Apparatus and method for event detection and duration determination
An asset class type of a new asset is predicted or determined based upon an evaluation of time series data from the new asset. A predicted asset type is used to identify sensors of the new asset to use for data collection. Using the readings of selected sensors from the new asset, states of the new asset are obtained. The duration at least one of these states of the new asset is determined. This information can be subsequently used to optimize the performance of the new asset.
1. A method, the method comprising:
storing a model in a memory, the model describing and predicting behavior of a new machine that is to be added to a group of currently operating machines;
sensing first time series data at the new machine with a plurality of sensors;
determining an activity value for the first time series data and discarding the first time series data when the activity value does not reach an activity threshold;
when the activity value reaches the activity threshold, determining a type of the new machine based at least in part upon a comparison of the model with the first time series data;
mapping the determined type to one or more sensors from the plurality of sensors, only the one or more sensors from the plurality of sensors sensing second time series data from the new machine;
determining one or more of an event, a state, or an event duration at the new machine based upon an analysis of the second time series data; and
responsively determining an action that improves performance of the new machine based upon an evaluation of one or more of the determined event, state, or event duration;
wherein the electrical signal is transmitted to the new machine to control an aspect of the operation of the new machine, and the electrical control signal is applied to the new machine.
2. The method of claim 1 , wherein the model is built previous to sensing the first time series data, the model being built based upon detected patterns in sensed third time series data.
3. The method of claim 1 , wherein determining an event utilizes a sliding time window that is applied to the first time series data.
4. The method of claim 1 , wherein the action is a maintenance action.
5. The method of claim 1 , wherein the one or more sensors are configured to sense speed, electrical current, movement, pressure or temperature.
6. The method of claim 1 , wherein a Hidden Markov Model (HMM) is used to filter the events based upon knowledge of a domain in which the new machine is operating.
7. A system, the system comprising:
a network;
a currently-operating machine that is coupled to the network;
a new machine that is to be added with the currently operating machine, the new machine having a plurality of sensors that sense first time series data at the new machine;
a control circuit including a memory that stores the model that describes and predicts behavior of the new machine, the control circuit being coupled to the network, the control circuit being configured to determine an activity value for the first time series data and discarding the first time series data when the activity value does not reach an activity threshold, the control circuit being configured to, when the activity value reaches the activity threshold, determine a type of the new machine based at least in part upon a comparison of the model with the first time series data, the control circuit further configured to map the determined type to one or more sensors from the plurality of sensors, only the one or more sensors from the plurality of sensors sensing second time series data from the new machine, the control circuit being further configured to determine one or more of an event, a state, or an event duration at the new machine based upon an analysis of the second time series data; and responsively determine an action that improves performance of the new machine based upon an evaluation of one or more of the determined event, state, or event duration;
wherein the electrical signal is transmitted to the new machine to control an aspect of the operation of the new machine, and the electrical control signal is applied to the new machine.
8. The system of claim 7 , wherein the control circuit builds the model previous to sensing the first time series data, the model being built based upon detected patterns in sensed third time series data.
9. The system of claim 7 , wherein the control circuit determines an event by utilizing a sliding time window that is applied to the first time series data.
10. The system of claim 7 , wherein the action is a maintenance action.
11. The system of claim 7 , wherein the one or more sensors are configured to sense speed, electrical current, movement, pressure or temperature.
12. The system of claim 7 , wherein the control circuit utilizes a Hidden Markov Model (HMM) is used to filter the events based upon knowledge of a domain in which the new machine is operating.