Equipment end wafer caching system for semiconductor production

Abstract

The invention discloses an equipment end wafer caching system for semiconductor production, and particularly relates to the field of semiconductor manufacturing equipment, which comprises a task initialization module, an intelligent scheduling module, a round box transportation module, a caching scheduling module, an exception handling module and a task feedback module, according to the task initialization module, a production management system MES generates a wafer box carrying task according to the production plan; according to the intelligent scheduling module, the core scheduling control unit generates an optimal carrying strategy based on the carrying task; the intelligent scheduling system MCS monitors the state and position of the wafer box and the requirements of a production line in real time, the optimal wafer box carrying path and storage strategy can be selected, the carrying task is issued, it is ensured that the wafer box efficiently flows among the cache groove, the STK storage area, the SMIF equipment and the OHS air shuttle vehicle, the waiting time and invalid carrying are reduced, and the production efficiency is improved. And the overall production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of semiconductor manufacturing equipment, and more particularly, to an equipment-side wafer cache system for semiconductor production. Background Art

[0002] Wafer cassette processing, handling, and storage are crucial components of semiconductor fab production. These processes require handling and managing large numbers of wafers in various states to ensure production line continuity and efficiency. In actual production, wafer cassettes frequently need to be exchanged between the buffer tank, the STK storage tower, and the SMIF equipment. Wafer cassettes can be transferred from the buffer tank via the NTB equipment-side wafer buffer system to an automated access point. This docking with a shuttle vehicle (OHS) or overhead transport system (OHS) allows for rapid transfer of wafer cassettes between different equipment. Alternatively, wafer cassettes can be transferred directly from the buffer tank to processing tools to meet immediate production line needs. Furthermore, the OHS aerial shuttle, a crucial transport tool, transfers the wafer cassettes it carries to the STK automated access point. The STK is responsible for delivering the received wafer cassettes to designated storage locations for storage. This storage method not only ensures efficient cassette management but also provides sufficient material support for subsequent production. Throughout the production process, the exchange, transportation, and storage of wafer cassettes require precise control and coordination.

[0003] However, practical challenges remain. Traditional storage methods, such as flat storage bins, suffer from low space utilization. Limited space in production workshops, restricting storage capacity expansion, increases storage pressure on wafer cassettes and impacts production continuity. During the OHS transport process, poor routing planning often leads to multiple vehicle conflicts, resulting in low transport efficiency and increased production cycle time. A lack of effective coordination between the NTB, STK, SMIF, and OHS makes it difficult to share operational information in real time. This leads to long equipment wait times and irrational task allocation during production, making it impossible to dynamically adjust to the actual needs of the production line, severely impacting overall production efficiency. Tasks are issued from the MES, with wafer cassettes being picked up from the STK, scheduled by the MCS, temporarily stored at the NTB, and finally transported to the processing area. This process is prone to congestion. When the NTB is undercapacity or scheduling is delayed, the OHS queues at the entrance. Multiple task conflicts can also hinder OHS transport. Unready process equipment can cause cargo to be stranded at the NTB, significantly reducing production efficiency.

[0004] Therefore, there is an urgent need for a device-side wafer cache system for semiconductor production to solve the problems of low storage efficiency, low transportation efficiency, cross-device collaboration and congestion in the existing technology; by adopting equipment and technical means such as NTB, STK, SMIF and OHS, the efficiency, safety and stability of the wafer box in each link can be ensured; at the same time, in conjunction with the software control system, intelligent scheduling of the entire production process can be realized. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an equipment-side wafer cache system for semiconductor production, which solves the problems raised in the above-mentioned background technology through the following scheme.

[0006] To achieve the above-mentioned object, the present invention provides the following technical solution: a wafer buffer system at the equipment end for semiconductor production, characterized by comprising an NTB mechanical structure, a transfer mechanism, a wafer storage library, and a scheduling control unit, specifically comprising: a task initialization module: a production management system MES generates a wafer box handling task according to a production plan;

[0007] Intelligent scheduling module: The core scheduling control unit generates the optimal transportation strategy based on the transportation task;

[0008] Wafer box transport module: Based on the optimal handling strategy STK, the wafer box is taken out according to the instructions and handed over to the OHS. The OHS is transported to the NTB automatic entrance and exit via the elevated track, and the NTB transfer mechanism receives the wafer box;

[0009] Cache Scheduling Module: The MCS issues access instructions to the NTB based on the needs of the process equipment. The NTB robotic arm grabs the wafer box from the cache slot and, through the coordination of the ceiling rail and lifting mechanism, delivers it to the SMIF temporary storage area or directly connects it to the process equipment.

[0010] Abnormal processing module: The system integrates advanced alarm functions to monitor the equipment status throughout the process;

[0011] Task feedback module: After the wafer box arrives at the destination, the equipment provides feedback to the NTB, which synchronizes the information to the MES via the MCS, and the MES updates the wafer status. The system records task data for optimized scheduling and equipment maintenance, completing the task closed loop.

[0012] Preferably, the NTB mechanical structure specifically includes a long track body, a built-in buffer slot, an adjustable speed lifting mechanism, an automatic entrance and exit seamlessly connected to the OHS, and a robotic arm with integrated sensors; the long track body has a smooth motion path and precise guiding function; the built-in buffer slot is used for wafer storage; the adjustable speed lifting mechanism is used to achieve rapid vertical positioning of the wafer box; the OHS is an aerial walking shuttle that can travel on an elevated track in the air, used to transport carriers between designated storage equipment, thereby driving the carriers for transportation, and seamlessly connecting with the NTB equipment and the STK equipment; the robotic arm with integrated sensors is used to detect the status of the wafer box, the existence of the storage location, and the position of the carrier, and transmit the signal to the scheduling control unit to complete the precise grasping and transportation of the wafer box between the buffer slot and the equipment port; the transfer mechanism is composed of a robotic arm and a lifting mechanism; the scheduling control unit generates access instructions through an algorithm, coordinates the operation of the transfer mechanism and the intelligent conveyor, and achieves seamless connection with the OHS and full process automation of wafer access and transmission.

[0013] Preferably, the STK is an important component of the wafer cache system, which can store a large number of wafer boxes and is equipped with automatic entrances and exits. It can seamlessly connect with the OHS and quickly transfer the wafer boxes to the designated location. Subsequently, the transport instruction will reach the NTB automatic entrance and exit and seamlessly connect with the OHS. After the OHS transports the wafer box to the NTB, the NTB's overhead rail, lifting mechanism, and robotic arm start to work, accurately placing the wafer box on the SMIF and continuing to perform the transport task; the SMIF stores 4-6 units under each NTB device as a temporary storage area. The SMIF temporary storage area provides a closed environment, and the NTB device seamlessly connects with the SMIF temporary storage area below for quick access.

[0014] Preferably, the smooth motion path is used in the access process, and the system adopts an intelligent lifting control strategy, drives the lifting mechanism through a high-precision servo motor, and quickly increases the lifting speed in the initial stage; when the lifting mechanism approaches the target position, the system automatically switches to deceleration mode, and adjusts the descent speed in real time through closed-loop feedback control; a progressive slow stop algorithm is adopted in the deceleration stage; at the same time, the system is equipped with a high-resolution position sensor and a vibration monitoring module to collect position, speed and vibration data in real time during the lifting process, and dynamically adjust the deceleration curve through an adaptive control algorithm.

[0015] Preferably, the initial stage quickly increases the lifting speed v=v max , where v represents the initial velocity, v max Indicates the maximum speed set by the system; the deceleration mode adopts a quadratic polynomial deceleration curve Wherein, d represents the distance when the lifting mechanism approaches the target position, d0 represents the starting distance of deceleration, and d∈[0,d0]; the progressive slow-down algorithm is specifically expressed as a(t)=-k×v(t), where k represents the damping coefficient, which is obtained through adaptive adjustment of the vibration monitoring module, and v(t) represents the real-time speed.

[0016] Preferably, the specific method of dynamically adjusting the deceleration curve is as follows: (1) calculating the speed based on the position data collected by the position sensor, using the difference method to obtain the position p(t) and p(Δt) at discrete time points t and t+Δt, and the speed (2) Calculate the acceleration based on the velocity data. Also use the difference method to obtain the velocity v(t) and v(Δt) at discrete time points t and t+Δt. The acceleration (3) The vibration monitoring module collects vibration data and uses the root mean square value (RMS) to evaluate the vibration intensity. Suppose the vibration sensor collects N vibration data points x in a period of time T. i , i=1,2…N; vibration root mean square value (4) Use the proportional-integral-differential PID control algorithm to dynamically adjust the deceleration curve and calculate the position error e(t) = p(target) - p(t), where p(target) represents the target position; the control quantity output by the PID controller is Where e(τ) represents the error value of the system at the time variable τ, dT represents the small increment of time, represents the derivative of the error with respect to time, Kp is the proportional coefficient, K i is the integral coefficient, Kd is the differential coefficient, and the control quantity u(t) is used to adjust the acceleration or speed of the device, thereby dynamically adjusting the deceleration curve. The proportional link adjusts the control quantity according to the current error, the integral link eliminates the accumulated error, and the differential link predicts the error change trend. The three work together to enable the device to decelerate smoothly and reach the target position.

[0017] Preferably, the optimal handling strategy generates and optimizes access instructions based on the hardware status, system load and wafer access requirements, and sends them to NTB. After receiving the instructions, the system realizes the optimal configuration and efficient execution of tasks, and records the historical data of wafer access in real time, including access time, location information, and equipment operation status. The handling tasks issued by the system will pass through STK, OHS, NTB, and SMIF to run across devices. STK has automatic entrances and exits and seamlessly connects with OHS. After receiving the goods, the handling instructions will reach the NTB automatic entrance and exit and seamlessly connect with OHS. After OHS arrives at the designated position, NTB receives the instructions, and the overhead rail, lifting mechanism, and robotic arm start to work, and finally arrive at the SMIF temporary storage area, and automatically dispatch the wafer box according to production needs; if there is an abnormal situation, it has an abnormality detection and processing function, and makes corresponding processing according to the scenario. The system will verify the handling instructions, and can automatically identify and handle abnormal situations in the transportation process, and continue to execute the handling tasks after processing repeated tasks.

[0018] Preferably, the abnormal situations include blocking situation 1: when the capacity of the NTB intermediate buffer reaches its limit or the scheduling system is delayed, the OHS cannot transport the wafer box from the STK inventory area to the NTB in time, causing the OHS to queue at the entrance; blocking situation 2: multiple OHSs are queued due to path intersection or resource competition; blocking situation 3: the process equipment fails to feedback the status in time, causing the OHS to wait for the release of goods at the NTB exit, causing a backlog of subsequent tasks; the abnormal detection and processing function includes real-time monitoring of the NTB status in case of blocking situation 1, triggering an early warning and automatically sending the NTB to the NTB when the NTB capacity threshold reaches 80%. Low-priority goods are temporarily stored in the spare elastic buffer, and the OHS task issuance frequency is dynamically adjusted according to the real-time capacity of the NTB to avoid concentrated congestion causing the overhead crane to queue at the entrance; for the blocking situation 2, when OHS-A reaches the translation mechanism, the wafer box is temporarily stored in the NTB cache slot, and the track lock is released to allow OHS-B to enter; at the same time, the NTB moves the OHS-B goods to the Process, and the MCS manages the task sequence through the priority queue, and high-priority tasks can preempt the lock resources; for the blocking situation 3, the device status is collected in real time and pushed to the MCS through the communication protocol, triggering the access operation of the NTB goods release instruction.

[0019] Preferably, the alarm function at the hardware level will monitor various key parameters and operating status in real time. Once an abnormality is detected, the hardware will immediately report relevant information to the software system. However, at this time, the hardware itself will usually not intervene directly, but wait for further instructions from the software system. After the software system receives the abnormal information reported by the hardware, it will immediately start the preset alarm mechanism; this mechanism will not only display the alarm information on the software interface, but also remind the operator through various means such as sound and light; at the same time, the system will suspend the current handling task to prevent the abnormality from further expanding and affecting the production process; for different types of abnormal scenarios, the software system will perform intelligent processing according to preset rules and strategies, including automatically starting the wafer transfer program, transferring the wafer to a safe location, choosing to terminate the current task and deleting related records, or marking the task as an abnormal end state.

[0020] Preferably, the scheduling control unit adopts a dynamic goods transfer strategy: priority is given to the nearest handling task; when transferring goods across regions, a goods transfer plan is generated according to the inventory status to reduce backlog and out-of-stock costs: (1) the task distance is calculated by Manhattan distance priority: D = |x2-x1| + |y2-y1| + |z2-z1|, where (x1, y1, z1) represents the current warehouse coordinates and (x2, y2, z2) represents the target device coordinates; (2) the inventory cost is optimized by the objective function minC = C hold ×I+C s ×(Ds-I), where C hold represents the inventory holding cost, C s represents the out-of-stock cost, I represents the current inventory, and Ds represents the expected demand.

[0021] Technical effects and advantages of the present invention:

[0022] 1. The present invention uses the intelligent scheduling system MCS to monitor the status, location and production line needs of wafer boxes in real time, select the optimal wafer box transportation path and storage strategy, and issue transportation tasks to ensure the efficient flow of wafer boxes between the buffer tank, STK storage area, SMIF equipment and OHS aerial shuttle, reducing waiting time and ineffective transportation, and improving overall production efficiency;

[0023] 2. The present invention significantly improves the storage density of wafer cassettes through the three-dimensional structural design of the STK storage tower. Through a precise mechanical transmission system and positioning technology, wafer cassettes can be accurately delivered to or retrieved from designated storage locations, achieving efficient storage and rapid retrieval of wafer cassettes. This design not only saves valuable production space, but also provides sufficient material support for the production line, enhancing production flexibility and responsiveness.

[0024] 3. This system utilizes the OHS aerial shuttle as the primary transport tool, in conjunction with the NTB equipment-side wafer buffer system, to achieve precise docking and rapid transfer of wafer cassettes between different devices. High-precision sensors and control systems ensure the stability and safety of wafer cassettes during transport, preventing wafer damage or production interruptions caused by operational errors or equipment failures. A unified software control platform enables seamless collaboration between NTB, STK, SMIF, and OHS equipment. Each device can share data and information in real time, dynamically adjusting and optimizing according to the actual needs of the production line, ensuring the continuity and efficiency of the entire production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0026] Figure 2 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0028] As attached Figure 1 The device-side wafer buffer system shown in FIG. 1 for semiconductor production includes an NTB mechanical structure, a transfer mechanism, a wafer storage library, and a scheduling control unit.

[0029] In this embodiment, it is specifically necessary to explain that: the NTB mechanical structure specifically includes a long track body, a built-in buffer slot, a speed-adjustable lifting mechanism, an automatic entrance and exit seamlessly connected to the OHS, and a robotic arm with integrated sensors; the long track body has a smooth motion path and precise guidance function; the built-in buffer slot is used for wafer storage; the speed-adjustable lifting mechanism is used to achieve rapid vertical positioning of the wafer box; the OHS is an aerial shuttle vehicle that can travel on an elevated track in the air, and is used to transport carriers between designated storage equipment, thereby driving the carriers to be transported, and is connected to the OHS. The NTB equipment and the STK equipment are seamlessly connected; the robot arm with integrated sensors is used to detect the status of the wafer box, the existence of the storage location and the position of the carrier, and transmits the signal to the scheduling control unit to complete the precise grasping and transportation of the wafer box between the buffer slot and the equipment port; the transfer mechanism is composed of a robot arm and a lifting mechanism; the scheduling control unit generates access instructions through an algorithm, coordinates the operation of the transfer mechanism and the intelligent conveyor, and realizes seamless connection with the OHS and full process automation of wafer access and transmission; the STK is an important part of the wafer cache system and can store a large number of wafers. The wafer box is equipped with an automatic entrance and exit, which can be seamlessly connected with the OHS to quickly transfer the wafer box to the designated location. Subsequently, the handling instruction will reach the NTB automatic entrance and exit to seamlessly connect with the OHS. After the OHS transports the wafer box to the NTB, the NTB's overhead rail, lifting mechanism, and robotic arm begin to work, accurately placing the wafer box on the SMIF and continuing to perform the handling task; the SMIF is stored under each NTB device as a temporary storage area. The SMIF temporary storage area provides a closed environment, and the NTB device is seamlessly connected to the SMIF temporary storage area below for rapid Access; The smooth motion path is used in the access process. The system adopts an intelligent lifting control strategy, drives the lifting mechanism through a high-precision servo motor, and quickly increases the lifting speed in the initial stage; when the lifting mechanism approaches the target position, the system automatically switches to deceleration mode and adjusts the descent speed in real time through closed-loop feedback control; a progressive slow-stop algorithm is used in the deceleration stage; at the same time, the system is equipped with a high-resolution position sensor and a vibration monitoring module to collect position, speed and vibration data in real time during the lifting process, and dynamically adjust the deceleration curve through an adaptive control algorithm; the initial stage quickly increases the lifting speed v = v max , where v represents the initial velocity, v max Indicates the maximum speed set by the system; the deceleration mode adopts a quadratic polynomial deceleration curve Wherein, d represents the distance when the lifting mechanism approaches the target position, d0 represents the starting distance of deceleration, and d∈[0,d0]; the progressive slow-down algorithm is specifically expressed as a(t)=-k×v(t), where k represents the damping coefficient, which is obtained by adaptive adjustment of the vibration monitoring module, and v(t) represents the real-time speed; the specific method of dynamically adjusting the deceleration curve is as follows: (1) the speed is calculated according to the position data collected by the position sensor, and the position p(t) and p(Δt) are obtained at discrete time points t and t+Δt by using the difference method, and the speed (2) Calculate the acceleration based on the velocity data. Also use the difference method to obtain the velocity v(t) and v(Δt) at discrete time points t and t+Δt. The acceleration (3) The vibration monitoring module collects vibration data and uses the root mean square value (RMS) to evaluate the vibration intensity. Suppose the vibration sensor collects N vibration data points x in a period of time T. i , i=1,2…N; vibration root mean square value (4) Use the proportional-integral-differential PID control algorithm to dynamically adjust the deceleration curve and calculate the position error e(t) = p(target) - p(t), where p(target) represents the target position; the control quantity output by the PID controller is Where e(τ) represents the error value of the system at the time variable τ, dT represents the small increment of time, represents the derivative of the error with respect to time, Kp is the proportional coefficient, K i is the integral coefficient, Kd is the differential coefficient, the control quantity u(t) is used to adjust the acceleration or speed of the equipment, and then dynamically adjust the deceleration curve. The proportional link adjusts the control quantity according to the current error, the integral link eliminates the cumulative error, and the differential link predicts the error change trend. The three work together to make the equipment decelerate smoothly and reach the target position; the scheduling control unit adopts a dynamic goods transfer strategy: give priority to the nearest handling task; when transferring goods across regions, generate a goods transfer plan based on the inventory status to reduce backlog and out-of-stock costs: (1) prioritize the task distance through Manhattan distance calculation: D = |x2-x1| + |y2-y1| + |z2-z1|, where (x1, y1, z1) represents the current warehouse coordinates, and (x2, y2, z2) represents the target equipment coordinates; (2) optimize the objective function minC = C through inventory cost hold ×I+C s ×(Ds-I), where C hold represents the inventory holding cost, C s represents the out-of-stock cost, I represents the current inventory, and Ds represents the expected demand.

[0030] As attached Figure 2 The device-side wafer cache system for semiconductor production shown in FIG. specifically includes:

[0031] Task initialization module: The production management system (MES) generates wafer box handling tasks based on the production plan;

[0032] Intelligent scheduling module: The core scheduling control unit generates the optimal transportation strategy based on the transportation task;

[0033] In this embodiment, it is specifically necessary to explain that: the optimal handling strategy generates and optimizes access instructions based on the hardware status, system load and wafer access requirements, and sends them to NTB. After receiving the instructions, the system realizes the optimal configuration and efficient execution of tasks, and records the historical data of wafer access in real time, including access time, location information, and equipment operation status. The handling tasks issued by the system will pass through STK, OHS, NTB, and SMIF to run across devices. STK has automatic entrances and exits and seamlessly connects with OHS. After receiving the goods, the handling instructions will reach the automatic entrance and exit of NTB and seamlessly connect with OHS. After OHS arrives at the designated position, NTB receives the instructions, and the overhead rail, lifting mechanism, and robotic arm start to work, and finally arrive at the SMIF temporary storage area, and automatically dispatch the wafer box according to production needs; if there is an abnormal situation, it has an abnormality detection and processing function, and makes corresponding processing according to the scenario. The system will verify the handling instructions, and can automatically identify and handle abnormal situations in the transfer process, and continue to execute the handling tasks after processing repeated tasks. The abnormal situations include blocking situation 1: when the capacity of the NTB intermediate buffer reaches its limit or the scheduling system is delayed, the OHS cannot transport the wafer box from the STK inventory area to the NTB in time, causing the OHS to queue at the entrance; blocking situation 2: multiple OHSs are queued due to path intersection or resource competition; blocking situation 3: the process equipment does not feedback the status in time, causing the OHS to wait for the release of goods at the NTB exit, causing a backlog of subsequent tasks; the abnormal detection and processing function includes real-time monitoring of the NTB status in case of blocking situation 1, triggering an early warning and automatically transferring the low-priority task to the NTB when the NTB capacity threshold reaches 80%. Priority goods are temporarily stored in the spare elastic buffer, and the OHS task issuance frequency is dynamically adjusted according to the real-time capacity of the NTB to avoid concentrated congestion causing the overhead crane to queue at the entrance; for the blocking situation 2, when OHS-A reaches the translation mechanism, the wafer box is temporarily stored in the NTB cache slot, and the track lock is released to allow OHS-B to enter; at the same time, the NTB moves the OHS-B goods to the Process, and the MCS manages the task sequence through the priority queue, and high-priority tasks can preempt the lock resources; for the blocking situation 3, the device status is collected in real time and pushed to the MCS through the communication protocol, triggering the access operation of the NTB goods release instruction.

[0034] Wafer box transport module: Based on the optimal handling strategy STK, the wafer box is taken out according to the instructions and handed over to the OHS. The OHS is transported to the NTB automatic entrance and exit via the elevated track, and the NTB transfer mechanism receives the wafer box;

[0035] Cache Scheduling Module: The MCS issues access instructions to the NTB based on the needs of the process equipment. The NTB robotic arm grabs the wafer box from the cache slot and, through the coordination of the ceiling rail and lifting mechanism, delivers it to the SMIF temporary storage area or directly connects it to the process equipment.

[0036] Abnormal processing module: The system integrates advanced alarm functions to monitor the equipment status throughout the process;

[0037] In this embodiment, it is specifically necessary to explain that: the alarm function will monitor various key parameters and operating status in real time at the hardware level. Once an abnormality is detected, the hardware will immediately report relevant information to the software system. However, at this time, the hardware itself will usually not intervene directly, but wait for further instructions from the software system. After the software system receives the abnormal information reported by the hardware, it will immediately start the preset alarm mechanism; this mechanism will not only display the alarm information on the software interface, but also remind the operator through various means such as sound and light; at the same time, the system will suspend the current handling task to prevent the abnormality from further expanding and affecting the production process; for different types of abnormal scenarios, the software system will perform intelligent processing according to preset rules and strategies, including automatically starting the wafer transfer program, transferring the wafer to a safe location, choosing to terminate the current task and deleting related records, or marking the task as an abnormal end state.

[0038] Task feedback module: After the wafer box arrives at the destination, the equipment provides feedback to the NTB, which synchronizes the information to the MES via the MCS, and the MES updates the wafer status. The system records task data for optimized scheduling and equipment maintenance, completing the task closed loop.

[0039] Secondly: The drawings of the embodiments disclosed in the present invention only involve structures related to the embodiments disclosed in the present invention. Other structures may refer to conventional designs. The same embodiment and different embodiments of the present invention may be combined with each other without conflict.

[0040] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A wafer cache system for semiconductor production equipment, characterized in that: It includes NTB mechanical structure, transfer mechanism, wafer storage library and scheduling control unit, specifically including: Task initialization module: The production management system (MES) generates wafer box handling tasks based on the production plan; Intelligent scheduling module: The core scheduling control unit generates the optimal transportation strategy based on the transportation task; Wafer box transport module: Based on the optimal handling strategy STK, the wafer box is taken out according to the instructions and handed over to the OHS. The OHS is transported to the NTB automatic entrance and exit via the elevated track, and the NTB transfer mechanism receives the wafer box; Cache Scheduling Module: The MCS issues access instructions to the NTB based on the needs of the process equipment. The NTB robotic arm grabs the wafer box from the cache slot and, through the coordination of the ceiling rail and lifting mechanism, delivers it to the SMIF temporary storage area or directly connects it to the process equipment. Abnormal processing module: The system integrates advanced alarm functions to monitor the equipment status throughout the process; Task feedback module: After the wafer box arrives at the destination, the equipment provides feedback to the NTB, which synchronizes the information to the MES via the MCS, and the MES updates the wafer status. The system records task data for optimized scheduling and equipment maintenance, completing the task closed loop.

2. The device-side wafer cache system for semiconductor production according to claim 1, characterized in that: The NTB mechanical structure specifically includes a long track body, a built-in buffer slot, an adjustable speed lifting mechanism, an automatic entrance and exit that seamlessly connects with the OHS, and a robotic arm with integrated sensors; the long track body has a smooth motion path and precise guiding function; the built-in buffer slot is used for wafer storage; the adjustable speed lifting mechanism is used to achieve rapid vertical positioning of the wafer box; the OHS is an aerial shuttle that can travel on an elevated track in the air, used to transport carriers between designated storage equipment, thereby driving the carriers for transportation, and seamlessly connecting with the NTB equipment and the STK equipment; the robotic arm with integrated sensors is used to detect the status of the wafer box, the existence of the storage location, and the position of the carrier, and transmit the signal to the scheduling control unit to complete the precise grasping and transportation of the wafer box between the buffer slot and the equipment port; the transfer mechanism is composed of a robotic arm and a lifting mechanism; the scheduling control unit generates access instructions through an algorithm, coordinates the operation of the transfer mechanism and the intelligent conveyor, and achieves seamless connection with the OHS and full process automation of wafer access and transmission.

3. The device-side wafer cache system for semiconductor production according to claim 2, characterized in that: The STK is an important part of the wafer cache system. It can store a large number of wafer boxes and is equipped with automatic entrances and exits. It can seamlessly connect with the OHS to quickly transfer the wafer boxes to the designated location. Subsequently, the handling instructions will reach the NTB automatic entrance and exit and seamlessly connect with the OHS. After the OHS transports the wafer box to the NTB, the NTB's overhead rail, lifting mechanism, and robotic arm start working, placing the wafer box accurately on the SMIF and continuing the handling task; the SMIF stores 4-6 units under each NTB device as a temporary storage area. The SMIF temporary storage area provides a closed environment, and the NTB device seamlessly connects with the SMIF temporary storage area below for quick access.

4. The device-side wafer cache system for semiconductor production according to claim 2, characterized in that: The smooth motion path is used in the access process. The system adopts an intelligent lifting control strategy, which drives the lifting mechanism through a high-precision servo motor to quickly increase the lifting speed in the initial stage. When the lifting mechanism approaches the target position, the system automatically switches to deceleration mode and adjusts the descent speed in real time through closed-loop feedback control. A progressive slow-down algorithm is used in the deceleration stage; at the same time, the system is equipped with a high-resolution position sensor and a vibration monitoring module to collect position, speed and vibration data in real time during the lifting process, and dynamically adjust the deceleration curve through an adaptive control algorithm.

5. The device-side wafer cache system for semiconductor production according to claim 4, characterized in that: The initial stage quickly increases the lifting speed v=v max , where v represents the initial velocity, v max Indicates the maximum speed set by the system; the deceleration mode adopts a quadratic polynomial deceleration curve Wherein, d represents the distance when the lifting mechanism approaches the target position, d0 represents the starting distance of deceleration, and d∈[0,d0]; the progressive slow-down algorithm is specifically expressed as a(t)=-k×v(t), where k represents the damping coefficient, which is obtained through adaptive adjustment of the vibration monitoring module, and v(t) represents the real-time speed.

6. The device-side wafer cache system for semiconductor production according to claim 4, characterized in that: The specific method of dynamically adjusting the deceleration curve is as follows: (1) calculating the speed based on the position data collected by the position sensor, using the difference method to obtain the position p(t) and p(Δt) at discrete time points t and t+Δt, and the speed (2) Calculate the acceleration based on the velocity data. Also use the difference method to obtain the velocity v(t) and v(Δt) at discrete time points t and t+Δt. The acceleration (3) The vibration monitoring module collects vibration data and uses the root mean square value (RMS) to evaluate the vibration intensity. Suppose the vibration sensor collects N vibration data points x in a period of time T. i , i=1,2…N; vibration root mean square value (4) Use the proportional-integral-differential PID control algorithm to dynamically adjust the deceleration curve and calculate the position error e(t) = p(target) - p(t), where p(target) represents the target position; the control quantity output by the PID controller is Where e(τ) represents the error value of the system at the time variable τ, dT represents the small increment of time, represents the derivative of the error with respect to time, Kp is the proportional coefficient, K i is the integral coefficient, Kd is the differential coefficient, and the control quantity u(t) is used to adjust the acceleration or speed of the device, thereby dynamically adjusting the deceleration curve. The proportional link adjusts the control quantity according to the current error, the integral link eliminates the accumulated error, and the differential link predicts the error change trend. The three work together to enable the device to decelerate smoothly and reach the target position.

7. The device-side wafer cache system for semiconductor production according to claim 1, characterized in that: The optimal handling strategy generates and optimizes access instructions based on the hardware status, system load and wafer access requirements, and sends them to NTB. After receiving the instructions, the system realizes the optimal configuration and efficient execution of tasks, and records the historical data of wafer access in real time, including access time, location information, and equipment operation status. The handling tasks issued by the system will pass through STK, OHS, NTB, and SMIF to run across devices. STK has automatic entrances and exits to seamlessly connect with OHS. After receiving the goods, the handling instructions will reach the NTB automatic entrance and exit to seamlessly connect with OHS. After OHS arrives at the designated position, NTB receives the instructions, and the overhead rail, lifting mechanism, and robotic arm start to work, and finally arrive at the SMIF temporary storage area, and automatically dispatch the wafer box according to production needs; if there is an abnormal situation, it has an abnormality detection and processing function, and makes corresponding processing according to the scenario. The system will verify the handling instructions and can automatically identify and handle abnormal situations in the transfer process. At the same time, it will continue to execute the handling tasks after processing repeated tasks.

8. The device-side wafer cache system for semiconductor production according to claim 7, characterized in that: The abnormal situations include blocking situation 1: when the capacity of the NTB intermediate buffer reaches its limit or the scheduling system is delayed, the OHS cannot transport the wafer box from the STK inventory area to the NTB in time, causing the OHS to queue at the entrance; blocking situation 2: multiple OHSs are queued due to path intersection or resource competition; blocking situation 3: the process equipment does not feedback the status in time, causing the OHS to wait for the release of goods at the NTB exit, causing a backlog of subsequent tasks; the abnormal detection and processing function includes real-time monitoring of the NTB status in case of blocking situation 1, triggering an early warning and automatically transferring the low-priority task to the NTB when the NTB capacity threshold reaches 80%. Priority goods are temporarily stored in the spare elastic buffer, and the OHS task issuance frequency is dynamically adjusted according to the real-time capacity of the NTB to avoid concentrated congestion causing the overhead crane to queue at the entrance; for the blocking situation 2, when OHS-A reaches the translation mechanism, the wafer box is temporarily stored in the NTB cache slot, and the track lock is released to allow OHS-B to enter; at the same time, the NTB moves the OHS-B goods to the Process, and the MCS manages the task sequence through the priority queue, and high-priority tasks can preempt the lock resources; for the blocking situation 3, the device status is collected in real time and pushed to the MCS through the communication protocol, triggering the access operation of the NTB goods release instruction.

9. The device-side wafer cache system for semiconductor production according to claim 1, characterized in that: At the hardware level, the alarm function will monitor various key parameters and operating status in real time. Once an abnormality is detected, the hardware will immediately report relevant information to the software system. However, at this time, the hardware itself will not usually intervene directly, but wait for further instructions from the software system. After the software system receives the abnormal information reported by the hardware, it will immediately activate the preset alarm mechanism; this mechanism will not only display the alarm information on the software interface, but also remind the operator through various means such as sound and light; at the same time, the system will suspend the current handling task to prevent the abnormality from further expanding and affecting the production process; for different types of abnormal scenarios, the software system will perform intelligent processing according to preset rules and strategies, including automatically starting the wafer transfer program, transferring the wafer to a safe location, choosing to terminate the current task and deleting related records, or marking the task as an abnormal end state.

10. The device-side wafer cache system for semiconductor production according to claim 1, characterized in that: The scheduling control unit adopts a dynamic goods transfer strategy: priority is given to the nearest handling task; when transferring goods across regions, a goods transfer plan is generated based on the inventory status to reduce backlog and out-of-stock costs: (1) the task distance is calculated by Manhattan distance priority: D = |x2-x1| + |y2-y1| + |z2-z1|, where (x1, y1, z1) represents the current warehouse coordinates and (x2, y2, z2) represents the target device coordinates; (2) the inventory cost is optimized by the objective function minC = C hold ×I+C s ×(Ds-I), where C hold represents the inventory holding cost, C s represents the out-of-stock cost, I represents the current inventory, and Ds represents the expected demand.

Images