Radio frequency matcher motor life prediction method, system and device based on neural network model and medium thereof

Abstract

The invention relates to the technical field of motor life prediction, in particular to a life prediction method, system and device for a radio frequency matcher motor based on a neural network model and a medium thereof, and the prediction method comprises the following steps: obtaining an absolute cycle index matrix of the radio frequency matcher motor; carrying out normalization processing on the obtained cycle index matrix; constructing a radio frequency matcher motor life prediction neural network model to obtain an optimal radio frequency matcher motor life prediction neural network model; and inputting the cycle index matrix of the radio frequency matcher motor to output the predicted deviation value of the radio frequency matcher motor, thereby obtaining the predicted life of the radio frequency matcher motor. According to the method, the service life prediction neural network model is constructed, and the service life of the radio frequency matcher motor is predicted by using the service life prediction neural network model, so that the method has strong nonlinear modeling capability, and the accuracy of service life prediction of the radio frequency matcher motor can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of motor life prediction, and in particular to a method, system, device and medium for predicting the life of a radio frequency matching motor based on a neural network model. Background Art

[0002] The RF matching box motor is mainly used to adjust the position or parameters of the adjustable components in the RF matching box. Through motor drive, the capacitance, inductance and other parameters in the matching network can be accurately changed, so that the RF circuit can achieve the best matching state under different working conditions, thereby improving power transmission efficiency, reducing reflected power, and reducing signal distortion. It is widely used in semiconductor manufacturing, photovoltaic cell production and other fields.

[0003] Life prediction for RF matching motors is not only crucial for improving their functionality and developing maintenance strategies, but also for enhancing the durability of their application systems. RF matching motors receive signals from the control system and drive internal transmission mechanisms (e.g., screws, gears, etc.) to fine-tune adjustable components. They offer advantages such as high precision, rapid response, strong anti-interference capabilities, and high reliability. Common failure modes during operation include brush wear, insulation aging, bearing wear, excitation loss, corrosion, and rust. Furthermore, they are affected by environmental conditions such as high humidity, high temperature, and high salt spray corrosion. Furthermore, RF matching motors frequently start and stop, increasing loads, and long-term operation can cause environmental damage (e.g., corrosion, aging, etc.) and mechanical damage (e.g., wear, fatigue, etc.). These damages exhibit significant coupling and cumulative effects, accelerating the occurrence of RF matching motor failures.

[0004] At present, the RF matching motor mainly predicts the changes in the system state at the current moment by establishing different degradation models based on the historical information of the system state changes. This method is highly dependent on the model. However, its application in nonlinear aspects is insufficient, and its life prediction ability in complex variable conditions has certain limitations. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: in order to solve the technical problems of insufficient application and large limitations of existing RF matching motors in terms of nonlinearity, the present invention provides a life prediction method, system, equipment and medium of RF matching motors based on a neural network model. By improving the life prediction method of RF matching motors, it has powerful nonlinear modeling capabilities to improve the accuracy of RF matching motor life prediction.

[0006] The technical solution adopted by the present invention to solve the technical problem is: a life prediction method of a radio frequency matching motor based on a neural network model, comprising the following steps:

[0007] S1. Obtain the absolute cycle number matrix of the RF matching motor ;

[0008] S2, the cycle number matrix obtained from S1 Perform normalization processing;

[0009] S3, build the RF matching motor life prediction neural network model, and use the cycle number matrix after S2 normalization Conduct training to obtain the optimal RF matching motor life prediction neural network model;

[0010] S4, input the cycle number matrix of the RF matching motor , and use the optimal RF matching motor life prediction neural network model in S3 to output the predicted deviation value of the RF matching motor, and then obtain the predicted life of the RF matching motor.

[0011] Therefore, by constructing a life prediction neural network model and using the life prediction neural network model to predict the life of the RF matching motor, compared with the existing method of establishing different degradation models, this method has a simple structure, is easy to operate, has powerful nonlinear modeling capabilities, and can improve the accuracy of the life prediction of the RF matching motor. In addition, the prediction neural network model can adapt to changes in different working environment factors, automatically adjust the prediction results to a certain extent, has good robustness and adaptability, and can meet the prediction needs of the RF matching motor life in different scenarios.

[0012] Furthermore, in S1, the number of cycles of the RF matching motor is Each corresponds to the deviation between the actual position and the set position of a RF matching motor .

[0013] Furthermore, the S1 includes the following steps:

[0014] S1-1, set the running speed of the RF matching device motor;

[0015] S1-2, set the number of cycles , to obtain the actual position and set position under the number of cycles, and calculate the deviation value ;

[0016] S1-3, repeat S1-2 to get the deviation value matrix ;

[0017] S1-4, the cycle number matrix in S1-3 Convert to absolute cycle count matrix ;

[0018] in: Represents the number of cycles, the cycle number matrix .

[0019] Furthermore, in S1-4, the absolute cycle number matrix and the cycle count matrix The expression is:

[0020] .

[0021] Furthermore, the actual position of the RF matching motor refers to: the actual position of the output shaft relative to the starting point at the set operating speed of the RF matching motor; the set position of the RF matching motor refers to: the set position of the output shaft relative to the starting point at the set operating speed of the RF matching motor; the deviation value refers to: the absolute value of the difference between the actual position and the set position.

[0022] Furthermore, in S4, the predicted deviation value of the RF matching motor is ,like , then the RF matching device motor is in normal condition. , then the RF matching device motor is damaged; where: To set the deviation value.

[0023] Furthermore, in S3, the RF matching motor life prediction neural network model includes: an input layer, a hidden layer, and an output layer; wherein: the activation function of the hidden layer adopts the Sigmoid function, the activation function of the output layer adopts the Purelin function, the mean square error function is used to calculate the network error, and the loss function is optimized by the gradient descent method.

[0024] A life prediction system for a radio frequency matching unit motor based on a neural network model includes: a data acquisition module, a controller, and a host computer. The data acquisition module is used to collect the number of operations of the radio frequency matching unit motor, the number of cycles of the radio frequency matching unit motor, and the operating status of the radio frequency matching unit motor. The controller is connected to the data acquisition module and is used to control the data acquisition module. The host computer is connected to the data acquisition module and executes a life prediction method for the radio frequency matching unit motor based on the neural network model.

[0025] A computer device comprises: a processor and a memory, wherein the memory is used to store executable instructions; wherein: the processor is used to read the executable instructions from the memory and execute the executable instructions to implement a life prediction method for a radio frequency matching device motor based on a neural network model.

[0026] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor implements a life prediction method for a radio frequency matching device motor based on a neural network model.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] By constructing a life prediction neural network model and using the life prediction neural network model to predict the life of the RF matching motor, compared with the existing method of establishing different degradation models, this method has a simple structure, is easy to operate, has powerful nonlinear modeling capabilities, and can improve the accuracy of the life prediction of the RF matching motor. In addition, the prediction neural network model can adapt to changes in different working environment factors, automatically adjust the prediction results to a certain extent, has good robustness and adaptability, and can meet the prediction needs of the RF matching motor life in different scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The present invention will be further described below with reference to the accompanying drawings and examples.

[0030] Figure 1 Flowchart of the life prediction method of the radio frequency matching motor based on the neural network model of the present invention;

[0031] Figure 2 This is a flow chart of S2 of the present invention. DETAILED DESCRIPTION

[0032] The present invention will now be described in further detail with reference to the accompanying drawings, which are simplified schematic diagrams that illustrate the basic structure of the present invention in a schematic manner.

[0033] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention 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 limiting the present invention. In addition, features defined as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0034] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0035] like Figures 1 to 2 As shown, a life prediction method for a radio frequency matching motor based on a neural network model includes the following steps:

[0036] S1. Obtain the absolute cycle number matrix of the RF matching motor ;

[0037] S2, the cycle number matrix obtained from S1 Perform normalization processing;

[0038] S3, build the RF matching motor life prediction neural network model, and use the cycle number matrix after S2 normalization Conduct training to obtain the optimal RF matching motor life prediction neural network model;

[0039] S4, input the cycle number matrix of the RF matching motor , and uses the optimal RF matching motor life prediction neural network model in S3 to output the predicted deviation value of the RF matching motor, thereby obtaining the predicted life of the RF matching motor. Therefore, by constructing a life prediction neural network model and using this life prediction neural network model to predict the life of the RF matching motor, compared with the existing method of establishing different degradation models, this method has a simple structure, is easy to operate, has powerful nonlinear modeling capabilities, and can improve the accuracy of RF matching motor life prediction. In addition, the prediction neural network model can adapt to changes in different working environment factors and automatically adjust the prediction results to a certain extent. It has good robustness and adaptability and can meet the prediction needs of RF matching motor life in different scenarios.

[0040] In other words, by constructing a life prediction neural network model, since the neural network model is composed of a large number of neurons, each neuron can receive multiple inputs and process the inputs through a nonlinear activation function, so that the neurons can perform nonlinear mapping on different input patterns, thereby processing complex nonlinear relationships; in addition, the life prediction neural network model has a multi-layer structure (for example: input layer, hidden layer and output layer). Through the combination of multiple layers of neurons, the life prediction neural network model can learn very complex nonlinear function relationships, thereby effectively modeling nonlinear problems such as RF matching motors whose service life is affected by multiple factors.

[0041] Specifically, a number of cycles is set so that the RF matching motor moves back and forth between the maximum value and the minimum value, and the current position of the RF matching motor is fed back in real time. As the prediction continues, the return position of the RF matching motor will gradually produce errors. When the cycle continues, until the deviation of the return position of the RF matching motor reaches a certain threshold, all the cycle numbers are defined as the service life of the RF matching motor.

[0042] In this embodiment, the number of cycles of the RF matching motor is Each corresponds to the deviation between the actual position and the set position of a RF matching motor ;

[0043] Said S1 comprises the following steps:

[0044] S1-1, set the running speed of the RF matching device motor;

[0045] S1-2, set the number of cycles , to obtain the actual position and set position under the number of cycles, and calculate the deviation value ;

[0046] S1-3, repeat S1-2 to get the deviation value matrix ;

[0047] S1-4, the cycle number matrix in S1-3 Convert to absolute cycle count matrix ;

[0048] in: Represents the number of cycles, the cycle number matrix ;

[0049] The actual position of the RF matching motor refers to the actual position of the output shaft relative to the starting point at the set operating speed of the RF matching motor. The set position of the RF matching motor refers to the set position of the output shaft relative to the starting point at the set operating speed of the RF matching motor. The deviation value refers to the absolute value of the difference between the actual position and the set position.

[0050] In S1-4, the absolute cycle number matrix and the cycle count matrix The expression is:

[0051] .

[0052] Specifically, in S1-2, the RF matching device motor switches back and forth between the maximum value and the minimum value, and the number of cycles is reached. Then, at this time, the position of the RF matching device motor is the actual position.

[0053] In this embodiment, in S3, the RF matching motor life prediction neural network model includes an input layer, a hidden layer, and an output layer. The hidden layer uses a Sigmoid function as its activation function, the output layer uses a Purelin function as its activation function, the network error is calculated using a mean square error function, and the loss function is optimized using the gradient descent method. Specifically, the RF matching motor life prediction neural network model uses a BP (Back Propagation) neural network model.

[0054] In this embodiment, in S4, the predicted deviation value of the RF matching motor is ,like , then the RF matching device motor is in normal condition. , then the RF matching device motor is damaged; where: To set the deviation value.

[0055] A life prediction system for a radio frequency matching unit motor based on a neural network model includes: a data acquisition module, a controller, and a host computer. The data acquisition module is used to collect the number of operations of the radio frequency matching unit motor, the number of cycles of the radio frequency matching unit motor, and the operating status of the radio frequency matching unit motor. The controller is connected to the data acquisition module and is used to control the data acquisition module. The host computer is connected to the data acquisition module and executes a life prediction method for the radio frequency matching unit motor based on the neural network model.

[0056] A computer device includes: a processor and a memory, the memory being used to store executable instructions; wherein: the processor is used to read the executable instructions from the memory and execute the executable instructions to implement a life prediction method for a radio frequency matching device motor based on a neural network model.

[0057] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor implements a life prediction method for a radio frequency matching device motor based on a neural network model.

[0058] In summary, the present invention constructs a life prediction neural network model and uses the life prediction neural network model to predict the life of the RF matching motor. Compared with the existing method of establishing different degradation models, this method has a simple structure, is easy to operate, has powerful nonlinear modeling capabilities, and can improve the accuracy of the life prediction of the RF matching motor. In addition, the prediction neural network model can adapt to changes in different working environment factors, automatically adjust the prediction results to a certain extent, has good robustness and adaptability, and can meet the prediction needs of the RF matching motor life in different scenarios.

[0059] The above description is intended to serve as a guide for the preferred embodiments of the present invention. Based on the above description, relevant personnel are fully capable of making various changes and modifications without departing from the technical scope of the present invention. The technical scope of the present invention is not limited to the contents of the specification and must be determined according to the scope of the claims.

Claims

1. A life prediction method for a radio frequency matching motor based on a neural network model, characterized in that: The following steps are involved: S1. Obtain the absolute cycle number matrix of the RF matching motor ; S2, the cycle number matrix obtained from S1 Perform normalization processing; S3, build the RF matching motor life prediction neural network model, and use the cycle number matrix after S2 normalization Conduct training to obtain the optimal RF matching motor life prediction neural network model; S4, input the cycle number matrix of the RF matching motor , and use the optimal RF matching motor life prediction neural network model in S3 to output the predicted deviation value of the RF matching motor, and then obtain the predicted life of the RF matching motor.

2. The life prediction method of the RF matching motor based on the neural network model according to claim 1 is characterized in that: In S1, the number of cycles of the RF matching motor Each corresponds to the deviation between the actual position and the set position of a RF matching motor .

3. The life prediction method of the RF matching motor based on the neural network model according to claim 1 is characterized in that: Said S1 comprises the following steps: S1-1, set the running speed of the RF matching device motor; S1-2, set the number of cycles , to obtain the actual position and set position under the number of cycles, and calculate the deviation value ; S1-3, repeat S1-2 to get the deviation value matrix ; S1-4, the cycle number matrix in S1-3 Convert to absolute cycle count matrix ; in: Represents the number of cycles, the cycle number matrix .

4. The life prediction method of a radio frequency matching motor based on a neural network model according to claim 1, characterized in that: In S1-4, the absolute cycle number matrix and the cycle count matrix The expression is: 。 5. The life prediction method of the RF matching motor based on the neural network model according to claim 1 is characterized in that: In S1, the actual position of the RF matching motor is: The actual position of the output shaft relative to the starting point at the set operating speed of the RF matching motor; The setting position of the RF matching motor is: At the set running speed of the RF matching motor, the output shaft is set to the position relative to the starting point; Deviation value means: The absolute value of the difference between the actual position and the set position.

6. The life prediction method of a radio frequency matching motor based on a neural network model according to claim 1, characterized in that: In S4, the predicted deviation value of the RF matching motor is ,like , then the RF matching device motor is in normal condition. , then the RF matching device motor is damaged; in: To set the deviation value.

7. The life prediction method of a radio frequency matching motor based on a neural network model according to claim 1, characterized in that: In S3, the RF matching motor life prediction neural network model includes: Input layer, hidden layer, and output layer; Among them, the activation function of the hidden layer adopts the Sigmoid function, the activation function of the output layer adopts the Purelin function, the mean square error function is used to calculate the network error, and the loss function is optimized by the gradient descent method.

8. The life prediction system of the RF matching motor based on the neural network model according to claim 1 is characterized in that: include: A data acquisition module, wherein the data acquisition module is used to collect the number of operations of the RF matching motor, the number of cycles of the RF matching motor, and the operating status of the RF matching motor; A controller connected to the data acquisition module and configured to control the data acquisition module; A host computer is connected to the data acquisition module, and the host computer executes the life prediction method of the radio frequency matching device motor based on the neural network model according to any one of claims 1 to 7.

9. A computer device, characterized in that: include: processor; a memory for storing executable instructions; Wherein: the processor is used to read the executable instructions from the memory and execute the executable instructions to implement the life prediction method of the radio frequency matching motor based on the neural network model as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the life prediction method of the radio frequency matching motor based on the neural network model as described in any one of claims 1 to 7.

Images