Multi-spectral optical detector with improved false alarm prevention

Abstract

The invention relates to a multispectral optical detector with improved misinformation prevention. A multi-spectral optical detector includes a housing having a window. The first channel is disposed within the housing adjacent the window and is sensitive to electromagnetic radiation within a first frequency range. The second channel is disposed within the housing adjacent the window and is sensitive to electromagnetic radiation within a second frequency range different from the first frequency range. A reference channel is also disposed within the housing and is substantially insensitive to optical electromagnetic radiation. The digitizer is operably coupled to the first channel, the second channel, and the reference channel, and is configured to provide digital indications of the first channel, the second channel, and the reference channel. A processor is coupled to the digitizer to receive the digital indication and generate a flame output signal based thereon. The processor is configured to calculate a correlation between the digital indication of the first channel and the digital indication of the second channel.

Description

Technical Field

[0001] This invention relates to a multi-spectral optical detector with improved false alarm prevention. Background Technology

[0002] The process control and monitoring industry supports a wide range of processing industries. Some processing industries may use or handle highly flammable or even explosive materials. Examples of such industries include chemical processing facilities as well as oil extraction and refining. In these environments, fire and explosion are significant hazards. In these highly unstable environments, the use of one or more optical detectors (e.g., optical flame detectors) (which detect any flame in the process environment so that it can be quickly extinguished) is useful and sometimes necessary. Summary of the Invention

[0003] A multi-spectral optical detector includes a housing with a window. A first channel is disposed within the housing adjacent to the window and is sensitive to electromagnetic radiation in a first frequency range. A second channel is disposed within the housing adjacent to the window and is sensitive to electromagnetic radiation in a second frequency range different from the first frequency range. A reference channel is also disposed within the housing and is substantially insensitive to optical electromagnetic radiation. A digitizer is operatively coupled to the first channel, the second channel, and the reference channel and configured to provide digital indications for the first channel, the second channel, and the reference channel. A processor is coupled to the digitizer to receive the digital indications and generate a flame output signal based thereon. The processor is configured to calculate a correlation between the digital indications for the first channel and the digital indications for the second channel, and is configured to compare the digital indications for the first and second channels with the digital indication for the reference channel to generate an EMI indication based on the correlation and comparison. A method for detecting electromagnetic interference using at least one multi-channel optical detector is also provided. Attached Figure Description

[0004] Figure 1 This is a system block diagram of a multi-channel optical sensor, in which the embodiments described herein are particularly useful.

[0005] Figure 2 This is a diagram illustrating the EMI mode and reference channel response according to an embodiment of the present invention.

[0006] Figure 3 This is a flowchart of a computer-implemented method for detecting flames using an optical sensor with improved false alarm prevention, according to an embodiment of the present invention.

[0007] Figure 4 This is a system block diagram of a pair of multi-channel optical sensors according to another embodiment of the present invention, operating in a cloud environment. Detailed Implementation

[0008] Figure 1 This is a system block diagram of a multi-channel optical sensor, in which the embodiments described herein are particularly useful. When used in the context of optical flame detection, the multi-channel optical sensor 10 is a device designed to detect the presence of a flame by analyzing emissions on multiple spectral bands or channels. Using multiple channels offers several advantages. First, detection accuracy is improved because different materials have different spectral characteristics when burning. Multi-channel sensors can detect various types of fires by analyzing different portions of the electromagnetic spectrum, including ultraviolet (UV), visible, and infrared (IR) bands. Second, by comparing the intensity of emissions on several channels, these sensors can distinguish actual flames from other emission sources (e.g., sunlight, artificial light, or reflections) that might cause false alarms, thus reducing the false alarm rate. Third, reliability is improved because multi-channel sensors can operate effectively under a variety of environmental conditions, thereby reducing the risk of detection failure due to factors such as dust, moisture, or other atmospheric obstructions.

[0009] While the embodiments described herein are applicable to multispectral optical sensors using any kind of wavelength detection, one embodiment will be described for multiple individual IR sensors. Sensor 10 includes a housing 12 with a lens through which a flame 16 can be viewed. The flame 16 emits broad-spectrum infrared radiation. Sensor 10 includes multiple individual IR sensors IR1, IR2, IR3, ..., IRn, wherein each individual sensor is sensitive to a specific frequency band or IR wavelength. Thus, each individual sensor is substantially tuned or otherwise focused to the associated flame emission wavelength. Each of the IR sensors IR1, IR2, IR3, ..., IRn is operatively coupled to a digitizer 18, which includes circuitry for converting the analog signals from the individual IR sensors into their digital representations. Digitizer 18 is coupled to a processor 20 and configured to provide the processor 20 with digital representations associated with the various IR sensors.

[0010] Processor 20 is any suitable device capable of executing programmed steps or functions to provide various characteristics of sensor 10. Examples of such devices include digital signal processors, microcontrollers, field-programmable gate arrays, and application-specific integrated circuits (ASICs). In some examples, processor 20 is a microprocessor. Digitizer 18 provides processor 20 with a digital representation of the IR sensor signals for signal processing. Processor 20 processes the digitized signals from the IR sensors and analyzes the signals from each IR sensor. Processor 20 attempts to identify specific patterns associated with flame flicker and intensity. To analyze the flame flicker frequency, processor 20 typically converts the signal from the time domain to a frequency fast Fourier transform (FFT). By comparing the outputs of multiple IR sensors, processor 20 can distinguish flames from other IR sources such as sunlight, thermomechanical sources, or artificial lighting. This characterization is achieved by analyzing the intensity and frequency response of each channel. A real flame is characterized by a low-frequency signal with a response of 1-5 Hz, a high-intensity signal at the signal channel, and a low-intensity signal at the reference channel. The expected correlation (frequency response) of the channels is high, but not perfectly correlated, which would indicate an artificial signal. This multi-spectral analysis reduces false alarms. Processor 20 executes the following method: applying this analysis to calculate the integrity of the signal corresponding to the flame intensity, calculating the ratio between channels to understand whether the signal channel is higher than the reference channel. The final step is to compare the frequency behavior between different channels to measure the correlation between them.

[0011] When the processor 20 confirms the presence of a flame, it generates an output 22, such as triggering an alarm and / or other appropriate action. The sensor 10 can also initiate automatic safety measures, such as shutting down equipment and / or activating a fire suppression system.

[0012] One limitation of existing multi-channel optical detectors is the degradation of sensor data quality when detecting flames due to electromagnetic interference (EMI), a problem that can be prevalent in industrial environments with numerous electronic devices and machinery. Traditional EMI filtering methods often struggle to distinguish between real sensor signals and EMI, leading to the loss of valuable data or the inclusion of corrupted data in the analysis. Furthermore, EMI signals can sometimes resemble real flame signals, resulting in false alarms.

[0013] The embodiments described herein typically provide multi-spectral optical detectors with improved false alarm prevention. The embodiments typically utilize a variety of different techniques, individually or in combination, to better process sensor data and reduce the level of false alarms, including those due to EMI.

[0014] In one embodiment, sensor responses are processed to analyze the correlation between multiple sensor channels to detect EMI. As mentioned above, artificial EMI signals will cause very high correlations between individual sensor channels, a correlation that is impossible with real fire. When analyzing real flames, the expected correlation is above the detection threshold, but certainly below 100%. For example, real flames will typically have channel correlations in the range of 40% to 80%. This is due to the different responses of flames at different wavelengths. When analyzing EMI signals, high correlations are observed between channels, meaning that the frequency responses are very similar. This is due to the fact that EMI affects all channels in the same way and to the same extent.

[0015] In another embodiment, a reference channel is used. The reference channel employs the same IR sensor and circuitry as the sensing channel, but is optically shielded from all illumination. The reference channel is unaffected by flame and is expected to provide a zero-level signal. During EMI, the response of the reference channel is similar to the output of the channel exposed to the external world. This strongly indicates that the signal does not reach the sensor through the optical window, but rather through physical effects caused by EMI. By identifying these signals and confirming their presence via the reference channel, the system can filter out or otherwise eliminate or reduce interference without discarding or altering legitimate sensor data. In some embodiments, the signal from the reference channel is analyzed in the time and / or frequency domains.

[0016] Digital signal processing is used to provide EMI filtering, addressing the different frequency mode characteristics of EMI. In one embodiment, unique EMI modes are identified by performing Fourier transform analysis on the incoming signal. When analyzing a real fire, the expected frequency response is 1 Hz to 5 Hz, while EMI typically has frequency components above 10 Hz, caused by modulation. This difference allows for the isolation of signal segments affected by EMI. Through this identification process, the embodiments described herein effectively distinguish and filter the effects of EMI on incoming signals. This method ensures signal integrity by filtering out interference while preserving the original data.

[0017] By comparing and analyzing a reference signal with signals from other channels, the embodiments described herein are able to identify differences attributable to EMI. This comparison allows for the precise detection and / or elimination of electromagnetic interference affecting the channels of the sensor. Therefore, signal integrity is enhanced by utilizing a reference channel as a baseline for identifying and mitigating EMI.

[0018] Compared to existing methods, this approach offers significant advantages, including enhanced accuracy of sensor readings in EMI-sensitive environments, preservation of the sensor's original sensitivity and specificity, and the ability to dynamically adapt to different EMI characteristics.

[0019] Figure 2 This is a graph illustrating the EMI mode and reference channel response according to an embodiment of the present invention. Figure 2 As shown, an EMI noise signal 52 can be detected on reference channel 50, and all sensor channels respond to the noise signal, with a very high correlation between the various sensor channels. Furthermore, it can be seen that the frequency behavior of the noise signal 52 exhibits specific frequency characteristics.

[0020] Figure 3 This is a flowchart of a computer-implemented method for detecting flames using an optical sensor with improved false alarm prevention, according to an embodiment of the present invention. Method 100 can be practiced on any suitable hardware, including processor 20, a remote processor (e.g., cloud resources), or a combination thereof. Method 100 begins at block 102. Next, at block 104, sensor data is sampled. In an embodiment where method 100 is performed within a single multi-channel optical detector, the sensor data includes data from all IR sensor channels of the detector. However, in other embodiments, sensor data from multiple multi-channel optical detectors can be sampled. Next, at block 106, processor 20 or a remote processor calculates the correlation between the individual channels whose data were sampled in block 104. The correlation calculation can employ any suitable technique, including but not limited to obtaining Pearson correlation coefficients, Spearman correlation coefficients, and / or Kendall correlation coefficients. Once the correlation has been calculated, or while it is being calculated, processor 20 or the remote processor performs a Fast Fourier Transform on the sampled data of each channel sampled during block 104. The Fast Fourier Transform (FFT) is the process of computing the Discrete Fourier Transform (DFT) of a sequence or its inverse (IDFT). Fourier analysis transforms a signal from its original domain (usually time or space) into a representation in the frequency domain, and vice versa. In this case, the FFT is performed on the sampled data from each channel to identify the frequency characteristics in the sampled data.

[0021] At block 110, processor 20 or a remote processor identifies and / or classifies EMI frequencies from the frequency data generated by the Fast Fourier Transform of block 108. For example, detected frequency components that may not indicate a flame (i.e., above 10 Hz) can be classified as indicating EMI. Next, at block 112, processor 20 or the remote processor performs data analysis on the relevant sensor channel data and the identified / classified frequency data. This allows for adjustment of the data filters when EMI is observed in a sensor channel. Superior false alarm prevention results are provided by leveraging the correlation between sensor channels within this device and / or other devices, as well as frequency mode characteristics, and distinguishing and using one or more reference channels. Unlike methods that apply general filtering (hardware, optical, or software-based filters) to all data (which may lose valuable information or fail to effectively filter EMI), the embodiments described herein utilize sophisticated processing to analyze specific interference patterns on different channels. This helps to selectively eliminate or reduce EMI, thereby significantly reducing the likelihood of false alarms.

[0022] Next, at box 114, the data analyzed in box 112 is compared with data from one or more reference channels. As described above, the reference channel includes the same electronic components as the sensing channel and similar connections (e.g., circuit traces or wiring). The reference channel is set up by ensuring that no illumination can reach it. This can be achieved through optical shielding, mechanical design, or both. Therefore, the presence of a signal on the reference channel is itself interference. At box 114, the signal from the sensor channel, along with the analyzed data (e.g., threshold data, frequency characteristic data, etc.), is compared with the reference channel to determine whether the signal is a valid flame signal or EMI. Furthermore, in the presence of both a valid flame signal and EMI, comparison with the reference channel allows determination of whether the flame signal is strong enough to generate a flame indication, as shown in box 120. If the signal is determined to be EMI, method 100 can iterate, as shown in line 122, where control returns to box 104. However, if the frequency characteristics of the EMI have been identified, as shown in box 110, these frequencies can be selectively attenuated in the time domain, frequency domain, or both, so that the next iteration of data sampling is less sensitive in that frequency band.

[0023] Figure 4 This is a system block diagram of a pair of multi-channel optical sensors according to another embodiment of the present invention, operating in a cloud computing environment. Figure 4 This is a system block diagram of a pair of multi-channel optical sensors according to another embodiment of the present invention, operating in a cloud environment. Figure 4Each sensor 250 shown includes a communication module 252 coupled to the processor 20 and operatively coupled to the cloud computing system 300. The communication module 252 allows the processor 20 to communicate with remote devices (e.g., the cloud computing system 300). This communication can take any suitable form, but is preferably wireless. Examples of wireless communication include, but are not limited to: the wireless HART process communication protocol (IEC 62591); cellular communication protocols such as GPRS, UMTS, CDMA2000, LTE, LTE-M, NB-IoT, WiMax, 5G NR; WiFi standards such as IEEE 802.11b / g / n / a / ac / ax / be; and the LoRaWAN protocol (ITU-T Y.4480). Advanced data analytics, predictive maintenance, and remote energy monitoring can be achieved using cloud-based analytics (e.g., analytics provided by the cloud computing resource 300). In one example, the cloud computing resource 300 can perform method 100, and because the resource 300 is coupled to multiple sensors 250, it may be able to identify EMI better than a single multi-channel optical sensor. Furthermore, if the location of sensor 250 is known, cloud resource 300 may even be able to locate one or more EMI sources for remediation.

[0024] Although the invention has been described with reference to preferred embodiments, those skilled in the art will recognize that modifications in form and detail may be made without departing from the spirit and scope of the invention.

Claims

1. A multi-spectral optical detector, comprising: a housing having a window; a first channel disposed within the housing proximate the window and sensitive to electromagnetic radiation in a first frequency range; a second channel disposed within the housing proximate the window and sensitive to electromagnetic radiation in a second frequency range different from the first frequency range; a reference channel disposed within the housing and substantially insensitive to optical electromagnetic radiation; a digitizer operably coupled to the first channel, the second channel, and the reference channel, the digitizer configured to provide digital indications of the first channel, the second channel, and the reference channel; a processor coupled to the digitizer to receive the digital indications and generate a flame output signal based thereon; and wherein the processor is configured to compute a correlation between the digital indications of the first channel and the second channel, and wherein the processor is further configured to compare the digital indications of the first channel and the second channel to the digital indication of the reference channel to generate an EMI indication based on the correlation and the comparison.

2. The multi-spectral optical detector of claim 1, and further comprising at least one additional channel disposed within the housing proximate the window and sensitive to electromagnetic radiation in at least one additional frequency range different from the first frequency range and the second frequency range.

3. The multi-spectral optical detector of claim 1, wherein, the processor configured to determine whether the computed correlation is above a flame detection threshold but below an electromagnetic interference threshold, and selectively provide a flame output signal based on the determination.

4. The multi-spectral optical detector of claim 1, wherein, the processor configured to generate a fast Fourier transform of the digital indications of each of the first channel and the second channel.

5. The multi-spectral optical detector of claim 4, wherein, the processor configured to identify frequency characteristics of electromagnetic interference in the fast Fourier transform data.

6. The multi-spectral optical detector of claim 5, wherein, the processor configured to generate an electromagnetic interference output based on the at least one frequency characteristic or identification indicative of electromagnetic interference.

7. The multi-spectral optical detector of claim 6, wherein, the at least one frequency characteristic comprises a frequency greater than 10 hertz.

8. The multi-spectral optical detector of claim 6, wherein, the processor configured to attenuate a sensor channel signal based on the identified at least one frequency characteristic indicative of electromagnetic interference.

9. The multi-spectral optical detector of claim 5, wherein, the processor configured to recognize a pattern in the digital indication of the reference channel.

10. A method of detecting electromagnetic interference using at least one multi-channel optical detector, the method comprising: sampling data from a plurality of sensing channels of a first multi-channel optical detector; sampling data from a reference channel of the first multi-channel optical detector; computing a correlation between the plurality of sensor channels; generating fast Fourier transform information for the data of each sensor channel; identifying at least one EMI frequency in the fast Fourier transform; analyzing the correlation and the identified EMI frequency, and comparing the correlation to reference data to determine whether the data sampled from the sensing channels is indicative of a flame or electromagnetic interference; and selectively providing a flame output based on the determination.

11. The method of claim 10, wherein, Computing the correlation between the plurality of sensor channels is performed by a processor of the first multi-channel optical detector.

12. The method of claim 10, wherein, Analyzing the correlation includes comparing the correlation to a range having a lower limit and an upper limit.

13. The method of claim 12, wherein, Correlations above the upper limit indicate electromagnetic interference.

14. The method of claim 10, wherein, The processor is configured to compute a correlation between the sensing channel and the reference channel.

15. The method of claim 10, wherein, Analyzing the correlation and identifying at least one EMI frequency is performed by a remote device.

16. The method of claim 10, wherein, The at least one multi-channel optical detector includes a plurality of multi-channel optical detectors.

17. The method of claim 16, wherein, Each of the plurality of multi-channel optical detectors is configured to communicate with a cloud computing resource.

18. The method of claim 17, wherein, The cloud computing resource is configured to perform at least one of: computing the correlation between the plurality of sensor channels; generating fast Fourier transform information for data of each sensor channel; identifying at least one EMI frequency in the fast Fourier transform; analyzing the correlation and the identified EMI frequency, and comparing the correlation to reference data to determine whether data sampled from the sensing channel is indicative of a flame or electromagnetic interference; and and selecting to provide a flame output based on the determination.

Images