Three-dimensional modeling device and method for semiconductor factory

Abstract

The invention discloses a three-dimensional modeling device and method for a semiconductor factory. The method comprises the steps that an acquisition module acquires interface layout information and connection element data of the semiconductor factory; the equipment modeling module receives semiconductor factory interface layout information, determines a three-dimensional modeling starting point and generates a semiconductor factory equipment model containing an equipment interface; the connection element modeling module determines an element connection path by using semiconductor factory starting equipment or factory service interfaces, combining element interface matching rules and logic relation information among elements, and fusing a three-dimensional touch automatic checking algorithm and an optimal path algorithm based on safe distance correction; and interface connection among elements in each semiconductor factory equipment model in the equipment modeling module is realized, and three-dimensional modeling of the semiconductor factory is completed. According to the method, the layout information and the connection element information are fused, and three-dimensional modeling of the semiconductor factory is efficiently and accurately achieved through the logic relation between the connection elements and a three-dimensional touch automatic verification algorithm.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of semiconductor factory modeling, and particularly relates to a three-dimensional modeling device and method for a semiconductor factory. BACKGROUND

[0002] A semiconductor factory is a complex building space interwoven by extremely complex equipment layout and highly precise pipeline system. In the semiconductor factory, a huge and complex pipeline network needs to be integrated into the equipment layout to realize various core power and materials for maintaining the operation of the machine table. These pipelines have almost stringent requirements for cleanliness, pressure stability, and material compatibility, and any slight leakage or pollution can affect the production efficiency of the semiconductor factory. Therefore, the reliability of the pipeline delivery system directly determines the efficiency and yield of the semiconductor factory production line.

[0003] How to construct the layout of the pipeline network and the factory equipment, realize global optimization of equipment positioning, pipeline routing, and space collision detection, and achieve the effects of pre-solution of installation conflicts, accurate material statistics, and effective construction guidance. Three-dimensional design is the key to avoiding rework and saving costs for a semiconductor factory, and is also a solid cornerstone to ensure the efficient and stable operation of a semiconductor factory.

[0004] Patent CN117077459A discloses a semiconductor manufacturing simulation model construction method and device fusing logistics, which includes the following steps: based on the semiconductor manufacturing process, obtaining simulation manufacturing parameters, wherein the simulation manufacturing parameters include process parameters and simulation parameters; constructing an initial simulation model of simulated semiconductor manufacturing, wherein the initial simulation model includes a production module and a logistics module; based on the initial simulation model, and in combination with the simulation manufacturing parameters, setting interaction parameters, associating the production module with the logistics module, and forming a cycle of the production module and the logistics module connected by the interaction parameters; iteratively training the initial simulation model in combination with the updated interaction parameters until the index condition is met, and completing the construction of the fusing logistics semiconductor manufacturing simulation model.

[0005] The current three-dimensional design cannot combine complex building modeling and pipeline modeling, the semiconductor machine equipment interface is numerous (hundreds), there is a lack of means for quickly and batch drawing of equipment interfaces, the efficiency is low, and the error rate is high. At the same time, the current standardized three-dimensional component library and design rule library are still not perfect and popular, and rely on the experience of engineers, which is easy to produce design inconsistency.

[0006] Therefore, how to combine complex building modeling and pipeline modeling to realize efficient and accurate three-dimensional modeling of a semiconductor factory and ensure efficient and stable operation of the semiconductor factory is a problem to be solved by those skilled in the art. SUMMARY

[0007] In view of the defects in the prior art, the application provides a three-dimensional modeling device and method for a semiconductor factory, which comprises: a collection module, configured to acquire interface layout information of the semiconductor factory and connection element data, wherein the connection element data comprises element interface matching rules, element attribute information, and logical relationship information between elements; a device modeling module, configured to receive the interface layout information of the semiconductor factory from the collection module, determine a three-dimensional modeling starting point, and generate a semiconductor factory device model comprising device interfaces; and a connection element modeling module, configured to, based on the interface layout information of the semiconductor factory and the three-dimensional modeling starting point received from the collection module, determine an element type and an element connection path by taking a starting device of the semiconductor factory or a factory interface as a starting point, combining the element interface matching rules and the logical relationship information between elements received from the collection module, and fusing a three-dimensional touch automatic checking algorithm and an optimal path algorithm based on a safety distance correction, and realize interface connection between elements in each semiconductor factory device model in the device modeling module by combining the element attribute information of the collection module, thereby completing three-dimensional modeling of the semiconductor factory. The application fuses layout information and connection element information, and efficiently and accurately realizes three-dimensional modeling of the semiconductor factory by means of logical relationship between connection elements and a three-dimensional touch automatic checking algorithm.

[0008] In a first aspect, the application provides a three-dimensional modeling device for a semiconductor factory, comprising: a collection module, configured to acquire interface layout information of the semiconductor factory and connection element data, wherein the connection element data comprises element interface matching rules, element attribute information, and logical relationship information between elements; a device modeling module, configured to receive the interface layout information of the semiconductor factory from the collection module, determine a three-dimensional modeling starting point, and generate a semiconductor factory device model comprising device interfaces; a connection element modeling module, configured to, based on the interface layout information of the semiconductor factory and the three-dimensional modeling starting point received from the collection module, determine an element type and an element connection path by taking a starting device of the semiconductor factory or a factory interface as a starting point, combining the element interface matching rules and the logical relationship information between elements received from the collection module, and fusing a three-dimensional touch automatic checking algorithm and an optimal path algorithm based on a safety distance correction, and realize interface connection between elements in each semiconductor factory device model in the device modeling module by combining the element attribute information of the collection module, thereby completing three-dimensional modeling of the semiconductor factory.

[0009] Further, the connection element data is acquired, specifically comprising the following steps: acquiring element interface matching rules and initial element attribute information; labeling the initial element attribute information, and constructing logical relationship between elements based on the labels of the elements, thereby forming logical relationship information between elements.

[0010] Further, the element interface matching rules and the logical relationship information between elements are combined, and a three-dimensional touch automatic checking algorithm and an optimal path algorithm based on a safety distance correction are fused to determine the element type and the element connection path, and specifically include the following steps: In response to the determination of the three-dimensional modeling starting point, the element interface matching rules of the semiconductor factory starting equipment model or plant interface are obtained; According to the semiconductor factory interface layout information, the element connection logic is determined; Based on the logical relationship information between elements, and in combination with the element connection logic, all feasible element connection path sets and corresponding element types are given through matching and screening of the connection element data; Fusing the safety distance, and determining the initial element connection path from all feasible element connection path sets through the corrected optimal path algorithm; Based on the three-dimensional touch automatic checking algorithm, the elements in the initial element connection path are checked to determine the element type and the corresponding element connection path.

[0011] Further, the safety distance is fused, and the initial element connection path is determined from all feasible element connection path sets through the corrected optimal path algorithm, and specifically includes the following steps: Set the safety distance to determine the path cost function; Through the optimal path algorithm, the path length and the obstacle distance are fused to form the first element connection path set; The comprehensive cost of each feasible element connection path in all feasible element connection path sets and the comprehensive cost of each first element connection path in the first element connection path set are given respectively; The reference cost is determined by the comprehensive cost of each first element connection path in the first element connection path set; Based on the reference cost, all feasible element connection path sets are screened to form the second element connection path set; The Manhattan distance of each second element connection path in the second element connection path set from the nearest obstacle is analyzed, and in combination with the safety distance, the safety coefficient function of the second element connection path is given; Based on the evaluation screening function, the path length, the safety coefficient function and the path cost function are fused to evaluate each second element connection path to determine the initial element connection path.

[0012] Further, through the optimal path algorithm, the path length and the obstacle distance are fused to form the first element connection path set, and specifically include the following steps: Based on the semiconductor factory building space information, a grid layout map composed of unit cells is formed, and the obstacle position is determined; According to the breadth-first search, the diffusion is performed on all obstacles, and the Manhattan distance of each cell to the nearest obstacle is calculated; The path length and the obstacle penalty term are fused to form a semiconductor factory equipment node weight relationship, a modified optimal path algorithm is given, and a first component connection path set is generated.

[0013] Further, the path length and the obstacle penalty term are fused to form an equipment node weight relationship, a modified optimal path algorithm is given, and the specific representation is:

[0014] Wherein, cost(i) is the weight of the semiconductor factory equipment node i, length(i) is the path length from the starting point of the three-dimensional modeling to the semiconductor factory equipment node i, λ is the obstacle weight parameter, d obs (i) is the distance from the semiconductor factory equipment node i to the nearest obstacle, d safe is the safety distance.

[0015] Further, the safety coefficient function is specifically represented as:

[0016] The evaluation screening function is specifically represented as:

[0017] Wherein, S(P) is the safety coefficient function of the second component connection path, d obs (i) is the distance from the semiconductor factory equipment node i to the nearest obstacle, d safe is the safety distance,∏ i is the relative safety degree product of all semiconductor factory equipment nodes on the second component connection path, P k is the screening function of the optimal k paths, F(P) is the path cost function of the second component connection path, and length(P) is the total length of the second component connection path.

[0018] Further, based on the three-dimensional touch automatic checking algorithm, the components in the initial component connection path are checked to determine the component type and the corresponding component connection path, which specifically includes the following steps: The initial component connection path is combined with the attribute extraction function to extract component attribute information to form a set of components to be checked; The set of components to be checked is checked by the checking function; Based on the checking screening rule, the component type and the corresponding component connection path are determined.

[0019] Further, the attribute set of the to-be-verified element is verified by a verification function, and the verification function specifically comprises:

[0020] Wherein, The verification function of the element between the semiconductor factory equipment nodes A and B, m is the number of a certain attribute of the element, is an arbitrary symbol, M is an attribute set of the element, f m is an attribute extraction function, f m (P A ) is a specified attribute of the semiconductor factory equipment node A, f m (P B ) is a specified attribute of the semiconductor factory equipment node B.

[0021] Further, the method further comprises a collaborative design module and an output management module; The collaborative design module is connected with the equipment modeling module and the connecting element modeling module, and is used for acquiring information changes of the semiconductor factory equipment model and the interface connection between elements in the semiconductor factory equipment model and updating the information in time to realize multi-stage collaborative design. The output management module is connected with the connecting element modeling module, and is used for visualizing the semiconductor factory three-dimensional modeling output by the connecting element modeling module.

[0022] In a second aspect, the application further provides a three-dimensional modeling method of a semiconductor factory, which adopts the three-dimensional modeling device of the semiconductor factory according to any one of the above. Acquire the interface layout information of the semiconductor factory and the connecting element data, wherein the connecting element data comprises element interface matching rules, element attribute information and logical relationship information between elements. Based on the interface layout information of the semiconductor factory, determine a three-dimensional modeling starting point, and generate a semiconductor factory equipment model comprising an equipment interface. Based on the interface layout information of the semiconductor factory and the three-dimensional modeling starting point, determine the element type and the element connection path by taking the starting equipment of the semiconductor factory or the factory interface as a starting point, combining the element interface matching rules and the logical relationship information between elements, and fusing a three-dimensional touch automatic verification algorithm and an optimal path algorithm based on a safety distance correction, realize the interface connection between elements in each semiconductor factory equipment model, and complete the three-dimensional modeling of the semiconductor factory.

[0023] The three-dimensional modeling device and method of the semiconductor factory provided by the application have at least the following beneficial effects: (1) By setting the acquisition module, the semiconductor factory interface layout information and the connection component data including component interface matching rules, component attribute information and logical relationship information between components are obtained, providing data basis for deepening model information dimension and reducing professional use threshold, and storing two-dimensional process pipeline and PID information, which is cross-referenced and checked with three-dimensional information to achieve data consistency; through the interaction of the equipment modeling module and the connection component modeling module, the complex building modeling and pipeline modeling are combined to realize the reasonable layout of equipment across floors, automatically generate semiconductor factory equipment models containing equipment interfaces, reduce personnel error rate and save time cost; through the connection component modeling module, three-dimensional pipeline design is completed in limited space to ensure the power demand of equipment in the semiconductor factory, and at the same time, the three-dimensional touch automatic checking algorithm and the optimal path algorithm based on safety distance correction are fused to reduce the learning difficulty, improve the three-dimensional design efficiency and design quality, and efficiently and accurately realize the three-dimensional modeling of the semiconductor factory.

[0024] (2) By labeling the initial attribute information of the component, it can be ensured that the initial attribute information of the component is effectively labeled, and the logical relationship between components based on these labels can be accurately reflected in three-dimensional design, thereby improving design efficiency and quality.

[0025] (3) The EID model ensures the compatibility and interoperability between devices by detailing the interface information, and the clear interface description helps to quickly locate and solve problems, improving the maintenance efficiency of the equipment.

[0026] (4) By obtaining component interface matching rules, determining component connection logic, matching and filtering component connection paths, and finally determining the final connection path through three-dimensional touch automatic checking algorithm and optimal path algorithm, the three-dimensional modeling of the semiconductor factory is judged to ensure the accuracy and reliability of the three-dimensional modeling.

[0027] (5) By introducing the obstacle penalty term, the path planning algorithm can tend to select paths that have longer straight-line distances but fewer obstacles, considering not only the straight-line distance but also the influence of obstacles, thereby ensuring the feasibility and efficiency of the paths in the first component connection path set.

[0028] (6) By using three-dimensional touch automatic checking algorithm and optimal path algorithm to check the components in the initial component connection path, determining the component type and corresponding component connection path, automatic and intelligent path planning can be realized in semiconductor three-dimensional modeling, improving the layout efficiency of the factory and the accuracy of equipment pipeline connection. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 A structural block diagram of a semiconductor factory three-dimensional modeling device provided in an embodiment of the present application; Figure 2 An effect diagram of a semiconductor factory equipment model containing an equipment interface provided in an embodiment of the present application; Figure 3 A flow chart for determining element types and element connection paths provided in an embodiment of the present application; Figure 4 An effect diagram for determining element connection paths provided in an embodiment of the present application; Figure 5 A flow chart for determining initial element connection paths provided in an embodiment of the present application; Figure 6 A flow chart for generating a first element connection path set provided in an embodiment of the present application; Figure 7 A flow chart of a semiconductor factory three-dimensional modeling method provided in an embodiment of the present application.

[0030] Wherein, 10, a semiconductor factory equipment model; 11, an equipment interface; 201, a collection module; 202, an equipment modeling module; 203, a connection element modeling module. DETAILED DESCRIPTION

[0031] In order to better understand the above technical solutions, the above technical solutions will be described in detail below in combination with the drawings of the specification and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0032] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two.

[0033] It should also be noted that the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the goods or devices including a series of elements not only include those elements, but also include other elements not explicitly listed, or include elements inherent to such goods or devices. Without more limitation, the element defined by the sentence "including a" does not exclude the presence of other identical elements in the goods or devices including the element.

[0034] Currently, the three-dimensional design of the conveying pipeline in the semiconductor factory still faces significant challenges and shortcomings, as follows: (1) there is no three-dimensional design software for the semiconductor industry that can combine complex building modeling and pipeline modeling, building system software cannot quickly and conveniently design complex pipeline modeling and lacks professionalism, and pipeline modeling software cannot reasonably arrange thousands of devices; (2) standards and specifications lag, the standardized three-dimensional component library and design rule library for semiconductor-specific requirements (such as special materials of ultra-pure systems, double-sealing requirements, and fine layout of interface elements) are still not perfect and popular, relying on the experience of engineers, and inconsistent design is prone to occur; (3) there are numerous interfaces of machine devices in the semiconductor factory, and there is a lack of means for quickly and batch drawing of device interfaces, which is low in efficiency and high in error rate; (4) the current three-dimensional design model lacks depth and information integrity, many models focus on geometry and physical collision, but lack sufficient engineering attribute data (such as accurate material specifications, welding process records, non-destructive testing requirements, and ultra-high purity certification documents) and operation and maintenance information (such as valve serial numbers and maintenance manual links), which are difficult to support fine construction, quality control and intelligent operation and maintenance (digital twin foundation is weak); (5) the learning difficulty of the current three-dimensional modeling software is high, the skill threshold is high, the user interaction logic performance is poor, and the use personnel are required to have semiconductor process, pipeline engineering and three-dimensional software skills, resulting in uneven design quality; (6) the two-dimensional data information and the three-dimensional data information in the semiconductor factory are not unified, which causes trouble to the construction personnel.

[0035] Therefore, the three-dimensional modeling device and method of the semiconductor factory are provided, the device comprises: a collection module configured to acquire semiconductor factory interface layout information and connection element data, wherein the connection element data comprises element interface matching rules, element attribute information and logical relationship information between elements; a device modeling module configured to receive the semiconductor factory interface layout information of the collection module, determine a three-dimensional modeling starting point, and generate a semiconductor factory device model comprising device interfaces; and a connection element modeling module configured to determine an element type and an element connection path based on the semiconductor factory interface layout information and the three-dimensional modeling starting point of the collection module, the starting device of the semiconductor factory or the factory interface, the element interface matching rules and the logical relationship information between elements received by the collection module, the three-dimensional touch automatic checking algorithm and the optimal path algorithm based on the safety distance correction, and the element attribute information of the collection module, realize the interface connection between elements in each semiconductor factory device model in the device modeling module, and complete the three-dimensional modeling of the semiconductor factory. The three-dimensional modeling of the semiconductor factory is efficiently and accurately realized by the layout information and the connection element information, the logical relationship between the connection elements and the three-dimensional touch automatic checking algorithm.

[0036] By setting the acquisition module, the semiconductor factory interface layout information and the connection element data including the element interface matching rule, the element attribute information and the logical relationship information between each element are acquired, which provides a data basis for deepening the model information dimension and reducing the professional use threshold, and at the same time, the two-dimensional process pipeline and instrument flow chart (process and instrumentation diagram, PID) information is stored, which is cross-referenced and checked with the three-dimensional information to achieve data consistency. Through the interaction of the equipment modeling module and the connection element modeling module, the complex building modeling and pipeline modeling are combined, the reasonable layout of the equipment across floors is realized, the semiconductor factory equipment model containing the equipment interface is automatically generated, the personnel error rate is reduced, and the time cost is saved. Through the connection element modeling module, the three-dimensional pipeline design in the limited space is completed, the power demand of the machine equipment in the semiconductor factory is ensured, at the same time, the three-dimensional touch automatic checking algorithm and the optimal path algorithm based on the safety distance correction are fused, the learning difficulty is reduced, the three-dimensional design efficiency and design quality are improved, and the three-dimensional modeling of the semiconductor factory is efficiently and accurately realized.

[0037] As shown in Figure 1 The embodiment of the application provides a three-dimensional modeling device of a semiconductor factory, which specifically comprises: An acquisition module is used to acquire semiconductor factory interface layout information and connection element data, wherein the connection element data includes element interface matching rules, element attribute information and logical relationship information between each element. The interface layout information includes a two-dimensional and / or three-dimensional standard layout library, construction drawings, three-dimensional modeling design requirement information and requirement information data standards. The requirement information data standards include secondary piping engineering three-dimensional design requirement data information standard templates, specification data requirements and data types. For example, the machine power demand information standard template must include machine code (unique identifier), machine type, machine interface position and connection system information, flow information, pressure information, interface information, equipment end valve information and the like. For example, the interface information includes code, associated delivery and distribution system, pipeline color for distinguishing delivery and distribution system, size, pipe material, spatial position, angle and the like. The drawing standards include two-dimensional / three-dimensional standard libraries, construction drawings (PID / MLD) standards. For example, drawing range, reference standards (GB / T 14689-2008, ASME Y14.5-2018 and the like), drawing templates (A3 A4), standard styles (font, font size, arrow style and the like), coordinate system, view and the like.

[0038] An equipment modeling module is used to receive the semiconductor factory interface layout information of the acquisition module, determine a three-dimensional modeling starting point, and generate a semiconductor factory equipment model containing equipment interfaces. The connection element modeling module is used to determine the element type and element connection path based on the received semiconductor factory interface layout information and three-dimensional modeling starting point of the acquisition module, the semiconductor factory starting device or plant interface, the received element interface matching rule of the acquisition module and the logical relationship information between elements, and the three-dimensional touch automatic checking algorithm and the optimal path algorithm based on the safety distance correction, determine the element type and element connection path, combine the element attribute information of the acquisition module, realize the interface connection between elements in each semiconductor factory device model in the device modeling module, and complete the three-dimensional modeling of the semiconductor factory.

[0039] Further, the connection element data is acquired, specifically including the following steps: Acquiring element interface matching rules and element initial attribute information; Labeling the element initial attribute information, and constructing the logical relationship between elements based on the labels of the elements to form the logical relationship information between elements.

[0040] In a specific embodiment, the element interface matching rule is a rule that limits the technical specification and operation specification of the corresponding element of the interface, including the interface associated attribute information in the two-dimensional drawing (i.e. two-dimensional PID drawing), the logical relationship between three-dimensional elements, and the corresponding relationship between the interface in the two-dimensional drawing and the three-dimensional element. The element interface matching rule also includes the design specification requirement of the technical rule, the piping logic, etc. to reduce the difficulty of three-dimensional design specialty and improve the efficiency of three-dimensional modeling. The technical rule includes the pipe diameter selection rule, the pipe turning rule, etc. For example, the element interface matching rule includes the flange end face form, pressure matching, screw inner and outer connection matching, etc. The design specification requirement of the technical rule includes the use of SSC material for the exhaust system acid exhaust, the use of SUS316L-BA material for bulk gas pure N2, etc. The pipe turning rule includes a pipe turning angle of 90°, a pipe length of 60mm, etc.

[0041] It should be understood that the semiconductor factory can be generally divided into 11 kinds of delivery and distribution systems (ultra-pure water, exhaust gas, wastewater, etc. systems), and each kind of delivery and distribution system is provided with a sub-class delivery and distribution system. For example, the delivery and distribution system is an exhaust gas system, and the sub-class delivery and distribution systems such as acid exhaust, alkaline exhaust, and general exhaust are provided. The pipe materials used in different sub-class delivery and distribution systems are different or the same.

[0042] The element initial attribute information includes three-dimensional element geometric information and three-dimensional element engineering attributes, and is used to ensure correct connection and drawing of three-dimensional elements (pipes) of a plant terminal pipe interface and a device terminal interface and a device-to-device interface. The three-dimensional element geometric information includes basic dimensions (such as length, width, height, diameter, radius, thickness, etc.), shape types (such as rectangle, circle, triangle, polygon, etc.), edge types (such as right angle, round angle, chamfer, etc.), surface features (such as groove, protrusion, hole, etc.), spatial positions (such as position coordinates, orientation, inclination angle, etc.), assembly features (such as assembly hole, assembly surface, assembly direction, etc.), fitting types (such as clearance fit, interference fit, etc.), complex geometric features (such as complex curved surface, complex structure), geometric constraints (such as parallel, perpendicular, tangent, etc.), and geometric analysis parameters (such as barycenter, moment of inertia, etc.). The three-dimensional element engineering attributes include physical attributes (such as size, weight, material, color, surface treatment, etc.), electrical attributes (such as voltage level, current level, insulation level, etc.), thermal attributes (such as heating temperature, heat resistance, etc.), compliance attributes (such as safety certification, environmental protection requirement, industry standard, etc.), manufacturer attributes (such as manufacturer name, manufacturer standard, etc.), and pressure attributes (such as Class 150, 300, etc.).

[0043] The collected element initial attribute information is sorted, and a plurality of tags are determined according to the three-dimensional element geometric information and / or the three-dimensional element engineering attributes. The plurality of tags are classified, for example, the tags can be classified into categories such as “material”, “pressure”, “system”, “function”, etc.

[0044] According to the plurality of tags, the element initial attribute information of each element is tagged, and each element is assigned a corresponding tag. According to the tags of the elements, potential relationships between the elements are identified. For example, if two elements both have “electrical system” tags, they can be connected by a power line. Based on the identified relationships, logical relationships between the elements are constructed, including the order, direction and type of connection, etc. The constructed logical relationships are recorded to form logical relationship information between the elements. The logical relationship information describes in detail how each element is connected to other elements.

[0045] In a specific example, the tags are classified according to the functions and materials of the pipes, for example, the tags include pipes, pipe fittings, valves, devices, and interfaces. In a system requiring polyvinylidene difluoride (PVDF) pipes, the pipe fittings include tees, elbows, reducers, etc., the valve is a diaphragm valve, the device is a pure water machine, and the interfaces include flanges, socket welds, and union joints, etc. In a system requiring Teflon-coated stainless steel air pipes, the pipe fittings include tees, elbows, pressure gauges, test ports, and gas detectors, the valve is a butterfly valve, the device includes an exhaust machine and an exhaust treatment device, and the interfaces include flanges, threads, and clamps, etc.

[0046] By tagging the initial attribute information of the elements, it can be ensured that the initial attribute information of the elements is effectively tagged, and the logical relationship between the elements based on the tags can be accurately reflected in the three-dimensional design, thereby improving the design efficiency and quality.

[0047] The three-dimensional modeling design requirement information in the semiconductor factory interface layout information is the interface information of different devices at each location in the semiconductor factory, including electrical interface, pure wastewater interface, gas interface, etc. According to the interface information, the starting interface can be determined, that is, the three-dimensional modeling starting point is determined, and the semiconductor factory device model 10 containing the device interface 11 is generated, as shown in Figure 2 The device interface 11 in the semiconductor factory device model 10 containing the device interface 11 is an equipment interface description model (EID). The EID model is a unique feature of semiconductor equipment, and its interface is quite complex. It is used to describe the interface information of each device in the semiconductor factory in detail. These interface information is crucial for the integration, debugging and maintenance of the equipment.

[0048] The EID model ensures the compatibility and interoperability between devices by describing the interface information in detail. At the same time, clear interface description helps to quickly locate and solve problems, improving the maintenance efficiency of the equipment. Using standardized interface description also facilitates the integration of equipment from different suppliers into a system.

[0049] In a specific example, a reference point or datum point is selected from the semiconductor factory interface layout information as the starting point of three-dimensional modeling. A coordinate system is defined in the three-dimensional modeling software to ensure matching with the actual factory layout. According to the collected device data, the geometric model of the device (i.e. the semiconductor factory device model) is created in the three-dimensional modeling software. The semiconductor factory device model is placed in the three-dimensional space according to the actual layout. Define attributes for each device interface, such as type, size, position, direction, etc. Create a geometric model of the interface in the three-dimensional modeling software and associate it with the corresponding semiconductor factory device model. Label the detailed information of the interface in the semiconductor factory device model, such as interface name, type, connection requirements, etc. Check the accuracy and completeness of the semiconductor factory device model and the EID model through the three-dimensional view. Through the above steps, the three-dimensional modeling starting point is determined and the detailed semiconductor factory device model containing the device interface is constructed, providing strong support for the design, layout and equipment integration of the semiconductor factory.

[0050] Further, the element interface matching rules and the logical relationship information between each element are combined, and the three-dimensional touch automatic checking algorithm and the optimal path algorithm based on safety distance correction are fused to determine the element type and the element connection path, as shown inFigure 3 , specifically comprising the following steps: In response to the determination of the three-dimensional modeling starting point, obtaining the element interface matching rules of the semiconductor factory starting equipment model or plant interface; According to the semiconductor factory interface layout information, determining the element connection logic; Based on the logical relationship information between each element, and combined with the element connection logic, through the matching and screening of the connection element data, all feasible element connection path sets and corresponding element types are given; Fuse the safety distance, and determine the initial element connection path from all feasible element connection path sets through the corrected optimal path algorithm; Based on the three-dimensional touch automatic checking algorithm, the elements in the initial element connection path are checked to determine the element type and the corresponding element connection path.

[0051] In a specific embodiment, according to the determined three-dimensional modeling starting point, the element interface matching rules of the semiconductor factory starting equipment model or plant interface corresponding to the three-dimensional modeling starting point can be obtained. In a specific example, in a semiconductor manufacturing equipment, it is necessary to match the output interface of the exhaust system of a cleaning equipment with the input interface of the plant system. The element interface matching rules may include interface type, pressure level, interface system, etc.

[0052] The element connection logic is the connection relationship between each element, which involves the physical position of the element, the interface direction and the connection sequence, for example, element A can be connected with element B, but element A cannot be connected with element C. For example, when drawing a 1 / 4-inch to 2-inch pipe, the corresponding size elbow and valve are automatically inserted to recommend different types of valves under the size, and the three-dimensional pipe is drawn. From the interface of the machine tool, the exhaust pipe is drawn, the connection point is positioned, and when the machine tool interface is connected, the three-dimensional touch automatic checking algorithm is used to check the pipe material, size, interface, etc. of the machine tool interface; The valve or pipe fitting is inserted into the pipe port or pipe, triggering the element connection logic, and inserting the pipe fitting under the corresponding size. Specifically, in the exhaust system of a semiconductor factory, the exhaust system pipe needs to connect the exhaust interface of the cleaning equipment to the plant system interface. The multiple exhaust interfaces of the cleaning equipment have certain interface requirements, including pressure requirements, interface type requirements, and interface system requirements. Among them, the pressure requirement defines the pressure matching rule of the selected interface and pipe type; the interface type requirement and the system requirement ensure the adaptation of the interface type and the correct connection of the interface subsystem; determine the correct connection of each interface, plan the element connection path from one point to another, including turning, avoiding, etc. Action, such as Figure 4 As shown in the result display diagram in the element connection path selection process, the red and blue dashed lines are other element connection paths, and the yellow pipe is the optimal element connection path. The element connection path corresponding to the red and blue dashed lines is far away.

[0053] It can be understood that there can be multiple connectable other elements for each element, and by sequentially matching and screening the connection element data of the connectable other elements, all feasible element connection path sets corresponding to each element and the element types of each element in each element connection path are obtained.

[0054] It can be understood that there is a corresponding safe distance between each element, and based on the safe distance, the safety coefficient of the corresponding element is determined and the optimal path algorithm is corrected accordingly. The corrected optimal path algorithm is used to determine the initial element connection path from all feasible element connection path sets. After determining the initial element connection path, a three-dimensional touch automatic checking algorithm is used to check the elements in the initial element connection path to determine the element type. According to the output management module, it is judged whether the connection between the elements established based on the initial element connection path in the three-dimensional environment exists collision, if there is collision, the connection relationship or element position of the element is adjusted, if there is no collision, the element position of the initial element connection path is retained. After completing the collision detection of each element in the initial element connection path, the final element type and the corresponding element connection path are determined. The three-dimensional touch automatic checking algorithm checks and gives the corresponding element type when the connection between the device interface and other device interfaces / plant interface occurs touch.

[0055] Through the above steps, from obtaining the element interface matching rule, to determining the element connection logic, then matching and screening the element connection path, and finally determining the final element connection path through the three-dimensional touch automatic checking algorithm, the three-dimensional modeling of the semiconductor factory is judged to ensure the accuracy and reliability of the three-dimensional modeling.

[0056] Further, the safety distance is fused, and the initial element connection path is determined from all feasible element connection path sets through the corrected optimal path algorithm, referring to Figure 5 , specifically comprising the following steps: Set the safety distance to determine the path cost function; Fuse the path length and obstacle distance through the optimal path algorithm to form the first element connection path set; Give the comprehensive cost of each feasible element connection path in all feasible element connection path sets and the comprehensive cost of each first element connection path in the first element connection path set, respectively; Determine the reference cost based on the comprehensive cost of each first element connection path in the first element connection path set; Based on the reference cost, screen all feasible element connection path sets to form the second element connection path set; The Manhattan distance between each second element connection path in the second element connection path set and the nearest obstacle is analyzed, and a safety coefficient function of the second element connection path is given in combination with the safety distance. Based on the evaluation screening function, the path length, the safety coefficient function and the path cost function are fused to evaluate each second element connection path, and an initial element connection path is determined.

[0057] The path cost function is a function with path length and path cost as independent variables, and the safety distance is a parameter in the path cost function. The relationship between the safety distance and the path length and the path cost is determined according to the specific scene. The path cost function can be expressed as:

[0058] Wherein, F(P) is the path cost function of the second element connection path, d safe is the safety distance, length(P) is the path length of the path P, and cost(P) is the path cost of the path P.

[0059] In a specific embodiment, after the first element connection path set is determined, the comprehensive cost of each first element connection path in the first element connection path set is analyzed in combination with the path cost function, and is used as a reference cost to judge and screen the comprehensive cost of each feasible element connection path in all feasible element connection path sets, to give a second element connection path set. On the basis of the second element connection path set, the Manhattan distance between each second element connection path in the second element connection path set and the nearest obstacle is analyzed, and a safety coefficient function of the second element connection path is given in combination with the safety distance. An evaluation screening function including the path length, the safety coefficient function and the path cost function is used to evaluate each second element connection path, and an initial element connection path is determined.

[0060] Further, by the optimal path algorithm, the path length and the obstacle distance are fused to form the first element connection path set, and the optimal path algorithm is referred to Figure 6 , which specifically includes the following steps: Based on the spatial information of the semiconductor factory building, a grid layout map composed of unit cells is formed, and the positions of the obstacles are determined; According to the breadth-first search, the diffusion of all obstacles is calculated, and the Manhattan distance of each unit cell to the nearest obstacle is calculated; The path length and the obstacle penalty term are fused to form the semiconductor factory equipment node weight relationship, and a modified optimal path algorithm is given to generate the first element connection path set.

[0061] In a specific embodiment, a grid layout map corresponding to the semiconductor factory is first created according to the semiconductor factory building space information, wherein the semiconductor factory building space information includes the positions of various equipment facilities in the semiconductor factory and the relative positions between them, and the semiconductor factory building space information can also be obtained by the acquisition module and provided to the connection element modeling module. The grid layout map is composed of a plurality of cells, which can be used to plan element connection paths and also can be used to analyze the distances of various equipment facilities or elements. For example, there are a plurality of equipment in the semiconductor factory, such as a lithography machine, an etching machine, and a cleaning device. The positions of each equipment on the grid layout map can be determined through the grid layout map, and a basis is provided for subsequent path planning. In addition to marking the positions of various facilities, equipment, and elements on the grid layout map, the positions of various obstacles in the semiconductor factory that affect path planning, such as support columns, walls, other equipment, etc., are also marked.

[0062] After the positions of the obstacles are determined, the distances from each cell to the nearest obstacle are calculated by diffusing all obstacles according to the breadth-first search algorithm. The Manhattan distance refers to the shortest path length from one point to another point in the cells of the grid layout map. For example, the path from the lithography machine to the etching machine on the grid layout map can be blocked by a plurality of obstacles. By calculating the Manhattan distance, the feasibility and priority of different paths can be evaluated. The breadth-first search (BFS) algorithm starts from an initial node and visits its adjacent nodes in order of level until all reachable nodes are visited.

[0063] In path planning, not only the straight-line length of the path, but also the influence of the obstacles on the path need to be considered. By introducing an obstacle penalty term, the path can be weighted, and then a modified optimal path algorithm is used for path planning to make the path planning more reasonable. The obtained semiconductor factory equipment node weight relationship reflects the advantages and disadvantages of different paths, so that the optimal first element connection path set can be determined. For example, in a semiconductor factory, there can be some paths that have a relatively short straight-line distance but need to bypass a plurality of obstacles. By introducing an obstacle penalty term, the path planning algorithm tends to select those paths that have a relatively long straight-line distance but fewer obstacles, not only considering the straight-line distance but also considering the influence of the obstacles, so as to ensure the feasibility and efficiency of the paths in the first element connection path set.

[0064] Further, the path length and the obstacle penalty term are fused to form the equipment node weight relationship, and a modified optimal path algorithm is given, which is specifically represented as:

[0065] wherein, cost(i) is the weight of the semiconductor factory equipment node i, length(i) is the path length from the three-dimensional modeling starting point to the semiconductor factory equipment node i, λ is the obstacle weight parameter, d obs (i) is the distance from the semiconductor factory equipment node i to the nearest obstacle, d safe is the safety distance.

[0066] Further, the safety coefficient function is specifically represented as:

[0067] The evaluation screening function is specifically represented as:

[0068] wherein, S(P) is the safety coefficient function of the second element connection path, d obs (i) is the distance from the semiconductor factory equipment node i to the nearest obstacle, d safe is the safety distance, Π i is the relative safety degree product of all semiconductor factory equipment nodes on the second element connection path, P k is the screening function of the optimal k paths, F(P) is the path cost function of the second element connection path, length(P) is the total length of the second element connection path, and argtopk() is the selection of the k initial element connection paths from the second element connection path.

[0069] Further, based on the three-dimensional touch automatic verification algorithm, the elements in the initial element connection path are verified to determine the element type and the corresponding element connection path, specifically including the following steps: The element attribute information is extracted from the initial element connection path to form a set of elements to be verified by combining the attribute extraction function. The set of elements to be verified is verified by the verification function. Based on the verification screening rule, the element type and the corresponding element connection path are determined.

[0070] Further, the set of elements to be verified is verified by the verification function, specifically including:

[0071] wherein, The verification function of the element between the semiconductor factory equipment nodes A and B, m is the number of a certain attribute of the element, is any symbol, M is the element attribute set, f m () is the attribute extraction function, f m (P A ) is the specified attribute of the semiconductor factory equipment node A, fm (P B ) is a specified attribute of a semiconductor factory equipment node B.

[0072] In a specific embodiment, after determining the initial component connection path, the attribute information of each component in the initial component connection path is also determined. Through the attribute extraction function, the attribute information of each component in the initial component connection path is retrieved, such as size, pressure, material, system, etc., to obtain a set of to-be-verified component attributes. For example, the lithography machine and the etching machine need to be connected through a pipeline, and the interface size, pipeline diameter, material attribute, etc. of the two devices are extracted to form a set of to-be-verified component attributes. Through the verification function, the set of to-be-verified component attributes is verified, and the verification content includes checking whether the component size matches, whether the material is compatible, whether the interface is correctly connected, etc. For example, the verification function is used to check whether the pipeline interface between the lithography machine and the etching machine matches, whether the pipeline material meets the process requirements, and whether the pipeline size matches the interface. Based on the verification screening rules and combined with the verification results, the component type and the corresponding component connection path are determined. The verification screening rules include the shortest path, the safest path, the lowest cost path, etc. For example, according to the verification results and the verification screening rules, a path that meets the technical requirements and has the highest cost-effectiveness is selected to connect the lithography machine and the etching machine. The path may need to bypass certain obstacles, or a specific pipeline material is selected to meet the process requirements. For example, M={material, interface type, pipe diameter, standard, pressure rating, etc.

[0073] By using the three-dimensional touch automatic verification algorithm to verify the initial component connection path, the component connection path is determined, which can realize automatic and intelligent path planning in semiconductor three-dimensional modeling, and improve the layout efficiency of the factory and the accuracy of equipment connection.

[0074] The three-dimensional modeling device of the semiconductor factory also includes a collaborative design module and an output management module. The collaborative design module is connected with the equipment modeling module and the connection component modeling module, and is used to obtain information changes of the semiconductor factory equipment model and the interface connection between components in the semiconductor factory equipment model and update the information in time to realize multi-section collaborative design. The output management module is connected with the connection component modeling module and is used to visually present the semiconductor factory three-dimensional modeling output by the connection component modeling module.

[0075] The collaborative design module is a system component used to support multiple people to design at the same time, aiming to improve design efficiency and reduce errors and conflicts, ensure multiple people to design at the same time, update information in time, and share in real time.

[0076] The collaborative design module allows multiple designers to work on the same design project simultaneously, each designer can independently make design modifications without interfering with each other. And any changes in design information will be updated in real time, ensuring that all designers can access the latest design data. The latest design data will be presented in a visual form, designers can real-time view the work progress and design results of other designers. Through real-time monitoring and intelligent prompting mechanism, reduce errors and conflicts in the design process, ensure the consistency and accuracy of the design.

[0077] Through the collaborative design module, collaborative work and information sharing can be achieved, improving the overall work efficiency of the design team and shortening the design cycle.

[0078] The output management module is responsible for converting the three-dimensional modeling of the semiconductor factory into executable construction drawings and bill of materials. When outputting the three-dimensional modeling of the semiconductor factory, potential collisions between pipes are automatically detected to ensure the feasibility of the design, while space management is performed to optimize the space layout and ensure the reasonable arrangement of equipment and pipes, improving the space utilization rate.

[0079] The output management module also automatically generates detailed construction drawings based on the three-dimensional modeling of the semiconductor factory, including equipment layout, pipe routing, etc., and guides the on-site construction through virtual reality (VR) / augmented reality (AR) technology. And generate material and material list, including quantity, specification, model, etc., to facilitate cost control and procurement. The output management module is an integrated carrier of data information, integrating design data, construction drawings and bill of materials for easy management and query. It can also be linked with the simulation system for more in-depth analysis and verification, such as fluid dynamics simulation, structural strength analysis, etc.

[0080] Referring to Figure 7 The embodiment of the present application provides a three-dimensional modeling method of a semiconductor factory, and the specific steps include: Obtain the interface layout information of the semiconductor factory and the connection element data, wherein the connection element data includes element interface matching rules, element attribute information and logical relationship information between elements; Based on the interface layout information of the semiconductor factory, determine the starting point of three-dimensional modeling, and generate a semiconductor factory equipment model containing equipment interfaces; Based on the three-dimensional design requirement information in the interface layout information of the semiconductor factory, starting with the semiconductor factory starting equipment or plant interface, combining the element interface matching rules and the logical relationship information between elements, and integrating the three-dimensional touch automatic checking algorithm and the optimal path algorithm based on the safety distance correction, determine the element type and the element connection path, realize the interface connection between elements in each semiconductor factory equipment model, and complete the three-dimensional modeling of the semiconductor factory.

[0081] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described steps can refer to the corresponding process in the foregoing device embodiments, and will not be described here.

[0082] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the preferred embodiments by those skilled in the art once they learn of the basic inventive concepts. Therefore, the appended claims are intended to encompass within their scope all possible variations and modifications of the preferred embodiments. It is apparent that those skilled in the art can, without departing from the spirit and scope of the application, make various changes and modifications of the application. Thus, if these modifications and variations fall within the scope of the claims and their equivalents, they are intended to be covered by the claims.

Claims

1. A three-dimensional modeling apparatus of a semiconductor factory, characterized by, Specifically comprising: The acquisition module is used for acquiring the semiconductor factory interface layout information and the connection element data, wherein the connection element data comprises element interface matching rules, element attribute information and logical relationship information between elements; The device modeling module is used for receiving the semiconductor factory interface layout information of the acquisition module, determining a three-dimensional modeling starting point, and generating a semiconductor factory device model comprising device interfaces; The connection element modeling module is used for determining element types and element connection paths based on the semiconductor factory interface layout information and the three-dimensional modeling starting point received by the acquisition module, starting with a semiconductor factory starting device or a factory interface, combining the element interface matching rules and the logical relationship information between elements received by the acquisition module, and fusing a three-dimensional touch automatic checking algorithm and an optimal path algorithm based on a safety distance correction, combining the element attribute information of the acquisition module, realizing interface connection between elements in each semiconductor factory device model in the device modeling module, and completing three-dimensional modeling of the semiconductor factory.

2. The apparatus for three-dimensional modeling of a semiconductor factory of claim 1, wherein The connection element data is acquired, specifically comprising the following steps: Acquiring element interface matching rules and initial element attribute information; Labeling the initial element attribute information and constructing logical relationships between elements based on the labels of the elements to form logical relationship information between the elements.

3. The apparatus for three-dimensional modeling of a semiconductor factory of claim 1, wherein Combining the element interface matching rules and the logical relationship information between elements received by the acquisition module and fusing a three-dimensional touch automatic checking algorithm and an optimal path algorithm based on a safety distance correction to determine element types and element connection paths, specifically comprising the following steps: In response to the determination of the three-dimensional modeling starting point, acquiring element interface matching rules of a semiconductor factory starting device model or a factory interface; Determining element connection logic according to the semiconductor factory interface layout information; Based on the logical relationship information between elements and combining the element connection logic, matching and screening the connection element data to give all feasible element connection path sets and corresponding element types; Fusing a safety distance and determining an initial element connection path from all feasible element connection path sets through a corrected optimal path algorithm; Based on a three-dimensional touch automatic checking algorithm, checking the elements in the initial element connection path to determine element types and corresponding element connection paths.

4. The apparatus for three-dimensional modeling of a semiconductor factory of claim 3, wherein Fusing a safety distance and determining an initial element connection path from all feasible element connection path sets through a corrected optimal path algorithm, specifically comprising the following steps: Setting a safety distance to determine a path cost function; Through an optimal path algorithm, fusing path length and obstacle distance to form a first element connection path set; Giving the comprehensive cost of each feasible element connection path in all feasible element connection path sets and the comprehensive cost of each first element connection path in the first element connection path set; Determining a reference cost based on the comprehensive cost of each first element connection path in the first element connection path set; Based on the reference cost, screening all feasible element connection path sets to form a second element connection path set; Analyzing the Manhattan distance between each second element connection path in the second element connection path set and the nearest obstacle, and combining the safety distance, a safety coefficient function of the second element connection path is given; Based on the evaluation screening function, the path length, the safety coefficient function and the path cost function are fused to evaluate each second element connection path and determine the initial element connection path.

5. The three-dimensional modeling apparatus for a semiconductor factory as described in claim 4, characterized in that, Through the optimal path algorithm, the path length and the obstacle distance are fused to form the first element connection path set, which specifically includes the following steps: Based on the semiconductor factory building space information, a grid layout map composed of unit cells is formed, and the obstacle positions are determined; According to the breadth-first search, all obstacles are diffused, and the Manhattan distance from each unit cell to the nearest obstacle is calculated; The path length and the obstacle penalty term are fused to form the device node weight relationship, and the modified optimal path algorithm is given to generate the first element connection path set.

6. The apparatus for three-dimensional modeling of a semiconductor factory of claim 5, wherein, The path length and the obstacle penalty term are fused to form the device node weight relationship, and the modified optimal path algorithm is given, which is specifically represented as: ； wherein cost(i) is a weight of the semiconductor factory equipment node i, length(i) is a path length from the three-dimensional modeling start point to the semiconductor factory equipment node i, λ is an obstacle weight parameter, d obs (i) is a distance from the semiconductor factory equipment node i to the nearest obstacle, d safe is a safety distance.

7. The apparatus for three-dimensional modeling of a semiconductor factory of claim 4, wherein, The safety coefficient function is specifically represented as: ； The evaluation screening function is specifically represented as: ； where S(P) is a safety factor function of the second element connection path, d obs (i) is the distance from the semiconductor factory equipment node i to the nearest obstacle, d safe is the safety distance, d i is the product of the relative safety degrees of all semiconductor factory equipment nodes on the second element connection path, P k is the screening function of the optimal k paths, F(P) is a path cost function of the second element connection path, and length(P) is the total length of the second element connection path.

8. The three-dimensional modeling apparatus for a semiconductor factory as described in claim 3, characterized in that, Based on the three-dimensional touch automatic verification algorithm, the initial element connection path is verified to determine the element type and the corresponding element connection path, which specifically includes the following steps: Based on the initial element connection path, the attribute extraction function is used to extract the element attribute information to form a set of elements to be verified; Through the verification function, the set of elements to be verified is verified, which specifically includes: ； in, Verification function for components between nodes A and B in a semiconductor factory, where m is the ID of a certain attribute of the component. Let f be any symbol, M be the set of component attributes, and f m () represents the attribute extraction function, f m (P) A ) is a specified attribute of semiconductor factory equipment node A, f m (P) B ) is a specified attribute of semiconductor factory equipment node B; Based on the verification screening rule, the element type and the corresponding element connection path are determined.

9. The apparatus of claim 1, wherein It also includes a collaborative design module and an output management module; The collaborative design module is connected with the device modeling module and the connection element modeling module, which is used to obtain the information change of the semiconductor factory device model and the interface connection between elements in the semiconductor factory device model and update the information in time to realize multi-section collaborative design; The output management module is connected with the connection element modeling module, which is used for visual presentation of the semiconductor factory three-dimensional modeling output by the connection element modeling module.

10. A method of three-dimensional modeling of a semiconductor factory, characterized by, The three-dimensional modeling device of the semiconductor factory according to any one of claims 1-9 is adopted, which specifically includes the following steps: Obtain the interface layout information of the semiconductor factory and the connection element data, wherein the connection element data includes element interface matching rules, element attribute information and logical relationship information between elements; Based on the semiconductor factory interface layout information, the starting point of three-dimensional modeling is determined to generate a semiconductor factory device model containing device interfaces; Based on the semiconductor factory interface layout information and the three-dimensional modeling starting point, the starting device or the factory interface of the semiconductor factory is taken as the starting point, combined with the element interface matching rules and the logical relationship information between elements, and fused with the three-dimensional touch automatic verification algorithm and the optimal path algorithm modified based on the safety distance, the element type and the element connection path are determined to realize the interface connection between elements in each semiconductor factory device model, and the three-dimensional modeling of the semiconductor factory is completed.

Images