Method for Developing Machine Operation Classifier Using Machine Learning
US-2016078363-A1 · Mar 17, 2016 · US
US9716422B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9716422-B2 |
| Application number | US-201615220901-A |
| Country | US |
| Kind code | B2 |
| Filing date | Jul 27, 2016 |
| Priority date | Jul 31, 2015 |
| Publication date | Jul 25, 2017 |
| Grant date | Jul 25, 2017 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A machine learning apparatus able to learn a point of connection of a ground wire or a shield wire optimal for reducing noise. It is a machine learning apparatus learning a point of connection of a ground wire or shield wire used in an electric motor apparatus provided with a status observing part and learning part. The status observing part observes a point of connection of a ground wire or shield wire and a feedback signal from an electric motor as status variables. The learning part learns a point of connection of a ground wire or a shield wire able to reduce noise included in the feedback signal in accordance with a training data set prepared based on the status variables.
Opening claim text (preview).
What is claimed is: 1. A machine learning apparatus learning a point of connection of a ground wire or a shield wire used in an electric motor apparatus, the machine learning apparatus comprising a status observing part observing the point of connection of the ground wire or the shield wire and a feedback signal from an electric motor as status variables and a learning part learning the point of connection of the ground wire or the shield wire reducing noise included in the feedback signal in accordance with a training data set prepared based on the status variables. 2. An electric motor control apparatus comprising: a machine learning apparatus according to claim 1 , a connecting part connecting the ground wire or the shield wire to the point of connection, a signal acquiring part acquiring a feedback signal, and a decision-making part using a result learned by the learning part as the basis to determine the point of connection of the ground wire or the shield wire. 3. The electric motor control apparatus according to claim 2 , wherein the learning part comprises a reward calculating part using noise included in the feedback signal as the basis to calculate a reward and a function updating part using the reward as the basis to update a function for determining the point of connection of the ground wire or the shield wire. 4. The electric motor control apparatus according to claim 3 , wherein the reward calculating part is configured to increase the reward when the noise is smaller than a predetermined threshold value and to decrease the reward when it is the threshold value or more. 5. The electric motor control apparatus according to claim 3 , wherein the function updating part is configured so as to update an action value table in accordance with the reward. 6. An electric motor apparatus comprising an electric motor control apparatus according to claim 2 , an electric motor controlled by the electric motor control apparatus, and an encoder outputting the feedback signal. 7. A machine learning method learning a point of connection of a ground wire or a shield wire used in an electric motor apparatus, the machine learning method comprising observing the point of connection of a ground wire or shield wire and a feedback signal from an electric motor as status variables and learning the point of connection of the ground wire or the shield wire reducing noise included in the feedback signal in accordance with a training data set prepared based on the status variables.
Structural association with grounding devices · CPC title
Physics · mapped topic
Drive circuits, e.g. power electronics (H02K11/38 takes precedence) · CPC title
for shielding from electromagnetic fields {, i.e. structural association with shields} (means for preventing or reducing eddy-current losses in the winding heads by shielding H02K3/42) · CPC title
for measuring, monitoring, testing, protecting or switching (rectifiers H02K11/04; power electronics H02K11/33) · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.