set different levels of wear state thresholds, and determine the wear state of the check valve according to the degree of wear of the check valve and the different levels of wear state thresholds.

[0052] In actual application, the microprocessor 4 extracts data features based on the injection molding process data and obtains an injection molding feature set in combination with the injection molding working condition information. The injection molding feature set is used as the injection molding input data, and the check valve wear state evaluation model is used to predict and output the check valve wear degree. The microprocessor 4 sets different levels of wear state thresholds, and determines whether the check valve of the current injection molding machine is worn according to the different levels of wear state thresholds, and can accurately estimate the level of wear of the current check valve, that is, determine the wear state of the check valve, and arrange inspection and maintenance in time. Among them, the data features include the maximum value, mean, covariance and skewness, and single injection power consumption.

[0053] In this embodiment, the terminal device 5 is used to display the wear status of the check valve. The terminal device 5 can be a desktop computer, a laptop computer, a smart phone, a PDA handheld terminal, a tablet computer or other terminals with a display function.

[0054] Example 2

[0055] like Figure 3 As shown, this embodiment provides a method for online monitoring of the wear state of a check valve of an injection molding machine, the method comprising the following steps:

[0056] Injection molding data collection steps: collect screw displacement information, screw force information and injection molding working condition information. In actual application, the screw displacement information is obtained through the linear displacement sensor 2, the screw force information is obtained through the load cell pressure sensor 3, and the injection molding working condition information of the current injection molding machine is obtained through the injection molding machine working condition collection device;

[0057] Injection molding data collection and processing steps: convert the screw displacement information and the screw force information from analog signals into digital signals, and perform fitting processing based on the screw displacement information and the screw force information to form injection molding process data, wherein the injection molding process data includes a load cell pressure curve and a screw displacement curve.

[0058] Injection molding feature set generation steps: extracting data features based on injection molding process data and quantizing features in combination with injection molding working condition information to obtain an injection molding feature set; wherein the data features include maximum value, mean, covariance and skewness and single injection power consumption.

[0059] Check valve wear state prediction step: prediction is performed based on the injection molding feature set combined with the check valve wear state evaluation model to determine the wear state of the check valve.

[0060] This embodiment uses a method for online monitoring of the wear state of the check valve of an injection molding machine to perform online monitoring of the wear state of the check valve to obtain the wear state of the check valve, and then performs predictive maintenance based on the wear state of the check valve.

[0061] In this embodiment, the step of generating the injection molding feature set specifically includes the following steps:

[0062] Extracting injection molding process data, extracting first data features, the first data features including maximum, mean, covariance, and skewness, forming a first injection molding data feature matrix based on the first data features, and performing feature vectorization on the first injection molding data feature matrix to obtain a first feature vector of a single injection molding sample;

[0063] Combined with the load cell pressure curve and screw displacement curve, the injection power consumption formula is used to calculate the power consumption of a single injection.

[0064] The specific injection molding power consumption formula is:

[0065] w=∫floadcell (t)·f' position (t)dt

[0066] Among them, f loadcell (t) represents the load cell pressure curve function, f′ position (t) represents the screw speed, w represents the screw work of a single injection, and t represents the time variable;

[0067] The screw work w of a single injection is added to the first characteristic vector of the injection molding sample to obtain the second characteristic vector of the injection molding sample, and the second characteristic vector of the injection molding sample and the injection molding working condition information are combined to perform characteristic vectorization to form a third characteristic vector of the injection molding sample;

[0068] The third feature vectors of all injection molding samples are integrated into an injection molding feature set.

[0069] This embodiment also provides a method for establishing a check valve wear state assessment model, the method comprising the following steps:

[0070] Wear state simulation steps: manufacturing check valves with different degrees of wear defects to simulate check valves with different degrees of wear states, and forming notches of different depths between the plug rubber ring and the injection meson to simulate check valves with different wear states;

[0071] Training data collection steps: Collect screw displacement information, screw force information and injection molding condition information under different wear conditions of the check valve.

[0072] Training data collection and processing steps: convert the screw displacement information and the screw force information from analog signals into digital signals, and perform fitting processing based on the screw displacement information and the screw force information to form injection molding process data, wherein the injection molding process data includes a load cell pressure curve and a screw displacement curve.

[0073] Training set generation steps: extract data features based on injection molding process data and perform feature vectorization based on injection molding working condition information to obtain an injection molding feature set, and use the wear degree of the check valve as a training label to form an injection molding training set;

[0074] Training steps: Use machine learning algorithm to establish a check valve wear state assessment model. When the preset training times are reached and the test accuracy value of the check valve wear degree reaches the preset test accuracy threshold, the training ends, and the trained check valve wear state assessment model is obtained.

[0075] In this implementation, the training data collection step specifically sets the injection molding machine to collect data of all samples under the same temperature, injection speed, same material and other process conditions, thereby obtaining the load cell pressure curve, screw displacement curve, and injection molding condition information. In actual application, the material used is the same grade of high-density polyethylene HDPE.

[0076] In this embodiment, the training set generation step specifically includes the following steps:

[0077] Extract data features for each group of injection molding process data. The data features include maximum, mean, covariance and skewness. Let j represent the variable and J represent the number of variables.

[0078] The maximum value is extracted using the first extraction formula:

[0079] σ=[σ j1 , σ j2 ]∈R J×2 ;

[0080] σ j1 =max x j (k), k=1, 2, ..., K;

[0081] σ j2 =min x j (k), k=1, 2, ..., K;

[0082] σ represents the maximum value of the injection molding process data;

[0083] The mean is extracted using the second extraction formula:

[0084] μ∈R J×1 ;

[0085]

[0086] μ represents the mean of the injection molding process data;

[0087] The covariance is extracted using the third extraction formula:

[0088] Σ∈R J×J ;

[0089]

[0090] Σ represents the covariance of the injection molding process data;

[0091] The skewness is extracted using the fourth extraction formula:

[0092] γ∈R J×1 ;

[0093]

[0094] γ represents the skewness of the injection molding process data;

[0095] Form the injection molding data feature matrix based on the data features:

[0096]

[0097] The injection molding data feature matrix is ​​feature vectorized. Flat Finally, the feature vector of the injection molding sample is formed:

[0098]

[0099] Combine the load cell pressure curve, screw displacement curve, injection molding condition information and use the injection molding power consumption formula to calculate the power consumption of a single injection:

[0100] w=∫f loadcell (t)·f' position (t)dt

[0101] Among them, f loadcell (t) represents the load cell pressure curve function, f′ position (t) represents the derivative of the screw displacement curve function, that is, the screw speed, and w represents the work done by the screw in a single injection;

[0102] Add the screw work w of a single injection to the feature vector of the injection molding sample In combination with the injection molding condition information, the feature vector P of each injection molding sample is formed. 1×m , the feature vectors of all injection molding samples are integrated into an injection molding feature set Add the label of injection molding feature set ω n×1 , where the label ω of the injection molding feature set is n×1 Specifically, it represents different levels of wear states, where n and m are positive integers;

[0103] Use machine learning algorithms to perform data mining on the injection molding feature set to form the injection molding training set D n×(m+1) .

[0104] In this embodiment, the machine learning algorithm preferably uses SVR, i.e., support vector regression algorithm. SVR uses ε-insensitive function and kernel function algorithm to construct a linear decision function in high-dimensional space to realize linear regression. In order to adapt to the nonlinearity of the training sample set, SVR uses kernel function to solve the risk of overfitting when the adjustable parameters are increased in the traditional fitting method. Replacing the linear terms in the linear equation with kernel function can make the original linear algorithm nonlinear, that is, it can perform nonlinear regression. At the same time, the introduction of kernel function can not only achieve the purpose of dimensionality increase, but also control overfitting when the adjustable parameters are increased.

[0105] In practical applications, the trained check valve wear state assessment model is used to monitor the check valve wear state online, so as to determine whether the check valve of the current injection molding machine is worn and accurately estimate the level of wear of the current check valve.

[0106] In this embodiment, the wear state of the check valve is determined by setting different levels of wear state thresholds, outputting the degree of wear of the check valve by a trained check valve wear state assessment model, and determining the wear state of the check valve according to the degree of wear of the check valve and different levels of wear state thresholds.

[0107] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for online monitoring of the wear status of a check valve of an injection molding machine, characterized in that: A device for online monitoring of the wear state of the non-return valve of the injection molding machine is provided, which is provided with a microprocessor, and also includes a linear displacement sensor, a load cell pressure sensor, an injection molding machine working condition collection device, and a terminal device. The microprocessor is respectively connected to the linear displacement sensor, the load cell pressure sensor, the injection molding machine working condition collection device, and the terminal device, and the injection molding machine working condition collection device is also connected to the injection molding machine; The linear displacement sensor, the load cell pressure sensor, and the injection molding machine working condition acquisition device are all arranged on the injection molding machine, the linear displacement sensor is used to acquire screw displacement information, the load cell pressure sensor is used to acquire screw force information, and the injection molding machine working condition acquisition device is used to acquire injection molding working condition information; The microprocessor is used to output the degree of wear of the check valve, set different levels of wear state thresholds, and determine the wear state of the check valve according to the degree of wear of the check valve and the different levels of wear state thresholds; The terminal device is used to display the wear status of the check valve; The following steps are involved: Injection molding data collection steps: collect screw displacement information, screw force information and injection molding working condition information; Injection molding data collection and processing step: converting the screw displacement information and the screw force information from analog signals into digital signals, performing fitting processing based on the screw displacement information and the screw force information to form injection molding process data, wherein the injection molding process data includes a load cell pressure curve and a screw displacement curve; Injection molding feature set generation step: extracting data features based on injection molding process data and quantizing features in combination with injection molding working condition information to obtain an injection molding feature set, wherein the data features include maximum, mean, covariance, skewness and power consumption of a single injection; The step of generating the injection molding feature set specifically comprises the following steps: Extracting injection molding process data, extracting first data features, wherein the first data features include maximum, mean, covariance, and skewness, forming a first injection molding data feature matrix based on the first data features, and performing feature vectorization on the first injection molding data feature matrix to obtain a first feature vector of a single injection molding sample; Combined with the load cell pressure curve and the screw displacement curve, the injection power consumption formula is used to calculate the power consumption of a single injection. The injection power consumption formula is specifically: w=∫f loadcell (t)·f’ position (t)dt Among them, f loadcell (t) represents the load cell pressure curve function, f' position (t) represents the screw speed, w represents the screw work of a single injection, and t represents the time variable; The screw work w of a single injection is added to the first characteristic vector of the injection molding sample to obtain the second characteristic vector of the injection molding sample, and the second characteristic vector of the injection molding sample and the injection molding working condition information are combined to perform characteristic vectorization to form a third characteristic vector of the injection molding sample; The third feature vectors of all injection molding samples are integrated into an injection molding feature set; Check valve wear state prediction step: predicting the check valve wear state evaluation model based on the injection molding feature set combined with machine learning training to determine the check valve wear state; In the check valve wear state prediction step, the check valve wear state assessment model trained by machine learning is established by the following steps: Wear state simulation steps: manufacturing check valves with different degrees of wear defects to simulate check valves with different degrees of wear states, and forming notches of different depths between the plug rubber ring and the injection meson to simulate check valves with different wear states; Training data collection steps: Collect screw displacement information, screw force information and injection molding working condition information under different wear conditions of the check valve; Training data collection and processing steps: converting the screw displacement information and the screw force information from analog signals into digital signals, performing fitting processing based on the screw displacement information and the screw force information to form injection molding process data, wherein the injection molding process data includes a load cell pressure curve and a screw displacement curve; Training set generation steps: extract data features based on injection molding process data and perform feature vectorization based on injection molding working condition information to obtain an injection molding feature set, and use the wear degree of the check valve as a training label to form an injection molding training set; Training steps: Use machine learning algorithm to establish a check valve wear state assessment model. When the preset training times are reached and the test accuracy value of the check valve wear degree reaches the preset test accuracy threshold, the training ends, and the trained check valve wear state assessment model is obtained.

2. The method for online monitoring of the wear state of the check valve of an injection molding machine according to claim 1 is characterized in that: The linear displacement sensor is arranged at one side of the end of the screw rod.

3. The method for online monitoring of the wear state of the check valve of an injection molding machine according to claim 1, characterized in that: The load cell pressure sensor is arranged at the end edge portion of the screw rod.

4. The method for online monitoring of the wear state of the check valve of an injection molding machine according to claim 1, characterized in that: The injection molding machine working condition acquisition device is arbitrarily arranged on the injection molding machine.

5. The method for online monitoring of the wear state of the check valve of an injection molding machine according to claim 1, characterized in that: The injection molding working condition information specifically includes injection speed, minimum padding, V / P conversion point and raw material process conditions.

6. The method for online monitoring of the wear state of the check valve of an injection molding machine according to claim 5, characterized in that: The raw material process conditions include melt index performance indicators.

7. The method for online monitoring of the wear state of the check valve of an injection molding machine according to claim 1, characterized in that: The training data collection step specifically sets the injection molding machine to collect data under the same temperature, injection speed, and the same material.

Images