General Electric Company

David W Tong

Christopher H Jolly

Kevin C Zalondek

Bantval J Baliga

Hsueh-Rong Chang

John Benjamin

Stefan J Rzad

Jerry H Carmichael

Edward N Diei

Paul A Delano

Kenneth M Martin

William R VanDell

A system and method for monitoring vibrations of a cutting tool uses a neural network for classifying signal features as break or non-break or, in another embodiment, as non-break or abnormal. A vibration signal is produced by an accelerometer, positioned to sense vibrations at the tool-workpiece interface. The signal is pre-processed to extract low frequency machining noise and detect the energy in a higher frequency band. The signal is then sampled and segments of the digitized signals are processed by digital logic into feature vectors for input to a trained neural net having two output nodes for classification. The use of a neural net provides performance improvement and economies over previously known heuristic methods of signal analysis.

System and method including neural net for tool break detection

An artificial neural network detects points in feature space outside of a boundary determined by a set of sample data. The network is trained using pseudo data which compensates for the lack of original data representing 'abnormal' or novel combinations of features. The training process is done iteratively using a net bias parameter to close the boundary around the sample data. When the neural net stabilizes, the training process is complete. Pseudo data is chosen using several disclosed methods.

Method for using a feed forward neural network to perform classification

A method and apparatus for automated generation of multi-branched diagnostic trees is described. The present invention generally is practiced on model-based diagnostic systems and utilizes a candidate generator, constraint propagator and best measurement generator, along with a model of a system to be diagnosed, to generate nodes and branches of a diagnostic tree.

Method and apparatus for generation of multi-branched diagnostic trees

A method and apparatus for constraint propagation control including determining whether more than one range is to be propagated through a selected component, propagating the one range if only one range is to be propagated, and propagated a composite range if more than one range is to be propagated and if the ranges are sufficiently identical.

Composite range constraint propagation control

A semiconductor power rectifier attains low forward voltage drop, low reverse leakage current and improved switching speed by utilizing Schottky contact regions in a p-i-n rectifier along with other means for reducing the required forward bias voltage. In a preferred embodiment, the other means for reducing the required forward bias voltage includes a respective trench between each respective pair of successively spaced current interruption means.

Power rectifier with trenches

A method for the efficient scheduling of a first plurality of jobs and the performance of the jobs on a second plurality of processing machines uses various heuristic rules in order to meet scheduling objectives. Initially, each job is scheduled on the machines without regard to the scheduling of any other job. Job scheduling conflicts are then identified and a priority index for each conflict is computed. For each job step involved in the highest priority conflict, a flexibility index is computed. Based on the flexibility index, the highest priority conflict is resolved by relaxing one or more steps of one or more jobs. If the conflicts have not been resolved at this stage, control returns to the step of identification of all remaining job scheduling conflicts. Upon resolution of all conflicts, an overall schedule of all jobs on all machines may be displayed. Additionally, the jobs are then performed on the machines based upon the final, conflict free schedule.

Job shop scheduling and production method and apparatus

A process for depositing oxygen doped semi-insulating polycrystalline silicon (SIPOS) as a passivation layer over the junction of a semiconductor silicon substrate in which the substrate is subjected to an oxygen removal step immediately prior to the creation of the SIPOS layer to thereby prevent the creation of an oxide layer at the interface between the SIPOS and the substrate.

Passivation with a low oxygen interface

A process for removing pollutants including sulfur dioxide and/or nitrogen oxides from effluent gas which relies on the use of corona discharge to enhance the efficiency of the process. In one embodiment, corona discharge is utilized in a conventional spray dryer. In another it is combined with an ammonia injection technique. In yet another embodiment corona discharge treatment is followed by exposing the acidic mist byproduct leaving the discharge treatment chamber to a neutralizing reagent directed into the path of the effluent. The reagent may be provided with an electrical charge opposite to that on the acidic mist byproduct emerging from the corona reaction chamber to enhance neutralization.

Growing community of inventors

Scotia, NY, United States of America

David W Tong