Method, system and equipment for detecting oil mist concentration in air by combining ultrasonic attenuation with optical refractive index and medium

Abstract

The invention discloses a method, a system, equipment and a medium for detecting the concentration of oil mist in air by combining ultrasonic attenuation with optical refractive index, and relates to the field of detection of the concentration of oil mist in air, comprising the following steps: synchronously exciting an acoustic signal and an optical signal in a detection cavity; the method comprises the following steps: acquiring a sound wave signal to obtain corresponding sound wave attenuation information, and detecting an interference light path to obtain corresponding interference fringe offset information; respectively calculating corresponding acoustic characteristic parameters and optical characteristic parameters based on the sound wave attenuation information and the interference fringe offset information; obtaining a concentration parameter based on the acoustic characteristic parameter, the optical characteristic parameter, the environment temperature parameter and the environment humidity parameter; a residual error is calculated and judged, and when the residual error exceeds a preset threshold value, the intelligent compensation model is called to correct the concentration parameter. According to the invention, when the residual error is abnormal, the intelligent compensation model is introduced to correct the result, and the problem that a single sensing mode is easily influenced by particle size distribution and environmental change is avoided.

Description

Technical Field

[0001] This invention relates to the field of air oil mist concentration detection, and in particular to a method, system, device and medium for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index. Background Technology

[0002] In the daily operation of large hydropower stations, due to the large number of auxiliary equipment and frequent start-up and shutdown operations, the frequent start-up and shutdown of equipment such as push bearings, water guide bearings, and air coolers may lead to equipment aging, valve shut-off, and pipeline loosening, which in turn can cause oil leakage.

[0003] Existing oil mist detection methods, such as the single acoustic attenuation method, achieve non-contact, real-time, and harsh-environment-resistant oil mist concentration detection by analyzing the scattering and absorption effects of sound wave energy by oil mist particles. However, their accuracy is greatly affected by particle size distribution and environmental parameters. At high concentrations, acoustic attenuation saturates, the linear range is limited, and it is sensitive to background noise, which severely restricts its application in hydropower station scenarios. Introducing optical refractive index to quantify the refractive index change in the oil mist-air mixture is an indirect measurement method based on the difference in refractive index between oil mist particles and the surrounding gas. By analyzing changes in the propagation characteristics of light in the oil mist-containing medium (such as phase delay, change in light speed, or shift in interference fringes), the oil mist concentration can be deduced. This method has the advantages of being miniaturized and having a fast response speed, which can compensate for the accuracy limitations of the acoustic attenuation method at high concentrations.

[0004] This patent integrates acoustic attenuation detection and optical refractive index detection technologies to achieve simultaneous analysis of oil mist concentration, particle size, and composition. Summary of the Invention

[0005] In view of the above-mentioned problems, the present invention provides a method, system, device and medium for detecting air oil mist concentration by combining ultrasonic attenuation with optical refractive index.

[0006] Therefore, the problem that this invention aims to solve is that the accuracy of existing methods is greatly affected by particle size distribution and environmental parameters, the sound wave attenuation saturates at high concentrations, the linear range is limited, and they are sensitive to background noise, which greatly restricts their application in actual hydropower station scenarios.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index, comprising: simultaneously exciting acoustic signals and optical signals in a detection cavity, causing the acoustic signals to pass through the oil mist-containing gas to be tested along a set path, while simultaneously causing the optical signals to pass through the detection cavity and form an interference optical path; receiving and acquiring the acoustic signals after propagation through the oil mist-containing gas to be tested, obtaining corresponding acoustic attenuation information, and simultaneously detecting the interference optical path to obtain corresponding interference fringe shift information; calculating corresponding acoustic characteristic parameters and optical characteristic parameters based on the acoustic attenuation information and the interference fringe shift information, and simultaneously acquiring ambient temperature parameters and ambient humidity parameters; inputting the acoustic characteristic parameters, optical characteristic parameters, and ambient temperature and humidity parameters into a fusion calculation model for joint inversion processing to obtain concentration parameters; calculating and judging the residual of the fusion calculation model, and when the residual exceeds a preset threshold, calling an intelligent compensation model to correct the concentration parameters to obtain corrected concentration parameters; uploading the corrected concentration parameters and processing them as an alarm basis.

[0008] As a preferred embodiment of the method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index as described in this invention, the step of synchronously exciting acoustic and optical signals within the detection cavity includes: using a unified clock source to synchronously trigger and control the acoustic excitation module and the optical excitation module to initiate the transmission of acoustic and optical signals under the same time reference; and timestamping the synchronous start time of the acoustic and optical signals to align the acoustic and optical data during the acquisition process according to corresponding time segments.

[0009] As a preferred embodiment of the method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index as described in this invention, the following steps are included: obtaining the corresponding acoustic attenuation information includes comparing the acquired acoustic signal with a preset reference acoustic signal to calculate acoustic attenuation information characterizing the degree of influence of oil mist on acoustic propagation; obtaining the corresponding interference fringe shift information includes dividing the incident beam into a reference optical path and a sample optical path, allowing the sample optical path to pass through the detection cavity and re-merge with the reference optical path to form interference fringes, performing image acquisition and analysis processing on the interference fringes, and calculating the corresponding interference fringe shift information based on the change in the displacement of the interference fringes.

[0010] As a preferred embodiment of the method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index as described in this invention, the calculation of the corresponding acoustic and optical characteristic parameters includes: calculating the acoustic attenuation coefficient based on the obtained acoustic attenuation information to form acoustic characteristic parameters; extracting the displacement change characteristics of the interference fringes based on the obtained interference fringe offset information, and calculating the optical path change corresponding to the sample optical path based on the displacement change, thereby calculating the effective refractive index of the mixed medium to form optical characteristic parameters.

[0011] As a preferred embodiment of the method for detecting airborne oil mist concentration using ultrasonic attenuation combined with optical refractive index as described in this invention, the method for obtaining concentration parameters includes: performing unified time alignment and scale normalization on the acoustic characteristic parameters, optical characteristic parameters, ambient temperature parameters, and ambient humidity parameters, and constructing a multi-dimensional input parameter set based on the parameter source category; calling a preset physical inversion model, and simultaneously solving the multi-dimensional input parameter set into the acoustic attenuation mechanism model and the refractive index hybrid model to obtain the concentration parameters calculated based on physical constraints.

[0012] The beneficial effects of this preferred technical solution are as follows: by inputting acoustic characteristic parameters, optical characteristic parameters and environmental parameters into the fusion calculation model, and solving the problem by combining the acoustic wave attenuation model and the refractive index hybrid model, the multi-physical quantity constraint inversion of the oil mist concentration parameter is realized, so that the result no longer depends on a single sensing channel, and the stability of the result under complex working conditions is improved.

[0013] As a preferred embodiment of the method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index as described in this invention, the method for correcting the concentration parameters includes: constructing a residual relationship between the fused output value and each input feature parameter based on the obtained concentration parameters, and calculating the corresponding residual value; comparing the residual value with a preset threshold to determine whether the output result of the current fusion calculation model is within a valid confidence interval; when the determination result indicates that the residual value exceeds the preset threshold, triggering an intelligent compensation model to correct the concentration parameters.

[0014] As a preferred embodiment of the method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index as described in this invention, the step of processing as an alarm basis includes: comparing the corrected concentration parameter with a preset safety threshold to determine whether the current air environment is in a safe state, a warning state, or an alarm state; when it is determined to be a warning state or an alarm state, outputting the corresponding alarm information to the alarm execution unit to trigger the corresponding alarm processing action; recording and archiving the alarm information, and updating the status during continuous monitoring.

[0015] The beneficial effects of this preferred technical solution are: it transforms oil mist detection results into actionable safety management actions, avoiding the problem of providing only numerical values ​​without providing decision support, thereby improving the operational safety level of the hydropower plant and realizing the upgrade from "monitoring equipment" to "safety management system".

[0016] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a system for detecting the concentration of oil mist in air using ultrasonic attenuation combined with optical refractive index, comprising: an excitation module, an acoustic excitation module, an optical excitation module, an acquisition module, a feature extraction module, a concentration parameter calculation module, a correction module, and an alarm module; the excitation module simultaneously excites the acoustic excitation module and the optical excitation module within the detection cavity to form acoustic and optical signals, causing the acoustic signal to pass through the oil mist-containing gas to be tested along a set path, while simultaneously causing the optical signal to pass through the detection cavity and form an interference optical path; the acquisition module receives and acquires the acoustic signal after it has propagated through the oil mist-containing gas to be tested, obtains the corresponding acoustic attenuation information, and simultaneously detects the interference optical path to obtain... The corresponding interference fringe offset information is obtained; the feature extraction module calculates the corresponding acoustic feature parameters and optical feature parameters based on the acoustic attenuation information and the interference fringe offset information, and simultaneously collects the ambient temperature parameters and ambient humidity parameters; the concentration parameter calculation module inputs the acoustic feature parameters, optical feature parameters, ambient temperature parameters, and ambient humidity parameters into the fusion calculation model for joint inversion processing to obtain the concentration parameters; the correction module calculates the residual of the fusion calculation model and makes a judgment. When the residual exceeds a preset threshold, the intelligent compensation model is called to correct the concentration parameters to obtain the corrected concentration parameters; the alarm module uploads the corrected concentration parameters and uses them as the basis for alarm processing.

[0017] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of a method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index as described above.

[0018] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for detecting airborne oil mist concentration using ultrasonic attenuation combined with optical refractive index as described above.

[0019] The beneficial effects of this invention are as follows: By integrating ultrasonic attenuation detection with optical refractive index detection, this invention incorporates acoustic characteristic parameters, optical characteristic parameters, and environmental parameters into a unified fusion calculation model for joint inversion. Furthermore, when residual anomalies occur, an intelligent compensation model is introduced to correct the results. This effectively avoids the problem of single-sensor methods being susceptible to the influence of particle size distribution and environmental changes, improving the accuracy and stability of oil mist concentration measurement. Simultaneously, through a graded alarm and linkage processing mechanism, the detection results are directly converted into executable safety management commands. This enables the system to achieve real-time, continuous, and high-precision monitoring of oil mist concentration, and to promptly issue warnings or alarms and trigger response measures when anomalies occur, fundamentally improving the safety assurance capabilities and intelligent management level of hydropower plant operation. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of a method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index in Example 1. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for detecting airborne oil mist concentration using ultrasonic attenuation combined with optical refractive index, comprising: S1: Simultaneously excite acoustic and optical signals within the detection cavity, causing the acoustic signal to pass through the oil mist-containing gas to be tested along a set path, while simultaneously causing the optical signal to pass through the detection cavity and form an interference optical path.

[0025] S2: Receive and collect the acoustic signal after it has propagated through the oil mist gas to be tested, obtain the corresponding acoustic attenuation information, and at the same time detect the interference optical path to obtain the corresponding interference fringe shift information.

[0026] S3: Based on the acoustic attenuation information and interference fringe shift information, calculate the corresponding acoustic characteristic parameters and optical characteristic parameters respectively, and simultaneously collect the ambient temperature parameters and ambient humidity parameters.

[0027] S4: Input the acoustic characteristic parameters, optical characteristic parameters, ambient temperature parameters, and ambient humidity parameters into the fusion calculation model, perform joint inversion processing, and obtain the concentration parameters.

[0028] S5: Calculate and judge the residual of the fusion calculation model. When the residual exceeds the preset threshold, call the intelligent compensation model to correct the concentration parameters and obtain the corrected concentration parameters.

[0029] S6: Upload the corrected concentration parameters and use them as the basis for alarm processing.

[0030] It should be noted that existing technologies in the field of air oil mist detection generally suffer from problems such as limited detection methods, poor adaptability, and restricted engineering applications. On the one hand, although laboratory analysis methods have high accuracy, they require sampling, transfer, and offline analysis, making the detection process complex and time-consuming, which is difficult to meet the requirements of real-time and continuous monitoring in scenarios such as hydropower plant buildings. On the other hand, field detection methods based on humidity, charge, or light scattering are easily affected by changes in ambient temperature, humidity, airflow disturbance, and particle size distribution, and are prone to accumulation of measurement errors or even inaccuracy under high concentrations or complex operating conditions.

[0031] Therefore, in response to the above problems, such as Figure 1 As shown, through steps S1-S6, the acoustic and optical channels in the detection environment are first synchronously initialized, so that ultrasound and laser act on the oil mist gas under the same time reference. Then, the energy change information generated during the propagation of the sound wave and the interference change characteristics of the light in the mixed medium are collected respectively. The above multi-source raw signals are characterized to form acoustic parameters and optical parameters. At the same time, ambient temperature and humidity data are introduced for collaborative correction. On this basis, various parameters are sent to the fusion calculation module for joint inversion to obtain the preliminary oil mist concentration result. When it is determined that there is a large deviation between the result and the model expectation, it is further corrected by the intelligent compensation model. Finally, reliable oil mist concentration information is output, and the status judgment and alarm processing are completed accordingly, realizing a complete closed-loop process from multi-source perception, fusion calculation to safety response.

[0032] Example 2, a second embodiment of the present invention, differs from the first embodiment in that: a method for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index further includes, in step S1, simultaneously exciting acoustic and optical signals within the detection cavity, comprising the following steps A1-A2: A1: A unified clock source is used to synchronously trigger and control the acoustic excitation module and the optical excitation module, so as to start the transmission of acoustic signals and optical signals under the same time base.

[0033] A2: Timestamp the synchronous start time of the acoustic signal and the optical signal to align the acoustic data and optical data during the acquisition process according to the corresponding time segments.

[0034] In this embodiment of the application, in step A1, the synchronization trigger control adopts a master-slave trigger synchronization mechanism, including the following steps A111-A113: A111: Set the acoustic excitation module as the master node and the optical excitation module as the slave node, with the clock module inside the master node outputting the reference trigger signal.

[0035] A112: At the same time as the master node sends a trigger signal, a synchronous start signal is sent to the slave node through the hardware trigger interface, so that the acoustic excitation module and the optical excitation module start working at the same time.

[0036] A113: At the same time as the acoustic module completes the transmission start-up, the optical module is forcibly triggered to enter the light-emitting state, and the corresponding start-up timestamp is recorded for subsequent data synchronization and alignment processing.

[0037] In an optional implementation, the synchronization trigger control may also employ a broadcast synchronization mechanism, including the following steps A121-A123: A121: Set up a unified synchronization control unit as a broadcast node to send start control signals to both the acoustic excitation module and the optical excitation module simultaneously.

[0038] A122: The synchronous control unit broadcasts the start command to each controlled module simultaneously, so that the acoustic excitation module and the optical excitation module enter a synchronous working state after receiving the start command.

[0039] A123: The time of sending the start command is recorded uniformly, and the corresponding unified time base is marked in each module to ensure the consistency of subsequent data collection.

[0040] In another alternative implementation, the synchronous trigger control may also employ a coordinator synchronization mechanism, including the following steps A131-A133: A131: Set up a synchronization and coordination module to send preparation commands to the acoustic excitation module and the optical excitation module respectively before starting the detection.

[0041] A132: When the synchronization coordination module receives the readiness feedback from the acoustic excitation module and the optical excitation module, it sends a unified execution command to each module to trigger the two modules to enter the working state simultaneously.

[0042] A133: When a unified execution command is issued, the current system time is marked and distributed to each module to establish a unified time base.

[0043] It should be noted that this step synchronizes the acoustic excitation module and the optical excitation module by setting a unified clock source and aligning the timestamps of the acoustic and optical signals. This ensures that acoustic and optical data are acquired and fused under the same time reference, avoiding time drift errors caused by asynchronous startup between different sampling channels. This improves the accuracy of multi-source data fusion from the source, establishes an effective synergy between acoustic and optical detection, and provides a unified and reliable data foundation for subsequent joint inversion and intelligent compensation, thereby ensuring the timeliness and stability of the overall detection results.

[0044] Furthermore, in step S2, obtaining the corresponding acoustic attenuation information includes comparing the collected acoustic signal with a preset reference acoustic signal and calculating acoustic attenuation information that characterizes the degree of influence of oil mist on acoustic propagation.

[0045] Obtaining the corresponding interference fringe offset information includes splitting the incident beam into a reference optical path and a sample optical path, allowing the sample optical path to pass through the detection cavity and re-merge with the reference optical path to form interference fringes, performing image acquisition and analysis processing on the interference fringes, and calculating the corresponding interference fringe offset information based on the change in the displacement of the interference fringes.

[0046] In this application embodiment, the comparison processing for obtaining the corresponding acoustic attenuation information employs a template matching comparison mechanism, including the following steps B111-B113: B111: Select the acoustic signal collected under oil-free or standard operating conditions as the reference template, and input the acoustic signal collected this time into the comparison module for alignment processing.

[0047] B112: By matching the correspondence between the acquired acoustic signal and the reference template in the time axis and frequency domain, the differences between the two in amplitude distribution and phase change are obtained.

[0048] B113: Based on the difference relationship, the attenuation characteristic quantity reflecting the degree of sound wave energy loss is calculated, and the attenuation characteristic quantity is output as sound wave attenuation information.

[0049] In an optional implementation, the comparison process may also employ a statistical baseline comparison method, including the following steps B121-B123: B121: Construct a statistical benchmark model of the acoustic signal based on multiple historical detection results, and use this statistical benchmark model as a reference acoustic wave.

[0050] B122: Perform deviation analysis between the currently acquired acoustic signal and the statistical baseline model to obtain the degree of change of the acoustic signal in terms of amplitude, phase and spectral distribution relative to the statistical baseline.

[0051] B123: Calculate the corresponding sound wave attenuation index based on the degree of change, and output it as sound wave attenuation information.

[0052] In another alternative implementation, the comparison process can also employ a sliding window dynamic comparison method, including the following steps B131-B133: B131: During the continuous acquisition of acoustic wave signals passing through the detection cavity, the acquired acoustic wave signals are divided into multiple continuous time windows according to a preset time length, and the acoustic wave signals in the previous time window are selected as the dynamic reference acoustic waves corresponding to the current time window.

[0053] B132: Align the acoustic wave signal acquired within the current time window with the dynamic reference acoustic wave point by point, and perform difference analysis on the aligned acoustic wave data in terms of amplitude variation trend, phase drift characteristics and spectral distribution changes.

[0054] B133: Based on the difference analysis results, calculate the degree of change in sound wave energy between adjacent time windows, and output the degree of change in sound wave energy as the sound wave attenuation information corresponding to the current window.

[0055] It should be noted that this step compares the acquired acoustic signal with a preset reference acoustic signal and integrates multi-dimensional features such as amplitude attenuation, phase change, and spectral change to obtain acoustic attenuation information, making the characterization of the influence of oil mist on acoustic propagation more comprehensive and accurate. At the same time, by forming an interference optical path through beam splitting and performing image acquisition and displacement calculation on the interference fringes, a highly sensitive measurement of the influence of oil mist on light propagation characteristics is achieved. This allows the system to acquire independent information sources in two different physical dimensions, acoustic and optical, simultaneously, thus avoiding the problem of inaccuracy of a single detection method in complex environments and significantly improving the reliability of the detection results.

[0056] Furthermore, in step S3, calculating the corresponding acoustic and optical characteristic parameters includes the following steps C1-C2: C1: Based on the obtained sound wave attenuation information, calculate the sound wave attenuation coefficient to form acoustic characteristic parameters.

[0057] C2: Based on the obtained interference fringe offset information, the displacement change characteristics of the interference fringes are extracted, and the optical path change corresponding to the sample optical path is calculated according to the displacement change. Then, the effective refractive index of the mixed medium is calculated to form optical characteristic parameters.

[0058] Specifically, acoustic attenuation detection of oil mist is an indirect measurement method based on the energy loss of sound waves propagating in oil mist-containing gas. Its core principle is the scattering and absorption of sound waves by oil mist particles, causing the sound signal intensity to decrease with increasing oil mist concentration. When sound waves propagate in a gas, encountering suspended particles (such as oil mist droplets) will result in the following effects: Scattering: Particles cause sound waves to deviate from their original propagation direction, resulting in a reduction of energy at the receiving end.

[0059] Absorption: The viscous effect between particles and sound waves, as well as heat conduction, cause sound energy to be converted into heat energy.

[0060] Dispersion: Differences in the propagation speed of sound waves at different frequencies may introduce phase changes.

[0061] Sound wave attenuation coefficient The relationship with oil mist concentration can be approximated by a classical theoretical model: in, This refers to the oil mist volume concentration. The average radius of the oil mist particles is... is the extinction coefficient.

[0062] Optical refractive index measurement of oil mist concentration is an indirect method based on the difference in refractive index between oil mist particles and the surrounding gas. By analyzing changes in the propagation characteristics of light in an oil mist-containing medium (such as phase delay, changes in light speed, or shifts in interference fringes), the oil mist concentration can be deduced. The refractive index of the oil mist particles (…) ) is usually related to the refractive index of air ( Significant differences exist (e.g., the refractive index of mineral oil). When light passes through an oil mist-containing gas, its propagation characteristics change due to the influence of particles as follows: Phase delay: Light travels at a slower speed in particles than in air, resulting in optical path difference and phase shift.

[0063] Effective refractive index: The overall refractive index of the oil mist-air mixture. It increases with increasing oil mist concentration.

[0064] Interference effect: By using an interferometer to detect the phase difference between two beams of light (reference light and sample light), the change in refractive index can be quantified.

[0065] Using a Michelson interferometer, the laser beam is split into a reference beam and a sample beam. The sample beam passes through an oil mist-containing gas chamber. After the two beams re-merge, they form interference fringes. Changes in the oil mist concentration cause a phase delay in the sample beam, resulting in a shift in the interference fringes. The amount of fringe shift ( ) Calculate the phase difference ( ), and then infer ( ) and concentration ( ).

[0066] in, Let L be the air refractive index and L be the optical path length.

[0067] Relationship between concentration and refractive index: Assuming the oil mist particles are a homogeneous monodisperse system, the effective refractive index of the mixed medium can be approximately expressed by the Lorentz-Lorentz formula: Furthermore, in step S4, obtaining the concentration parameter includes the following steps D1-D2: D1: Perform unified time alignment and scale normalization on acoustic feature parameters, optical feature parameters, ambient temperature parameters, and ambient humidity parameters, and construct a multi-dimensional set of input parameters based on the parameter source category.

[0068] D2: Call the preset physical inversion model, substitute the multi-dimensional input parameter set into the acoustic attenuation mechanism model and the refractive index hybrid model for simultaneous solution, and obtain the concentration parameters calculated based on physical constraints.

[0069] Specifically, multimodal data fusion, based on simultaneous equations of a physical model, eliminates environmental interference: in, For temperature, For humidity, For the sound wave attenuation mechanism model, This is a refractive index mixing model.

[0070] By measurement Combined with known Solvable , These are collectively referred to as concentration parameters.

[0071] Furthermore, in step S5, the correction process for the concentration parameter includes the following steps E1-E3: E1: Based on the obtained concentration parameters, construct the residual relationship between the fusion output value and each input feature parameter, and calculate the corresponding residual value.

[0072] E2: Compare the residual value with a preset threshold to determine whether the output of the current fusion computing model is within a valid confidence interval.

[0073] E3: When the judgment result shows that the residual value exceeds the preset threshold, the intelligent compensation model is triggered to correct the concentration parameter.

[0074] In this embodiment of the application, in step E3, the intelligent compensation model adopts an intelligent compensation model based on a convolutional neural network (CNN), including the following steps E311-E313: E311: The output concentration parameter is combined with the acoustic feature parameters, optical feature parameters, ambient temperature parameters, and ambient humidity parameters at the same time to form a standardized input vector, which is then fed into the input layer of the convolutional neural network model according to the preset feature arrangement rules.

[0075] E312: The convolutional neural network model performs convolutional feature extraction and nonlinear mapping operations to obtain the error prediction (compensation) for the current working condition. This compensation represents the systematic deviation of the physical model under this working condition.

[0076] E313: Based on the compensation amount, the concentration parameters are superimposed or weighted and corrected, the corrected oil mist concentration parameters are output, and the original value, compensation amount and correction value are recorded for model self-learning and updating.

[0077] Specifically, the CNN model training process includes: Step 1: Collect historical running data and form a training sample set. Each sample contains: Input features: acoustic feature parameters, optical feature parameters, ambient temperature parameters, ambient humidity parameters.

[0078] Supervision label: "calibrated oil mist concentration" or "high confidence reference concentration" corresponding to the same time point. Missing values ​​were handled, outliers were removed, and dimensions were standardized and normalized for all samples.

[0079] Step 2: Establish a network structure consisting of multiple convolutional layers and nonlinear mapping layers, and perform feature learning on the input samples; use the "deviation between predicted concentration and calibrated concentration" as the loss function, and iteratively optimize the network parameters through backpropagation and gradient update methods until the prediction error on the validation set converges to within the target threshold.

[0080] Step 3: Solidify the trained model parameters and download them to the field system; when the cumulative number of new samples reaches the update condition or when a drift in the working condition distribution is detected, trigger incremental training or full retraining to maintain the compensation accuracy of the model under the new working conditions.

[0081] In an optional implementation, the intelligent compensation model may also employ an intelligent correction mechanism based on a regression model, including the following steps E321-E323: E321: Establish a multivariate regression model to fit the mapping relationship between "acoustic characteristic parameters, optical characteristic parameters, environmental parameters" and "concentration error" using historical data, and perform initial calibration of the model parameters.

[0082] E322: Input various characteristic parameters under the current operating conditions into the regression model to obtain the corresponding error prediction.

[0083] E323: Based on the error prediction, the initial oil mist concentration result is linearly or nonlinearly weighted to generate the corrected oil mist concentration parameter, and the regression model parameter is updated on a rolling basis to adapt to long-term drift.

[0084] In another alternative implementation, the intelligent compensation model may also employ a rule-engine-based compensation inference mechanism, including the following steps E331-E333: E331: Construct a rule base. Rule items are used to characterize the correspondence between "temperature and humidity change range, acoustic attenuation range, optical offset range" and "known deviation mode", and a correction weight is configured for each rule.

[0085] E332: Matches the current input features with the rule base and outputs a set of rules that meet the conditions and their corresponding corrected weights.

[0086] E333: The correction weights of the hit rules are weighted and fused to form compensation parameters, and the initial oil mist concentration results are corrected, outputting the corrected oil mist concentration parameters.

[0087] It should be noted that this step involves residual judgment on the output results of the fusion model, and triggering the intelligent compensation model to correct it when the residual exceeds the threshold. This enables the system to automatically identify the failure risk of the physical model under abnormal working conditions, and correct the error through data-driven model, thereby avoiding distortion of the detection results under extreme or abnormal conditions. This constructs a dual-layer protection mechanism of "physical model + intelligent model", improves the overall system's adaptability, and enables the detection system to have long-term stability and reliability.

[0088] Furthermore, in step S6, the processing based on the alarm includes the following steps G1-G3: G1: Compare the corrected concentration parameter with the preset safety threshold to determine whether the current air environment is in a safe state, a warning state, or an alarm state.

[0089] G2: When the alarm or warning state is determined, the corresponding alarm information is output to the alarm execution unit to trigger the corresponding alarm processing action.

[0090] G3: Records and archives alarm information and updates the status during continuous monitoring.

[0091] Specifically, the determination of oil mist status is based on a graded process using the corrected concentration parameter. When the oil mist concentration parameter is less than or equal to the safety threshold, it is determined to be in a safe state. When the oil mist concentration parameter is greater than the safety threshold but less than the alarm threshold, or shows an abnormal growth trend within a unit of time, it is determined to be in a warning state. When the oil mist concentration parameter reaches or exceeds the alarm threshold, or shows a sudden increase exceeding the preset change threshold within a short period of time, it is determined to be in an alarm state, thus providing a clear basis for subsequent alarm processing.

[0092] In this embodiment of the application, in step G2, the alarm processing action adopts the local audible and visual alarm execution mode, including the following steps G211-G213: G211: The generated alarm information is sent to the alarm execution unit set up at the testing site, and the alarm execution unit formats and encapsulates the alarm information, including the alarm level, trigger time, current location identifier and corresponding oil mist concentration parameters.

[0093] G212: The alarm execution unit drives the audible and visual alarm device to perform graded responses according to the alarm level. When it is in the warning state, it outputs an intermittent audible and visual prompt signal, and when it is in the alarm state, it outputs a continuous audible and visual alarm signal.

[0094] G213: During the alarm process, the on-site sensor data is continuously collected. When the oil mist concentration parameter falls back to the safe threshold range or a manual confirmation signal is received, the audible and visual alarm is automatically terminated, and a record of the alarm handling is generated.

[0095] In an optional implementation, the alarm handling action can also adopt a remote system push alarm mode, including the following steps G221-G223: G221: The generated alarm information is encapsulated into a communication data packet according to a preset communication protocol and sent to the remote monitoring system via a wired or wireless network.

[0096] G222: The remote monitoring system highlights the corresponding alarm points on the monitoring interface and simultaneously pushes alarm notifications to the operation and maintenance terminal to prompt relevant personnel to take timely action.

[0097] G223: During the alarm period, the remote monitoring system periodically receives resumed data and dynamically refreshes the alarm status. When the monitoring data recovers to a safe range, it sends an alarm cancellation notification to the operation and maintenance terminal and archives the alarm process data.

[0098] In another optional implementation, the alarm handling action can also adopt a control-linked alarm execution mode, including the following steps G231-G233: G231: The generated alarm information is sent to the linkage control unit, which then performs priority parsing on the alarm information to match the corresponding linkage strategy set.

[0099] G232: The linkage control unit triggers the corresponding control equipment to perform linkage actions according to the linkage strategy in order to adjust the plant operating environment.

[0100] G233: Collects and provides feedback on the execution status of the linkage action. When the execution fails or the expected adjustment effect is not achieved, an abnormal linkage alarm is generated and reported to the monitoring system.

[0101] It should be noted that this step uses oil mist concentration as the basis for alarm classification and judgment, and triggers the alarm execution unit to respond and process in the early warning or alarm state, so that the system not only has detection function, but also risk warning and handling capabilities.

[0102] Example 3, the third embodiment of the present invention, differs from the previous two embodiments in that it is a system for detecting air oil mist concentration using ultrasonic attenuation combined with optical refractive index. The system includes an excitation module, an acoustic excitation module, an optical excitation module, a data acquisition module, a feature extraction module, a concentration parameter calculation module, a correction module, and an alarm module. The excitation module simultaneously excites the acoustic and optical excitation modules within the detection cavity, generating acoustic and optical signals. The acoustic signal passes through the oil mist-containing gas to be tested along a predetermined path, while the optical signal passes through the detection cavity and forms an interference optical path. The data acquisition module receives and acquires the acoustic signal after propagation through the oil mist-containing gas to be tested, obtaining corresponding acoustic attenuation information, and simultaneously calculates the concentration parameter calculation, correction, and alarm modules for the interference optical path. The system performs detection to obtain corresponding interference fringe shift information. The feature extraction module calculates the corresponding acoustic and optical feature parameters based on the acoustic attenuation information and the interference fringe shift information, while simultaneously collecting ambient temperature and humidity parameters. The concentration parameter calculation module inputs the acoustic, optical, ambient temperature, and ambient humidity parameters into the fusion calculation model for joint inversion processing to obtain the concentration parameters. The correction module calculates the residual of the fusion calculation model and makes a judgment. When the residual exceeds a preset threshold, the intelligent compensation model is invoked to correct the concentration parameters, obtaining the corrected concentration parameters. The alarm module uploads the corrected concentration parameters and uses them as the basis for alarm processing.

[0103] Specifically, the acoustic excitation module integrates a bidirectional piezoelectric ceramic transducer and a temperature and humidity sensor, extending the acoustic path and enhancing detection sensitivity through a Z-shaped reflection channel. The optical excitation module is based on a Michelson interferometer design, using a 650nm vertical-cavity surface-emitting laser as the light source. The sample optical path penetrates the detection cavity, and interference fringe shifts are captured by a high-sensitivity photodiode. The detection cavity adopts an orthogonal dual-channel structure for acoustic and optical paths to avoid signal interference, and an anti-turbulence honeycomb grid is installed at the inlet to stabilize airflow. A multi-channel data acquisition card simultaneously captures acoustic attenuation, phase, and optical interference signals, as well as temperature and humidity. An embedded AI chip runs a fusion algorithm, combining a physical model and a convolutional neural network (CNN) to achieve concentration inversion and interference compensation. This architecture, through the synergy of acoustic dynamic response and optical refractive index calibration, balances high accuracy (±4%) with adaptability to all operating conditions, supporting real-time monitoring of oil mist concentration in the air under complex environments.

[0104] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0105] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0106] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0107] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented in combination with any of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0108] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for detecting air-borne oil mist concentration using ultrasonic attenuation in combination with optical refractive index, characterized in that: The method comprises the following steps: synchronously exciting an acoustic signal and an optical signal in a detection cavity, making the acoustic signal pass through the oil mist gas to be detected along a set path, and making the optical signal pass through the detection cavity and form an interference light path; receiving and collecting the acoustic signal after propagating through the oil mist gas to be detected, obtaining corresponding acoustic attenuation information, and detecting the interference light path to obtain corresponding interference fringe shift information; based on the acoustic attenuation information and the interference fringe shift information, calculating corresponding acoustic characteristic parameters and optical characteristic parameters, and synchronously collecting environmental temperature parameters and environmental humidity parameters; inputting the acoustic characteristic parameters, optical characteristic parameters, environmental temperature parameters and environmental humidity parameters into a fusion calculation model for joint inversion processing to obtain concentration parameters; calculating the residual error of the fusion calculation model and judging whether the residual error exceeds a preset threshold value; if yes, calling an intelligent compensation model to correct the concentration parameters to obtain corrected concentration parameters; uploading the corrected concentration parameters and processing them as alarm basis.

2. The method for detecting the concentration of air-borne oil mist using ultrasonic attenuation combined with optical refractive index as claimed in claim 1, wherein: The method comprises the following steps: synchronously exciting an acoustic signal and an optical signal in a detection cavity, making the acoustic signal pass through the oil mist gas to be detected along a set path, and making the optical signal pass through the detection cavity and form an interference light path; synchronously exciting an acoustic signal and an optical signal in a detection cavity, making the acoustic signal pass through the oil mist gas to be detected along a set path, and making the optical signal pass through the detection cavity and form an interference light path; 3. A method for detecting the concentration of air-borne oil mist using ultrasonic attenuation in combination with optical refractive index as claimed in claim 2, wherein: The method comprises the following steps: comparing the collected acoustic signal with a preset reference acoustic signal, calculating acoustic attenuation information representing the influence of oil mist on acoustic propagation, and obtaining corresponding acoustic attenuation information; 4. The method for detecting the concentration of air-borne oil mist using ultrasonic attenuation combined with optical refractive index as claimed in claim 3, wherein: The method comprises the following steps: comparing the collected acoustic signal with a preset reference acoustic signal, calculating acoustic attenuation information representing the influence of oil mist on acoustic propagation, and obtaining corresponding acoustic attenuation information; The method comprises the following steps:

5. A method for detecting the concentration of air-borne oil mist using ultrasonic attenuation in combination with optical refractive index as claimed in claim 4, wherein: based on the obtained acoustic attenuation information, calculating the acoustic attenuation coefficient to form the acoustic characteristic parameters; based on the obtained interference fringe shift information, extracting the displacement change characteristics of the interference fringes, calculating the optical path change of the sample light path according to the displacement change, and then calculating the effective refractive index of the mixed medium to form the optical characteristic parameters. The method comprises the following steps:

6. A method for detecting the concentration of air-borne oil mist using ultrasonic attenuation in combination with optical refractive index as claimed in claim 5, wherein: performing unified time alignment and scale normalization on the acoustic characteristic parameters, optical characteristic parameters, environmental temperature parameters and environmental humidity parameters, and constructing a multi-dimensional input parameter set according to the parameter source categories; calling a preset physical inversion model, substituting the multi-dimensional input parameter set into the acoustic attenuation mechanism model and the refractive index mixing model for simultaneous solution to obtain the concentration parameters calculated based on the physical constraint conditions. The method comprises the following steps: based on the obtained concentration parameters, constructing a residual error relationship between the fusion output value and each input characteristic parameter, and calculating the corresponding residual error value; The residual value is compared with a preset threshold to determine whether the current fusion calculation model output result is within an effective confidence interval; When the judgment result shows that the residual value exceeds the preset threshold, the intelligent compensation model is triggered to correct the concentration parameter.

7. A method for detecting the concentration of air-borne oil mist using ultrasonic attenuation in combination with optical refractive index as claimed in claim 6, wherein: The processing as an alarm basis includes, The corrected concentration parameter is compared with a preset safety threshold to determine which of the safety state, the early warning state and the alarm state the current air environment is in; When the early warning state or the alarm state is determined, the corresponding alarm information is output to an alarm execution unit to trigger corresponding alarm processing actions; The alarm information is recorded and archived, and the state is updated in the continuous monitoring process.

8. A system for detecting the concentration of oil mist in air using ultrasonic attenuation combined with optical refractive index, applying a method for detecting the concentration of oil mist in air using ultrasonic attenuation combined with optical refractive index according to any one of claims 1 to 7, characterized in that: The system comprises an excitation module, a sound wave excitation module, an optical excitation module, a collection module, a feature extraction module, a concentration parameter calculation module, a correction module and an alarm module. The excitation module synchronously excites the sound wave excitation module and the optical excitation module in the detection cavity to form sound wave signals and optical signals, so that the sound wave signals pass through the oil mist gas to be measured along a set path, and the optical signals pass through the detection cavity and form an interference light path; The collection module receives and collects the sound wave signals after the sound wave signals pass through the oil mist gas to be measured, obtains corresponding sound wave attenuation information, and detects the interference light path to obtain corresponding interference fringe shift information; The feature extraction module calculates corresponding acoustic feature parameters and optical feature parameters based on the sound wave attenuation information and the interference fringe shift information, and synchronously collects environmental temperature parameters and environmental humidity parameters; The concentration parameter calculation module inputs the acoustic feature parameters, the optical feature parameters, the environmental temperature parameters and the environmental humidity parameters into a fusion calculation model for joint inversion processing to obtain the concentration parameter; The correction module calculates the residual of the fusion calculation model and performs judgment, and when the residual exceeds a preset threshold, an intelligent compensation model is called to correct the concentration parameter to obtain a corrected concentration parameter; The alarm module uploads the corrected concentration parameter and processes it as an alarm basis. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to implement the steps of the method for detecting air oil mist concentration by combining ultrasonic wave attenuation with optical refractive index according to any one of claims 1 to 7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the method for detecting air oil mist concentration by combining ultrasonic wave attenuation with optical refractive index according to any one of claims 1 to 7.

Images