Fault detection method and system for plastic pipe extruding machine

Abstract

The invention relates to the technical field of data processing, in particular to a fault detection method and system for a plastic pipe extruding machine. The method comprises the following steps: acquiring the temperature, pressure and extrusion rate of an extrusion opening of an extruding machine according to a preset acquisition frequency so as to obtain a temperature sequence, a pressure sequence and an extrusion rate sequence; the variance of the temperature sequence is calculated, and if the variance is larger than a preset variance threshold value, it is preliminarily judged that noise data exist in the temperature sequence; obtaining suspected noise data; obtaining a first Pearson's correlation coefficient; and obtaining a fault prediction result of the plastic extruding machine. By adopting the method, the accuracy of the monitoring result of the plastic extruding machine can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing. More specifically, the present invention relates to a fault detection method and system for a plastic pipe extrusion machine. Background Art

[0002] With the development of technology, the application of plastic pipes in life and work has become more and more popular. Especially in the fields of construction engineering, agriculture, industry, and medical care, plastic pipes play an irreplaceable and important role. In the water supply system, plastic pipes are commonly used in the tap water supply system of buildings, can operate stably for a long time, and will not be corroded by chemical substances in water; plastic pipes in the drainage system include drainage, exhaust, and sewage sanitary pipes, etc., can quickly and effectively discharge wastewater and waste gas from buildings, and will not be eroded by chemical substances in wastewater; plastic pipes can also be used as conduit pipes for wire installation, protecting wires and cables from damage.

[0003] Plastic pipes are generally tubular materials made of synthetic resin (polyester) as the main raw material, added with additives such as stabilizers, lubricants, and plasticizers, and processed by a plastic pipe extrusion machine. In order to ensure the quality of the produced plastic pipes, it is necessary to ensure that the extrusion machine is in a normal operating state when producing plastic pipes. Therefore, it is crucial to monitor and predict the operating state of the extrusion machine so as to stop the machine for maintenance in time when a fault occurs.

[0004] With the popularization of deep learning, a trained neural network model can be used in combination with the operating parameters of the extrusion machine collected to monitor the extrusion machine to predict the fault state of the extrusion machine. For example, Figure 1 The method for monitoring the extrusion machine using a neural network shown is as follows: First, use sensors to collect the physical parameters at the extrusion port of the extrusion machine, and then input the collected physical parameters into the neural network model, so as to obtain a fault prediction result and display it on the display interface; however, since the sensors used to collect the operating parameters of the extrusion machine will be interfered and generate noise, the collected operating parameters of the extrusion machine usually contain noise data, resulting in poor accuracy of the monitoring results of the extrusion machine. Summary of the Invention

[0005] To solve the technical problem of poor accuracy of the monitoring results when monitoring the extrusion machine in the prior art, the present invention provides solutions in the following aspects.

[0006] In the first aspect, the present invention provides a fault detection method for a plastic pipe extrusion machine, including: collecting the temperature, pressure, and extrusion rate at the extrusion port of the extrusion machine according to a preset collection frequency, so as to obtain a temperature sequence, a pressure sequence, and an extrusion rate sequence; calculating the variance of the temperature sequence, and if the variance is greater than a preset variance threshold, initially determining that there is noise data in the temperature sequence; Fit the temperature data in the temperature sequence to obtain the first fitting curve, and calculate the residuals corresponding to each temperature data; determine the temperature data with residuals greater than the preset residual threshold as suspected noise data; Fit the pressure data corresponding to the temperature data in the neighborhood range of the suspected noise data to obtain the second fitting curve, and calculate the first Pearson correlation coefficient between the temperature data points in the neighborhood range of the suspected noise data and the pressure data points on the second fitting curve; If the value of the first Pearson correlation coefficient is -1, determine that the suspected noise data is non-noise data, and input the temperature sequence, pressure sequence, and extrusion rate sequence into the preset neural network model to obtain the fault prediction result of the extruder; otherwise, determine that the suspected noise data is real noise data, and perform median filtering on it; and input the filtered temperature sequence, pressure sequence, and extrusion rate sequence into the preset neural network model to obtain the fault prediction result of the extruder.

[0007] In the fault detection method of the plastic pipe extruder of the present invention, after collecting the temperature sequence at the extrusion port, first preliminarily judge whether there is noise in the temperature sequence based on the variance corresponding to the temperature sequence. Under the condition of the existence of noise, further screen out the suspected noise data based on the fitting curve of the temperature data and the residuals of the curve data points; after screening out the suspected noise data, considering the relationship between the temperature and pressure at the extrusion port during the plastic extrusion process, further determine whether the suspected noise data is real noise in combination with the value of the first Pearson correlation coefficient, and filter and denoise the real noise, thereby greatly improving the accuracy of the collected temperature data at the extrusion port.

[0008] In addition, since the temperature data is further processed only when the preliminary judgment result is that there is noise in the temperature sequence, and when the preliminary judgment result is that there is no noise in the temperature sequence, the collected temperature sequence, pressure sequence, and extrusion rate sequence are directly input into the preset neural network model to obtain the fault prediction result. The neural network model can automatically learn the relationship between the physical parameters of the extrusion port collected and the fault detection result, thereby greatly improving the efficiency and accuracy of the extruder monitoring; furthermore, when using the neural network model to detect the fault state of the extruder, the temperature, pressure, and extrusion rate of the extrusion port are comprehensively considered, rather than considering a single physical parameter, thereby avoiding misjudgment and further improving the accuracy of the monitoring result of the extruder.

[0009] Preferably, the pressure acquisition method at the extrusion port includes: Arrange pressure sensors at the extrusion port, and calculate the pressure at the extrusion port according to the values collected by the pressure sensors. The calculation expression is: ; In the formula, Represents the pressure of the extrusion die Represents the value collected by the pressure sensor, k represents the compensation coefficient, and its value range is 0.05 to 0.1.

[0010] When collecting the pressure of the extrusion die, the pressure collection method of the extrusion die of the present invention does not directly use the value collected by the pressure sensor as the pressure of the extrusion die, but takes into account the friction between the plastic pipe and the inner wall of the extrusion die, corrects the value collected by the pressure sensor, and uses the corrected value as the collected pressure of the extrusion die, thereby improving the accuracy of the pressure collection of the extrusion die.

[0011] Preferably, the neighborhood range is a range composed of 5 data points centered on the suspected noise data point.

[0012] Preferably, it further includes: calculating the noise performance degree of each pressure data point in the pressure sequence, and performing median filtering on the pressure data points whose noise performance degree is greater than the noise performance degree threshold, and the calculation expression is: ; In the formula, Represents the noise performance degree of the i-th pressure data point, Represents the exponential function with e as the base, Represents the mean value of the pressure data points within the neighborhood range of the i-th pressure data point, Represents the mean value between the i-th pressure data point and its adjacent pressure data points before and after, Represents the second Pearson correlation coefficient between the pressure data points within the neighborhood range of the i-th pressure data point and the corresponding temperature data points; Represents the pressure data point with the largest value within the neighborhood range of the i-th pressure data point.

[0013] When calculating the noise performance degree of the pressure data point, the noise performance degree calculation method of the pressure data point of the present invention comprehensively considers the volatility of the i-th pressure data point and the change relationship between the temperature and pressure of the extrusion die, thereby more accurately calculating the noise performance degree of the pressure data point.

[0014] Preferably, the training method of the neural network model includes: Collecting the temperature sequence, pressure sequence and extrusion rate sequence of the extruder at historical moments, and labeling them to form a training set; Inputting the data in the training set into the neural network model to obtain a predicted value; Comparing the predicted value with the true value, and calculating the loss value using the mean squared error function; Calculate the gradient of the loss function with respect to each parameter using the chain rule, and then backpropagate these gradients back through each layer of the network; and update the weights and biases of each neuron according to the gradient values; Iteratively update the weights and biases of each neuron until the loss value converges or a preset number of iterations is reached.

[0015] Preferably, it further includes: real-time collecting the rotation speed of the extruder screw; Call the target extrusion rate according to the real-time extrusion rate and the thickness requirement of the plastic pipe, and calculate the difference between the target extrusion rate and the real-time extrusion rate; Perform PID operation on the difference to obtain the target rotation speed of the extruder screw; Perform PID control on the rotation speed of the extruder screw according to the difference between the target rotation speed and the real-time rotation speed of the extruder screw.

[0016] In the process of monitoring the extrusion port of the present invention, further considering the influence of the real-time extrusion rate on the thickness of the plastic pipe, automatically call the target extrusion rate according to the thickness requirement of the plastic pipe, and perform PID control on the rotation speed of the extruder screw according to the target extrusion rate, which helps to make the thickness of the plastic pipe meet the thickness requirement.

[0017] Preferably, it further includes: adjusting the target rotation speed of the extruder screw according to the temperature of the extrusion port, and the calculation expression of the target rotation speed of the adjusted screw is: ; In the formula, represents the target rotation speed of the adjusted screw, represents the target rotation speed of the screw before adjustment, represents the real-time temperature of the extrusion port, represents the standard temperature of the extrusion port.

[0018] When the real-time temperature of the extrusion port increases, it will cause the fluidity of the plastic to become better and the thickness of the extruded product to become thinner. When the real-time temperature decreases, it will cause the fluidity of the plastic to become worse and the thickness of the extruded product to increase. Therefore, in the present invention, after calculating the target rotation speed of the extruder screw, further adjust the target rotation speed of the extruder screw according to the difference between the real-time temperature and the standard temperature, which helps to keep the thickness of the produced plastic pipe constant and further improves the control accuracy of the thickness of the produced plastic pipe.

[0019] Preferably, it further includes: correcting the target rotation speed of the adjusted screw according to the pressure of the extrusion port, and the calculation expression of the screw rotation speed correction value is: ; In the formula, represents the screw rotation speed correction value, Represents the pressure value of the extrusion die. Represents the standard pressure value of the extrusion die. Represents the exponential function with base e.

[0020] After adjusting the target speed of the screw according to the temperature of the extrusion die, further considering the influence of the pressure of the extrusion die on the thickness of the plastic pipe, the target speed of the screw is further corrected according to the pressure of the extrusion die, thereby further improving the control accuracy of the thickness of the produced plastic pipe.

[0021] In a second aspect, the present invention provides a fault detection system for a plastic pipe extruder, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the fault detection method of the plastic pipe extruder of the present invention is implemented.

[0022] The beneficial effect of the present invention is that the method of the present invention can effectively improve the efficiency and accuracy of the monitoring of the extruder. Brief Description of the Drawings

[0023] By referring to the accompanying drawings and reading the following detailed description, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 Is a schematic flowchart showing a method for fault prediction of an extruder using a neural network model in the prior art; Figure 2 Is a schematic flowchart showing a fault detection method for a plastic pipe extruder according to an embodiment of the present invention; Figure 3 Is a schematic diagram showing a force analysis of the extrusion die according to an embodiment of the present invention; Figure 4 Is a schematic structural diagram showing a fault detection system for a plastic pipe extruder according to an embodiment of the present invention. Detailed Embodiments

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0025] Next, the specific embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0026] Embodiment of the Fault Detection Method for a Plastic Pipe Extrusion Machine: As Figure 2 shown, the fault detection method for a plastic pipe extrusion machine according to the present invention includes: S101. Obtain the temperature sequence, pressure sequence, and extrusion rate sequence at the extrusion port of the extrusion machine. Specifically: Collect the temperature, pressure, and extrusion rate at the extrusion port of the extrusion machine according to a preset acquisition frequency, so as to obtain the temperature sequence, pressure sequence, and extrusion rate sequence; The temperature at the extrusion port can be collected by setting a resistance temperature sensor, a thermocouple temperature sensor, or a radiation pyrometer at the extrusion port. The pressure at the extrusion port can be collected by setting a pressure sensor inside the extrusion port; A flow meter can be installed at the extrusion port to collect the plastic extrusion rate.

[0027] S102. Calculate the variance of the temperature sequence. If the variance is greater than a preset variance threshold, it is preliminarily determined that there is noise data in the temperature sequence; If the variance of the temperature sequence is too large, it indicates that the temperature data fluctuates greatly, and it is very likely that noise data appears due to interference of the temperature sensor.

[0028] S103. Obtain the suspected noise data. Specifically: Fit the temperature data in the temperature sequence to obtain a first fitting curve, and calculate the residuals corresponding to each temperature data; The temperature data with a residual greater than the preset residual threshold is determined as suspected noise data; The fitting method can adopt the polynomial fitting method. The value of the preset residual threshold can be determined based on experiments. In this embodiment, the residual threshold can be set to 3 degrees Celsius.

[0029] S104. Obtain the first Pearson correlation coefficient. Specifically: Fit the pressure data corresponding to the temperature data in the neighborhood range of the suspected noise data to obtain a second fitting curve, and calculate the first Pearson correlation coefficient between the temperature data points in the neighborhood range of the suspected noise data and the pressure data points on the second fitting curve; Since the temperature data and the pressure data are collected at the same acquisition frequency, the pressure data corresponding to the temperature data refers to the pressure data collected at the moment when the temperature data is collected.

[0030] In this embodiment, the neighborhood range is a range composed of 5 data points centered on the suspected noise data point. In other embodiments, the neighborhood range can also be other suitable ranges.

[0031] During the extrusion process, the change in temperature will directly affect the melting state and fluidity of the plastic. When the temperature rises, the melting degree of the plastic increases and the fluidity becomes better, which will lead to a decrease in the pressure at the extrusion die. On the contrary, when the temperature drops, the melting degree of the plastic decreases and the fluidity becomes worse, which will lead to an increase in the pressure at the extrusion die. Therefore, there is a linear relationship between the temperature data and the pressure data at the extrusion die, and the change trend of the temperature data at the extrusion die is opposite to that of the pressure data, that is, if the change trend of the temperature data is upward, the corresponding change trend of the pressure data is downward. The value of the first Pearson correlation coefficient can more accurately reflect the correlation between the temperature data points in the neighborhood range of the noise data and the pressure data points on the second fitting curve. When the value of the first Pearson correlation coefficient is 0, it indicates that there is no linear relationship between the temperature data points and the pressure data points. When the value of the first Pearson correlation coefficient is 1, when one of the temperature data points and the pressure data points increases, the other also increases. When the value of the first Pearson correlation coefficient is -1, when one of the temperature data points and the pressure data points increases, the other decreases.

[0032] S105. Obtain the fault prediction result of the extruder, specifically: if the value of the first Pearson correlation coefficient is -1, determine that the suspected noise data is non-noise data, and input the temperature sequence, pressure sequence and extrusion rate sequence into a preset neural network model to obtain the fault prediction result of the extruder; otherwise, determine that the suspected noise data is real noise data, and perform median filtering on it; and input the filtered temperature sequence, pressure sequence and extrusion rate sequence into a preset neural network model to obtain the fault prediction result of the extruder.

[0033] In this embodiment, the neural network model includes a first feature extraction model, a second feature extraction model, a third feature extraction model, a feature splicing model and a classification model. The first feature extraction model is used to extract the temporal features of the temperature sequence to obtain temperature features; the second feature extraction model is used to extract the temporal features of the pressure sequence to obtain pressure features; the third feature extraction model is used to extract the temporal features of the extrusion rate sequence to obtain extrusion rate features; the feature splicing model is used to splice the temperature features, pressure features and extrusion rate features together to obtain spliced features; the classification model is used to perform dimensionality transformation on the spliced features and output the fault prediction result.

[0034] The fault detection method of the plastic pipe extrusion machine in this embodiment, after collecting the temperature sequence of the extrusion port, first preliminarily determines whether there is noise in the temperature sequence based on the variance corresponding to the temperature sequence. Under the condition that there is noise, further filters out the suspected noise data based on the fitting curve of the temperature data and the residuals of the curve data points; after filtering out the suspected noise data, considering the relationship between the temperature and pressure at the extrusion port during the plastic extrusion process, further determines whether the suspected noise data is real noise based on the value of the first Pearson correlation coefficient, and filters and removes the real noise, thereby greatly improving the accuracy of the collected temperature data of the extrusion port; in addition, since the temperature data is further processed only under the condition that the preliminary judgment result is that there is noise in the temperature sequence, and when the preliminary judgment result is that there is no noise in the temperature sequence, the collected temperature sequence, pressure sequence and extrusion rate sequence are directly input into the preset neural network model to obtain the fault prediction result. The neural network model can automatically learn the relationship between the physical parameters of the extrusion port collected and the fault detection result, thereby greatly improving the efficiency and accuracy of the extrusion machine monitoring; furthermore, when using the neural network model to detect the fault state of the extrusion machine, comprehensively considers the temperature, pressure and extrusion rate of the extrusion port, rather than considering a single physical parameter, thereby avoiding misjudgment and further improving the accuracy of the monitoring result of the extrusion machine.

[0035] In one embodiment, the pressure acquisition method of the extrusion port includes: Install a pressure sensor at the extrusion port, and calculate the pressure of the extrusion port according to the value collected by the pressure sensor. The calculation formula is: ; In the formula, represents the pressure of the extrusion port, represents the pressure value collected by the pressure sensor, and k represents the compensation coefficient, and its value range is 0.05 - 0.1. The value range of the compensation coefficient can be obtained based on experiments.

[0036] As Figure 3 shown, the pressure at the extrusion port includes the pressure F1 perpendicular to the surface of the extrusion port and the frictional force f parallel to the surface of the extrusion port. The resultant force F of the two is the pressure at the extrusion port. The pressure sensor installed at the extrusion port can only collect the pressure F1 perpendicular to the surface of the extrusion port, and the actual pressure at the extrusion port will be slightly greater than the force collected by the pressure sensor. Therefore, multiply the pressure value collected by the pressure sensor by a correction coefficient to obtain the actual pressure at the extrusion port. Using the pressure acquisition method of the extrusion port in this embodiment can collect the pressure at the extrusion port more accurately, thereby further improving the accuracy of the monitoring result of the extrusion machine.

[0037] In one embodiment, it further includes: calculating the noise performance degree of each pressure data point in the pressure sequence, and performing median filtering on the pressure data points whose noise performance degree is greater than the noise performance degree threshold. The calculation expression is: ; In the formula, represents the noise performance degree of the i-th pressure data point, represents the exponential function with e as the base, represents the mean value of the pressure data points within the neighborhood range of the i-th pressure data point, represents the mean value between the i-th pressure data point and its adjacent pressure data points before and after, represents the second Pearson correlation coefficient between the pressure data points within the neighborhood range of the i-th pressure data point and the corresponding temperature data points; represents the pressure data point with the largest value within the neighborhood range of the i-th pressure data point.

[0038] can reflect the degree of difference between the i-th pressure data point and the pressure data points within its neighborhood range. The larger this value is, the greater the volatility of the i-th pressure data point and the more likely it is to be noise; the second Pearson correlation coefficient between the pressure data points within the neighborhood range of the i-th pressure data point and the corresponding temperature data points can reflect the linear relationship between the pressure data points within the neighborhood range of the i-th pressure data point and the corresponding temperature data points. Since normally, when the temperature at the extrusion die increases, the pressure decreases. If the second Pearson correlation coefficient is -1, it means that within the time period corresponding to the neighborhood range of the i-th pressure data point, as the pressure data increases, the temperature data decreases, indicating that the i-th pressure data point is very likely not noise data; otherwise, it means that when the pressure data changes, the temperature data may increase or remain unchanged, and the pressure value of the i-th pressure data point may be noise generated by interference to the pressure sensor rather than the true pressure value of the extrusion die. In this embodiment, when calculating the noise performance degree of the pressure data point, the volatility of the i-th pressure data point and the change relationship between the temperature and pressure at the extrusion die are comprehensively considered, so as to more accurately calculate the noise performance degree of the pressure data point. In one embodiment, the training method of the neural network model includes: S201. Collect the temperature sequence, pressure sequence, and extrusion rate sequence of the extruder at historical moments, and label them to form a training set; Multiple groups of operating data sequences of the extruder at historical moments can be collected. One group of operating data sequences of the extruder includes a corresponding temperature sequence, a pressure sequence, and an extrusion rate sequence; the collected operating data sequences of the extruder include both the operating data sequences under normal conditions of the extruder and the operating data sequences under abnormal conditions of the extruder.

[0039] S202. Input the data in the training set into the neural network model to obtain predicted values; S203. Compare the predicted values with the true values, and calculate the loss value using the mean squared error function; S204. Use the chain rule to calculate the gradients of the loss function with respect to each parameter, and then backpropagate these gradients back to each layer in the network; and update the weights and biases of each neuron according to the gradient values; S205. Iteratively update the weights and biases of each neuron until the loss value converges or reaches a preset number of iterations.

[0040] In one embodiment, it further includes: real-time collecting the rotation speed of the extruder screw; S301. Call the target extrusion rate according to the real-time extrusion rate and the thickness requirement of the plastic pipe, and calculate the difference between the target extrusion rate and the real-time extrusion rate; Generally, under the rotation of the screw, the plastic is forced by the thread towards the head direction. During the pushing process of the screw, the plastic is heated by the heat source of the barrel, and also generates a large amount of heat due to the friction between the barrel and the screw and the friction between plastic molecules, making the plastic gradually plasticized. By controlling the rate at which the plastic is extruded from the extrusion port, the heating duration of the plastic in the barrel and the degree of friction between plastic molecules can be controlled, thereby indirectly controlling the temperature of the plastic. The change in the temperature of the plastic will directly affect the meltability and fluidity of the plastic, and further affect the thickness of the extrudate. Therefore, for the same extruder, there is a corresponding relationship between the thickness of the plastic pipe and the extrusion rate. Through experiments, the extrusion rates corresponding to the thicknesses of various plastic pipes can be obtained respectively.

[0041] S302. Perform PID operation on the difference to obtain the target rotation speed of the extruder screw; S303. Perform PID control on the rotation speed of the extruder screw according to the difference between the target rotation speed and the real-time rotation speed of the extruder screw.

[0042] In this embodiment, during the monitoring of the extrusion port, further considering the influence of the real-time extrusion rate on the thickness of the plastic pipe, automatically call the target extrusion rate according to the thickness requirement of the plastic pipe, and perform PID control on the rotation speed of the extruder screw according to the target extrusion rate, which helps to make the thickness of the plastic pipe meet the thickness requirement.

[0043] In one embodiment, it further includes: adjusting the target rotation speed of the extruder screw according to the temperature of the extrusion port. The calculation expression for the adjusted target rotation speed of the screw is: ; In the formula, represents the adjusted target rotation speed of the screw, represents the target rotation speed of the screw before adjustment, represents the real-time temperature of the extrusion die, represents the standard temperature of the extrusion die.

[0044] When the real-time temperature of the extrusion die increases, it will cause the fluidity of the plastic to become better and the thickness of the extrudate to become thinner. When the real-time temperature decreases, it will cause the fluidity of the plastic to become worse and the thickness of the extrudate to increase. Therefore, in this embodiment, after calculating the target rotation speed of the extruder screw, the target rotation speed of the extruder screw is further adjusted according to the difference between the real-time temperature and the standard temperature, which helps to keep the thickness of the produced plastic pipe constant and further improves the control accuracy of the thickness of the produced plastic pipe.

[0045] In one embodiment, it further includes: correcting the target rotation speed of the adjusted screw according to the pressure of the extrusion die. The calculation expression of the screw rotation speed correction value is: ; In the formula, represents the screw rotation speed correction value, represents the pressure value of the extrusion die, represents the standard pressure value of the extrusion die, represents the exponential function with e as the base.

[0046] The pressure during the extrusion process will affect the fluidity of the plastic and thus affect the thickness of the extrudate. When the pressure increases, the fluidity of the plastic becomes better, which may cause the extrudate to become thinner. Conversely, when the pressure decreases, the extrudate may become thicker. In this embodiment, after adjusting the target rotation speed of the extruder screw, the influence of the pressure of the extrusion die on the thickness of the produced plastic pipe is further considered. When the pressure of the extrusion die is greater than the standard pressure, the target rotation speed of the adjusted screw is increased, so as to reduce the temperature of the plastic and increase the thickness of the plastic pipe, thereby compensating for the influence of the pressure of the extrusion die on the thickness of the produced plastic pipe. When the pressure of the extrusion die is less than the standard pressure, the target rotation speed of the adjusted screw is decreased, so as to increase the temperature of the plastic and decrease the thickness of the plastic pipe, thereby compensating for the influence of the pressure of the extrusion die on the thickness of the produced plastic pipe. In this embodiment, after adjusting the target rotation speed of the screw according to the temperature of the extrusion die, the influence of the pressure of the extrusion die on the thickness of the plastic pipe is further considered, and the target rotation speed of the screw is further corrected according to the pressure of the extrusion die, thereby further improving the control accuracy of the thickness of the produced plastic pipe.

[0047] Embodiment of the fault detection system for a plastic pipe extruder: The present invention also provides a fault detection system for a plastic pipe extruder. As Figure 4As shown, the fault detection system of the plastic pipe extrusion machine includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the fault detection method of the plastic pipe extrusion machine described in the above embodiments.

[0048] The fault detection system of the plastic pipe extrusion machine further includes a communication bus, a communication interface, and other components well-known to those skilled in the art. Their settings and functions are known in the art, so they will not be elaborated here.

[0049] In the present invention, the aforementioned memory can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or apparatus. For example, a computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium can be part of the device or accessible or connectable to the device. Any application or module described in the present invention can be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.

[0050] In the description of this specification, the meanings of "a plurality of" and "several" are at least two, such as two, three, or more, unless otherwise clearly and specifically defined.

[0051] Although this specification has shown and described multiple embodiments of the present invention, it is obvious to those skilled in the art that such embodiments are provided only by way of example. Those skilled in the art will think of many changes, alterations, and alternative ways without departing from the spirit and idea of the present invention. It should be understood that various alternative solutions to the embodiments of the present invention described herein can be adopted in the process of practicing the present invention.

Claims

1. A fault detection method for a plastic pipe extrusion machine, characterized in that, Including: Collect the temperature, pressure, and extrusion rate at the extrusion port of the extruder according to a preset acquisition frequency, so as to obtain a temperature sequence, a pressure sequence, and an extrusion rate sequence; Calculate the variance of the temperature sequence. If the variance is greater than a preset variance threshold, it is preliminarily determined that there is noise data in the temperature sequence; Fit the temperature data in the temperature sequence to obtain a first fitting curve, and calculate the residuals corresponding to each temperature data; Determine the temperature data with residuals greater than the preset residual threshold as suspected noise data; Fit the pressure data corresponding to the temperature data in the neighborhood range of the suspected noise data to obtain a second fitting curve, and calculate the first Pearson correlation coefficient between the temperature data points in the neighborhood range of the suspected noise data and the pressure data points on the second fitting curve; If the value of the first Pearson correlation coefficient is -1, it is determined that the suspected noise data is non-noise data, and the temperature sequence, pressure sequence, and extrusion rate sequence are input into a preset neural network model to obtain the fault prediction result of the extruder; otherwise, it is determined that the suspected noise data is real noise data and median filtering is performed on it; And input the filtered temperature sequence, pressure sequence, and extrusion rate sequence into a preset neural network model to obtain the fault prediction result of the extruder.

2. The fault detection method of the plastic pipe extrusion machine according to claim 1, characterized in that, The method for collecting the pressure at the extrusion port includes: Arrange a pressure sensor at the extrusion port, and calculate the pressure at the extrusion port according to the value collected by the pressure sensor. The calculation expression is: ； In the formula, represents the pressure of the extrusion die,[ represents the value collected by the pressure sensor, and k represents the compensation coefficient, whose value range is 0.05 to 0.

1.

3. The fault detection method of the plastic pipe extrusion machine according to claim 2, characterized in that, The neighborhood range is a range composed of 5 data points centered on the suspected noise data point.

4. The fault detection method of the plastic pipe extrusion machine according to claim 1, characterized in that, Also including: Calculate the noise manifestation degree of each pressure data point in the pressure sequence, and perform median filtering on the pressure data points with a noise manifestation degree greater than the noise manifestation degree threshold. The calculation expression is: ； In the formula, represents the noise performance degree of the i-th pressure data point, represents the exponential function with base e, represents the mean value of the pressure data points within the neighborhood range of the i-th pressure data point, represents the mean value between the i-th pressure data point and its adjacent pressure data points before and after, represents the second Pearson correlation coefficient between the pressure data points within the neighborhood range of the i-th pressure data point and the corresponding temperature data points; represents the pressure data point with the largest value within the neighborhood range of the i-th pressure data point.

5. The fault detection method of the plastic pipe extrusion machine according to claim 1, characterized in that, The training method of the neural network model includes: Collect the temperature sequence, pressure sequence, and extrusion rate sequence of the extruder at historical moments, and label them to form a training set; Input the data in the training set into the neural network model to obtain a predicted value; Compare the predicted value with the real value, and calculate the loss value using the mean squared error function; Use the chain rule to calculate the gradient of the loss function with respect to each parameter, and then backpropagate these gradients back to each layer in the network; and update the weights and biases of each neuron according to the gradient values; Iteratively update the weights and biases of each neuron until the loss value converges or reaches a preset number of iterations.

6. The fault detection method of the plastic pipe extrusion machine according to claim 1, characterized in that Also including: Collect the rotation speed of the screw of the extruder in real time; Call the target extrusion rate according to the real-time extrusion rate and the thickness requirement of the plastic pipe, and calculate the difference between the target extrusion rate and the real-time extrusion rate; Perform PID operation on the difference to obtain the target rotation speed of the screw of the extruder; Perform PID control on the rotation speed of the screw of the extruder according to the difference between the target rotation speed and the real-time rotation speed of the screw of the extruder.

7. The fault detection method of the plastic pipe extrusion machine according to claim 1, characterized in that, Also including: Adjust the target rotation speed of the screw of the extruder according to the temperature at the extrusion port. The calculation expression for the adjusted target rotation speed of the screw is: ； Wherein, represents the target rotational speed of the adjusted screw, represents the target rotational speed of the screw before adjustment, represents the real-time temperature of the extrusion die, represents the standard temperature of the extrusion die.

8. The fault detection method of the plastic pipe extrusion machine according to claim 7, characterized in that, Also including: Correct the adjusted target rotation speed of the screw according to the pressure at the extrusion port. The calculation expression for the screw rotation speed correction value is: ； In the formula, represents the correction value of the screw rotation speed, represents the pressure value at the extrusion die, represents the standard pressure value at the extrusion die, represents the exponential function with base e.

9. A fault detection system for a plastic pipe extrusion machine, comprising a processor and a memory, wherein the memory stores computer program instructions, characterized in that, When the computer program instructions are executed by the processor, a fault detection method of the plastic pipe extrusion machine according to any one of claims 1 to 8 is implemented.

Images