support_k3ljvfns

The old school performance system is based on daily reporting by the ship’s crew (noon data).

The intention of the reporting system is that the crew sends in average values of parameters over a certain p

With relation to uncertainty in measurements of data it is necessary to analyse each sensor with respect to manufacture and type.

The manufacturer of the sensor provides information about the expected

The data used for the performance analysis:

may be with an interval i.e. 10 seconds. This interval is not sufficient to describe the full dynamics of all logged parameters in the system, and it therefore

Another method for detecting changes in a signal is based on change detection algorithms.

An expected behaviour of a system is compared to the actual behaviour and a change or a fault is detected by the

When setting up rules for removal of spikes it will be necessary to evaluate the mean and variance in a certain time frame.

If a spike is detected, it can be replaced by an interpolated value.

To identify

If the ship’s performance has to be simulated under all conditions it is necessary to find a method for running the series from point A to point B including series with variations over time.

It is needed to f

More than 20 parameters are logged through the ship’s automation system.

The timestamp for the logging has to be the same for each parameter in order to have the data logged at the same time.

The l

When measuring and analysing the data it is desirable that the data has statistical properties that are invariant over time, i.e. stationary data.

In practice, this is not the case and the data will often be

The data is sampled at an interval given by the software system.

According to the Nyquist sampling theorem, Jerri (1977), if, to retain the information that a continuous signal contains, the sampling rate

Logged parameters Data acquisition sampling rate stationarityData transformationData evaluationPre-processing spikes drop-outsAnalysisStatisticsDissemination

Loading conditionsDraught – trim Loading – wind area Position of point of gravityOperational conditions Heading Speed through water (speed log) Speed over ground (GPS) Rudder angle Stabilisers in/out Bow ver

The logged data from the sensors on board the ship is picked up by software designed for the purpose.

The ship will normally be equipped with an automation system which is used to monitor all input from

Draught information is given as manual input.

At departure from port the draught forward and aft is taken from draught marks on the side of the ship.

During sailing the draught information is updated

The sea water temperature is measured at the sea chest suction on the starboard side.

The water temperature at this level more or less follows the air temperature, except on occasions at high latitudes in the

The bow motions are measured by an accelerometer which is of the low-frequency type and placed in the bow section.

The motion sensor measures the vertical acceleration and velocity is calculated from

The gyrocompass measures the ship’s heading with reference to true north.

The heading is used to calculate true wind speed and directions based on relative measurements from the wind anemometer in the f

The air temperature is measured by a thermometer and the air pressure is measured by a barometer, both placed outside on the aft part of the navigation bridge.

The temperature changes during day and night,

The GPS gives information about the speed of the ship.

The speed is measured above ground whereas the speed measured by the ship’s log is speed through water.

In cases where the ship is not subject to a