Acousto-optic-magnetic intelligent detection method based on ecological loss comprehensive index

Abstract

The invention discloses an acousto-optic magnetic intelligent detection method based on an ecological loss comprehensive index, and the method comprises the steps: carrying out the derivation and solving of an acousto-optic magnetic equation according to an acousto-optic magnetic basic theory; assembling an acousto-optic magnetic comprehensive acquisition instrument and installing the acousto-optic magnetic comprehensive acquisition instrument on the mobile carrying device; according to different characteristics of the acousto-optic and magnetic signals, acousto-optic and magnetic characteristics generated on the construction site are extracted; establishing physical models of a sound wave field, a light wave field and an electromagnetic field based on an acousto-optic magnetic theory and field test data; and proposing an ecological loss index of acousto-optic magnetism generated in the construction process, proposing an ecological loss comprehensive index in combination with weight coefficients of different physical fields, and evaluating the influence of acousto-optic magnetism generated in the construction process on the environment. According to the invention, the detection technologies of sound, light and magnetic signals are combined, through advanced technical means such as multi-signal fusion, intelligent data analysis and machine learning optimization, the detection precision and reliability are greatly improved, the adaptability is strong, the anti-interference capability is strong, and the real-time feedback capability is good.

Description

Technical Field

[0001] The present invention relates to the field of intelligent perception technology involving acoustic-optical-magnetic signal acquisition, intelligent algorithm analysis, and physical field model construction, specifically to the field of construction environment monitoring and intelligent analysis technology, and is widely used in construction environment detection and environmental impact assessment of construction. Background Art

[0002] The construction process consumes a lot of resources and has a certain degree of impact on the ecological environment. Therefore, the implementation of green construction in engineering construction is crucial for energy conservation, emission reduction and protection of the ecological environment. For bridges, their construction technology is relatively cumbersome, and the construction process will inevitably cause vegetation destruction, water pollution, noise pollution, light pollution, air pollution and other phenomena, which will have a certain impact on the surrounding flora and fauna. Studies have shown that organisms are more sensitive to the intensity and frequency of construction noise, and light wave radiation and electromagnetic radiation phenomena are also directly related to the death, migration, reproduction and other behaviors of organisms in the construction area. Therefore, in terms of the specifications for the limits of sound-light-magnetic physical quantities, my country has promulgated a number of relevant technical specifications. Among them, the "General Specification for Building Environment" GB55016-2021 sets sound pressure level limits for different time periods for the exterior and interior of buildings based on five different functions such as sleep and daily life. GB12523-2011, the "Environmental Noise Emission Standard for Construction Sites," specifies that sound pressure levels within construction areas must not exceed 70dB at night and 55dB during the day. Regarding light and electromagnetic radiation, GBZ 2.1-2019, the "Occupational Exposure Limits for Hazardous Factors in the Workplace," specifies limits for various parameters related to light and electromagnetic radiation within the workplace, including light intensity, exposure duration, electromagnetic intensity, exposure duration, and exposure level. Accurate measurement of acoustic, optical, and magnetic quantities within construction areas is crucial for the effective evaluation and technical optimization of green bridge construction processes.

[0003] Existing non-destructive testing technologies mainly rely on single physical field detection methods, such as ultrasonic testing, optical imaging or magnetic field detection. However, when faced with complex materials or structures, single physical field methods have problems such as insufficient sensitivity, limited resolution, and detection blind spots, making it difficult to meet the needs of high-precision and high-reliability detection. In addition, there is a lack of integrated and efficient measurement equipment. Compared with other measurement spaces, the bridge construction area is more intricate and complex, and it is necessary to develop accurate and effective measurement technologies for the construction area; in addition, existing research mainly focuses on acoustic-optical-magnetic monitoring in a single space or a relatively short period of time, and rarely considers their spatiotemporal evolution laws and construction effects. In recent years, with the development of sensing technology, artificial intelligence and data fusion technology, multi-physical field coupling detection methods have gradually attracted attention, but there is still a lack of unified and efficient detection methods. The present invention aims to solve the above problems and achieve higher precision and wider range of intelligent detection through the organic combination of acoustic, optical and magnetic multimodal detection technologies. Summary of the Invention

[0004] The present invention provides an intelligent acoustic, optical and magnetic detection method based on the comprehensive ecological loss index. Through multimodal data fusion and intelligent algorithm processing, it solves the problems of insufficient sensitivity, limited resolution and incomplete detection range of traditional single detection methods, and realizes accurate detection and analysis of target objects.

[0005] The above-mentioned purpose of the present invention is achieved through the following solutions:

[0006] An intelligent acoustic, optical and magnetic detection method based on a comprehensive ecological loss index comprises the following steps:

[0007] Step 1: Based on the basic theory of acousto-optic magnetism, derive and solve the acousto-optic magnetism equations;

[0008] Step 2: Assemble the acoustic, optical and magnetic integrated data acquisition instrument and install it on the mobile carrying device;

[0009] Step 3: Extract the characteristics of the acoustic, optical and magnetic signals generated at the construction site based on their different characteristics.

[0010] Step 4: Use IoT technology to transmit the extracted acousto-optical-magnetic characteristic values to a cloud database for secondary development and utilization. At the same time, develop a mobile phone app to enable real-time viewing of the acousto-optical-magnetic characteristic data.

[0011] Step 5: Arrange the acoustic, optical and magnetic perception test equipment at and around the construction site based on the location of the acoustic, optical and magnetic field.

[0012] Step 6: Based on the acousto-optic-magnetic theory and field test data, establish the physical models of the acoustic wave field, optical wave field and electromagnetic field;

[0013] Step 7: Propose the ecological loss index of sound, light and magnetism generated during construction: , and combined with the weight coefficients of different physical fields, a comprehensive ecological loss index is proposed to evaluate the environmental impact of the acoustic, optical and magnetic effects generated during the construction process.

[0014] Furthermore, the basic theories of acousto-optical-magnetic theory described in step 1 include structural vibration theory, basic acoustic theory, acoustic-vibration coupling theory, acoustic radiation theory and basic light wave theory, and the theory of sound wave generation and propagation is derived based on these theories; the light radiation equation generated during arc welding; and the Maxwell equations are used to describe the electromagnetic phenomena generated during the power-on working process of power equipment, and the Maxwell equations are solved by analytical methods.

[0015] Furthermore, the acoustic, optical and magnetic integrated data acquisition instrument described in step 2 includes an acoustic wave sensor, an ultraviolet radiometer and a three-axis magnetic field probe. The three are connected to a small data acquisition instrument through GeoCOM serial port technology and installed on a mobile carrying device to facilitate layout and data collection at the construction site.

[0016] Furthermore, step 3 includes extracting characteristic values of the collected sound based on the MFCC feature extraction method, and processing and analyzing them; extracting the intensity, reflectivity and polarization characteristics of the light wave field based on the spectrometer and polarization optical instrument, and processing and analyzing them; extracting the frequency band interval energy characteristics of the db4 wavelet packet decomposition based on the energy feature extraction method of wavelet packet decomposition, and processing and analyzing them.

[0017] Furthermore, the MFCC feature extraction method is used to extract the feature value of the collected sound, including:

[0018] 1. Preprocessing:

[0019] (1) Pre-emphasis: A first-order high-pass filter is used, and the sound signal of the construction area after pre-emphasis processing is set to y(n). Its mathematical expression is as follows:

[0020]

[0021] in: is the sound signal of the current frame, is the sound signal of the previous frame, is the weighting factor;

[0022] (2) Endpoint detection: Endpoint detection determines the starting point and ending point of the input sound signal, and uses short-time energy and short-time zero-crossing rate for detection. The short-time energy calculation formula is as follows:

[0023]

[0024] in: is a short-term sound signal with a length of N;

[0025] The calculation formula of short-time zero-crossing rate is as follows:

[0026]

[0027] in: is a symbolic function, namely:

[0028]

[0029] (3) Framing: Framing is to intercept a 10ms-30ms sound signal from a stable sound signal as a frame. To ensure the continuity and smoothness between two consecutive frames, each frame is shifted forward by 10ms to ensure the continuity and smoothness between frames.

[0030] (4) Windowing: There will be a truncation effect between the frames of the sound signal. In order to avoid causing a sharp change at the beginning and end of the frame signal, the windowing steps are introduced as follows:

[0031] Assume the sound frame signal is , the window function is , the sound signal after windowing is :

[0032]

[0033] The Hamming window is selected as the window function, and its formula is as follows:

[0034]

[0035] 2. Fast Fourier Transform

[0036] Perform fast Fourier transform on each frame of sound and calculate its spectral line energy:

[0037]

[0038]

[0039] 3. Filtering

[0040] Calculate the energy of the spectrum of each frame in the filter when it passes through the filter, and convert the spectrum line energy into The corresponding frequency domain of the filter Multiply and add, the specific calculation is as follows:

[0041]

[0042] 4. Discrete Cosine Transform

[0043] The spectral coefficients are converted to the time domain using discrete cosine transform, and finally the standard frequency-based cepstral coefficients are obtained;

[0044]

[0045] Where: m is the mth TS filter, and M represents the number of TS filters.

[0046] Furthermore, the extracting of the intensity, reflectivity and polarization characteristics of the light wave field includes:

[0047] Light wave field intensity feature extraction: Analyze the intensity distribution of the light wave field to extract surface topography information of reflecting or scattering objects; use two-dimensional or three-dimensional imaging technology to obtain the spatial intensity distribution of the light wave field and perform edge detection and region segmentation analysis;

[0048] The intensity feature extraction method of the light wave field is as follows:

[0049] 1. Data preprocessing

[0050] Denoising: using filters to remove noise from the signal;

[0051] Normalization: Standardize the intensity data for subsequent analysis;

[0052] 2. Time domain feature extraction

[0053] Calculate the peak value, mean value and variance of the signal, extract the instantaneous amplitude change and the envelope characteristics of the signal;

[0054] 3. Frequency domain feature extraction

[0055] Use Fourier transform to analyze the spectrum distribution and extract the main frequency components and their corresponding intensities; analyze the frequency domain energy distribution and determine the energy proportion of a specific frequency band;

[0056] 4. Time-frequency domain analysis

[0057] Use short-time Fourier transform or wavelet transform to extract the joint features of the signal in the time domain and frequency domain, and identify the instantaneous change points or local intensity mutations in the signal;

[0058] 5. Pattern recognition and feature classification

[0059] Use feature selection methods to reduce dimensionality and extract key features; perform classification or pattern recognition on the extracted intensity and reflectivity features based on machine learning;

[0060] Reflectivity feature extraction: By analyzing the relationship between the reflection intensity of the light wave and the incident light, the reflectivity feature is extracted;

[0061] Polarization feature extraction: Polarization optical instruments are used to extract the polarization characteristics of light waves. The change in polarization state is related to the microscopic structure of the object surface.

[0062] Furthermore, the algorithm for extracting the frequency band energy features of the db4 wavelet packet decomposition is as follows:

[0063] 1. Perform wavelet packet decomposition of electromagnetic signals based on the wavelet analysis tool in MATLAB;

[0064] The obtained signal data is decomposed into four layers, and the signal obtained after reconstruction is:

[0065] .

[0066] 2. Measure the wavelet packet coefficients and calculate their energy, coefficient mean, and standard deviation;

[0067] A total of 16 frequency bands are obtained, and the energy of the reconstructed signal within the frequency band is :

[0068]

[0069] The formula for calculating the mean of the wavelet decomposition coefficients is as follows:

[0070]

[0071] 3. Combine the three statistical quantities of energy parameter, coefficient mean, and wavelet packet coefficient standard deviation to represent the characteristics of signal data; classify the three statistical quantities of energy parameter, coefficient mean, and wavelet packet coefficient standard deviation into a vector parameter, which is used as the energy parameter feature of the data for subsequent cluster analysis and identification; use the Min-Max method for normalization:

[0072]

[0073] The data is collected based on the magnetic field intensity and its frequency distribution, the wavelength and intensity of light wave radiation, and the electromagnetic field intensity and frequency distribution range generated by different power equipment.

[0074] Furthermore, the acoustic wave field modeling process in step 6 is as follows:

[0075] In the COMSOL Multiphysics acoustics module, the acoustic-structure interaction element was used for modeling. Based on the size of the construction site, an air domain of similar size was established. A solid model was built based on the construction environment, and perfectly matched layers were set to simulate radiation boundaries. Material properties were defined based on the materials of the different structures. A mesh size of 5 elements per wavelength was used to divide the model. Finally, the solution time step and output step size were set to generate the output acoustic field contour map and time-domain sound data, and post-process the acoustic-frequency data.

[0076] The modeling process of the light wave field is as follows:

[0077] Firstly, the radiation transfer process of argon plasma in arc welding is analyzed, and a relationship model between effective irradiance and local emission in the arc welding process is established; the radiation characteristics of argon plasma in the optical wave line, UVA and blue light regions are calculated, and the relative contributions of optical wave line, UVA and blue light radiation to the total radiation are analyzed; the effective emission coefficients of the three spectral regions are calculated, and their relationship with temperature is fitted, laying the foundation for subsequent arc modeling; secondly, the modeling parameters are set: the spatial size of the welding analysis is set according to actual needs; the diameter of the tungsten electrode is set to 3.2 mm, the cone angle is set to 60°, and its end is flattened to a radius of 0.2 mm; the thickness of the copper plate is set to 2 mm and is perpendicular to the tungsten electrode; the shielding gas of argon is set to flow out from a nozzle with a radius of 6 mm at a rate of 10 L / min; the temperature of the top cathode is set to 2200 K, and the ambient temperature during welding is set to 300 K; Finally, a coupled partial differential equation involving conservation of mass, energy, momentum, current, and magnetic potential is established, the electrode mesh is divided using the "simplified unified model" method, and the computational solution is achieved using the finite volume method and the SIMPLE algorithm of the FLUENT solver; standard space discretization is used for pressure items, and second-order upwind is used for energy items;

[0078] The electromagnetic field modeling process is as follows:

[0079] First, the method of establishing the electromagnetic field is based on the Maxwell-Abe equation, Faraday's law, Gauss's law, and Gauss's magnetic law. The specific equations are as follows:

[0080] Maxwell-Abe's law:

[0081] Faraday's Law:

[0082] Gauss's law, Gauss's magnetic law: ,

[0083] In COMSOL Multiphysics, select the three-dimensional dimension, the magnetic field module, and the transient study model to analyze the time-varying changes of the electromagnetic field. Second, set the spatial dimensions of the electromagnetic field and establish a geometric model. Third, define the material parameters of the geometric model. Fourth, set the boundary conditions. Fifth, use the physical field to control the mesh to implement meshing. Finally, solve the electromagnetic field and output the magnetic flux density change data at different distances.

[0084] Furthermore, step 7 includes: based on the threshold and tolerance of the acoustic-optical-magnetic sensitivity index, a unit space biological loss coefficient for the acoustic-optical-magnetic pollution generated by the construction is given, and the acoustic / optical / magnetic ecological loss index is calculated according to the following formula: :

[0085]

[0086]

[0087]

[0088] Where: ELI s ELI is the ecological loss index of acoustic signals g is the light signal ecological loss index, ELI c is the magnetic signal ecological loss index, n is the number of acoustic / optical / magnetic signal sensitivity indicators, is the impact area of the i-th sensitivity index of the acoustic / optical / magnetic signal, T i is the impact time of the i-th sensitivity index of the acoustic / optical / magnetic signal, The biological loss coefficient per unit space of the i-th sensitivity index of acoustic / optical / magnetic signals;

[0089] Finally, based on the principal component analysis method, the weights of the acoustic, optical, and magnetic ecological loss indices were determined, and a comprehensive ecological loss index considering acoustic, optical, and magnetic was proposed. After determining the weights of different physical fields, the comprehensive ecological loss index was calculated using the following formula: :

[0090]

[0091] Where: is the sum of ecological loss indices, α s is the weight of the acoustic ecological loss index; α g is the weight of the light ecological loss index; α c is the weight of the magnetic ecological loss index;

[0092] At the same time, based on the comprehensive ecological loss index, the environmental impact of the sound, light and magnetism generated during the construction process is evaluated.

[0093] The invention has the following beneficial effects:

[0094] (1) This invention proposes an integrated acoustic-optical-magnetic detection method that integrates the detection technologies of three different physical quantities: acoustic, optical, and magnetic field signals. This overcomes the limitations of single-signal detection methods. By simultaneously collecting multiple physical signals, the physical state of the object being detected can be comprehensively assessed from multiple perspectives. This method is particularly suitable for precision detection in complex environments.

[0095] (2) This invention innovatively proposes a comprehensive ecological loss index for the impact of acoustics, light, and magnetism on ecology. , which can be used to evaluate the impact of engineering projects on the environment during construction, and can significantly improve the accuracy of test results and the reliability of decision-making.

[0096] (3) This invention innovates in data processing and feedback mechanisms. By collecting, processing, and analyzing acoustic, optical, and magnetic signals in real time, the system can quickly respond to test results and provide real-time feedback. In particular, when an anomaly or defect is detected, the system will immediately issue an alarm, prompting the user to take appropriate measures. This rapid feedback mechanism can significantly improve the timeliness of detection and reduce the occurrence of potential failures.

[0097] (4) The invention also innovatively integrates acoustic, optical, and magnetic sensors and their associated data processing systems into a single platform. This integrated design not only reduces system costs but also simplifies equipment installation and maintenance. Furthermore, the integrated system enables unified management and processing of data from various sensors, thereby improving data interoperability and system controllability. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] Figure 1 A technical roadmap for the specific implementation of the present invention;

[0099] Figure 2 Schematic diagram of MFCC coefficient extraction steps;

[0100] Figure 3 Schematic diagram of light wave field feature extraction;

[0101] Figure 4 Data collection and display method for acoustic-optical-magnetic intelligent sensing technology;

[0102] Figure 5 This is the arc welding radiation correlation coefficient diagram;

[0103] Figure 6 Diagram of the process of modeling electromagnetic fields;

[0104] Figure 7 A comprehensive ecological loss index process considering acoustic-optical-magnetic physical fields; DETAILED DESCRIPTION

[0105] The following is a clear and complete description of the technical solutions used in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all inspired embodiments obtained by ordinary technicians in this field without making any creative work are within the scope of protection of the present invention.

[0106] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0107] See also Figure 1This embodiment provides an acousto-optic-magnetic intelligent detection method based on a comprehensive ecological loss index, the method comprising:

[0108] Step 1: Based on the basic theory of acousto-optic magnetism, derive the equations of acousto-optic magnetism and establish the connection between basic theory and practical application.

[0109] The basic theories of acousto-optical-magnetic theory include structural vibration theory, basic acoustic theory, acoustic-vibration coupling theory, acoustic radiation theory and basic light wave theory. Based on these theories, the theory of sound wave generation and propagation is derived; the light radiation equation generated during arc welding is used; the Maxwell equations are used to describe the electromagnetic phenomena generated during the power-on working process of power equipment, and the Maxwell equations are solved by analytical methods.

[0110] The specific derivation of the basic theory of acousto-optical magnetism is as follows:

[0111] Fourier analysis is a mathematical method that decomposes complex periodic or non-periodic signals into sine waves (or cosine waves). For non-periodic signals, Fourier transform can be used to convert the signal from the time domain to the frequency domain to reveal its frequency components and amplitude. Most of the sounds at the construction site are irregular and their period is not fixed, so the Fourier transform formula is used. For non-periodic functions , Fourier transform transforms the time domain signal Convert to frequency domain signal , the formula is:

[0112]

[0113] in: It is a frequency domain representation that reflects the frequency components of the signal; is the angular frequency, is the frequency.

[0114] The inverse Fourier transform is used to recover the time domain signal from the frequency domain:

[0115]

[0116] The light radiation generated during arc welding mainly includes visible light, ultraviolet (UV), and infrared (IR). The welding arc is approximately blackbody radiation. The high-temperature metal in the welding area releases energy in the form of electromagnetic waves within a certain wavelength range. The energy distribution of blackbody radiation can be calculated using Planck's formula:

[0117]

[0118] in: is the wavelength The radiation intensity at ); is Planck's constant ( ); is the speed of light ( ); is the wavelength (unit: m); is the Boltzmann constant ( ); is the temperature of the welding arc (unit: Kelvin).

[0119] The temperature of the welding arc is usually between 3000 K and 20000 K, and its radiation energy is concentrated in the ultraviolet, visible and near-infrared bands.

[0120] Ultraviolet radiation estimation: During welding, ultraviolet radiation is a high-risk radiation, and its intensity can be estimated by experimental fitting equation:

[0121]

[0122] in: is the intensity of ultraviolet radiation; Empirical constants (related to welding materials and arc type); : Index (measured by experiment, usually between 2-4).

[0123] Maxwell's equations are the core theory of electromagnetism. They systematically describe the generation, changes, and interactions of electric and magnetic fields, revealing the nature of electromagnetic waves. The following is a calculation based on Maxwell's equations:

[0124] Gauss's law (electric field):

[0125] Gauss's law (magnetic field):

[0126] Faraday's law of electromagnetic induction:

[0127] Ampere-Maxwell Law:

[0128] Step 2: Assemble the acoustic, optical and magnetic integrated data acquisition instrument and install it on the mobile carrying device.

[0129] The acousto-optical-magnetic integrated data acquisition instrument includes an acoustic wave sensor, an ultraviolet radiation meter and a three-axis magnetic field probe. The three are connected to a small data acquisition instrument through GeoCOM serial port technology and installed on a mobile carrying device, which is convenient for layout and data collection at the construction site.

[0130] Specifically, the acoustic wave sensor uses a fiber optic Fabry-Pérot interferometer acoustic wave sensor. This sensor does not require a reference arm, so it can maintain high sensitivity while achieving miniaturization; the light wave radiation of arc welding is mainly ultraviolet light wave radiation, which is recorded by a TS180 ultraviolet radiometer; the three-axis magnetic field probe of the electromagnetic field uses a TM-195 three-axis radio frequency field strength meter high-frequency electromagnetic radiation detector, which can detect the electromagnetic field radiation intensity in three-dimensional space.

[0131] GeoCOM is an efficient and reliable data exchange protocol designed specifically for Geographic Information Systems (GIS). It is widely used in remote sensing data transmission, mobile geographic information services, and other fields. By defining a series of standardized commands and message formats, GeoCOM ensures compatibility and accurate data exchange across different platforms and devices.

[0132] Step 3: Based on the different characteristics of the three signals, the characteristics of the acoustic, optical and magnetic signals generated at the construction site are extracted.

[0133] It includes extracting the characteristic values of the collected sound based on the MFCC feature extraction method, and processing and analyzing it; extracting the intensity, reflectivity and polarization characteristics of the light wave field based on a spectrometer and polarization optical instrument, and processing and analyzing it; and extracting the frequency band energy characteristics of the db4 wavelet packet decomposition based on the energy feature extraction method of wavelet packet decomposition, and processing and analyzing it.

[0134] Furthermore, the MFCC coefficient extraction steps are as follows Figure 2 As shown in the figure, 1. First, the sound signal must be preprocessed. The preprocessing part includes the following: (1) Pre-emphasis: In order to ensure the smoothness of the signal spectrum, the signal needs to be emphasized. A first-order high-pass filter is used. The sound signal of the construction area after pre-emphasis is set to , the specific calculation is as follows:

[0135]

[0136] in: is the sound signal of the current frame, is the sound signal of the previous frame, is the weighting coefficient, which is generally 0.94-0.97. In the present invention, it is 0.95 according to the on-site conditions.

[0137] (2) Endpoint detection: It can determine the starting and ending points of the input sound signal, monitor the effective range of the input sound, and determine the sound signal by detecting the short-time energy and short-time zero-crossing rate.

[0138] Suppose a short-duration sound signal is , the length is N, and the calculation formula of its short-time energy E is:

[0139]

[0140] The calculation formula of its short-time zero-crossing rate z is:

[0141]

[0142] in: is a symbolic function, namely:

[0143]

[0144] (3) Framing: The sound signals under various working conditions are often time-varying signals that may change at any time. However, it can be considered that the sound signals are stable in a very short time. Framing is to intercept a 10ms-30ms sound signal from a characteristically stable sound signal as a frame. The sound signal can be considered stable within this frame. To ensure the continuity and smoothness between two consecutive frames, each frame is shifted forward by 10ms to ensure continuity and smoothness between frames.

[0145] (4) Windowing:

[0146] There is a truncation effect between the frames of the sound signal. In order to avoid a sharp change at the beginning and end of the frame signal, the frame signal is multiplied by the window function. The present invention uses a Hamming window for windowing processing, and its formula is:

[0147]

[0148] Assume the sound frame signal is , the window function is , the sound signal after windowing is :

[0149]

[0150] 2. Fast Fourier Transform

[0151] Perform fast Fourier transform on each frame of sound and calculate its spectral line energy through fast Fourier transform:

[0152]

[0153]

[0154] 3. Filtering

[0155] Calculate the energy of the spectrum of each frame in the filter when it passes through the filter, and convert the spectrum line energy into The corresponding frequency domain of the filter Multiply and add, the specific calculation is as follows:

[0156]

[0157] 4. Discrete Cosine Transform

[0158] The spectral coefficients are converted to the time domain using discrete cosine transform, and finally the standard frequency-based cepstral coefficients are obtained.

[0159]

[0160] Where: m is the mth TS filter, and M represents the number of TS filters.

[0161] Furthermore, the intensity, reflectivity and polarization characteristics of the light wave field are extracted as follows:

[0162] Intensity feature extraction of light wave fields: Analyze the intensity distribution of light wave fields to extract surface topography information of reflecting or scattering objects. Typically, two-dimensional or three-dimensional imaging techniques (such as laser confocal scanning, interferometric imaging, and speckle imaging) are used to obtain the spatial intensity distribution of light wave fields, and perform edge detection, region segmentation, and other analyses.

[0163] Reflectivity feature extraction: By analyzing the relationship between the reflection intensity of the light wave and the incident light, characteristic quantities such as reflectivity are extracted. This is particularly important for monitoring surface roughness and morphology changes.

[0164] Polarization feature extraction: Polarization optical instruments are used to extract the polarization characteristics of light waves. The change in polarization state is related to the microstructure of the object surface (such as surface roughness and stress distribution).

[0165] Specifically, the intensity feature extraction method of the light wave field is as follows:

[0166] 1. Data preprocessing

[0167] Denoising: Use filters (such as low-pass filters, wavelet transforms, etc.) to eliminate noise in the signal.

[0168] Normalization: Normalize the intensity data for subsequent analysis.

[0169] 2. Time domain feature extraction

[0170] Calculate the peak value, mean value, variance and other statistical quantities of the signal. Extract the instantaneous amplitude change and the envelope characteristics of the signal.

[0171] 3. Frequency domain feature extraction

[0172] Use Fourier transform to analyze the spectrum distribution and extract the main frequency components and their corresponding intensities. Analyze the frequency domain energy distribution and determine the energy proportion of specific frequency bands.

[0173] 4. Time-frequency domain analysis

[0174] Use short-time Fourier transform (STFT) or wavelet transform to extract joint features of the signal in the time and frequency domains. Identify instantaneous change points or local intensity mutations in the signal.

[0175] 5. Pattern recognition and feature classification

[0176] Feature selection methods (such as principal component analysis (PCA)) are used to reduce dimensionality and extract key features. Machine learning (such as support vector machines (SVMs) and neural networks) are used to classify or perform pattern recognition on the extracted features, such as intensity and reflectivity.

[0177] Furthermore, wavelet packet decomposition is used for electromagnetic wave feature extraction. Previous studies have shown that the correlation coefficient between electromagnetic radiation signal data and the fourth order of Daubechies wavelet (abbreviated as db4) is the largest. The frequency band interval energy feature of db4 wavelet packet decomposition is used for extraction.

[0178] Specifically, the algorithm is as follows:

[0179] 1. Perform wavelet packet decomposition of electromagnetic signals based on the wavelet analysis tool in MATLAB;

[0180] The obtained signal data is decomposed into four layers, and the signal obtained after reconstruction is:

[0181] .

[0182] 2. Measure the wavelet packet coefficients and calculate their energy, coefficient mean, and standard deviation;

[0183] A total of 16 frequency bands are obtained, and the energy of the reconstructed signal within the frequency band is :

[0184]

[0185] The formula for calculating the mean of the wavelet decomposition coefficients is as follows:

[0186]

[0187] 3. Combine the three statistics of energy parameter, coefficient mean, and wavelet packet coefficient standard deviation to represent the characteristics of signal data. The three statistics of energy parameter, coefficient mean, and wavelet packet coefficient standard deviation are classified into a vector parameter as the energy parameter feature of the data for subsequent cluster analysis and identification. We use the Min-Max method for normalization:

[0188]

[0189] The data is collected based on the magnetic field intensity and its frequency distribution, the wavelength and intensity of light wave radiation, and the electromagnetic field intensity and frequency distribution range generated by different power equipment.

[0190] Step 4: Use the Internet of Things technology to transmit the extracted acousto-optical-magnetic characteristic values to the cloud database for secondary development and utilization. At the same time, develop a mobile phone APP to realize real-time viewing of the acousto-optical-magnetic characteristic data.

[0191] Step 5: Arrange the acoustic, optical and magnetic perception test equipment at and around the construction site based on the location of the acoustic, optical and magnetic signals generated on site.

[0192] Specifically, the layout of the acoustic, optical and magnetic perception test equipment should be focused on the locations where more acoustic, optical and magnetic signals are generated on site. Acoustic, optical and magnetic perception test equipment should also be arranged around the construction site to record the impact of the construction site on the surrounding environment. At the same time, the data inside the site and the data outside the site can be compared to analyze the attenuation law of acoustic, optical and magnetic propagation, so as to provide protection and reduce the impact of acoustic, optical and magnetic signals generated during construction on the surrounding environment.

[0193] Step 6: Based on the acousto-optic-magnetic theory and field test data, establish the physical models of the acoustic wave field, optical wave field, and electromagnetic field.

[0194] Specifically, the acoustic wave field modeling process is as follows:

[0195] In the acoustics module of COMSOL Multiphysics, acoustic-structure coupling interaction elements were selected for modeling. Based on the size of the construction site area, air domains of similar size were established. Based on the construction environment, a solid model was built, and perfectly matched layers were set to simulate radiation boundaries. Material properties were defined based on the materials of different structures. Considering that the grid size affects calculation accuracy and time, a grid size of 5 elements per wavelength, based on the research results of existing literature, can achieve a good balance between accuracy and time consumption. Therefore, a grid size of 5 elements per wavelength was used to divide the model. Finally, the solution time step and output step size were set to solve the output sound field cloud map and time-domain sound data, and the sound-frequency domain data was obtained through post-processing.

[0196] Specifically, the modeling process of the light wave field is as follows:

[0197] First, the radiative transfer process of argon plasma during arc welding was analyzed, and a model for the relationship between effective irradiance and local emission during arc welding was established. The radiative characteristics of argon plasma in the optical, UVA, and blue regions were calculated, and the relative contributions of optical, UVA, and blue radiation to the total radiation were analyzed. The effective emission coefficients in these three spectral regions were calculated and their relationships with temperature were fitted, laying the foundation for subsequent arc modeling. Second, the modeling parameters were set: the spatial dimensions of the welding analysis were set according to actual needs; the tungsten electrode diameter was set to 3.2 mm, the cone angle was set to 60°, and its end was flattened to a radius of 0.2 mm; the copper plate thickness was set to 2 mm and perpendicular to the tungsten electrode; the argon shielding gas flowed at a rate of 10 L / min from a nozzle with a radius of 6 mm; the top cathode temperature was set to 2200 K, and the ambient temperature during welding was set to 300 K. Finally, coupled partial differential equations involving conservation of mass, energy, momentum, current, and magnetic potential were established. The electrodes were meshed using the "Simplified Unified Model" approach, and the computational solution was achieved using the finite volume method and the SIMPLE algorithm of the Fluent solver. Standard space discretization was used for the pressure term, and a second-order upwind was used for the energy term.

[0198] Specifically, the electromagnetic field modeling process is as follows:

[0199] First, the method of establishing the electromagnetic field is based on the Maxwell-Abe equation, Faraday's law, Gauss's law, and Gauss's magnetic law. The specific equations are as follows:

[0200] Maxwell-Abe's law:

[0201] Faraday's Law:

[0202] Gauss's law, Gauss's magnetic law: ,

[0203] In COMSOL Multiphysics, select the three-dimensional dimension, the magnetic field module, and the transient study model to analyze the time-varying changes of the electromagnetic field. Second, set the spatial dimensions of the electromagnetic field and establish a geometric model. Third, define the material parameters of the geometric model. Fourth, set the boundary conditions. Fifth, use the physical field to control the mesh to implement meshing. Finally, solve the electromagnetic field and output the magnetic flux density change data at different distances.

[0204] Step 7: Propose the ecological loss index of sound, light and magnetism generated during construction: , and combined with the weight coefficients of different physical fields, a comprehensive ecological loss index is proposed to evaluate the environmental impact of the acoustic, optical and magnetic effects generated during the construction process.

[0205] Specifically, the project first examined numerous scientific research reports and academic papers on the effects of sound waves, light waves, and electromagnetic fields on organisms. The project analyzed the growth, division, and death of various organisms in cell-based experiments, gene expression analysis, and proteomics experiments. Furthermore, the project summarized the impact of construction activities on biological behaviors, such as migration and mortality, from research reports on similar construction projects both domestically and internationally. By integrating these findings, the project determined the sensitivity indicators of different organisms to sound waves, light waves, and electromagnetic fields, clarified the sensitivity thresholds for these indicators, and established a method for assessing biological sensitivity using acoustic, optical, and magnetic methods.

[0206] Secondly, combining field test data and numerical simulation results, the impact range and duration of the acoustic-optical-magnetic sensitivity indicators generated by bridge construction were determined. Furthermore, based on the thresholds and tolerance levels of the acoustic-optical-magnetic sensitivity indicators, and referring to relevant guidelines such as the "Green GDP (GGDP / EDP) Accounting Technical Guide" (Trial Use) and the "Manual on Pollution Emission Statistical Survey and Production and Discharge Accounting Methods and Coefficients", a unit space biological loss coefficient for the acoustic-optical-magnetic pollution generated by construction was given, and the acoustic / optical / magnetic ecological loss index was calculated according to the following formula:

[0207]

[0208]

[0209]

[0210] Where: ELI s ELI is the ecological loss index of acoustic signals g is the light signal ecological loss index, ELI c is the magnetic signal ecological loss index, n is the number of acoustic / optical / magnetic signal sensitivity indicators, is the impact area of the i-th sensitivity index of the acoustic / optical / magnetic signal, T i is the impact time of the i-th sensitivity index of the acoustic / optical / magnetic signal, The biological loss coefficient per unit space of the i-th sensitivity index of acoustic / optical / magnetic signals.

[0211] Finally, based on the principal component analysis method, the weights of the acoustic, optical, and magnetic ecological loss indices were determined, and a comprehensive ecological loss index considering acoustic, optical, and magnetic was proposed. The steps of the principal component analysis method include standardization, correlation coefficient matrix calculation, eigenvalue and eigenvector calculation, and comprehensive evaluation value calculation. After determining the weights of different physical fields, the comprehensive ecological loss index is calculated using the following formula :

[0212]

[0213] Where: is the sum of ecological loss indices, αs is the weight of the acoustic ecological loss index; α g is the weight of the light ecological loss index; α c is the weight of the magnetic ecological loss index.

[0214] The environmental impacts of the construction process on acoustics, light, and magnetism were also evaluated based on the comprehensive ecological loss index. Principal component analysis was used to determine the weights of different sensitivity indicators, and an environmental impact assessment method for the construction process of the animal passage bridge was established based on the comprehensive ecological loss index.

[0215] The innovation of this invention lies in its integration of acoustic, optical, and magnetic signal detection technologies. Through advanced technologies such as multi-signal fusion, intelligent data analysis, and machine learning optimization, this significantly improves detection accuracy and reliability. The system is highly adaptable, robust against interference, and offers excellent real-time feedback. Furthermore, the system's integrated design and adaptive processing methods enable it to adapt to complex environments and diverse application scenarios, demonstrating its strong practical application value and broad market prospects.

[0216] Table 1 shows the noise monitoring data and results during the construction of a bridge

[0217] Detection time Noise level (dB) Detection location Type of construction activity Noise characteristics Compliance (day / night) 2024-10-08 07:30 82 Main pier drilling platform Rotary drilling rig drilling Low-frequency vibration + mechanical roar Exceeding the limit during the day (>70) 2024-10-07 09:15 88 Pier construction area Pile driving ship sinking piles Impact noise (peak 105dB) Seriously exceeded during the day 2024-10-08 11:00 75 Prefabricated beam yard Rebar cutting machine operation High-frequency harsh noise Daytime critical 2025-03-05 14:20 68 Bridge deck pavement area Concrete pump truck pouring Continuous medium-frequency noise Daytime compliance 2024-01-08 16:45 92 Main span cantilever section Bridge erection machine steel beam hoisting Metal collision sound + motor noise Exceeding the limit during the day 2024-9-15 22:30 63 Shore cofferdam Nighttime pumping and drainage operations Water pump runs at low frequency Exceeding the standard at night (>55) 2024-10-08 06:00 58 Construction access road Dump trucks pass Intermittent traffic noise Night Critical

[0218] Table 2 shows the monitoring data and results of light and electromagnetic radiation during the construction of a bridge

[0219] Monitoring time Monitoring Type Monitoring value Monitoring location pollution sources National standard limit Exceeding the standard 2024-10-08 20:30 light pollution 85 lux (vertical illuminance) Outside the east wall of the construction site Tower crane dysprosium lamp (2000W) ≤500 lux during the day, ≤50 lux at night Exceeding the limit at night 2024-10-08 16:15 light pollution 1200 cd / m² (brightness) Welding work area Arc welding of steel structures Brightness ≤500 cd / m² Seriously exceeding the standard 2024-10-08 16:00 Ultraviolet radiation 12 μW / cm² Steel plate cutting area plasma cutting machine ≤10 μW / cm² (8h exposure) Exceeding the standard by 20% 2024-10-08 16:30 electromagnetic radiation 0.8 μT (magnetic induction intensity) Transformer surroundings Temporary distribution box ≤100 μT (power frequency electric field) Compliance 2024-10-08 16:45 ionizing radiation 1.2 μSv / h Weld inspection area X-ray flaw detector ≤2.5 μSv / h (instantaneous) Compliance 2024-10-08 23:00 light pollution 65 lux (towards residential areas) Residential building on the north side of the construction site Night construction lighting ≤25 lux at night Exceeding the standard by 160%

Claims

1. An intelligent acoustic, optical and magnetic detection method based on the comprehensive index of ecological loss, characterized in that: The following steps are involved: Step 1: Based on the basic theory of acousto-optic magnetism, derive and solve the acousto-optic magnetism equations; Step 2: Assemble the acoustic, optical and magnetic integrated data acquisition instrument and install it on the mobile carrying device; Step 3: Extract the characteristics of the acoustic, optical and magnetic signals generated at the construction site based on their different characteristics. Step 4: Use IoT technology to transmit the extracted acousto-optical-magnetic characteristic values to a cloud database for secondary development and utilization. At the same time, develop a mobile phone app to enable real-time viewing of the acousto-optical-magnetic characteristic data. Step 5: Arrange the acoustic, optical and magnetic perception test equipment at and around the construction site based on the location of the acoustic, optical and magnetic field. Step 6: Based on the acousto-optic-magnetic theory and field test data, establish the physical models of the acoustic wave field, optical wave field and electromagnetic field; Step 7: Propose the ecological loss index of sound, light and magnetism generated during construction: , and combined with the weight coefficients of different physical fields, a comprehensive ecological loss index is proposed to evaluate the environmental impact of the acoustic, optical and magnetic effects generated during the construction process.

2. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 1, characterized in that: The basic theories of acousto-optical-magnetic theory described in step 1 include structural vibration theory, basic acoustic theory, acoustic-vibration coupling theory, acoustic radiation theory, and basic light wave theory. Based on these theories, the theory of sound wave generation and propagation is derived. The light radiation equation generated during arc welding; Maxwell's equations are used to describe the electromagnetic phenomena generated during the power supply process of power equipment, and the Maxwell equations are solved by analytical methods.

3. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 1 is characterized in that: The acoustic, optical and magnetic integrated data acquisition instrument described in step 2 includes an acoustic wave sensor, an ultraviolet radiometer and a three-axis magnetic field probe. The three are connected to a small data acquisition instrument through GeoCOM serial port technology and installed on a mobile device to facilitate layout and data collection at the construction site.

4. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 1, characterized in that: Step 3 includes extracting the characteristic values of the collected sound based on the MFCC feature extraction method, and processing and analyzing them; extracting the intensity, reflectivity and polarization characteristics of the light wave field based on the spectrometer and polarization optical instrument, and processing and analyzing them; extracting the frequency band interval energy characteristics of the db4 wavelet packet decomposition based on the energy feature extraction method of wavelet packet decomposition, and processing and analyzing them.

5. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 4 is characterized in that: The MFCC feature extraction method is used to extract the feature value of the collected sound, including:

1. Preprocessing: (1) Pre-emphasis: A first-order high-pass filter is used, and the sound signal of the construction area after pre-emphasis processing is set to y(n). Its mathematical expression is as follows: in: is the sound signal of the current frame, is the sound signal of the previous frame, is the weighting factor; (2) Endpoint detection: Endpoint detection determines the starting point and ending point of the input sound signal, and uses short-time energy and short-time zero-crossing rate for detection. The short-time energy calculation formula is as follows: in: is a short-term sound signal with a length of N; The calculation formula of short-time zero-crossing rate is as follows: in: is a symbolic function, namely: (3) Framing: Framing is to intercept a 10ms-30ms sound signal from a stable sound signal as a frame. To ensure the continuity and smoothness between two consecutive frames, each frame is shifted forward by 10ms to ensure the continuity and smoothness between frames. (4) Windowing: There will be a truncation effect between the frames of the sound signal. In order to avoid causing a sharp change at the beginning and end of the frame signal, the windowing steps are introduced as follows: Assume the sound frame signal is , the window function is , the sound signal after windowing is : The Hamming window is selected as the window function, and its formula is as follows:

2. Fast Fourier Transform Perform fast Fourier transform on each frame of sound and calculate its spectral line energy:

3. Filtering Calculate the energy of the spectrum of each frame in the filter when it passes through the filter, and convert the spectrum line energy into The corresponding frequency domain of the filter Multiply and add, the specific calculation is as follows:

4. Discrete Cosine Transform The spectral coefficients are converted to the time domain using discrete cosine transform, and finally the standard frequency-based cepstral coefficients are obtained; Where: m is the mth TS filter, and M represents the number of TS filters.

6. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 4, characterized in that: The extracting of the intensity, reflectivity and polarization characteristics of the light wave field includes: Light wave field intensity feature extraction: Analyze the intensity distribution of the light wave field to extract surface topography information of reflecting or scattering objects; use two-dimensional or three-dimensional imaging technology to obtain the spatial intensity distribution of the light wave field and perform edge detection and region segmentation analysis; The intensity feature extraction method of the light wave field is as follows:

1. Data preprocessing Denoising: using filters to remove noise from the signal; Normalization: Standardize the intensity data for subsequent analysis; 2. Time domain feature extraction Calculate the peak value, mean value and variance of the signal, extract the instantaneous amplitude change and the envelope characteristics of the signal; 3. Frequency domain feature extraction Use Fourier transform to analyze the spectrum distribution and extract the main frequency components and their corresponding intensities; analyze the frequency domain energy distribution and determine the energy proportion of a specific frequency band; 4. Time-frequency domain analysis Use short-time Fourier transform or wavelet transform to extract the joint features of the signal in the time domain and frequency domain, and identify the instantaneous change points or local intensity mutations in the signal; 5. Pattern recognition and feature classification Use feature selection methods to reduce dimensionality and extract key features; perform classification or pattern recognition on the extracted intensity and reflectivity features based on machine learning; Reflectivity feature extraction: By analyzing the relationship between the reflection intensity of the light wave and the incident light, the reflectivity feature is extracted; Polarization feature extraction: Polarization optical instruments are used to extract the polarization characteristics of light waves. The change in polarization state is related to the microscopic structure of the object surface.

7. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 4, characterized in that: The algorithm for extracting the frequency band energy features of the db4 wavelet packet decomposition is as follows:

1. Perform wavelet packet decomposition of electromagnetic signals based on the wavelet analysis tool in MATLAB; The obtained signal data is decomposed into four layers, and the signal obtained after reconstruction is: .

2. Measure the wavelet packet coefficients and calculate their energy, coefficient mean, and standard deviation; A total of 16 frequency bands are obtained, and the energy of the reconstructed signal within the frequency band is : The formula for calculating the mean of the wavelet decomposition coefficients is as follows:

3. Combine the three statistical quantities of energy parameter, coefficient mean, and wavelet packet coefficient standard deviation to represent the characteristics of signal data; classify the three statistical quantities of energy parameter, coefficient mean, and wavelet packet coefficient standard deviation into a vector parameter, which is used as the energy parameter feature of the data for subsequent cluster analysis and identification; use the Min-Max method for normalization: The data is collected based on the magnetic field intensity and its frequency distribution, the wavelength and intensity of light wave radiation, and the electromagnetic field intensity and frequency distribution range generated by different power equipment.

8. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 1, characterized in that: The acoustic wave field modeling process in step 6 is as follows: In the COMSOL Multiphysics acoustics module, the acoustic-structure interaction element was used for modeling. Based on the size of the construction site, an air domain of similar size was established. A solid model was built based on the construction environment, and perfectly matched layers were set to simulate radiation boundaries. Material properties were defined based on the materials of the different structures. A mesh size of 5 elements per wavelength was used to divide the model. Finally, the solution time step and output step size were set to generate the output acoustic field contour map and time-domain sound data, and post-process the acoustic-frequency data. The modeling process of the light wave field is as follows: First, the radiation transfer process of argon plasma in arc welding is analyzed, and a relationship model between effective irradiance and local emission during arc welding is established. The radiation characteristics of argon plasma in the optical wave line, UVA, and blue light regions are calculated, and the relative contributions of optical wave line, UVA, and blue light radiation to the total radiation are analyzed. The effective emission coefficients of the three spectral regions were calculated and their relationship with temperature was fitted, laying the foundation for subsequent arc modeling. Secondly, the modeling parameters were set: the spatial dimensions of the welding analysis were set according to actual needs; the diameter of the tungsten electrode was set to 3.2 mm, the cone angle was set to 60°, and its end was flattened to a radius of 0.2 mm; the thickness of the copper plate was set to 2 mm and perpendicular to the tungsten electrode; the argon shielding gas was set to flow at a rate of 10 L / min from a nozzle with a radius of 6 mm; the temperature of the top cathode was set to 2200 K, and the ambient temperature during welding was set to 300 K; finally, a coupled partial differential equation containing the conservation of mass, energy, momentum, current, and magnetic potential was established, the "simplified unified model" method was used to mesh the electrodes, and the computational solution was achieved using the finite volume method and the SIMPLE algorithm of the FLUENT solver; standard space discretization was used for pressure items, and second-order upwind was used for energy items; The electromagnetic field modeling process is as follows: First, the method of establishing the electromagnetic field is based on the Maxwell-Abe equation, Faraday's law, Gauss's law, and Gauss's magnetic law. The specific equations are as follows: Maxwell-Abe's law: Faraday's Law: Gauss's law, Gauss's magnetic law: , In COMSOL Multiphysics, select the three-dimensional dimension, the magnetic field module, and the transient study model to analyze the time-varying changes of the electromagnetic field. Second, set the spatial dimensions of the electromagnetic field and establish a geometric model. Third, define the material parameters of the geometric model. Fourth, set the boundary conditions. Fifth, use the physical field to control the mesh to implement meshing. Finally, solve the electromagnetic field and output the magnetic flux density change data at different distances.

9. The acousto-optic-magnetic intelligent detection method based on the comprehensive index of ecological loss according to claim 1, characterized in that: Step 7 includes: based on the threshold and tolerance of the acoustic-optical-magnetic sensitivity index, giving the unit space biological loss coefficient for the acoustic-optical-magnetic pollution generated by construction, and calculating the acoustic / optical / magnetic ecological loss index according to the following formula: : Where: ELI s ELI is the ecological loss index of acoustic signals g is the light signal ecological loss index, ELI c is the magnetic signal ecological loss index, n is the number of acoustic / optical / magnetic signal sensitivity indicators, is the impact area of the i-th sensitivity index of the acoustic / optical / magnetic signal, T i is the impact time of the i-th sensitivity index of the acoustic / optical / magnetic signal, The biological loss coefficient per unit space of the i-th sensitivity index of acoustic / optical / magnetic signals; Finally, based on the principal component analysis method, the weights of the acoustic, optical, and magnetic ecological loss indices were determined, and a comprehensive ecological loss index considering acoustic, optical, and magnetic was proposed. After determining the weights of different physical fields, the comprehensive ecological loss index was calculated using the following formula: : Where: is the sum of ecological loss indices, α s is the weight of the acoustic ecological loss index; α g is the weight of the light ecological loss index; α c is the weight of the magnetic ecological loss index; At the same time, based on the comprehensive ecological loss index, the environmental impact of the sound, light and magnetism generated during the construction process is evaluated.

Images