To test the ultrasound unit and to collect data for testing, we needed an experimental setup to continuously collect information from boats passing by the transducer. Once the transducer, pulser/receiver, oscilloscope, GPIB and the PC was set up to collect echo signals from two transducers using MatLab, the transducer was submerged into the tank and fixed on one side. In order to have boats to pass by the transducer’s beam automatically, they needed to be pulled around the tank continuously.
The boat passing test was conducted using an X-Y plotter driven by a signal generator. Figure 31 shows the setup.
Figure 31: Test setup using X-Y plotter
The pulser/receiver is connected to two ultrasound transducers through the multiplexer. The two transducers are submerged underwater in the tank and fixed at a width of 3cm apart facing sideways to the motion of the X-Y plotter arm. The arm on the X-Y plotter is connected to a bronze arm which extends down into the water where it has a sponge attached to the end. The sponge was used as a model boat and passed back and forth in front of the transducers by inputting a sine-wave into the control of the X-Y plotter. The data obtained from this setup is shown in Figure 32 and Figure 33.
Figure 32: Sample Arrival Time
Figure 33: Sample Correlation Amplitude
The green line shows the echo arrival time obtained from one transducer and the blue line shows the echo arrival time obtained from the other transducer. The first pass was first observed by the green transducer, then the blue transducer. A few samples later, the object has passed the green transducer and later passes the blue transducer. Since this sample is a clear sample showing that an object of considerable size has passed in front of the set of transducers coming from one way and going out the other, this is considered as a boat. The second pass comes in faster, noticeable by the time in between the first detection by each transducer, from blue transducer side. The third pass comes from green transducer side at about the same speed as first pass. All of this information is processed by the detection algorithm and a log is created. The log is shown in Figure 34.
|
Timestamp:
Fri Apr 04
Velocity: -0.031 km/h Length: 0.064 m Timestamp:
Fri Apr 04 Velocity: 0.072 km/h Length: 0.090 m Timestamp:
Fri Apr 04 Velocity: -0.031 km/h Length: 0.090 m |
Figure 34: Sample log file
The data shown in the log file in Figure 34 shows that the first and third objects logged have the same velocity while the second pass is faster and in the opposite direction. This is all consistent with what is shown in the transducer data input plot in Figure 32. Also, the tested object was measured to be 9.0 cm (0.090m), which is also consistent with the logged information except at the first transducer. This inconsistency is due to the fact that the object accelerated after being detected by both transducers. Since the system calculates the object speed at the instant when the object is detected by both transducers, and the length is calculated after the boat has cleared the first transducer it was detected by, when the speed changes between those time intervals, the length calculation can be thrown off.
The testing results show that the basic detection of passing objects works as desired in the controlled environment. The test was done using the setup for the first generation. More testing results obtained using the first generation are included on the CD-ROM in the folder “Testing and Results.”