Wafer thinning and grinding control method and system, computing equipment and storage medium

Abstract

The invention discloses a wafer thinning and grinding control method and system, computing equipment and a storage medium, the wafer thinning and grinding control method is executed in the computing equipment, and the method comprises the steps that feature extraction is conducted on a sensor signal from a sensor assembly, and multiple feature parameters are obtained; the characteristic parameters are input into a grinding state prediction model to be processed, and state parameters indicating the grinding state are obtained; according to the state parameters, whether the current technological parameters of the thinning lathe need to be optimized or not is judged; if yes, technological parameters for optimization are determined according to the current grinding state, and the technological parameters comprise the grinding wheel rotating speed, the grinding wheel feeding speed and the rotating speed of a rotary table of the thinning machine tool; and controlling the thinning lathe to process the wafer workpiece based on the optimized process parameters. According to the method, the processing efficiency and the processing quality of the wafer workpiece are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a wafer thinning and grinding control method, system, computing device and storage medium. Background Art

[0002] Wafer thinning is the basis of semiconductor manufacturing. The processing surface quality and surface accuracy will directly affect subsequent processing and thus affect the performance of semiconductor devices.

[0003] At present, the main method for thinning large-size wafers is wafer self-rotation grinding, which can achieve good surface accuracy. The principle is as follows: the spindle and turntable rotate around their own axes respectively, and the grinding wheel feeds in the axial direction at the same time; the wafer workpiece is adsorbed on the turntable through the suction cup, and its center passes through the circumference of the grinding wheel. At the same time, the suction cup is usually designed to be conical or inverted conical, and there is a certain angle between the two axes of the spindle and the turntable to achieve semi-contact grinding between the wafer and the grinding wheel.

[0004] This grinding method has high grinding efficiency. However, if the grinding is controlled to achieve the desired effect only by adjusting the grinding inclination angle and process parameters, it is necessary to adjust the inclination angle multiple times and conduct a large number of process tests to determine the optimal grinding process parameters. In addition, the grinding effect is affected by many factors, and it is difficult to ensure the consistency of the results. Summary of the invention

[0005] In view of the above problems, the present invention is proposed to provide a wafer thinning and grinding control method, system, computing device and storage medium that overcome the above problems or at least partially solve the above problems.

[0006] According to one aspect of the present invention, there is provided a wafer thinning and grinding control method, which is executed in a computing device, and comprises: extracting features from a sensor signal from a sensor assembly to obtain a plurality of feature parameters, wherein the sensor assembly comprises at least a vibration sensor, an acoustic emission sensor and a force sensor, wherein the vibration sensor is used to collect vibration signals of a spindle of a thinning machine tool, the acoustic emission sensor is used to collect elastic wave signals of a grinding wheel of the thinning machine tool and a wafer workpiece, and the force sensor is used to collect pressure signals of a grinding wheel of the thinning machine tool on the wafer workpiece; inputting feature parameters into a grinding state prediction model for processing to obtain state parameters indicating the grinding state; judging whether current process parameters of a thinning lathe need to be optimized based on the state parameters; if necessary, determining process parameters to be optimized based on the current grinding state, wherein the process parameters include a grinding wheel speed, a grinding wheel feed speed and a turntable speed of the thinning machine tool; and controlling the thinning lathe to process the wafer workpiece based on the optimized process parameters.

[0007] Optionally, in the wafer thinning and grinding control method according to the present invention, the optimized process parameters are determined according to the current grinding state, including: determining the characteristic parameters corresponding to the optimized grinding state according to the current grinding state; and matching the corresponding process parameters from the process parameter sample database using the determined characteristic parameters.

[0008] Optionally, in the wafer thinning and grinding control method according to the present invention, the characteristic parameters include vibration amplitude, acoustic emission amplitude, acoustic emission effective voltage and force amplitude.

[0009] Optionally, in the wafer thinning and grinding control method according to the present invention, feature extraction is performed on the sensor signal collected by the current sensor component to obtain multiple feature parameters, including: filtering the vibration signal; performing time-frequency conversion processing on the filtered vibration signal using fast Fourier transform to obtain a signal spectrum; extracting the spectrum amplitude from the signal spectrum as the vibration amplitude.

[0010] Optionally, in the wafer thinning and grinding control method according to the present invention, feature extraction is performed on the sensor signal collected by the current sensor component to obtain multiple feature parameters, including: filtering and amplifying the elastic wave signal; performing time-frequency analysis on the elastic wave signal after filtering and amplification using wavelet transform to obtain an amplitude spectrum and a power spectrum; and extracting the acoustic emission amplitude and the acoustic emission effective voltage from the amplitude spectrum and the power spectrum, respectively.

[0011] Optionally, in the wafer thinning and grinding control method according to the present invention, feature extraction is performed on the sensor signal collected by the current sensor component to obtain multiple feature parameters, including: filtering the pressure signal; generating a pressure time domain graph using the filtered pressure signal; and extracting the force amplitude from the pressure time domain graph.

[0012] Optionally, the wafer thinning and grinding control method according to the present invention further includes: if the current process parameters of the thinning lathe do not need to be optimized, controlling the thinning lathe to still process the wafer workpiece with the current process parameters.

[0013] Optionally, in the wafer thinning grinding control method according to the present invention, the grinding state prediction model adopts a feedforward neural network model, and the state parameters include the grinding wheel wear level and the roughness value of the wafer workpiece surface.

[0014] Optionally, in the wafer thinning and grinding control method according to the present invention, each process parameter and the corresponding characteristic parameter are associated and stored in a process parameter sample database, and the corresponding process parameters are matched from the process parameter sample database using the determined characteristic parameters, including: using the adjusted characteristic parameters as indexes to match the corresponding process parameters from the process parameter sample database.

[0015] Optionally, in the wafer thinning and grinding control method according to the present invention, if the current process parameters of the thinning lathe do not need to be optimized, the thinning lathe is controlled to still process the wafer workpiece with the current process parameters.

[0016] Optionally, the wafer thinning and grinding control method according to the present invention further includes: associating the characteristic parameters with current process parameters, and storing the associated characteristic parameters and current process parameters in a process parameter sample database.

[0017] According to another aspect of the present invention, there is provided a wafer thinning and grinding control system, comprising: a thinning lathe, comprising a grinding assembly including a spindle, a grinding wheel arranged at the bottom of the spindle, and a turntable arranged opposite to the grinding wheel, suitable for grinding a wafer workpiece, wherein the wafer workpiece is fixed on the turntable during grinding, and the grinding wheel and the wafer workpiece abut against each other; a sensor assembly, comprising a vibration sensor, an acoustic emission sensor and a plurality of force sensors, wherein the vibration sensor is adsorbed on the side wall of the spindle, the acoustic emission sensor and the force sensor are arranged on the turntable, the vibration sensor is used to collect vibration signals of the spindle of the thinning machine tool, the acoustic emission sensor is used to collect elastic wave signals of the grinding wheel of the thinning machine tool and the wafer workpiece, and the force sensor is used to collect pressure signals of the grinding wheel of the thinning machine tool on the wafer workpiece; a computing device, respectively coupled to the sensor assembly and the thinning lathe, for acquiring sensor signals from the sensor assembly, and executing the above method to obtain process parameters for controlling the thinning machine tool, and controlling the thinning lathe to process the wafer workpiece based on the process parameters.

[0018] Optionally, the wafer thinning and grinding control system according to the present invention also includes: a vibration signal acquisition card connected to the vibration sensor and the computing device, respectively, for sending the vibration signal collected by the vibration sensor to the computing device; an acoustic emission signal acquisition card connected to the acoustic emission sensor and the computing device, respectively, for sending the elastic wave signal collected by the acoustic emission sensor to the computing device; a force signal acquisition card connected to the force sensor and the computing device, respectively, for sending the pressure signal collected by the force sensor to the computing device.

[0019] Optionally, in the wafer thinning and grinding control system according to the present invention, it also includes: a carrying box arranged on the turntable, and a workbench set on the carrying box; the acoustic emission signal acquisition card is arranged in the carrying box, and the force sensor, the acoustic emission sensor and the force signal acquisition card are arranged in the workbench; the shell surfaces of the carrying box and the workbench are both arranged with routing holes, and the acoustic emission sensor and the acoustic emission signal acquisition card are connected, and the force sensor and the force signal acquisition card are connected through the routing holes.

[0020] Optionally, the wafer thinning and grinding control system according to the present invention further comprises: a grinding tape, which is arranged on the workbench and abuts against the wafer workpiece to grind the wafer workpiece.

[0021] Optionally, the wafer thinning and grinding control system according to the present invention further includes: a power supply, which is arranged in the carrying box and is used to provide power to various components in the system.

[0022] Optionally, the wafer thinning and grinding control system according to the present invention further includes: a water cooling device for controlling the temperature of the grinding wheel and the wafer workpiece during the thinning and grinding process.

[0023] According to another aspect of the present invention, there is provided a computing device, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be suitable for being executed by the at least one processor, and the program instructions include instructions for executing the above method.

[0024] According to yet another aspect of the present invention, there is provided a computer program product, comprising a computer program / instruction, wherein the computer program / instruction implements the above method when executed by a processor.

[0025] According to another aspect of the present invention, a readable storage medium storing program instructions is provided. When the program instructions are read and executed by a computing device, the computing device executes the above method.

[0026] According to the solution of the present invention, by acquiring relevant data of the wafer workpiece during the grinding process and performing feature extraction on the relevant data, the turntable parameters indicating the current grinding state are obtained according to the extracted feature parameters, the process parameters are determined according to the state parameters, and then the thinning machine tool is controlled according to the process parameters to process the wafer workpiece. Through the solution of the present invention, the process parameters of the grinding machine tool can be determined according to the grinding state monitored during the grinding process, without adjusting the inclination angle and conducting a large number of process tests to determine the optimal grinding process parameters, thereby improving the processing efficiency and processing quality of the wafer workpiece.

[0027] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0029] Figure 1 A schematic structural diagram of a thinning machine tool 100 is shown;

[0030] Figure 2 A schematic structural diagram of a wafer thinning and grinding control system 300 according to an embodiment of the present invention is shown;

[0031] Figure 3 A block diagram of the physical components (ie, hardware) of a computing device 400 is shown;

[0032] Figure 4 A flowchart of a wafer thinning and grinding control method 600 according to an embodiment of the present invention is shown;

[0033] Figure 5 A schematic diagram of a feedforward neural network model according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0034] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0035] Wafer thinning and grinding mainly relies on abrasive grinding technology, which achieves thinning by removing a layer of material on the surface of the wafer. Wafer thinning and grinding requires the use of a thinning machine.

[0036] Figure 1 A schematic diagram of the structure of a thinning machine tool 100 is shown. Figure 1 The thinning machine tool 100 mainly includes a spindle 110, a grinding wheel 120 and a turntable 130. When the thinning machine tool 100 thins and grinds the wafer workpiece 200, the spindle 110 and the turntable 130 rotate around their own axes respectively, and the grinding wheel 120 feeds in the axial direction at the same time. The wafer workpiece 200 is adsorbed on the turntable 130 through a suction cup, and its center passes through the circumference of the grinding wheel 120. At the same time, the suction cup is usually designed to be conical or inverted conical. There is also a certain angle between the two axes of the spindle 110 and the turntable 130 to achieve semi-contact grinding of the wafer workpiece 200 and the grinding wheel 120.

[0037] The thinning machine 100 has a compact structure and high grinding efficiency. However, the thinning machine 100 needs to continuously adjust the grinding inclination angle and process parameters to control the grinding to achieve the desired effect during the thinning and grinding process, so that the optimal grinding process parameters can be determined after multiple adjustments of the inclination angle and a large number of process tests. Moreover, the grinding effect is affected by many factors, and it is difficult to ensure the consistency of the results.

[0038] In order to solve the problems existing in the prior art, the solution of the present invention is proposed. The present application proposes a wafer thinning and grinding control system 300. Figure 2 A schematic structural diagram of a wafer thinning and grinding control system 300 according to an embodiment of the present invention is shown.

[0039] like Figure 2 As shown, the system 300 includes the above-mentioned thinning machine tool 100 , a sensor assembly 310 , a computing device 400 , a carrying box 320 , a workbench 330 , and a grinding tape 340 .

[0040] The sensor assembly 310 includes a vibration sensor 311, an acoustic emission sensor 312, a plurality of force sensors 313, a vibration signal acquisition card 314, an acoustic emission signal acquisition card 315, and a force signal acquisition card 316. The vibration sensor 311 is connected to the computing device 400 via the vibration signal acquisition card 314, the acoustic emission sensor 312 is connected to the computing device 400 via the acoustic emission signal acquisition card 315, and the force sensor 313 is connected to the computing device 400 via the force signal acquisition card 316.

[0041] In this embodiment, the grinding wheel 120 is disposed below the spindle base 111 of the spindle 110 .

[0042] The carrying box 320 is fixed on the turntable 132 , the acoustic emission signal acquisition card 315 is arranged in the carrying box 320 , and the force sensor 313 , the acoustic emission sensor 312 and the force signal acquisition card 316 are arranged in the workbench 330 .

[0043] The workbench 330 is fixed on the carrying box 320. The shell surfaces of the workbench 330 and the carrying box 320 are arranged with wiring holes, through which the acoustic emission sensor 312 and the acoustic emission signal acquisition card 315 are connected, and the force sensor 313 and the force sensor acquisition card 316 are connected.

[0044] The vibration sensor 311 is adsorbed on the side of the main shaft 110 .

[0045] The vibration sensor 311 is used to collect vibration signals of the spindle 110 , the acoustic emission sensor 312 is used to collect elastic wave signals of the grinding wheel 120 and the wafer workpiece 200 , and the force sensor 313 is used to collect pressure signals of the grinding wheel 120 on the wafer workpiece 200 .

[0046] The computing device 400 starts to execute after receiving the sensor signal sent by the sensor assembly 310. The computing device 400 is coupled to the sensor assembly 310 and the thinning lathe 110 respectively, and is used to obtain the sensor signal from the sensor assembly 310, and execute the wafer thinning and grinding control method 600, obtain the process parameters for controlling the thinning machine 110, and control the thinning lathe 110 to process the wafer workpiece 200 based on the process parameters. For the specific description of the wafer thinning and grinding control method 600, please refer to the following, and this application will not repeat it here.

[0047] The grinding tape 340 is arranged on the workbench 330 and abuts against the wafer workpiece 200 to grind the wafer workpiece 200 .

[0048] In some embodiments, the system 300 further includes a water cooling device 350 and a battery 360. The water cooling device 350 is used to control the temperature of the grinding wheel 120 and the wafer workpiece 200 during the thinning and grinding process, and the battery 360 is used to provide power to various electrical devices (such as various sensors, acquisition cards, etc.).

[0049] The computing device 400 may be implemented as a server, such as an application server, a Web server, etc.; it may also be implemented as a desktop computer, a laptop computer, a processor chip, a tablet computer, etc., but is not limited thereto. Figure 3 4 shows a block diagram of the physical components (i.e., hardware) of a computing device 400. In a basic configuration, the computing device 400 includes at least one processing unit 402 and a system memory 404. According to one aspect, depending on the configuration and type of the computing device, the system memory 404 includes, but is not limited to, volatile storage (e.g., random access memory), non-volatile storage (e.g., read-only memory), flash memory, or any combination of such memories.

[0050] According to one aspect, system memory 404 includes operating system 405. System memory 404 also includes program modules 450. According to one aspect, operating system 405, for example, is suitable for controlling the operation of computing device 400. Furthermore, examples are practiced in conjunction with graphics libraries, other operating systems, or any other application programs, and are not limited to any particular application or system. Figure 3 This basic configuration is illustrated in FIG. 4 by those components within dashed line 408. According to one aspect, computing device 400 has additional features or functionality. For example, according to one aspect, computing device 400 includes additional data storage devices (removable and / or non-removable), such as magnetic disks, optical disks, or tapes. Such additional storage Figure 3 4 is illustrated by a removable storage device 409 and a non-removable storage device 410.

[0051] As stated above, according to one aspect, a program module 450 is stored in the system memory 404. According to one aspect, the program module 450 can be implemented as one or more computer program products. The present application does not limit the type of computer program products, and can include, for example, email, word processing application, spreadsheet application, database application, slide show application, painting or computer-aided application, web browser, etc. In some embodiments according to the present application, the computer program / instructions related to the wafer thinning and grinding control method 600 are packaged into a computer program product. When these computer programs / instructions are executed by the processor (i.e., the processing unit 402), the wafer thinning and grinding control method 600 according to the present application is implemented.

[0052] According to one aspect, the examples may be practiced on a circuit comprising discrete electronic components, a packaged or integrated electronic chip containing logic gates, a circuit utilizing a microprocessor, or a single chip containing electronic components or a microprocessor. Figure 3 Each or many components shown in can be integrated into a system on chip (SOC) on a single integrated circuit to practice examples. According to one aspect, such a SOC device may include one or more processing units, a graphics unit, a communication unit, a system virtualization unit, and various application functions, all of which are integrated (or "burned") into a chip substrate as a single integrated circuit. When operated via SOC, the functions described in this article can be operated via a dedicated logic integrated with other components of the computing device 400 on a single integrated circuit (chip). Other technologies that can perform logical operations (such as AND, OR, and NOT) can also be used to practice embodiments of the present invention, including but not limited to mechanical, optical, fluid, and quantum technologies. In addition, embodiments of the present invention can be practiced in a general-purpose computer or in any other circuit or system.

[0053] According to one aspect, the computing device 400 may also have one or more input devices 412, such as a keyboard, a mouse, a pen, a voice input device, a touch input device, etc. Output devices 414 may also be included, such as a display, a speaker, a printer, etc. The aforementioned devices are examples and other devices may also be used. The computing device 400 may include one or more communication connections 416 that allow communication with other computing devices 518, which may be printing devices, such as printers. Examples of suitable communication connections 416 include, but are not limited to: RF transmitters, receivers and / or transceiver circuits; Universal Serial Bus (USB), parallel and / or serial ports.

[0054] The term computer-readable medium as used herein includes computer storage media. Computer storage media may include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (e.g., computer-readable instructions, data structures, or program modules). System memory 404, removable storage device 409, and non-removable storage device 510 are all examples of computer storage media (i.e., memory storage). Computer storage media may include random access memory (RAM), read-only memory (ROM), electrically erasable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage, cassettes, tapes, disk storage or other magnetic storage devices, or any other products that can be used to store information and can be accessed by computing device 400. According to one aspect, any such computer storage medium can be a part of computing device 400. Computer storage media does not include carrier waves or other propagated data signals.

[0055] According to one aspect, communication media is implemented by computer readable instructions, data structures, program modules, or other data in a modulated data signal (e.g., a carrier wave or other transport mechanism), and includes any information delivery media. According to one aspect, the term "modulated data signal" describes a signal that has one or more characteristics set or changed in a manner that encodes information in the signal. By way of example and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0056] In an embodiment according to the present invention, the computing device 400 is configured to execute the wafer thinning and grinding control method 600 according to the present invention. The computing device 400 includes one or more processors and one or more readable storage media storing program instructions. When the program instructions are configured to be executed by one or more processors, the computing device executes the wafer thinning and grinding control method 600 in the embodiment of the present invention.

[0057] Figure 4 The flowchart of a wafer thinning and grinding control method 600 according to one embodiment of the present invention is shown. The method 600 is suitable for being executed in a computing device (such as the aforementioned computing device 400).

[0058] like Figure 4 As shown, the purpose of method 600 is to implement a method for determining the process parameters of a grinding machine tool according to the grinding state monitored during the grinding process, without adjusting the inclination angle and conducting a large number of process tests to determine the optimal grinding process parameters, thereby improving the processing efficiency and processing quality of wafer workpieces.

[0059] Method 600 begins with step 602. In step 602, feature extraction is performed on a sensor signal from a sensor assembly to obtain multiple feature parameters. The sensor assembly includes at least a vibration sensor, an acoustic emission sensor, and a force sensor. The vibration sensor is used to collect vibration signals of a spindle of a thinning machine tool. The acoustic emission sensor is used to collect elastic wave signals of a grinding wheel of a thinning machine tool and a wafer workpiece. The force sensor is used to collect pressure signals of a grinding wheel of a thinning machine tool on a wafer workpiece.

[0060] In some embodiments, the characteristic parameters include vibration amplitude, acoustic emission amplitude, acoustic emission effective voltage, and force amplitude.

[0061] In some embodiments, the vibration amplitude is extracted in the following manner: first, the vibration signal is filtered.

[0062] Then, the filtered vibration signal is subjected to time-frequency conversion using fast Fourier transform to obtain the signal spectrum.

[0063] Finally, the spectrum amplitude is extracted from the signal spectrum as the vibration amplitude.

[0064] The vibration sensor filters the vibration signal collected in the direction of the spindle rotation of the thinning machine tool and transmits it to the computing device. The vibration signal is then converted into a signal spectrum using the fast Fourier transform. The spectrum diagram in the rotation axis direction and the amplitude characteristics of the spectrum are displayed on the display interface. The vibration amplitude reflects the spindle state during grinding and indirectly reflects the surface processing state of the workpiece.

[0065] In some embodiments, the acoustic emission amplitude and the acoustic emission effective voltage are extracted in the following manner: first, the elastic wave signal is filtered and amplified.

[0066] Then, wavelet transform is used to perform time-frequency analysis on the elastic wave signal after filtering and amplification to obtain the amplitude spectrum and power spectrum.

[0067] Finally, the AE amplitude and AE effective voltage are extracted from the amplitude spectrum and power spectrum, respectively.

[0068] The acoustic emission sensor collects the elastic waves generated by the contact grinding between the grinding wheel and the workpiece. The collected acoustic emission signals are filtered and amplified and then transmitted to the computing device. The wavelet transform is then used for time-frequency analysis. The amplitude spectrum and power spectrum characteristics are displayed on the display interface, and the acoustic emission amplitude and acoustic emission effective voltage are extracted. The acoustic emission amplitude and acoustic emission effective voltage reflect the wear state of the grinding wheel and the surface processing state of the workpiece.

[0069] In some embodiments, the force amplitude is extracted by first filtering the pressure signal.

[0070] Then, the pressure time domain diagram is generated using the filtered pressure signal.

[0071] Finally, the force amplitude is extracted from the pressure-time domain diagram.

[0072] It should be noted that each sensor in the sensor assembly of the present application collects data once every predetermined period, and each collected data is sent to the computing device in real time.

[0073] The force sensor collects the force in the normal direction when the grinding wheel grinds the workpiece. The collected force signal is filtered and transmitted to the computing device. The time domain diagram of the force recorded by multiple force sensors is displayed on the display interface, and the force amplitude is extracted. The force amplitude reflects the wear state of the grinding wheel and the surface processing state of the workpiece.

[0074] Then, in step 604, the characteristic parameter is input into the grinding state prediction model for processing to obtain the state parameter indicating the grinding state.

[0075] In some embodiments, the grinding state prediction model adopts a feedforward neural network model (BP model), and the state parameters include the grinding wheel wear level and the roughness value of the wafer workpiece surface.

[0076] In a specific example, Figure 5 A schematic diagram of a feedforward neural network model according to an embodiment of the present invention is shown.

[0077] like Figure 5 As shown in Figure 1, the structure of the BP neural network includes an input layer, an output layer, and several hidden layers. Each layer has several nodes, and the connection status of the nodes between layers is reflected by weights.

[0078] Among them, the number of nodes in the input layer is equal to the dimension of the input, the number of nodes in the output layer is equal to the dimension of the output, and the number of nodes in the hidden layer can be set according to actual conditions.

[0079] In a BP neural network, each node contains a perceptron (i.e., a single neuron), which contains input items, weights, biases, activation functions, and outputs. During the forward propagation process, the input data is calculated by the perceptron node and then processed by the activation function to obtain the output result; during the backward propagation process, the result is compared with the expected result, and the weights of each node on the network are continuously adjusted through multiple iterations.

[0080] The vibration amplitude, acoustic emission amplitude, acoustic emission effective voltage and force amplitude are used as the input layer, and the output signal is generated through the forward propagation of the BP neural network. If the actual output does not match the expected output, it will enter the error direction propagation process.

[0081] In this embodiment, the expected outputs include two, namely, the wear level of the grinding wheel and the surface processing state of the wafer workpiece, wherein the state parameters corresponding to the two expected outputs are the wear level of the grinding wheel and the roughness value of the surface of the wafer workpiece.

[0082] The wear level of the grinding wheel is manually graded after observation under a microscope: A, slight (no dressing required), B, medium (dressing may be considered depending on the situation), C, severe (dressing required).

[0083] The surface processing state of the wafer workpiece is replaced by the measured roughness value Ra value of the wafer workpiece surface.

[0084] During training, the output layer will transmit gradient information in reverse, and adjust the weights and bias items of each layer to minimize the error between the predicted results and the actual results, thereby improving the accuracy of the model. After the training is completed, the parameters can be input from the input layer to obtain the predicted grinding wheel wear level and the roughness value of the wafer workpiece surface.

[0085] Then, in step 606, it is determined whether the current process parameters of the thinning lathe need to be optimized according to the state parameters. For example, if the grinding wheel wear is level A and the roughness value Ra of the wafer workpiece surface is within the threshold range, it means that no optimization is required; if the grinding wheel wear is level B or C, or the roughness value Ra of the wafer workpiece surface is not within the threshold range, it means that optimization is required.

[0086] Then, in step 608, if the current process parameters of the thinning lathe need to be optimized, the process parameters to be optimized are determined according to the current grinding state, and the process parameters include the grinding wheel speed, grinding wheel feed speed and turntable speed of the thinning machine.

[0087] Specifically, according to the current grinding state, characteristic parameters corresponding to the optimized grinding state are determined, and corresponding process parameters are matched from a process parameter sample database using the determined characteristic parameters.

[0088] The process parameter sample database associates and stores each process parameter with the corresponding characteristic parameter. When matching the corresponding process parameter from the process parameter sample database, the adjusted characteristic parameter is used as an index to match the corresponding process parameter from the process parameter sample database.

[0089] Furthermore, the process parameter sample database contains a large amount of data obtained from single-factor experiments (the experiment is conducted by controlling the change of one process parameter, and the results of the experiment are recorded as four characteristic parameters of the BP neural network input layer: vibration amplitude, acoustic emission amplitude, acoustic emission effective voltage, and force amplitude), and the relationship between each process parameter and the four characteristic parameters is obtained.

[0090] Generally speaking, the smaller the four characteristic parameters are, the better (the smaller the surface processing state is, the smaller the grinding wheel wear is). According to the four characteristic parameters obtained in this grinding, the computing device will give the corresponding characteristic parameter reduction plan (how much the three process parameters are increased / decreased respectively). The adjusted characteristic parameters are used to match the corresponding process parameters from the process parameter sample database. In addition, the value of the process parameter will set the upper and lower limits according to the processing efficiency and actual conditions, so as to ensure both processing efficiency and processing quality).

[0091] Finally, in step 610, the thinning lathe is controlled to process the wafer workpiece based on the process parameters.

[0092] In addition, if the current process parameters of the thinning lathe do not need to be optimized, the thinning lathe is controlled to still process the wafer workpiece with the current process parameters.

[0093] In some embodiments, if the current process parameters of the thinning lathe do not need to be optimized, the characteristic parameters are associated with the current process parameters, and the associated characteristic parameters and current process parameters are stored in a process parameter sample database.

[0094] The method provided by the present invention obtains relevant data of a wafer workpiece during the grinding process, extracts features of the relevant data, obtains turntable parameters indicating the current grinding state according to the extracted feature parameters, determines process parameters according to the state parameters, and then controls the thinning machine tool to process the wafer workpiece according to the process parameters. The scheme of the present invention can determine the process parameters of the grinding machine tool according to the grinding state monitored during the grinding process, without adjusting the inclination angle and conducting a large number of process tests to determine the optimal grinding process parameters, thereby improving the processing efficiency and processing quality of the wafer workpiece.

[0095] A7. The method described in A2, wherein it further includes: if the current process parameters of the thinning lathe do not need to be optimized, then the thinning lathe is controlled to still process the wafer workpiece with the current process parameters. A8. The method described in A2, wherein the grinding state prediction model adopts a feedforward neural network model, and the state parameters include the grinding wheel wear level and the roughness value of the wafer workpiece surface. A9. The method described in A2, wherein the process parameter sample database stores each process parameter and the corresponding characteristic parameter in association, and uses the determined characteristic parameter to match the corresponding process parameter from the process parameter sample database, including: using the adjusted characteristic parameter as an index to match the corresponding process parameter from the process parameter sample database. A10. The method described in A9, wherein if the current process parameters of the thinning lathe do not need to be optimized, then the thinning lathe is controlled to still process the wafer workpiece with the current process parameters. A11. The method described in A10, wherein it further includes: associating the characteristic parameter with the current process parameter, and storing the associated characteristic parameter and the current process parameter in the process parameter sample database. B12. The system as described in B11, further comprising: a vibration signal acquisition card connected to the vibration sensor and the computing device, respectively, for sending the vibration signal collected by the vibration sensor to the computing device; an acoustic emission signal acquisition card connected to the acoustic emission sensor and the computing device, respectively, for sending the elastic wave signal collected by the acoustic emission sensor to the computing device; a force signal acquisition card connected to the force sensor and the computing device, respectively, for sending the pressure signal collected by the force sensor to the computing device. B13. The system as described in B12, further comprising: a carrying box arranged on the turntable, and a workbench set on the carrying box; the acoustic emission signal acquisition card is arranged in the carrying box, and the force sensor, the acoustic emission sensor and the force signal acquisition card are arranged in the workbench; the shell surfaces of the carrying box and the workbench are both arranged with wiring holes, through which the acoustic emission sensor and the acoustic emission signal acquisition card are connected, and the force sensor and the force signal acquisition card are connected. B14. The system as described in B13, further comprising: a grinding tape arranged on the workbench, abutting against the wafer workpiece, and used to grind the wafer workpiece. B15. The system as described in any one of B11-B14, further comprising: a power supply arranged in the carrying box for providing power to various components in the system. B16. The system as described in B11, further comprising: a water cooling device for controlling the temperature of the grinding wheel and the wafer workpiece during the thinning and grinding process.

[0096] The various techniques described herein may be implemented in combination with hardware or software, or a combination thereof. Thus, the method and apparatus of the present invention, or some aspects or parts of the method and apparatus of the present invention may be in the form of program codes (i.e., instructions) embedded in a tangible medium, such as a removable hard disk, a USB flash drive, a floppy disk, a CD-ROM, or any other machine-readable storage medium, wherein when the program is loaded into a machine such as a computer and executed by the machine, the machine becomes a device for practicing the present invention.

[0097] In the case where the program code is executed on a programmable computer, the computing device generally includes a processor, a storage medium readable by the processor (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. The memory is configured to store the program code; the processor is configured to execute the method of the present invention according to the instructions in the program code stored in the memory.

[0098] By way of example and not limitation, readable media include readable storage media and communication media. Readable storage media stores information such as computer readable instructions, data structures, program modules or other data. Communication media generally embody computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and include any information transfer medium. Any combination of the above is also included within the scope of readable media.

[0099] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to implement the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the preferred embodiment of the present invention.

[0100] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0101] Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be divided into multiple submodules.

[0102] In addition, some of the embodiments are described herein as methods or combinations of method elements that can be implemented by a processor of a computer system or by other devices that perform the functions. Therefore, a processor with necessary instructions for implementing the method or method elements forms a device for implementing the method or method elements. In addition, the elements described herein of the device embodiments are examples of devices for implementing the functions performed by the elements for the purpose of implementing the invention.

[0103] As used herein, unless otherwise specified, the use of ordinal numbers "first," "second," "third," etc. to describe common objects merely indicates that different instances of similar objects are involved, and is not intended to imply that the objects so described must have a given order in time, space, order, or in any other manner.

[0104] Although the present invention has been described in terms of a limited number of embodiments, it will be apparent to those skilled in the art, having benefit of the above description, that other embodiments are contemplated within the scope of the invention thus described. Furthermore, it should be noted that the language used in this specification is selected primarily for readability and didactic purposes, rather than for the purpose of explaining or defining the subject matter of the present invention. Therefore, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims.

Claims

1. A wafer thinning and grinding control method, executed in a computing device, the method comprising: Extracting features from sensor signals from a sensor assembly to obtain a plurality of feature parameters, wherein the sensor assembly comprises at least a vibration sensor, an acoustic emission sensor and a force sensor, wherein the vibration sensor is used to collect vibration signals of a spindle of a thinning machine tool, the acoustic emission sensor is used to collect elastic wave signals of a grinding wheel of a thinning machine tool and a wafer workpiece, and the force sensor is used to collect pressure signals of a grinding wheel of a thinning machine tool on a wafer workpiece; Inputting the characteristic parameter into a grinding state prediction model for processing to obtain a state parameter indicating a grinding state; According to the state parameters, determining whether the current process parameters of the thinning lathe need to be optimized; If necessary, according to the current grinding state, the optimized process parameters are determined, wherein the process parameters include the grinding wheel speed, the grinding wheel feed speed and the turntable speed of the thinning machine tool; The thinning lathe is controlled to process the wafer workpiece based on the optimized process parameters.

2. The method of claim 1, wherein: According to the current grinding status, determine the optimized process parameters, including: According to the current grinding state, determining the characteristic parameters corresponding to the optimized grinding state; The determined characteristic parameters are used to match corresponding process parameters from a process parameter sample database.

3. The method of claim 1, wherein: The characteristic parameters include vibration amplitude, acoustic emission amplitude, acoustic emission effective voltage and force amplitude.

4. The method of claim 3, wherein: Feature extraction is performed on the sensor signal collected by the current sensor component to obtain multiple feature parameters, including: Performing filtering processing on the vibration signal; The filtered vibration signal is subjected to time-frequency conversion by using fast Fourier transform to obtain the signal spectrum; A spectrum amplitude is extracted from the signal spectrum as the vibration amplitude.

5. The method of claim 3, wherein: Feature extraction is performed on the sensor signal collected by the current sensor component to obtain multiple feature parameters, including: filtering and amplifying the elastic wave signal; Wavelet transform is used to perform time-frequency analysis on the elastic wave signal after filtering and amplification to obtain the amplitude spectrum and power spectrum. The acoustic emission amplitude and the acoustic emission effective voltage are extracted from the amplitude spectrum and the power spectrum respectively.

6. The method of claim 3, wherein: Feature extraction is performed on the sensor signal collected by the current sensor component to obtain multiple feature parameters, including: Performing filtering processing on the pressure signal; Generate a pressure time domain diagram using the filtered pressure signal; The force amplitude is extracted from the pressure time domain graph.

7. A wafer thinning and grinding control system, comprising: A thinning lathe, comprising a grinding assembly including a spindle, a grinding wheel arranged at the bottom of the spindle, and a turntable arranged opposite to the grinding wheel, suitable for grinding a wafer workpiece. When grinding, the wafer workpiece is fixed on the turntable, and the grinding wheel and the wafer workpiece abut against each other; A sensor assembly, comprising a vibration sensor, an acoustic emission sensor and a plurality of force sensors, wherein the vibration sensor is adsorbed on the side wall of the spindle, the acoustic emission sensor and the force sensor are arranged on the turntable, the vibration sensor is used to collect vibration signals of the spindle of the thinning machine tool, the acoustic emission sensor is used to collect elastic wave signals of the grinding wheel of the thinning machine tool and the wafer workpiece, and the force sensor is used to collect pressure signals of the grinding wheel of the thinning machine tool on the wafer workpiece; A computing device is coupled to the sensor assembly and the thinning lathe, respectively, for acquiring a sensor signal from the sensor assembly, and executing the method described in any one of claims 1 to 6 to obtain process parameters for controlling the thinning machine, and controlling the thinning lathe to process the wafer workpiece based on the process parameters.

8. A computing device comprising: at least one processor; and A memory storing program instructions, wherein the program instructions are configured to be executed by the at least one processor, and the program instructions include instructions for executing the method according to any one of claims 1 to 6.

9. A computer program product comprising a computer program / instructions, wherein: When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A readable storage medium storing program instructions, when the program instructions are read and executed by a computing device, the computing device executes the method according to any one of claims 1 to 6.

Images