The invention relates to a method and device to measure the surface velocity of a fluid flowing in a confined space. More specifically, the present invention relates to a non-invasive method and device with a modified microwave patch antenna.
Non-invasive methods for measuring the flow velocity of a fluid in a channel or sewer, i.e. methods wherein there is no contact between the probe and the fluid, are becoming more and more popular. Among the techniques used we can find acoustic methods, optical methods, laser methods and microwave methods, the last one being the most popular.
Most sewer pipes and channels as well as most industrial waste water channels are located underground and consequently are in confined space. The non-invasive microwave technology is ideal to measure the fluid velocity in those applications. Microwave signals can be generated in different ways. Pulsed radars with horn antennas were historically used for fluid velocity measurements. Horn antennas used in Pulsed Radar systems have one horn (one antenna) being used for both transmitting and receiving the signals. The microwave transducer is mounted on the base of the horn antenna. Horn antennas have the advantage to be very directional with low energy side lobes which don't interact with the narrow environment in a confined space.
Over time due to traffic control and applications in the automobile industry another type of antenna has become very popular, the strip antenna or patch antenna or more exactly the patch array antenna 1 using a number of patches 2 which are interconnected (
The present invention aims to provide an improved non-invasive method and device for measuring the velocity of a fluid flowing in a confined space such as a sewer or an underground channel, using a patch antenna as microwave transmitter and receiver. To this end, at least the transmitting area of the patch antenna is mounted at one end of a reflector tube. The other end of the tube is left open or equipped with a microwave lens. This device with the reflector tube allows to encapsulate the side lobes of the microwave pattern (
The reflector tube has a constant cross section over its length or a cross section expanding gradually over its length. This geometry allows to reduce or even annihilate the side lobes but does not affect significantly the directivity of the signal.
The invention relates to a non-invasive method and a device for measuring the surface velocity of a fluid flowing more specifically in a confined space. However, the present invention does not exclude to use this device for measurements in open air for rivers, irrigation channels and other large man-made channels. The device 7 comprises a patch antenna 1 with a set of interconnected patches acting as a microwave transmitter 1a and another set of interconnected patches acting as a microwave receiver 1b (
The dimensions from the tube section can be equal or larger than the dimensions of the whole patch antenna including the transmitter and receiver as shown in
Depending on the dimensions and shape from the patch antenna the shape of the reflector tube can be square with parallel faces (
Preferably, the length of the reflector tube is a multiple of the wavelength (3, 6, or 12 times lambda) from the transmitting microwave frequency for a better guidance of the wave through the tube and to prevent the formation of side lobes at the exit of the tube.
Using the device according to the invention, the non-invasive method for measuring the fluid velocity flowing through a pipe or channel comprises the following steps:
Then different steps known from the prior art may be carried out to determine the mean velocity of the fluid into the channel based on the surface velocity of the fluid. Indeed, within a pipe traveled by a fluid, there is a velocity gradient both in the horizontal direction and the vertical direction of the wet straight section, the velocity being theoretically close to zero at the walls of the pipe. The steps may consist in applying multiplier factors on the measured surface velocity, the multiplier factors depending on the level of fluid in the pipe and being determined based on previous calibration or mathematical models. The last ones may be based on finite element models as described in the patent EP 0 681 683 B1. It simulates a set of flow-velocity distributions in a channel having a known profile, for several liquid levels in the channel. Based on a measurement of the liquid level and a velocity measurement, it next selects the simulated velocity distribution that is appropriate for the measured liquid level. On this based, a mean velocity is determined.
Preferred alternative steps for converting a surface velocity into a mean velocity are described in the document EP 3 011 278. They are the following. Each reflected pulse generates a measurement datum. The number of reflected pulses in a sequence of measurements will generate a complex mosaic of discrete data expressed in amplitude as a function of time. The spectrum of data expressed in the temporal domain is transformed into a frequency domain via a discrete Fourier transform (DFT), and preferably, a fast Fourier transform (FFT). Then a Gaussian curve is fitted on the spectrum of discrete data expressed in the frequency domain and the parameters of the Gaussian curve, namely the mean p and the standard deviation σ, are calculated. The frequency of the mean p and the standard deviation σ allow respectively to calculate the mean surface velocity over the illuminated zone and the velocity distribution over that same zone. It has been shown that the velocity distribution at the free surface of the liquid is representative of the vertical velocity distribution in the wet section. The mean velocity within the wet section can therefore be deduced from the mean velocity and thus the mean p at the surface of the liquid. However, the illuminated zone may not have a sufficient size to be representative of the entire velocity distribution at the free surface of the liquid. Thus, depending on the size of the illuminated zone, the mean velocity within the pipe may be deduced directly from the mean velocity measured at the surface or indirectly via corrections or extrapolations as further explained in this document EP 3 011 278 incorporated by reference.
On this basis, the flow rate Q in the pipe or channel can be determined. It is equaled to the wet area area A multiplied by the average velocity V. To calculate the wet area, the level in the pipe or channel is measured and associated with the channel or pipe shape.
Number | Date | Country | Kind |
---|---|---|---|
18204121.0 | Nov 2018 | EP | regional |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP2019/079835 | 10/31/2019 | WO | 00 |