519
As mentioned, the smartphone was installed in the operator’s cabin (see Fig. 3). For data
safety reasons, after one cycle of the haul truck, data acquisition was stopped for data saving
and restarted again to obtain separate data files.
In fact, several experiments have been done for various types of LHDs, Haul Trucks
and Self-propelled Transport Trucks (for transport of people). In Fig. 4 there are an
exemplary signals acquired on the machine from onboard monitoring system as well as
4 segments related to measurements of accelerations from a smartphone. For consistent,
graphical presentation all data have been normalized to obtain 0–1 range data. Some of
the data (mostly temperatures) are useless for the considered context of analysis, so they
will not be taken into account in the further part of the paper. An important variable
is Hyd_Oil_P that informs about unloading, so one may easily identify the loaded and
unloaded Haul Truck.
3 METHODOLOGY
Dynamics of external excitations during acceleration, deceleration, loading or unloading
processes are “natural” and expected. These parts of the process could be extracted and parametrized in the sense of number of events, length, maximum of excitation, etc. However, in
Figure 5. Proposed algorithm for data analysis.
On board
monitoring
V1brat1 on
acquisition using
smartphone
Synchronized
resampled data
Gear number
based
segmentation
Loaded machine
Statistical analysis
comparison
•
reasoning
Unloaded machine
Statistical analysis
Segment extraction
Gear no & speed
""""'
..,....,
Variance
analysis
As mentioned, the smartphone was installed in the operator’s cabin (see Fig. 3). For data
safety reasons, after one cycle of the haul truck, data acquisition was stopped for data saving
and restarted again to obtain separate data files.
In fact, several experiments have been done for various types of LHDs, Haul Trucks
and Self-propelled Transport Trucks (for transport of people). In Fig. 4 there are an
exemplary signals acquired on the machine from onboard monitoring system as well as
4 segments related to measurements of accelerations from a smartphone. For consistent,
graphical presentation all data have been normalized to obtain 0–1 range data. Some of
the data (mostly temperatures) are useless for the considered context of analysis, so they
will not be taken into account in the further part of the paper. An important variable
is Hyd_Oil_P that informs about unloading, so one may easily identify the loaded and
unloaded Haul Truck.
3 METHODOLOGY
Dynamics of external excitations during acceleration, deceleration, loading or unloading
processes are “natural” and expected. These parts of the process could be extracted and parametrized in the sense of number of events, length, maximum of excitation, etc. However, in
Figure 5. Proposed algorithm for data analysis.
On board
monitoring
V1brat1 on
acquisition using
smartphone
Synchronized
resampled data
Gear number
based
segmentation
Loaded machine
Statistical analysis
comparison
•
reasoning
Unloaded machine
Statistical analysis
Segment extraction
Gear no & speed
""""'
..,....,
Variance
analysis