Method Of Monitoring Machine Condition
First Claim
1. A method of monitoring a machine condition, comprising:
- modeling a normal signal model performed by detecting a signal for monitoring a condition of a normal machine and converting the detected signal to data of a normal signal model in a time domain using a hidden Markov model (HMM) algorithm;
calculating a probability value data of the monitoring signal at a subject machine performed by detecting a signal for monitoring a condition of the subject machine in real-time and converting the detected signal to the probability value data relative to the normal signal model using the HMM algorithm; and
determining a section having a deficiency where the probability value of the monitoring signal at the subject machinery is not maintained constantly relative to the normal signal model.
2 Assignments
0 Petitions
Accused Products
Abstract
Various embodiments of a method for monitoring a machine condition are provided. An embodiment of the present invention provides a method of monitoring a machine condition, comprising the following steps: modeling a normal signal model performed by detecting a signal for monitoring condition of a normal machine and converting the detected signal to a normal signal model in time domain using a hidden Markov model (HMM) algorithm; calculating a probability value data of the monitoring signal at a subject machine performed by detecting a signal for monitoring condition of the subject machine in real-time and converting the detected signal to the probability value data relative to the normal state signal model using the HMM algorithm; and determining a section having deficiency where the probability value data of the monitoring signal at the subject machinery is not maintained constantly relative to the normal signal model.
6 Citations
20 Claims
-
1. A method of monitoring a machine condition, comprising:
-
modeling a normal signal model performed by detecting a signal for monitoring a condition of a normal machine and converting the detected signal to data of a normal signal model in a time domain using a hidden Markov model (HMM) algorithm; calculating a probability value data of the monitoring signal at a subject machine performed by detecting a signal for monitoring a condition of the subject machine in real-time and converting the detected signal to the probability value data relative to the normal signal model using the HMM algorithm; and determining a section having a deficiency where the probability value of the monitoring signal at the subject machinery is not maintained constantly relative to the normal signal model. - View Dependent Claims (2)
-
-
3. A method of monitoring a welding deficiency, comprising
modeling a normal welding signal model performed by detecting welding signals at a welding portion in a time domain during a welding operation, checking normal welding sections without a deficiency, extracting a normal welding signal data from the normal welding sections, and converting the normal welding signal data to the normal welding signal model of the welding operation using a hidden Markov model (HMM); -
calculating probability value data of the welding signal at a subject welding portion relative to the normal welding signal model performed by detecting the welding signal at the subject welding portion in real-time and converting the detected signal at the subject welding portion to the probability value data using the HMM algorithm; and determining a section having a deficiency where the probability value data of the welding signal at the subject welding portion is not maintained constantly relative to the normal welding signal model. - View Dependent Claims (4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
-
19. A method of monitoring a welding deficiency, comprising:
-
modeling a normal welding voltage signal model performed by detecting welding voltage signals at a welding portion in a time domain during a welding operation, checking normal welding sections without a deficiency, extracting a normal welding voltage signal data from the normal welding sections, and converting the normal welding voltage signal data to the normal welding voltage signal model of the welding operation using a hidden Markov model (HMM); calculating probability value data of the welding voltage signal at a subject welding portion relative to the normal welding voltage signal model performed by detecting the welding voltage signal at the subject welding portion in real-time and converting the detected signal at the subject welding portion to the probability value data using the HMM algorithm; and determining a section having a deficiency where the probability value data of the welding voltage signal at the subject welding portion is not maintained constantly relative to the normal welding voltage signal model, wherein the normal welding voltage signal data is pre-processed to a time function and the pre-processed normal welding voltage signal data is converted to a characteristic vector array by a transformation, and wherein the welding voltage signal detected at the subject welding portion is pre-processed to a time function, and the pre-processed welding voltage signal at the subject welding portion is converted to a characteristic vector array by the transformation.
-
-
20. A method of monitoring a welding deficiency, comprising:
-
modeling a normal welding current signal model performed by detecting welding current signals at a welding portion in a time domain during a welding operation, checking normal welding sections without a deficiency, extracting a normal welding current signal data from the normal welding sections, and converting the normal welding current signal data to the normal welding current signal model of the welding operation using a hidden Markov model (HMM); calculating probability value data of the welding current signal at a subject welding portion relative to the normal welding current signal model performed by detecting the welding current signal at the subject welding portion in real-time and converting the detected signal at the subject welding portion to the probability value data using the HMM algorithm; and determining a section having a deficiency where the probability value data of the welding current signal at the subject welding portion is not maintained constantly relative to the normal welding current signal model, wherein the normal welding current signal data is pre-processed to a time function and the pre-processed normal welding current signal data is converted to a characteristic vector array by a transformation, and wherein the welding current signal detected at the subject welding portion is pre-processed to a time function, and the pre-processed welding current signal at the subject welding portion is converted to a characteristic vector array by the transformation.
-
Specification