National Cheng-Kung University

Metal Industries Research & Development Centre

Haw-Ching Yang

Fan-Tien Cheng

Yu-Lung Lo

Chih-Hung Huang

Hung-Chang Hsiao

Min-Chun Hu

Shyh-Hau Wang

Hao Tieng

Min-Nan Wu

Wen-Chieh Wu

Chia-Ming Jan

Cheng-Yen Chen

Min-Hsiung Hung

Chun-Fang Chen

Chia-Ying Tsai

Hong-Chuong Tran

Ying-Liang Chen

An additive manufacturing (AM) method is provided. The method includes performing a laser powder bed fusion (L-PBF) process on the powder layer. Then, a first surface roughness value of the powder layer after the L-PBF process is obtained to generate a first surface profile. An absorptivity and a set of re-melting process parameters data are used to perform a heat transfer simulation. A second surface profile of the powder layer after laser re-melting is obtained by using the first surface profile and a low-pass filter. Then, the set of re-melting process parameters data is adjusted iteratively to perform the heat transfer simulation until a second surface roughness value predicted from the second surface profile is smaller than or equal to a surface roughness threshold, thereby obtaining optimal values of re-melting process parameters for performing a re-melting process to reduce a surface roughness of a powder layer after the L-PBF process.

Additive manufacturing method

An additive manufacturing (AM) method includes using an AM tool to fabricate a plurality of workpiece products; measuring qualities of the first workpiece products respectively; performing a temperature measurement on each of the melt pools on the powder bed during a fabrication of each of the workpiece products; performing photography on each of the melt pools on the powder bed during the fabrication of each of the workpiece products; extracting a length and a width of each of the melt pools; performing a melt-pool feature processing operation; building a conjecture model by using a plurality of sets of first process data and the actual metrology values of the first workpiece products in accordance with a prediction algorithm; and predicting a virtual metrology value of the second workpiece product by using the conjecture model based on a set of second process data.

Additive manufacturing system and method and feature extraction method

An additive manufacturing (AM) method includes using an AM tool to fabricate a plurality of workpiece products; measuring qualities of the first workpiece products respectively; performing a temperature measurement on each of the melt pools on the powder bed; performing photography on each of the melt pools on the powder bed; extracting a length and a width of each of the melt pools; performing a melt-pool feature processing operation; first converting each of the workspace images to a gray level co-occurrence matrix (GLCM); building a conjecture model by using a plurality of sets of first process data and the actual metrology values of the first workpiece products in accordance with a prediction algorithm; and predicting a virtual metrology value of the second workpiece product by using the conjecture model based on a set of second process data.

An additive manufacturing (AM) system, an AM method, and an AM feature extraction method are provided. The AM system includes an AM tool, a product metrology system, an in-situ metrology system, a virtual metrology (VM) system, a compensator, a track planner, a controller, a simulator and an augmented reality (AR) device. The simulator is used to find feasible parameter ranges, while the AR device is used to support operations and maintenance of the AM tool. The product metrology system, the in-situ metrology system and the VM system are integrated to estimate the variation of material on a powder bed of the AM tool. The compensator is used for compensating the process variation by adjusting process parameters. The product metrology system is used to measure the quality of products. The in-situ metrology system is used to collect features of melt pools on the powder bed.

A tool wear monitoring and predicting method is provided, and uses a hybrid dynamic neural network (HDNN) to build a tool wear prediction model. The tool wear prediction model adopts actual machining (cutting) conductions, sensing data detected at the current tool run of operation and the predicted tool wear value at the previous tool run of operation to predict a predicted tool wear value at the current tool run. A cyber physical agent (CPA) is adopted for simultaneously monitoring and predicting tool wear values of plural machines of the same machine type.

Tool wear monitoring and predicting method

A method for predicting precision of an electrical discharge machine is provided. In the method, plural sets of process data are obtained while the electrical discharge machine processes workpiece samples. Process features are established based on the process data. Each of the workpiece samples with respect to each of at least one measurement item is measured by using a metrology tool, thereby obtaining measurement values of the workpiece samples with respect to each measurement item. A correlation analysis operation is performed to obtain correlation coefficients. At least one key feature is selected from the process features as representative according to the correlation coefficients. The measurement values of the workpiece samples with respect to each measurement item, and the sets of process data corresponding to the key features are used to build a predictive model for predicting the precision of the electrical discharge machine.

Method for predicting precision of electrical discharge machine

A product quality prediction method for mass customization is provided. When a production system has a status change, data of sets of process parameters and actual measurement values of workpiece samples processed before the status change occurs, and data of sets of process parameters and actual measurement values of few workpiece samples processed after the status change occurs are used for build or retrain a prediction model, thereby predicting a metrology value of a next workpiece.

Product quality prediction method for mass customization

A computer-implemented method is provided. First, channel data is obtained, and whether the channel data complies with a predetermined condition of an event is determined. If the channel data complies with the predetermined condition, the channel data is recorded into a database according to a sampling frequency and a recording duration. The channel data corresponding to the event is obtained from the database, and is displayed according to a user operation. A replay frequency and stepping time interval corresponding to the user operation are obtained. The recording duration is adjusted according to the replay frequency, and the sampling frequency is adjusted according to the stepping time interval.

Computer-implemented method for monitoring machine tool based on user behavior

A virtual metrology based method for predicting machining quality of a machine tool is provided. In this method, each product accuracy item is correlated with operation paths of the machine tool. During a modeling stage, the machine tool is operated to process workpiece samples, and sample sensing data of the workpiece samples associated with the operation paths are collected during the operation of the machine tool. The sample sensing data of each workpiece sample is de-noised and converted into the sample feature data corresponding to each feature type. The workpiece samples are measured with respect to the product accuracy item and integrated into the feature data for building a predictive model, thereby obtaining quality predicted data for each product accuracy item. During a usage stage, accuracy item values of a workpiece are predicted using the feature data during processing the workpiece in accordance with the predictive models.

Method for predicting machining quality of machine tool

A virtual production control system (VPCS), and a virtual production control method and a computer program product thereof are provided. At first, the VPCS processes historical work-in-process (WIP) information and a current shipping plan sent from a supplier side, thereby obtaining a plurality of sets of WIP input/output historical data and a goods output schedule. Then, the VPCS performs an integer programming (IP) method to find the latest output schedule in accordance to the current shipping plan; uses a genetic algorithm (GA) to fit the historical distributed-parameters; adopts a neural network (NN) method to predict the future distributed-parameters of production; and finally utilizes a Petri Nets to simulate and obtain a latest feasible input schedule and a latest feasible output schedule.

Virtual production control system and method and computer program product thereof

A generic service management system is disclosed. The generic service management system comprises a registration scheme; a search-and-execution scheme; and a detection-and-replacement scheme, used for detecting and replacing the invalid service provider, such as a semiconductor equipment manager. The present invention provides a GEV (Generic Evaluator) having the capabilities of error-detecting and data backup, and further combines Jini infrastructure and the programming technology of design by contract. The GEV archives the credit values of all the service providers for letting a client (such as a factory manager) to select a service provider having a higher credit value. The GEV periodically backups the execution status and parameters of service providers, so that, in case a service provider has errors, its backup information of execution status and parameters can be assigned to another service provider for continuously executing the unfinished job, thereby enhancing the reliability of distributed object-oriented systems. The present invention can detect abnormal behaviors, such as system crash, errors in transmitting information and degradation of performance, etc., and is suitable for use in production systems that are connected by networks, such as semiconductor manufacturing systems, optical telecommunication production systems, etc.

Growing community of inventors

Tainan, Taiwan

Haw-Ching Yang