Directional ultrasonic gas leak detector

Abstract

An embodiment of a directional ultrasonic gas leak detector includes an array of spaced-apart MEMS microphones, each responsive to incident airborne ultrasonic energy from a source of a gas leak to generate a microphone signal. A beamforming processor is responsive to the microphone signals from the array to generate a processor output signal indicative of an estimated angle of arrival of the ultrasonic energy incident on the array. The array can be disposed in an explosion-proof housing structure for operation in a hazardous location, or implemented as an intrinsically safe device. In another embodiment, a display is responsive to the processor output signal to generate an image representative of a scene being monitored, with an estimated magnitude of incident energy in the direction of the beam superimposed on the image.

Description

[0001] Related applications

[0002] This application claims a divisional application of the invention application filed on September 15, 2015, with a priority date of September 24, 2014, application number 201580051390.8, entitled "Directional Ultrasonic Gas Leak Detector". Background Technology

[0003] Ultrasonic gas leak detectors measure the sound pressure waves generated by turbulence when gas escapes from higher pressure into the ambient atmosphere. These detectors are used as industrial safety devices to monitor for the unintended or accidental release of flammable or toxic gases into the atmosphere. Leaks need to be identified quickly before their amplitude increases further to allow for timely remedial action. Ultrasonic gas leak detectors have the advantage of not requiring the gas to reach the detector; leaks can be detected even if the leaking gas is dispersed by wind. Optical signals are also used.

[0004] Conventional ultrasonic gas leak detectors are omnidirectional and, while providing useful information about the size and duration of pressurized gas leaks, offer no information to maintenance personnel about the location of the leak. Conventional open-path detectors, as line-of-sight optical detectors, also do not provide leak location information. Conventional point detectors (such as catalytic, infrared, or electrochemical detectors) require additional cost to be placed close to each other to provide more precise leak location information. Infrared gas cloud imaging cameras are expensive, their sensitivity varies greatly depending on the gas being monitored, and their performance is largely dependent on the difference between the leaking gas cloud and the background temperature. For these reasons, infrared gas cloud imaging cameras are not readily adopted in industrial stationary gas detection systems. Therefore, there is a need to add the benefit of leak location, or at least leak direction, to ultrasonic gas leak detection equipment. Summary of the Invention

[0005] An embodiment of a directional ultrasonic gas leak detector includes a spaced-out array of MEMS microphones, each microphone generating a microphone signal in response to incident airborne ultrasonic energy from a gas leak source. A beamforming processor, in response to the microphone signals from the array, generates a processor output signal indicating the estimated angle of arrival of the ultrasonic energy incident on the array. The array can be housed in an explosion-proof enclosure for operation in hazardous locations or implemented as an intrinsically safe device. In another embodiment, a display, in response to the processor output signal, generates an image representing the monitored scene, wherein the estimated amplitude of the incident energy in the beam direction is superimposed on the image.

[0006] The original application's claims were as follows:

[0007] 1. A directional ultrasonic gas leak detector, comprising: circuitry including a spaced-apart MEMS microphone array, each of the microphones generating a microphone signal in response to incident airborne ultrasonic energy from a gas leak source disposed within the array; and a beamforming processor, in response to the microphone signals from the MEMS microphone array, generating a processor output signal indicating an estimated angle of arrival of ultrasonic energy incident on the array from the gas leak source within the array.

[0008] 2. The detector of claim 1, wherein the array is a two-dimensional array, and the microphones of the array are uniformly and equally spaced in two orthogonal directions to form at least two linear arrays.

[0009] 3. The detector of claim 2, wherein the beamforming processor is configured to independently perform beam control processing on each of the first and second uniform linear arrays.

[0010] 4. The detector according to claim 2, wherein the microphone of the array is spaced apart from the adjacent MEMS microphone in the array by a distance not greater than 5 mm.

[0011] 5. The detector of claim 1, wherein the processor is configured to indicate the estimated angle of arrival of the ultrasonic energy only when the sound pressure level of the incident ultrasonic energy exceeds a threshold.

[0012] 6. The detector of claim 1, further comprising: a display for displaying an image of a scene monitored by the detector within the range, including a device for delivering or storing gas therethrough; the display responding to an output signal from the processor to depict the location of a gas leak source superimposed on the image.

[0013] 7. The detector of claim 6, wherein the display further responds to the processor output signal to depict the intensity of the gas leak source.

[0014] 8. The detector according to claim 1, wherein the processor output signal includes signals representing the azimuth and elevation coordinates of the detected gas leak.

[0015] 9. The detector of claim 1, wherein the beamforming processor is configured to perform time delay and summation beamforming on the array signal.

[0016] 10. The detector of claim 1, wherein the detector array is mounted to a circuit board and the microphone covers an area of ​​no more than about 10 square centimeters.

[0017] 11. The detector according to claim 1, wherein the opening size of the port is no greater than about 10 square centimeters.

[0018] 12. The detector according to claim 1, further comprising: an explosion-proof housing structure, wherein the array and the processor are disposed within the housing structure, the housing structure including a port through which incident ultrasonic energy is transmitted from a gas leak source to the array.

[0019] 13. The detector of claim 12 further includes a flame arrester structure that covers the port of the housing structure.

[0020] 14. The detector of claim 12 further includes a hydrophobic membrane disposed in or covering the port.

[0021] 15. The detector of claim 1, wherein the circuitry includes a blocking circuit connected to the microphone array, the blocking circuitry being configured to limit the power applied to the microphone array to prevent ignition of an explosive gas mixture.

[0022] 16. The detector of claim 1, wherein the processor responds to the microphone signal from the MEMS microphone array to process a plurality of beam directions and generate a processor output signal indicating an estimated angle of arrival of ultrasonic energy incident on the array in the beam direction in which the maximum response is generated.

[0023] 17. The detector of claim 1, wherein the beamforming processor responds to broadband incident airborne ultrasonic energy from a gas leak source located within 50 meters of the array to generate the processor output signal.

[0024] 18. A directional ultrasonic energy detector, comprising: a two-dimensional array of spaced-apart MEMS microphones, each of the microphones generating a microphone signal in response to incident airborne ultrasonic energy from a source disposed within a range of the array; a processor, in response to the microphone signals from the MEMS microphone array, processing a plurality of beam directions and generating a processor output signal indicating an estimated amplitude of ultrasonic energy incident on the array from each of the beam directions; and a display, in response to the processor output signal and for generating an image representing a monitored scene, wherein the estimated amplitudes of the incident energy in the beam directions are superimposed on the image.

[0025] 19. The detector of claim 18, wherein the detector array is mounted to a circuit board and the microphone covers an area of ​​no more than about 10 square centimeters.

[0026] 20. The detector of claim 18 further includes an explosion-proof housing structure, the array being disposed within the housing structure, the housing structure including a port through which incident ultrasonic energy is transmitted from the monitored scene to the array.

[0027] 21. The detector of claim 18, further comprising a blocking circuit connected to the microphone array, the blocking circuit being configured to limit the power applied to the microphone array to prevent ignition of an explosive gas mixture.

[0028] 22. The detector of claim 1, further comprising: an artificial neural network (ANN) responsive to the received ultrasonic energy and configured to distinguish ultrasonic waves generated by pressurized gas leakage from noisy ultrasonic waves.

[0029] 23. A directional ultrasonic gas leak detector, comprising: circuitry including a spaced-apart MEMS microphone array, each of the microphones generating a microphone signal in response to incident airborne ultrasonic energy from a gas leak source disposed within the array; and a beamforming processor, in response to the microphone signals from the MEMS microphone array, generating a processor output signal indicating an estimated angle of arrival of ultrasonic energy incident on the array from the gas leak source within the array.

[0030] 24. The detector of claim 23, wherein the array is a two-dimensional array, and the microphones of the array are uniformly and equally spaced in two orthogonal directions to form at least two linear arrays.

[0031] 25. The detector of claim 24, wherein the beamforming processor is configured to independently perform beam control processing on each of the first and second uniform linear arrays.

[0032] 26. The detector according to claim 24 or 25, wherein the microphone of the array is spaced apart from the adjacent MEMS microphone in the array by a distance not greater than 5 mm.

[0033] 27. The detector according to any one of claims 23 to 26, wherein the processor is configured to indicate the estimated angle of arrival of the ultrasonic energy only when the sound pressure level of the incident ultrasonic energy exceeds a threshold.

[0034] 28. The detector according to any one of claims 23 to 27, further comprising: a display for displaying an image of a scene monitored by the detector within the range, including a device for delivering or storing gas therethrough; the display responding to the processor outputting a signal to depict the location of a gas leak source superimposed on the image.

[0035] 29. The detector of claim 28, wherein the display further responds to the processor output signal to depict the intensity of the gas leak source.

[0036] 30. The detector according to any one of claims 23 to 29, wherein the processor output signal includes signals representing the azimuth and elevation coordinates of the detected gas leak.

[0037] 31. The detector according to any one of claims 23 to 30, wherein the beamforming processor is configured to perform time delay and summation beamforming on the array signal.

[0038] 32. The detector according to any one of claims 23 to 31, wherein the detector array is mounted to a circuit board and the microphone covers an area of ​​no more than about 10 square centimeters.

[0039] 33. The detector according to any one of claims 23 to 32, further comprising: an explosion-proof housing structure, wherein the array and the processor are disposed within the housing structure, the housing structure including a port through which incident ultrasonic energy is transmitted from a gas leak source to the array.

[0040] 34. The detector of claim 33 further includes a flame arrester structure that covers the port of the housing structure.

[0041] 35. The detector according to any one of claims 23 to 33, wherein the circuitry includes a blocking circuit connected to the microphone array, the blocking circuitry being configured to limit the power applied to the microphone array to prevent ignition of an explosive gas mixture.

[0042] 36. The detector according to any one of claims 23 to 35, wherein the processor responds to the microphone signal from the MEMS microphone array to process a plurality of beam directions and generate a processor output signal indicating an estimated angle of arrival of ultrasonic energy incident on the array in one beam direction that generates the maximum response in the beam directions.

[0043] 37. The detector according to any one of claims 23 to 36, wherein the beamforming processor generates the processor output signal in response to broadband incident airborne ultrasonic energy from a gas leak source disposed within 50 meters of the array.

[0044] 38. The detector according to any one of claims 23 to 37, further comprising: an artificial neural network (ANN) responsive to the received ultrasonic energy and configured to distinguish ultrasonic waves generated by pressurized gas leakage from noisy ultrasonic waves. Attached Figure Description

[0045] When read in conjunction with the accompanying drawings, those skilled in the art will readily understand the features and advantages of this disclosure based on the following specific embodiments, wherein:

[0046] Figure 1 This is a schematic block diagram of an exemplary embodiment of a 2D array of MEMS microphones and associated electronics.

[0047] Figure 2 This is a diagram of a linear array of MEMS microphones in which sound is incident at an angle θ to the microphone array axis.

[0048] Figure 3 This is an exemplary embodiment of a software algorithm used in time-delay-and-sum beamforming of an ultrasonic microphone array.

[0049] Figure 4 An exemplary explosion-proof enclosure structure and system for a gas leak detector system are depicted in cross-section. Figure 4A The diagram illustrates the use of reception. Figure 4 Exemplary housing structure and system of the system described. Figure 4B Another embodiment of the gas leak detector is illustrated, in which the sensing head is located away from the main system housing. Figure 4CAnother embodiment of an acoustic detector system in an explosion-proof enclosure structure is illustrated.

[0050] Figure 5 The diagram above illustrates the information about... Figures 1 to 3 A simplified circuit diagram of the intrinsic safety circuit for a microphone array is described.

[0051] Figure 6 yes Figures 1 to 3 A schematic block diagram illustrating the features of a directional ultrasonic gas leak detector.

[0052] Figure 7 This is a functional block diagram of the features of an embodiment of a directional ultrasonic gas leak detector employing a microphone array.

[0053] Figure 8 This is a functional block diagram showing the features of another embodiment of a directional ultrasonic gas leak detector.

[0054] Figure 9 Is using Figure 1 An exemplary embodiment of a laboratory ultrasound source is provided, wherein the location of the laboratory ultrasound source is superimposed on a visible image.

[0055] Figure 10 A display representing a monitored scene is schematically depicted, in which a directional ultrasonic detector has a field of view centered on the device. Detailed Implementation

[0056] In the following specific embodiments and in several figures in the accompanying drawings, the same elements are identified using the same reference numerals. The figures are not drawn to scale, and for illustrative purposes, relevant feature dimensions may be exaggerated.

[0057] Commercially available ultrasonic gas leak detectors can utilize a single pre-polarized pressure microphone, such as those manufactured by GRASSound and Vibration of Holte (Denmark), Microtech Gefell GmbH of Gefell (Germany), or Bruel Kjaer of Naerum (Denmark). The ultrasonic range is defined as the frequency range beyond human hearing, starting at approximately 20 kHz for a healthy young adult. Higher ultrasonic frequencies attenuate more quickly in air than lower frequencies, and practical applications of ultrasonic gas leak detection systems typically use frequencies below 100 kHz.

[0058] In an exemplary embodiment, the directional ultrasonic gas leak detector includes a spaced-out array of microphones. Each microphone generates a microphone signal in response to incident broadband ultrasonic energy from a gas leak source positioned within the array's range. A beamforming processor generates a processor output signal in response to the microphone signals from the array, indicating the estimated angle of arrival of the ultrasonic energy incident on the array from a gas leak source positioned within the array's range (e.g., within 30 to 50 meters of the array). The array may be housed in an explosion-proof enclosure, or the detector may be designed as an intrinsically safe device to meet operational requirements in hazardous locations defined by a regulatory body. In this scenario, a hazardous location is an area containing or potentially containing flammable concentrations of flammable gases, vapors, or dust. The microphones are preferably spaced apart from adjacent microphones in the array by a distance not exceeding 5 mm.

[0059] In an exemplary embodiment, the ultrasonic microphone utilized in the array can be a miniature microphone based on MEMS (Micro-Electro-Mechanical Systems) technology, which operates well in an audible range exceeding 15 kHz and an ultrasonic frequency range exceeding 100 kHz. The MEMS microphone can be mounted on a printed circuit board (PCB) and housed in an environmentally durable mechanical housing approved for hazardous locations, allowing ultrasonic energy to be transferred to the sensing element. U.S. Patent Publications 2009 / 0060246A1 and 2014 / 0000347A1 detail implementations of such MEMS microphones in industrial ultrasonic gas leak detectors, the entire contents of which are incorporated herein by reference.

[0060] Discrete prepolarized stainless steel ultrasonic microphones offer excellent ultrasonic performance, but are too large and expensive to package into arrays for industrial gas leak detectors. MEMS microphones are more miniaturized and are themselves mounted on a circuit board to form a microphone array. In an exemplary embodiment, a two-dimensional array of a total of nine MEMS microphones (1 to 9) is mounted on circuit board 10. Figure 1 The microphone arrays are evenly and equally spaced in two sets of five linear arrays intersecting and perpendicular to each other on the circuit board. To achieve uniform and equal spacing in the two perpendicular directions, the arrays are positioned in a "T" shape: this is specified by the MEMS microphone packaging, which is rectangular rather than square or circular. In an exemplary embodiment of ultrasonic detection of gas leaks, the microphone array area on the circuit board will typically not exceed 10 square centimeters. The signal generated by the microphone is digitized by ADC 11 and processed by a processor (typically indicated as 12) with embedded software. For microphones that produce digital output, the processor can process such digital signals without ADC 11.

[0061] In an exemplary embodiment, N omnidirectional MEMS microphones are evenly spaced in a column, wherein N is at least 2. Figure 2 This is a diagram of a linear array 20 of five MEMS microphones (1, 2, 3, 4, and 5), where the distance between the microphones is d, and the sound is incident at an angle θ to the axis of the microphone array. Beamforming is a signal processing technique used in sensor arrays for directional signal transmission or reception, the latter being the case of microphones. Some rules applicable to beamforming using a uniform linear array (ULA) are:

[0062] a) Increasing the number of microphones can enhance the signal-to-noise ratio of the array, which is limited to array gain, and is given in dB as 10log(N) under conditions of coherent signal and incoherent noise, where N is the number of microphones. Array gain can help increase detection distance. Increasing N also results in a physically larger array.

[0063] b) Increasing the total array length D by using a larger number of microphones improves spatial resolution. D is called the aperture size, and when N microphones are equally spaced apart by a distance d, D = (N-1)d. For the wide side direction, the half-power beamwidth is proportional to λ / D, where λ is the wavelength of the incident energy and is therefore a function of wavelength.

[0064] c) The microphone spacing d between microphones determines the highest frequency fmax that can be controlled without aliasing. The acoustic wavelength of the highest frequency fmax is the shortest wavelength λmin, and the microphone spacing d must satisfy the criterion d < λmin / 2 to prevent spatial aliasing. For a sound speed in air of 340 m / sec and a microphone spacing d = 3.4 mm, λmin is 6.8 mm, and fmax = 50 kHz; such a beamformer can be used for ultrasonic frequencies below 50 kHz without aliasing that causes copies of the main lobe of the directional pattern. For a MEMS microphone with a size of approximately 3 mm, the minimum possible microphone spacing is also approximately 3 mm, resulting in an fmax value of around 50 kHz.

[0065] d) Atmospheric attenuation of ultrasound is a function of frequency, increasing from approximately 1 dB / m at 40 kHz to approximately 10 dB / m at 170 kHz. Therefore, in practice, ultrasonic gas leak detector designs are limited to frequencies below approximately 75 kHz; the same frequency limitation may apply to beamforming due to atmospheric attenuation, further limiting it to 50 kHz due to the aforementioned practical constraints on microphone spacing. Because of the industry-standard leak rate on the order of 0.1 kg / sec, the gas leak range of ultrasonic gas leak detectors is at most tens of meters.

[0066] Figure 2This is a diagram of a linear array 20 of five MEMS microphones (1, 2, 3, 4, and 5), with a microphone spacing of d and sound incident at an angle θ to the array axis. The rightmost MEMS microphone 1 receives sound earlier than its leftmost MEMS microphone 2; this difference is called the time difference of arrival (TDOA). In the far-field approximation, the sound source is farther away compared to the microphone spacing and wavelength. This approximation is valid even if a gas leak is several meters away. In the far field, the sound wavefront received within the array's range can be considered planar rather than spherical; for a planar wavefront, the difference in time δt for receiving sound between consecutive microphones is:

[0067] δt=(d*cosθ) / v (1)

[0068] Here, v is the speed of sound in air, typically 340 m / sec. In the frequency domain, this delay causes a phase shift between the signals received by the microphone. The delay is directly related to the angle of incidence and geometry of the microphone array. Given the geometry of the microphone array, the delay, or phase difference, can be used to estimate the angle of arrival of the incident energy.

[0069] A technique known as time-delay summation beamforming can be used to estimate the angle of incidence θ. If a time delay is added to the recorded signal from each microphone, equal to and relative to the delay caused by the extra travel time, signals that are perfectly in phase with each other will be produced. Summing these in-phase signals will cause constructive interference, which will amplify the result through several microphones in the array and produce the main lobe in the directivity pattern. This is called time-delay summation beamforming. For DOA (Direction of Arrival) estimation (also known as AOA (Angle of Arrival)), the time delay for all possible directions can be iteratively tested. If the guess is wrong, the signals will cancel each other out, resulting in a reduced output signal, while a correct guess will result in the signal amplification described above. In practice, the time delay is swept continuously, resulting in beamsteering from an initial angle (typically 0 degrees) to a final angle (typically 180 degrees). The control angle is obtained by inverting Equation 1:

[0070] θ=cos⁻¹(δt*v / d) (2)

[0071] In an exemplary embodiment, the control angle increases in 2.5-degree increments over 72 steps, thereby providing 180-degree beam control. Figure 1Each of the two vertical ULAs independently performs this beam control. The azimuth and elevation (Ф, θ) angular coordinates generated by the two vertical ULAs referencing the axis perpendicular to the planar array produce a cone of sound sweep. In an exemplary embodiment, at each stride angle, for n data samples from Figure 1 The signal from each MEMS microphone in the array is sampled at 150 kHz, where n can be 256.

[0072] Figure 3 An exemplary embodiment of a software implementation of time delay and summation beamforming is shown. Signals (101 to 105) from a MEMS microphone are sampled for n data samples, where n can be 256, at a sampling rate of 150 kHz; Figure 1 and Figure 2 Five microphones are shown here as a ULA, but for beamforming, a ULA can have any number of microphones, greater than the minimum of two. The sampled data stream is split into even and odd values ​​by multiplying by an even number of 111s or an odd number of 112s. The even (I-in-phase) and odd (Q-out-of-phase or quadrature) data are combined at 113 to produce a complex number, which is more readily used by subsequent software processing to generate peak amplitudes corresponding to constructive interference along the direction of arrival (DOA). This in-phase or quadrature technique is well-known in digital signal processing. The complex number generated at 113 is multiplied by a scan factor 114, which is a complex number composed of the cosine and sine of the control angle θ (Equation 2).

[0073] Scan factor (n) = complex number (cos(θn), (sin(θn))(3)

[0074] Where n is the control angle direction between 0 degrees and 180 degrees (e.g., in 72 steps).

[0075] Data from each of the other MEMS microphones is processed similarly 115, and the processed signals from all MEMS microphones are summed 116. The summed data 116 is averaged 117 and normalized 118 to the maximum value calculated over all scan angles. This process is repeated 119 for each scan angle 120 from 0 degrees to 180 degrees. The direction of the maximum normalized value obtained over all control angles is the DOA of the ultrasound.

[0076] Continuous execution Figure 3The calculations described above, and in an exemplary embodiment, for the parameters described above (sampling rate 150 kHz, 256 data samples per stride, 72 strides), the time for a complete 0-degree to 180-degree scan is approximately 0.1 seconds. Using the beamforming parameters described above, the time delay and summing beamformer can therefore generate a DOA plot 121 once every tenth of a second. The beamforming plot can be averaged over ten scans, giving a response time of approximately one second. It should be noted that if there is more than one ultrasonic source, the beamforming plot can indicate more than one DOA vector; providing the user with the relative intensity of the local maximum peaks throughout the beamforming scan, thereby enabling appropriate action. The amplitude of the main lobe peak in the beamforming scan indicates the intensity of the ultrasonic source. A single planar microphone array calculates the direction of arrival without calculating the distance from the ultrasonic source; thus, a strong ultrasonic source at a distance can generate the same SPL and main lobe peak size as a weaker ultrasonic source closer to the microphone array.

[0077] In an exemplary embodiment, the DOA is meaningful only when the ultrasonic SPL is above a specified threshold, for example, 65 dB can be used as a threshold below which a warning can be generated instead of an alarm.

[0078] For example Figure 1 The exemplary 2D array shown performs DOA calculations independently for each ULA, providing two independent angles (Ф, θ) that provide the azimuth and elevation directions of the pressurized gas leak relative to the axis perpendicular to the planar microphone array. In a far-field approximation where the gas leak source to be detected is several meters away, Figure 1 The fact that the shape of the “T” in a MEMS microphone array does not cause any significant error in a more natural and centered MEMS microphone array in the form of a “+”, where the intersection of the two linear arrays is the origin of the coordinate system.

[0079] Figure 3 This is a flowchart illustrating an exemplary embodiment of time delay and summation beamforming for a directional gas leak detector. Other beamforming techniques are available and are known to those skilled in the art. These beamforming techniques include several types of frequency- or spectrum-based beamforming detailed in the references and can be considered within the scope and spirit of the invention. However, in Figure 1The planar array embodied here consists of two vertical linear arrays. Beamforming can also be performed using square, rectangular, or circular arrays on a plane or contoured surface. The trade-off for a larger number of analog or digital MEMS microphones is additional electronic circuitry, mathematically complex beamforming algorithms, and increased onboard computing power and memory. Various modifications and alterations can be made by those skilled in the art without departing from the scope and spirit of the invention.

[0080] For the use of MEMS microphones in gas leak detectors (industrial applications), the microphone is preferably encapsulated to meet operational requirements in hazardous locations defined by regulatory bodies. A globally recognized method of protection for gas detectors is the explosion-proof method (Ex d), which ensures that any explosive conditions are contained within the housing, preventing ignition of the surrounding environment, and utilizes a flame arrester as a protective element in front of the sensing element. The flame arrester prevents the transmission of accidentally ignited flames or explosions while allowing the flow of air and gas. Some of the standards widely accepted by industrial and government regulatory bodies for explosion-proof design are CSA C22.2 No. 30-M1986 from the Canadian Standards Association, FM 3600 and 3615 from Factory Mutual, and IEC 60079-0 and 60079-1 from the International Electrotechnical Commission.

[0081] Another protection method is intrinsic safety (Ex ia), which is covered by the International Electrotechnical Commission's IEC 60079-11. In intrinsic safety protection, an intrinsic safety circuit is used in conjunction with the sensing element; this blocking circuit limits the power of the sensing element so as not to generate a spark that could ignite an explosive gas mixture.

[0082] In practical implementations of MEMS microphones or microphone arrays for industrial applications, the array may be placed behind an acoustic transmission protective film, as detailed in U.S. Patent Application 2009 / 0060246A1 with explosion-proof (Ex d) enclosures or intrinsically safe (Ex ia) protection methods.

[0083] Figure 4 An exemplary embodiment of an explosion-proof acoustic system 50 is illustrated. This exemplary embodiment includes a housing 52, which in this embodiment may be made of aluminum, stainless steel, or other industrial metals with suitable tensile strength. In this example, the housing 52 may have a generally cylindrical configuration with a hollow, open internal space. The outer and inner surfaces of the housing may be machined or manufactured using various shoulder and stepped surfaces, and the outer surface may include threaded portions. For example, Figure 4The system 50 depicted includes a housing 52 having a hollow internal space, typically depicted as 52A. At the transducer end 52E of the housing, the hollow internal space is formed by a chamfer or inlet 52B, providing a port or window for acoustic energy to impact a transducer mounted within the housing. The hollow or open internal region 52A of the housing may be formed by support shoulder surfaces 52C and 52D, which can be aligned with the positions of specific components of the system 50.

[0084] The exemplary system 50 includes a porous metal sintered disk 53, which in an exemplary embodiment may be made of type 316L stainless steel. The disk 53 is disposed on the front of a housing 52 or transducer end 52E aligned in place by a shoulder 52C and serves as a flame arrester. The sintered disk 53 may be press-fitted into the housing, attached to the housing, or integrally formed with the housing. The latter method may be used if both the sintered material and the housing material are stainless steel. In an exemplary embodiment, the thickness of the sintered metal disk 53 may be one-eighth of an inch, with a maximum aperture of 250 micrometers, determined according to the design guidelines and testing of the regulatory body to meet the requirements of the Ex d protection method. The exemplary disk diameter is 1.0 inch. A microphone array 54 is mounted on the front side of a circuit board 55 and positioned close to but not in contact with the sintered metal disk; the exemplary spacing between the disk and the transducer is 0.015 inches.

[0085] If the MEMS microphone is of a type known as "bottom port" or "zero height," the microphone array can alternatively be mounted on the rear side of board 55; in this case, ultrasonic waves reach the MEMS microphone through vias or holes in the board. The Knowles MEMS microphone SPU0410LR5H is an exemplary bottom port analog microphone that can be used. The Knowles MEMS microphone SPH0641LU4H is an exemplary bottom port digital microphone that can be used alternatively. Figure 4C An alternative embodiment of the explosion-proof acoustic system 50' is shown, wherein the array 54' is mounted on the back side of the circuit board 55, away from the flame arrester 53, and the ultrasonic energy passes through holes or openings 64 in the circuit board 55. Figure 4C The embodiments are similar in other respects Figure 4 Examples of implementations.

[0086] For system 50 to operate in a humid environment, a hydrophobic membrane 56 can be placed between the sintering tray 53 and the external environment. Membrane 56 can be selected for its excellent acoustic transmission properties; an example of such a membrane is the Versapor 5000R manufactured by Pall Corporation in Port Washington, New York. Other membranes suitable for this purpose are manufactured by WLGore & Associates, Inc. in Elkton, Maryland.

[0087] Preferably, the porosity and thickness of the sintered metal disk are selected such that the disk does not significantly degrade the transmission of acoustic waves to the microphone within the desired frequency range (e.g., ultrasonic frequencies). Therefore, the sintered disk 53 provides protection not only for operation in hazardous environments but also for dust and water, while still allowing excellent acoustic wave transmission. If desired, a hydrophobic membrane 56 provides additional environmental protection. It also prevents dust and moisture from reaching the sintered metal disk 53, thereby preventing the porous metal disk from becoming clogged.

[0088] Still refer to Figure 4 Wires 57 for the bias, ground, and signal lines of the acoustic transducer extend from circuit board 55 to the back of the housing. The typical bias voltage for a MEMS microphone is 2.5V dc. A retainer 58 holds the components in place within the hollow area of ​​the housing, while the remaining portion of the hollow area in housing 52 is sealed from the rear using an electrically insulating potting compound 59. An example of an agency-compliant potting compound suitable for this purpose is the Stycast 2850FT from Emerson & Cuming.

[0089] In an exemplary embodiment, the acoustic system 50 with an explosion-proof microphone array can provide a complete sensor for ultrasonic detection. With its encapsulated rear end and sintered metal disc front end, it is suitable for operation in explosively hazardous locations, either as a remotely mounted standalone sensor or as a sensor attached, for example, by threaded engagement, to another housing also suitable for hazardous locations. Threads 60 on the housing 52 of the system 50 allow the sensor housing to be screwed into such a second housing. For example, Figure 4A Depicting installation in housing 70 Figure 4 The sensor 50, the housing 70 contains electronics for conditioning and processing the sensor microphone array signal and subsequently generating an output for the user.

[0090] Reference Figure 4A The ultrasonic array transducer system 50 is shown screwed into the housing 70, which also meets the requirements for operation in explosive hazardous environments. The acoustic system 50 can be, for example, system 50 (…). Figure 4 In cases where the housing 70 itself is explosion-proof, the acoustic system 50 of this application may omit the fire-resistant sealing structure 59 in some embodiments. In this exemplary embodiment, the housing 70 includes a main housing or bottom portion 72 and a cover 74; the cover 74 is shown to have an optical window 76 for observing the displayed signal proportional to the acoustic sound pressure level. The bottom 72 of the housing has features that enable the housing to be mounted on a suitable mounting bracket.

[0091] The second housing 70 may include electronics necessary for powering the microphone, processing electrical signals generated by the microphone array, and providing output to the user to monitor and record acoustic signals.

[0092] Furthermore, in other embodiments, the microphone array system 50 can be mounted remotely from the housing 70, and the connection between the remote housing and the housing can meet the requirements for operation in explosive hazardous environments. Figure 4B The diagram illustrates an exemplary embodiment of a remotely mounted microphone array. A communication link, such as cable 70A in a conduit, provides signal connectivity between the microphone array 50 and the housing 70.

[0093] MEMS microphones are low-power, low-voltage devices that typically operate at 2.5V and draw less than 1mA of current. Therefore, MEMS microphones are well-suited for intrinsically safe protection methods. Figure 5 An exemplary intrinsic safety circuit 150 is shown connecting a MEMS microphone array 160 and a power supply 158 in a hazardous area. Circuit 150 utilizes a current-limiting resistor 156 and Zener diodes 152A and 152B; two Zener diodes are used redundantly in this exemplary embodiment. A Zener diode is a diode that not only allows current to flow in the forward direction in the same manner as an ideal diode, but also permits current to flow in the reverse direction when the voltage exceeds a specific, well-defined value known as the breakdown voltage. If an unexpected high voltage is generated inside the instrument's safe area, the high reverse current through the Zener diode will cause fuse 154 to blow before electrical energy is delivered to the microphone array 160 located in the hazardous area.

[0094] like Figure 6As shown, in an exemplary embodiment of the gas leak detector system 150 employing a MEMS microphone array, the outputs of nine MEMS microphones (152a to 152i) are signal conditioned (153a to 153i), then digitized (153) if the microphone output is analog, and further processed by the ultrasonic gas leak detector 150, which includes an electronic controller 155 (e.g., a digital signal processor (DSP)), an ASIC, or a microcomputer-based or microprocessor-based system. For cases where the microphones provide digitized output, signal conditioning 153a to 153i includes analog-to-digital conversion and does not require an internal ADC 153. In the exemplary embodiment, the signal processor 155 may include a DSP, although other devices or logic circuitry may alternatively be used for other applications and embodiments. In an exemplary embodiment, the signal processor 155 also includes dual universal asynchronous receiver / transmitter (UART) 151 as a serial communication interface (SCI), a serial peripheral interface (SPI) 152, an internal ADC 153 (if needed), an external memory interface (EMIF) 154 for external memory (SRAM) 21, and a non-volatile memory (NVM) 156 for on-chip data storage. The Modbus 91 or HART 92 protocol can be used as the interface for serial communication via the UART 151. These two protocols are well-known in the manufacturing industry and, along with other protocols such as PROFIbus, Fieldbus, and CANbus, are used to interface field instruments to a user's computer or programmable logic controller (PLC). In an exemplary embodiment, the signal processor 155 is connected to several other interfaces via the SPI 152. These interfaces may include an external NVM 22, an alarm relay 23, a fault relay 24, a display 25, and an analog output 26.

[0095] In an exemplary embodiment, analog output 26 may generate an indicative current level between 0 mA and 20 mA, which can be used to trigger remedial actions, such as, by way of example only, shutting down a processing device in accordance with established facility protocols. A first current level at analog output 26 (e.g., between 4 mA and 20 mA) may indicate a gas leak, a second current level at analog output 26 (e.g., 4 mA) may indicate normal operation (e.g., when there is no gas leak), and a third current level at analog output 26 (e.g., 0 mA) may indicate a system failure that may be caused by conditions such as electrical faults. In other embodiments, other current levels may be selected to represent various conditions.

[0096] In an exemplary embodiment, as discussed more fully below, the signal processor 155 is programmed to perform signal preprocessing and artificial neural network (ANN) processing. As described more fully below, the signal processor 155 performs beamforming functionality. Upon detection of ultrasonic waves from a gas leak, the calculated azimuth and elevation angles (Φ, θ) of the ultrasonic waves, along with the measured sound pressure level (SPL), can be transmitted to the user via display 25, Modbus, or HART (91 or 92). In another exemplary embodiment, to be described more fully, this information can be used to overlay the gas leak SPL and direction coordinates onto an image of the monitored scene.

[0097] US 2014 / 0000347 details how artificial neural networks (ANNs) can be used in ultrasonic gas leak detectors to distinguish ultrasonic waves generated by pressurized gas leaks from noisy ultrasonic waves generated by, for example, mechanical or biological sources. ANNs can be used in conjunction with a sound pressure level (SPL) threshold (US2014 / 0000347A1). Figure 2 Use in combination, or avoid using the SPL threshold (US2014 / 0000347A1). Figure 5 In this exemplary embodiment, where beamforming is used to determine the DOA of the received ultrasonic waves, the teachings of US2014 / 0000347A1 can be used to determine whether the received ultrasonic energy originates from a pressurized gas leak (threat) or from a noise source (false alarm). SPL measurements and ANN calculations can be performed using signals from any MEMS microphone belonging to either of the two ULAs. In an example where a single ultrasonic source is illustrated in the DOA plot, the direction of either the pressurized gas leak or the noise ultrasonic wave can be identified. Signal processing based on non-ANN can also be used to distinguish the presence of a gas leak from the noise ultrasonic wave. Alternatively, the ultrasonic gas leak detector can be a simple threshold device without means of distinguishing between gas leak ultrasonic waves and noise ultrasonic waves. Therefore, beamforming using ULA and DOA calculations is independent of the gas leak discrimination capability of the ultrasonic gas leak detector. As mentioned earlier, an SPL threshold (e.g., 65 dB) can be used in conjunction with DOA information to determine the need for user action: such a threshold eliminates the triggering of alarms for low-level detected gas leaks or noise ultrasonic waves.

[0098] Figure 7 It can be achieved through the digital signal processor 155 ( Figure 6 An exemplary functional block diagram 200 of an exemplary directional ultrasonic gas detector is implemented using appropriate programming. For example, [the following is an example of a directional ultrasonic gas detector]. Figure 1The outputs of the nine analog (in this exemplary embodiment) MEMS microphones (152a to 152i) of the two vertical ULAs are signal conditioned (153a to 153i) and then digitized 211, typically indicated as 210. A processing algorithm 220 is then applied to the sensor data, including signal preprocessing 221, ANN verification function 222, and post-processing 224, as detailed in U.S. Publication US2014 / 0000347A1, and sound pressure calculation 223. For example, in Figure 3 The beamforming 100 described in the flowchart provides the angle of arrival and peak value of the incident ultrasonic wave. In an exemplary embodiment, the calculated sound pressure level (SPL) 223 is compared with a preset threshold 227, while a post-processed ANN determines whether the microphone signal is generated by an actual gas leak 225. In an exemplary embodiment, the combination of decision blocks 225 and 227 produces four combinations:

[0099] Output state 228A is for the combination of: (1) gas leak (yes); and (2) SPL > threshold (yes).

[0100] Output state 228B is for the combination of: (1) gas leak (no); and (2) SPL > threshold (yes).

[0101] Output state 228C is for the combination of: (1) gas leak (yes); and (2) SPL > threshold (no).

[0102] Output state 228D is for the combination of: (1) gas leak (no); and (2) SPL > threshold (no).

[0103] Output state 228A corresponds to a real gas leak, where the ultrasonic waves exceed the SPL threshold 227 and are emitted from the calculated (Ф, θ) azimuth and elevation angles. Threshold 227 can be considered the gas detection threshold; the user can choose to set a higher alarm threshold for alarm relay 23 in output block 230. Output state 228B corresponds to a situation where a large measured SPL has been diagnosed not as being caused by a gas leak, but by a false alarm source located at the calculated (Ф, θ) azimuth and elevation angles. Output state 228C corresponds to the detection of a real gas leak, but with an amplitude small enough to produce an SPL less than threshold 127. Output state 228C can be considered a minor leak, or a warning of a larger, urgent leak to the user. The user typically does not take corrective action but is advised to monitor the facility more closely. Output state 228D corresponds to a situation where nothing has happened; there is no evidence of a gas leak, no peaks in the beamforming plot, and the background SPL value is considered irrelevant. Output State 228D is typically used in quiet industrial environments, such as remote onshore wellheads.

[0104] Now refer to Figure 8 This describes features of another exemplary embodiment of the ultrasonic gas leak detector 250, which depicts a functional block diagram of the gas leak detector. This embodiment is similar to... Figure 7 The embodiment depicted herein. However, in this exemplary embodiment, the signal processor 155 is programmed to implement processing algorithm 220', wherein, as... Figure 7 As shown in block 227, the SPL calculated based on sound pressure calculation 223 is not compared with a preset threshold. Instead, the calculated SPL 229 is sent directly to output block 230. Simultaneously, the post-processed ANN determines via decision block 225 whether the ultrasound is generated by a real gas leak as indicated by output state 232 or a false alarm as indicated by output state 231. Then, output block 230 notifies the user of the presence of a real gas leak (according to output state 232), AOA direction 27 (Ф, θ), and severity (SPL, in dB) (according to signal 229) via the output functions of alarm relay 23, display 25, analog output 26, and external communication interfaces (such as Modbus 91 and HART 92). If the calculated SPL is shown as generated by a false alarm via output state 231 according to decision block 225, output block 230 can similarly notify the user of the false alarm event and its severity (in dB) via display 25, analog output 26 and external communication interfaces (such as Modbus 91 and HART 92); however, in the case of a false alarm event indicated by output state 231, alarm relay 23 may not be activated.

[0105] In an exemplary embodiment, the direction of the gas leak can be transmitted to the user via angular coordinates of the azimuth and elevation angles (Φ, θ) of the central axis of the sweeping sound cone. The measured beamforming peak size and the direction of the gas leak can also be conveyed via a graphical representation depicting the direction of the gas leak in the sound cone, for example, via discrete LEDs or a dot matrix display.

[0106] To better communicate the direction and size of leaking gas to users visually, practically, and quantitatively, more sophisticated embodiments can display the angle of arrival (AoA) of ultrasonic energy in a cone scan by overlaying visual information onto an image display using a consistent optical field of view. Figure 7 or Figure 8(27). Therefore, the user can see the actual location of the gas leak and its SPL value, visually overlaid on an image of the device, which may be the source of a gas leak generating strong ultrasonic energy detectable several meters away. While depth or distance information is not provided by a single 2D planar microphone array, it is clear from the image that the ultrasonic energy is emanating from the suspected device, not from the air space in between. This visual image also has the advantage that known, friendly gas leaks (such as pressurized air releases for maintenance purposes) can be zoned out by the user. Increases in mechanically generated ultrasonic noise can also be easily monitored, while using techniques such as the previously described ANN to distinguish this ultrasonic noise from actual gas leaks. Further benefits include recording and playback of ultrasonic overlay images of events that cause alarms (or false alarms), including highlighting equipment malfunctions, the temporal evolution of events, and safety aspects of the industrial facility. This ultrasonic overlay image can be continuously monitored via webcams or on an internal network of security cameras.

[0107] Superimposing ultrasonic energy onto visible images for industrial applications in hazardous environments typically involves low-cost MEMS microphone arrays, analog and digital electronics housed in enclosures suitable for hazardous locations, and industrial imagers. This ultrasonic imaging of real-world leaks of flammable and toxic gases offers significant benefits to stationary gas detection systems through conventional point and open-path gas detectors, as well as infrared gas imaging solutions. Infrared gas cloud imaging cameras are expensive, their sensitivity varies greatly depending on the gas being monitored, and their performance is largely dependent on the difference between the leaking gas cloud temperature and the background temperature. For these reasons, infrared gas cloud imaging cameras are not readily adopted in industrial stationary gas detection systems, although they can perform remote gas cloud detection over hundreds of meters.

[0108] The ultrasonic approach described above does not suffer from many of the drawbacks associated with infrared gas cloud imaging. The main requirements for ultrasonic gas leak detection, location, and imaging are that the leaking gas is under pressure, regardless of whether it is flammable, toxic, or inert, and that the leak is at most a few tens of meters away (typically less than 30 or 50 meters). Under these conditions, pressurized gas leaks of large quantities of hydrocarbon gases, toxic gases, and even inert gases (such as helium) can be easily detected using ultrasonic gas leak detectors. Highly flammable gases (such as hydrogen) that cannot be detected by optical or infrared devices can be easily detected using ultrasonic gas leak detectors, and beamforming arrays can be used to present leak location information even when ultrasonic waves are superimposed on the visible imaging scheme described above.

[0109] Figure 9A screenshot depicts an acoustic source 300 generating ultrasonic waves, positioned in the center of a visible camera image displayed using LabVIEW from National Instruments. Circle 302 represents the vector position of the sound source, calculated by the time delay and summation beamformer 100 described above and superimposed on the visible image. The circle follows the sound source as it moves to different positions. Pointers 304 and 306 indicate the horizontal and vertical angular coordinates of the sound source. The horizontal direction is 90 degrees, and therefore on the MEMS microphone array axis. The vertical direction is 95 degrees, and therefore 5 degrees below the MEMS microphone array axis. The measured SPL is shown as 69 dB, which is greater than the threshold set to 65 dB.

[0110] In another embodiment, the amplitude of the ultrasound waves measured by the beamformer for each scan angular direction can be superimposed on the corresponding visible image pixels to provide a continuous ultrasound map of the observed scene. For an exemplary embodiment, for each of the 72 azimuth scan angular directions, there are also 72 elevation scan angular directions that generate a matrix of 72×72 scan angular directions, where the beamformer calculates the ultrasound amplitude for each of the 72×72 directions, totaling 5184 directions in this example. These ultrasound amplitudes can be superimposed on the corresponding visible image pixel areas, for example, using a color-coding scheme as used for temperature measurements using a thermal imager. Alternatively, the numerical form of the ultrasound amplitude at each location can be displayed superimposed on the image of the observed scene.

[0111] Figure 10 A display representing a monitored scene is schematically shown, wherein a directional acoustic detector has a field of view centered on equipment 500 (e.g., a compressor) in a hazardous location. A beamforming array generates signals indicating the ultrasonic SPL generated by different parts of the equipment. In this example, a part of the equipment generates a stronger ultrasonic SPL than other parts, resulting in local maxima or peaks as the beamformer continuously scans in the azimuth and elevation directions. Figure 10 This shows several peak amplitudes measured when the beamformer continuously scans in the azimuth and elevation directions. The computer-generated SPL amplitudes are superimposed on an image of the monitored scene; Figure 10 In an exemplary embodiment, only SPL values ​​greater than or equal to a threshold of 63 dB are displayed on the monitor. Therefore, the continuous scanning beamforming array is able to monitor the compressor and provide a record of ultrasonic emissions from different parts of the compressor. This information can be useful to maintenance and process personnel.

[0112] Although specific embodiments of the subject matter have been described and illustrated above, various modifications and changes can be made to it by those skilled in the art without departing from the scope and spirit of the invention.

Claims

1. A directional ultrasonic gas leak detector, comprising: The circuit includes a spaced-out MEMS microphone array, each microphone responding to incident airborne ultrasonic energy from a gas leak source positioned within the array to generate a microphone signal; The array is a two-dimensional array, and the microphones of the array are evenly and equally spaced in two orthogonal directions to form at least two linear arrays. A beamforming processor, in response to a microphone signal from the MEMS microphone array, generates a processor output signal indicating an estimated angle of arrival of ultrasonic energy incident on the array from the gas leak source within the array's range, wherein the processor output signal includes signals representing the azimuth and elevation coordinates of the detected gas leak; The beamforming processor is configured to: independently perform beam control processing from an initial control angle to a final control angle for each of the first and second uniform linear arrays; and for each control angle, sum the data corresponding to each MEMS microphone in the corresponding array, average the summed data, and normalize the summed data to the maximum value calculated at all control angles, wherein the direction of the maximum normalized value obtained at all control angles is the estimated angle of arrival of the ultrasonic energy; The housing structure includes a spaced-apart MEMS microphone array and a processor disposed within the housing structure. The housing structure includes a port through which incident ultrasonic energy is transmitted from a gas leak source to the array. The processor is configured to indicate the estimated angle of arrival of the ultrasonic energy only when the sound pressure level of the incident ultrasonic energy exceeds a threshold.

2. The detector according to claim 1, wherein, The microphones in the array are spaced apart from adjacent MEMS microphones in the array by a distance not exceeding 5 mm.

3. The detector according to claim 1, further comprising: A display for displaying images of a scene monitored by the detector within the range, including a device for delivering or storing gas therethrough; The display responds to a signal output by the processor to depict the location of the gas leak source superimposed on the image.

4. The detector according to claim 3, wherein, The display also responds to the processor output signal to depict the intensity of the gas leak source.

5. The detector according to claim 1, wherein, The beamforming processor is configured to perform time-delay and summation beamforming on the microphone signal.

6. The detector according to claim 1, wherein, The array is mounted on a circuit board, and the microphone covers an area of ​​no more than 10 square centimeters.

7. The detector according to claim 1, wherein, The opening size of the port is no greater than 10 square centimeters.

8. The detector according to claim 1, wherein, The outer shell structure is an explosion-proof outer shell structure.

9. The detector of claim 8 further includes a flame arrester structure that covers the port of the housing structure.

10. The detector of claim 8 further includes a hydrophobic membrane disposed in or covering the port.

11. The detector according to claim 1, wherein, The circuit includes a blocking circuit connected to the microphone array, the blocking circuit being configured to limit the power applied to the microphone array to prevent ignition of an explosive gas mixture.

12. The detector according to claim 1, wherein, The processor responds to the microphone signal from the MEMS microphone array to process multiple beam directions and generate a processor output signal indicating the estimated angle of arrival of the ultrasonic energy incident on the array in the beam direction that generates the maximum response.

13. The detector according to claim 1, wherein, The beamforming processor responds to broadband incident airborne ultrasonic energy from a gas leak source positioned within 50 meters of the array to generate the processor output signal.

14. A directional ultrasonic energy detector, comprising: A two-dimensional array of spaced-apart MEMS microphones, each microphone responding to incident airborne ultrasonic energy from a gas leak source positioned within the range of the array to generate a microphone signal; A processor, in response to the microphone signals from the MEMS microphone array, processes multiple beam directions and generates a processor output signal indicating an estimated amplitude of ultrasonic energy incident on the array from each beam direction, wherein the processor output signal includes signals representing the azimuth and elevation coordinates of a detected gas leak; The housing structure includes a spaced-apart MEMS microphone array and a processor disposed within the housing structure. The housing structure includes a port through which incident ultrasonic energy is transmitted from the gas leak source to the array. and A display, in response to a processor output signal and for generating an image representing a monitored scene, wherein the estimated amplitude of the incident energy in the beam direction is superimposed on the image; The processor is configured to indicate the estimated angle of arrival of the ultrasonic energy only when the sound pressure level of the incident ultrasonic energy exceeds a threshold.

15. The detector according to claim 14, wherein, The array is mounted on a circuit board, and the microphone covers an area of ​​no more than 10 square centimeters.

16. The detector according to claim 15, wherein, The outer shell structure is an explosion-proof outer shell structure.

17. The detector of claim 14, further comprising a blocking circuit connected to the microphone array, the blocking circuit being configured to limit the power applied to the microphone array to prevent ignition of an explosive gas mixture.

18. The detector of claim 14, further comprising: Artificial neural networks (ANNs) respond to received ultrasonic energy and are configured to distinguish between ultrasonic waves generated by pressurized gas leaks and noisy ultrasonic waves.

19. A directional ultrasonic gas leak detector, comprising: The circuit includes a spaced-out MEMS microphone array, each microphone responding to incident airborne ultrasonic energy from a gas leak source positioned within the array to generate a microphone signal; The array is a two-dimensional array, and the microphones of the array are evenly and equally spaced in two orthogonal directions to form at least two linear arrays. A beamforming processor, in response to a microphone signal from the MEMS microphone array, generates a processor output signal indicating the estimated angle of arrival of ultrasonic energy incident on the array from the gas leak source within the array's range; The beamforming processor is configured to iteratively perform beam control processing from an initial control angle to a final control angle, and for each control angle, sum the data corresponding to each MEMS microphone in the corresponding array and determine the control angle that provides the maximum sum value as the corresponding azimuth and elevation coordinates of the detected leaking gas. The housing structure includes a spaced-apart MEMS microphone array and a processor disposed within the housing structure. The housing structure includes a port through which incident ultrasonic energy is transmitted from a gas leak source to the array. The processor is configured to indicate the estimated angle of arrival of the ultrasonic energy only when the sound pressure level of the incident ultrasonic energy exceeds a threshold.

20. The detector according to claim 19, wherein, The beamforming processor is configured to perform beam control processing independently for each of the first and second uniform linear arrays.

21. The detector according to claim 19 or 20, wherein, The microphones in the array are spaced apart from adjacent MEMS microphones in the array by a distance not exceeding 5 mm.

22. The detector according to claim 19 or 20, further comprising: A display for displaying images of a scene monitored by the detector within the range, including a device for delivering or storing gas therethrough; The display responds to a signal output by the processor to depict the location of the gas leak source superimposed on the image.

23. The detector according to claim 22, wherein, The display also responds to the processor output signal to depict the intensity of the gas leak source.

24. The detector according to any one of claims 19 to 20 and 23, wherein, The processor output signal includes signals representing the azimuth and elevation coordinates of the detected gas leak.

25. The detector according to any one of claims 19 to 20 and 23, wherein, The beamforming processor is configured to perform time-delay and summation beamforming on the microphone signal.

26. The detector according to any one of claims 19 to 20 and 23, wherein, The array is mounted on a circuit board, and the microphone covers an area of ​​no more than 10 square centimeters.

27. The detector according to any one of claims 19 to 20 and 23, wherein, The outer shell structure is an explosion-proof outer shell structure.

28. The detector according to any one of claims 19 to 20 and 23, further comprising a flame arrester structure covering the port of the housing structure.

29. The detector according to any one of claims 19 to 20 and 23, wherein, The circuit includes a blocking circuit connected to the microphone array, the blocking circuit being configured to limit the power applied to the microphone array to prevent ignition of an explosive gas mixture.

30. The detector according to any one of claims 19 to 20 and 23, wherein, The processor responds to the microphone signal from the MEMS microphone array to process multiple beam directions and generate a processor output signal indicating the estimated angle of arrival of the ultrasonic energy incident on the array in the beam direction that generates the maximum response.

31. The detector according to any one of claims 19 to 20 and 23, wherein, The beamforming processor responds to broadband incident airborne ultrasonic energy from a gas leak source positioned within 50 meters of the array to generate the processor output signal.

32. The detector according to any one of claims 19 to 20 and 23, further comprising: Artificial neural networks (ANNs) respond to received ultrasonic energy and are configured to distinguish between ultrasonic waves generated by pressurized gas leaks and noisy ultrasonic waves.

33. A method for detecting gas leaks, comprising: The sensor is positioned in the monitored space, the sensor comprising an array of spaced-apart MEMS microphones, each microphone generating a microphone signal in response to incident airborne ultrasonic energy from a gas leak source located within the array. The array is a two-dimensional array, and the microphones of the array are evenly and equally spaced in two orthogonal directions to form at least two linear arrays. The microphone signals from the array of MEMS microphones are processed to generate a processor output signal indicating the estimated angle of arrival of ultrasonic energy incident on the array from the gas leak source within the range of the array, wherein the processor output signal includes signals representing the azimuth and elevation coordinates of the detected gas leak; The process includes independently performing beam control processing from an initial control angle to a final control angle for each of the first and second uniform linear arrays, and for each control angle, summing the data corresponding to each MEMS microphone in the corresponding array, averaging the summed data, and normalizing the summed data to the maximum value calculated over all control angles, wherein the direction of the maximum normalized value obtained over all control angles is the estimated angle of arrival of the ultrasonic energy. The array is housed within a housing structure, which includes a port through which incident ultrasonic energy is transmitted from a gas leak source to the array. The processing is configured to indicate the estimated angle of arrival of the ultrasonic energy only when the sound pressure level of the incident ultrasonic energy exceeds a threshold.

Images