Environment pollution intelligent evaluation monitoring platform based on industrial big data

Abstract

The invention discloses an intelligent environmental pollution assessment monitoring platform based on industrial big data, and relates to the technical field of data management. An attenuation abnormal feature acquisition module analyzes the attenuation characteristics of echo signals, and an angle deviation feature acquisition module calculates the influence of atmospheric refraction on a laser propagation path; the measurement precision calculation module evaluates the pollutant measurement precision by using a data prediction model, and dynamically adjusts the laser power and the emission angle in the measurement error correction module to improve the measurement precision; according to the technology, interference of meteorological conditions on laser radar measurement can be effectively reduced, the accuracy and stability of pollutant concentration monitoring are improved, and more reliable data support is provided for environmental pollution monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and particularly to an intelligent evaluation and monitoring platform for environmental pollution based on industrial big data. Background Art

[0002] The intelligent evaluation and monitoring of environmental pollution based on industrial big data refers to the use of Internet of Things, sensor technology, and big data analysis means to collect, store, analyze, and predict pollution data generated during industrial production processes in real time, so as to achieve accurate monitoring and intelligent evaluation of environmental pollution. Through methods such as AI algorithms, machine learning, and cloud computing, this technology can identify pollution sources, predict pollution trends, and provide scientific decision-making support for environmental protection supervision departments and enterprises, improving the efficiency of pollution control. Currently, environmental pollution evaluation and monitoring technologies include remote sensing monitoring technology, online monitoring systems, and pollution source emission modeling analysis, etc. For example, lidar monitoring widely used in air pollution monitoring can scan pollutants in the air through laser beams and obtain the concentration distribution of pollutants such as PM2.5, NO2, and SO2 in real time.

[0003] The prior art has the following deficiencies:

[0004] In the prior art, when a laser beam passes through the air, it undergoes Mie scattering with suspended particulate matter (such as PM2.5) and is partially absorbed by certain gaseous pollutants (such as NO2, SO2). By selecting different wavelengths, the differentiation and detection of different pollutants can be achieved. However, atmospheric environmental factors (such as temperature gradients, atmospheric turbulence, air pressure changes, etc.) will affect the propagation path of the laser in the air, resulting in beam deflection (refraction phenomenon), which in turn affects the measurement accuracy of pollutants. Under different meteorological conditions, such as strong winds, precipitation, or temperature changes, the monitoring data of the same pollution source may fluctuate greatly, affecting the consistency and accuracy of the measurement. For example, strong wind weather may accelerate the diffusion of pollutants, causing the measured pollutant concentration to decrease, while the actual emissions have not decreased, resulting in misjudgment of the monitoring results. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent evaluation and monitoring platform for environmental pollution based on industrial big data to solve the deficiencies in the background art.

[0006] To achieve the above purpose, the present invention provides the following technical solution: An intelligent evaluation and monitoring platform for environmental pollution based on industrial big data, including an attenuation anomaly feature acquisition module, an angle offset feature acquisition module, a measurement accuracy calculation module, and a measurement error correction module;

[0007] Attenuation anomaly feature acquisition module: Determine the laser wavelength and emit a laser beam into the atmosphere by the lidar system. The detector of the lidar receives the echo signal after being scattered and absorbed by pollutants. The detector converts the optical signal into an electrical signal and records the attenuation anomaly feature of the laser echo signal;

[0008] Angle offset feature acquisition module: Install multiple temperature sensors around the lidar site, which are set at different heights. By recording the temperatures at each height, calculate the temperature change rate per unit height, and use the atmospheric refraction formula to calculate the refractive index gradient. Based on the refractive index gradient, obtain the angle offset feature of the laser propagation path;

[0009] Measurement accuracy calculation module: Input the attenuation anomaly feature of the acquired laser echo signal and the angle offset feature of the laser propagation path into a pre-constructed data prediction model, and determine the measurement accuracy of pollutants according to the model calculation result;

[0010] Measurement error correction module: According to the measurement accuracy of pollutants, divide the measurement results of pollutants into accurate measurement results and inaccurate measurement results. For inaccurate measurement results, adjust the laser power and emission angle to reduce the measurement error of pollutants.

[0011] Preferably, in the attenuation anomaly feature acquisition module, the detector of the lidar receives the echo signal after being affected by pollutants and measures its power P receive , record the time delay of the echo signal to calculate the spatial position where the pollutants are located. The detector converts the received optical signal into an electrical signal, and calculates the attenuation of the laser signal according to the Beer-Lambert law: Where: α is the attenuation coefficient, d is the distance that the laser propagates, P emit is the emission power; Set the normal attenuation threshold range. If the actual attenuation value exceeds the range, it is determined that there is an abnormal attenuation phenomenon.

[0012] Preferably, after analyzing the attenuation anomaly feature of the acquired laser echo signal, generate an echo signal attenuation anomaly index. The acquisition method of the echo signal attenuation anomaly index is:

[0013] Let the received power of the echo signal be P receive (t), where t is the time, and calculate the normalized attenuation signal: Where: P emit is the emission power, P receive (t) is the received power at time t, and S(t) represents the attenuation signal; Perform empirical mode decomposition on the attenuation signal S(t). S(t) is decomposed into multiple IMF components and a residual term, and the expression is: Among them, n is the number of IMFs obtained by decomposition, rn(t) is the remaining trend term, and the energy of each IMF is calculated. The expression is: E i = Σ t IMFi 2 (t); where: E i is the energy of the i-th IMF component, and IMFi(t) is the i-th IMF signal; define the energy ratio R HF of the high-frequency IMF components. The expression is: where: k is the number of high-frequency IMF components. Calculate the echo signal attenuation anomaly index EAI. The expression is: EAI = -log(1 - R HF ).

[0014] Preferably, the angle offset feature acquisition module includes: installing multiple temperature sensors around the lidar station, setting them at different heights respectively, and regularly recording the temperature data at each height. The temperature at height hi is Ti, forming a data set; calculate the temperature gradient between adjacent heights: where: Thigh and Tlow are the temperatures at high and low positions, and hhigh and hlow are the corresponding heights; calculate the average value of the temperature gradients at multiple measurement points, record the calculated temperature gradient values, and calculate the refractive index gradient based on the atmospheric refraction formula The expression is: where: r is the refraction sensitivity coefficient, is the temperature gradient.

[0015] Preferably, after analyzing the angle offset feature of the laser propagation path, generate a laser propagation angle offset index. The method for obtaining the laser propagation angle offset index is: calculate the temperature gradient and use the atmospheric refraction formula to calculate the laser propagation angle offset θ(t). The expression is: where: r is the refraction sensitivity coefficient, d is the length of the laser propagation path. Analyze the transient change of the angle offset θ(t) at different scales through wavelet transform, and extract the high-frequency anomaly feature. The wavelet transform formula is: WT(θ)(a, b) = ∫θ(t)ψ a,b (t)dt; where: ψ a,b (t) is the wavelet basis function, a is the control frequency scale, b is the control time offset, and WT(θ)(a, b) represents the wavelet transform result of the angle offset signal θ(t) at different control frequency scales a and control time offsets b;

[0016] To quantify the transient offset of the laser propagation angle, calculate the wavelet energy at different frequency scales, calculate the wavelet energy E(a) of each control frequency scale a. The expression is: E(a) = Σ b |WT(θ)(a, b)|2 ; where: E(a) represents the wavelet energy corresponding to the control frequency scale a, WT(θ)(a,b) is the wavelet transform coefficient of the angular offset signal. By summing over all control time offsets b, the total energy at the control frequency scale a is obtained, and the proportion of the total energy in the high-frequency part is calculated, that is, the laser propagation angular offset index DFC is calculated. The expression is: where: A is the set of all scales, H is the high-frequency scale range.

[0017] Preferably, the measurement accuracy calculation module specifically includes: normalizing the echo signal attenuation anomaly index and the laser propagation angular offset index so that they are both within [0,1], and calculating the measurement accuracy value of the pollutant according to the normalized echo signal attenuation anomaly index and the laser propagation angular offset index.

[0018] Preferably, the measurement error correction module compares the obtained measurement accuracy value of the pollutant with the pre-set measurement accuracy standard threshold according to historical data. If the measurement accuracy value of the pollutant is greater than or equal to the pre-set measurement accuracy standard threshold, it indicates that the measurement accuracy of the pollutant is high, and the measurement result of the pollutant is classified as an accurate measurement result; if the measurement accuracy value of the pollutant is less than the pre-set measurement accuracy standard threshold, it indicates that the measurement accuracy of the pollutant is low, and the measurement result of the pollutant is classified as an inaccurate measurement result.

[0019] Preferably, the transmit power P emit and the received power P receive are related as: P receive = P emit ·e -zd ; where: z is the attenuation coefficient, d is the laser propagation path length;

[0020] To reduce the echo signal attenuation anomaly index EAI, the new transmit power P emit,new is adjusted: P emit,new = P emit ×(1 + β1·EAI norm ); where: β1 is the power adjustment coefficient, EAI norm is the normalized echo signal attenuation anomaly index; if EAI norm is greater than 0.5, it indicates that the signal attenuation anomaly is serious, then the transmit power P emit is increased to enhance the echo signal; if EAI norm is less than or equal to 0.5, the original power is maintained;

[0021] To compensate for the angular offset, the new transmit angle θ emit,new is adjusted, and the expression is: θ emit,new = θ emit - β2·DFCnorm · θ(t); where: β2 is the angle adjustment coefficient, and DFC norm is the normalized laser propagation angle offset exponent; if DFC norm is greater than 0.5, it indicates that the angle offset is severe, then adjust the emission angle θ emit to compensate for the refraction effect; if DFC norm is less than or equal to 0.5, then keep the original emission angle;

[0022] After adjusting the laser power and emission angle, recalculate the measurement accuracy value of the pollutant. If it is greater than or equal to the pre-set measurement accuracy standard threshold, it indicates that the adjustment is effective and the measurement accuracy is improved; otherwise, continue to optimize the parameters.

[0023] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0024] The present invention provides an intelligent evaluation and monitoring platform for environmental pollution based on industrial big data. Aiming at the measurement error problem caused by the influence of atmospheric environmental factors (such as temperature gradient, atmospheric turbulence, air pressure change, etc.) on traditional lidar monitoring, a complete set of optimization solutions is proposed. The attenuation anomaly feature acquisition module analyzes the attenuation anomaly of the echo signal, the angle offset feature acquisition module calculates the influence of atmospheric refraction on the laser propagation path, the measurement accuracy calculation module comprehensively evaluates the measurement accuracy of the pollutant using the data prediction model, and in the measurement error correction module, the laser power and emission angle are adjusted based on the measurement accuracy value, thereby reducing the measurement error. The present invention accurately extracts the attenuation and angle offset features of the echo signal through advanced algorithms such as the Beer-Lambert law, empirical mode decomposition, and wavelet transform, and combines the normalization calculation and dynamic adjustment mechanism to effectively improve the accuracy and stability of pollutant concentration monitoring. This technology can be adaptively optimized under different meteorological conditions to ensure the reliability of the monitoring data, providing strong technical support for environmental pollution monitoring and precise treatment. Brief Description of the Drawings

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0026] Figure 1 It is a module diagram of an intelligent evaluation and monitoring platform for environmental pollution based on industrial big data. Detailed Embodiments

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] For the embodiments, please refer to Figure 1 As shown in the figure, the intelligent environmental pollution assessment and monitoring platform based on industrial big data in this embodiment includes an attenuation anomaly feature acquisition module, an angle deviation feature acquisition module, a measurement accuracy calculation module, and a measurement error correction module;

[0029] Attenuation anomaly feature acquisition module: Determine the laser wavelength and emit a laser beam into the atmosphere by the lidar system. Based on the detector of the lidar, receive the echo signal after being scattered and absorbed by pollutants. The detector converts the optical signal into an electrical signal and records the attenuation anomaly feature of the laser echo signal;

[0030] Angle deviation feature acquisition module: Install multiple temperature sensors around the lidar site, which are respectively set at different heights. Calculate the temperature change rate per unit height by recording the temperatures at each height, and calculate the refractive index gradient using the atmospheric refraction formula. Based on the refractive index gradient, obtain the angle deviation feature of the laser propagation path;

[0031] Measurement accuracy calculation module: Input the attenuation anomaly feature of the acquired laser echo signal and the angle deviation feature of the laser propagation path into a pre-constructed data prediction model, and determine the measurement accuracy of pollutants according to the model calculation results;

[0032] Measurement error correction module: According to the measurement accuracy of pollutants, divide the measurement results of pollutants into accurate measurement results and inaccurate measurement results. For inaccurate measurement results, adjust the laser power and emission angle to reduce the measurement error of pollutants.

[0033] The attenuation anomaly feature acquisition module specifically includes: Select a suitable laser wavelength (such as ultraviolet, visible light, or infrared) according to the optical characteristics of the target pollutant. The lidar system emits a pulsed or continuous-wave laser beam into the atmosphere to ensure that the beam can interact with pollutants in the air. Record the initial power P of the emitted laser emit As a reference value.

[0034] When the laser propagates in the air, it scatters and absorbs with pollutants (such as PM2.5, NO2, SO2), resulting in partial energy loss. The detector of the lidar receives the echo signal after being affected by pollutants and measures its power Preceive . Record the time delay of the echo signal to calculate the spatial position where the pollutant is located.

[0035] The detector converts the received optical signal into an electrical signal and transmits it to the data processing unit. Remove environmental noise through signal filtering and noise reduction algorithms to ensure data reliability. Record the power change curve of the echo signal and extract the attenuation characteristics of the optical signal.

[0036] Calculate the attenuation of the laser signal according to the Beer-Lambert law: where: α is the attenuation coefficient, d is the distance that the laser propagates; set the normal attenuation threshold range. If the actual attenuation value exceeds this range, it is determined that there is an abnormal attenuation phenomenon. Analyze the possible causes of abnormal attenuation in combination with historical data, such as high humidity, strong turbulence, or interference from unknown pollutants.

[0037] After analyzing the abnormal attenuation characteristics of the obtained laser echo signal, generate an echo signal attenuation anomaly index. The method for obtaining the echo signal attenuation anomaly index is:

[0038] Let the received power of the echo signal be P receive (t), where t is time, and calculate the normalized attenuation signal: where: P emit is the transmitted power, P receive (t) is the received power at time t, and S(t) represents the relative attenuation signal.

[0039] Perform empirical mode decomposition (EMD) on the attenuation signal S(t). EMD decomposes S(t) into multiple intrinsic mode functions (IMFs) through an iterative method, satisfying the following conditions:

[0040] The IMF has local symmetry, and the number of extreme points is close to the number of zero crossings. The IMFs are arranged from high-frequency components to low-frequency components, reflecting different time-frequency characteristics in the signal.

[0041] Identify all local maxima and local minima of S(t). Use spline interpolation to fit the maxima and minima respectively to obtain the upper envelope U(t) and the lower envelope L(t). Calculate the mean m(t) of the upper and lower envelopes. Calculate the first-layer IMF component h1(t): h1(t) = S(t) - m(t);

[0042] If h1(t) satisfies the IMF condition, set it as IMFi(t), otherwise continue to iterate until the IMF condition is satisfied, remove the IMF and repeat the decomposition, and calculate the new residual signal r1(t). The expression is: r1(t) = S(t) - IMF1(t); S(t) is decomposed into multiple IMF components and a residual term: Among them, n is the number of IMFs obtained by decomposition, and rn(t) is the remaining trend term. The echo signal attenuation anomaly index can be calculated based on the energy ratio of IMF components. Calculate the energy of each IMF, and the expression is: E i = Σ t IMFi 2 (t); where: E i is the energy of the i-th IMF component, and IMFi(t) is the i-th IMF signal; since abnormal signals usually contain higher frequency components, define the energy ratio R HF , and the expression is: where: k is the number of high-frequency IMF components (usually take the first 3-5 IMFs), calculate the echo signal attenuation anomaly index EAI, and the expression is: EAI = -log(1 - R HF ); if R HF is larger, it indicates that the high-frequency anomaly is more obvious, then the larger the EAI, the higher the degree of attenuation anomaly.

[0043] The angle offset feature acquisition module includes: installing multiple temperature sensors around the lidar site, setting them at different heights (such as 1m, 10m, 50m) respectively, ensuring uniform distribution of the sensors to accurately reflect the temperature gradient.

[0044] Regularly record the temperature data at each height. Let the temperature at height hi be Ti, and form a data set: (h1, T1), (h2, T2),..., (hn, Tn); use a filtering algorithm (such as the moving average method) to smooth the collected data, remove accidental errors, and improve data stability.

[0045] Calculate the temperature gradient (temperature change rate per unit height) between adjacent heights: where: Thigh and Tlow are the temperatures at high and low positions, and hhigh and hlow are the corresponding heights; calculate the average value of the temperature gradients at multiple measurement points, record the calculated temperature gradient values, and use them as the basic data for subsequent calculation of the refractive index gradient.

[0046] Calculate the refractive index gradient (the change rate of the refractive index with height) based on the atmospheric refraction formula The expression is: where: r is the refraction sensitivity coefficient, depending on the wavelength and meteorological conditions, is the temperature gradient. Calculate the change trend of the refractive index in the entire monitoring area and draw the refractive index distribution curve to analyze the refraction characteristics under different weather conditions.

[0047] After analyzing the angle offset characteristics of the laser propagation path, generate the laser propagation angle offset index. The method for obtaining the laser propagation angle offset index is:

[0048] Install multiple temperature sensors around the lidar site to obtain temperature data at different heights and calculate the temperature gradient And calculate the laser propagation angle offset θ(t) using the atmospheric refraction formula. The expression is as follows: Where: r is the refraction sensitivity coefficient, d is the laser propagation path length. Analyze the transient changes of the angle offset θ(t) at different scales through wavelet transform to extract high-frequency anomaly features. The wavelet transform formula is: WT(θ)(a,b) = ∫θ(t)ψ a,b (t)dt; Where: ψ a,b (t) is the wavelet basis function, a is the control frequency scale (scale factor, large values correspond to low frequencies, small values correspond to high frequencies), b is the control time offset (translation factor), and WT(θ)(a,b) represents the wavelet transform result of the angle offset signal θ(t) under different control frequency scales a and control time offsets b;

[0049] Select a suitable wavelet basis function ψ(t) (such as Morlet wavelet or Daubechies wavelet), calculate the WT coefficients WT(θ)(a,b) at different control frequency scales a to form a wavelet coefficient matrix, and identify the high-frequency part (smaller a values) for detecting sudden offsets.

[0050] To quantify the transient offset of the laser propagation angle, calculate the wavelet energy at different frequency scales. Calculate the wavelet energy E(a) for each control frequency scale a. The expression is: E(a) = ∑ b |WT(θ)(a,b)| 2 ; Where: E(a) represents the wavelet energy corresponding to the control frequency scale a, WT(θ)(a,b) is the wavelet transform coefficient of the angle offset signal. By summing over all control time offsets b, obtain the total energy at the control frequency scale a, and calculate the proportion of the total energy of the high-frequency part, that is, calculate the laser propagation angle offset index DFC. The expression is: Where: A is the set of all scales, H is the high-frequency scale range (usually select the first 3 - 5 smallest a values). DFC reflects the proportion of high-frequency energy in the total energy. The larger it is, the stronger the transient offset.

[0051] Measurement accuracy calculation module: Input the attenuation anomaly features of the acquired lidar echo signal and the angle offset features of the laser propagation path into a pre-constructed data prediction model, and determine the measurement accuracy of the pollutant according to the model calculation results. Specifically, it includes: normalizing the echo signal attenuation anomaly index and the laser propagation angle offset index so that they are both within [0,1], and calculating the measurement accuracy value of the pollutant based on the normalized echo signal attenuation anomaly index and the laser propagation angle offset index.

[0052] For example, the present invention can calculate the measurement accuracy value of pollutants using the following formula, and the calculation expression is: In the formula, MP is the measurement accuracy value of pollutants, EAI is the echo signal attenuation anomaly index, DFC is the laser propagation angle deviation index, f1 and f2 are the weight coefficients of the echo signal attenuation anomaly index and the laser propagation angle deviation index (which can be optimized according to experimental experience or machine learning), and both f1 and f2 are greater than 0.

[0053] Measurement error correction module: According to the measurement accuracy of pollutants, the measurement results of pollutants are divided into accurate measurement results and inaccurate measurement results. For inaccurate measurement results, the laser power and emission angle are adjusted to reduce the measurement error of pollutants.

[0054] Compare the obtained measurement accuracy value of pollutants with the pre-set measurement accuracy standard threshold based on historical data. If the measurement accuracy value of pollutants is greater than or equal to the pre-set measurement accuracy standard threshold, it indicates that the measurement accuracy of pollutants is high, and the measurement results of pollutants are divided into accurate measurement results; if the measurement accuracy value of pollutants is less than the pre-set measurement accuracy standard threshold, it indicates that the measurement accuracy of pollutants is low, and the measurement results of pollutants are divided into inaccurate measurement results.

[0055] Since the attenuation of the echo signal is mainly caused by atmospheric absorption and scattering, the relationship between the transmitted power P emit and the received power P receive can be expressed as: P receive = P emit ·e -zd ; where: z is the attenuation coefficient (related to atmospheric absorption and scattering), and d is the laser propagation path length;

[0056] To reduce the echo signal attenuation anomaly index EAI, adjust the new transmitted power P emit,new : P emit,new = P emit ×(1 + β1·EAI norm ); where: β1 is the power adjustment coefficient (which can be optimized through experiments or machine learning), and EAI norm is the normalized echo signal attenuation anomaly index (range [0, 1]); if EAI norm is greater than 0.5, it indicates that the signal attenuation is extremely abnormal, then increase the transmitted power P emit to enhance the echo signal. If EAI norm is less than or equal to 0.5, then maintain the original power to avoid increased energy consumption.

[0057] Since atmospheric refraction will cause the laser path to deviate, resulting in measurement errors in the target area.

[0058] To compensate for the angular offset, adjust the new emission angle θ emit,new , and the expression is: θ emit,new = θ emit - β2·DFC norm ·θ(t); where: β2 is the angle adjustment coefficient (optimized according to experiments or machine learning), DFC norm is the normalized laser propagation angle offset index (range [0, 1]); if DFC norm is greater than 0.5, it indicates that the angular offset is serious, then adjust the emission angle θ emit to compensate for the refraction effect; if DFC norm is less than or equal to 0.5, then keep the original emission angle to avoid unnecessary adjustments.

[0059] After adjusting the laser power and emission angle, recalculate the measurement accuracy value of the pollutant. If it is greater than or equal to the pre-set measurement accuracy standard threshold, it indicates that the adjustment is effective and the measurement accuracy is improved; otherwise, continue to optimize the parameters.

[0060] It should be noted here that the measurement accuracy standard threshold is an important parameter for judging whether the measurement result of the pollutant is accurate, and it is usually obtained through historical data analysis, experimental testing and statistical optimization.

[0061] As described above, this is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application.

Claims

1. An intelligent environmental pollution assessment and monitoring platform based on industrial big data, characterized by: It includes an attenuation anomaly feature acquisition module, an angle offset feature acquisition module, a measurement accuracy calculation module and a measurement error correction module; Attenuation anomaly feature acquisition module: The laser wavelength is determined and the laser radar system emits a laser beam into the atmosphere. The laser radar-based detector receives the echo signal after being scattered and absorbed by the pollutants. The detector converts the optical signal into an electrical signal and records the attenuation anomaly feature of the laser echo signal. Angle offset feature acquisition module: multiple temperature sensors are installed around the lidar site at different heights. The temperature change rate per unit height is calculated by recording the temperature at each height, and the refractive index gradient is calculated using the atmospheric refraction formula. The angle offset feature of the laser propagation path is obtained based on the refractive index gradient. Measurement accuracy calculation module: input the acquired attenuation anomaly characteristics of the laser echo signal and the angle deviation characteristics of the laser propagation path into the pre-built data prediction model, and determine the measurement accuracy of the pollutant based on the model calculation results; Measurement error correction module: According to the measurement accuracy of pollutants, the measurement results of pollutants are divided into accurate measurement results and inaccurate measurement results. For inaccurate measurement results, the laser power and emission angle are adjusted to reduce the measurement error of pollutants.

2. The environmental pollution intelligent assessment and monitoring platform based on industrial big data according to claim 1 is characterized by: In the attenuation anomaly feature acquisition module, the laser radar detector receives the echo signal after being affected by the pollutant and measures its power P receive , record the time delay of the echo signal to calculate the spatial position of the pollutant, the detector converts the received light signal into an electrical signal, and calculates the attenuation of the laser signal according to the Beer-Lambert law: Where: α is the attenuation coefficient, d is the distance the laser propagates, P emit is the transmission power; set the normal attenuation threshold range. If the actual attenuation value exceeds the range, it is determined that abnormal attenuation occurs.

3. The environmental pollution intelligent assessment and monitoring platform based on industrial big data according to claim 2 is characterized by: After analyzing the attenuation abnormality characteristics of the acquired laser echo signal, an echo signal attenuation abnormality index is generated. The method for obtaining the echo signal attenuation abnormality index is as follows: Assume the received power of the echo signal is P receive (t), where t is time, calculate the normalized attenuation Signal: Where: P emit is the transmission power, P receive (t) is the received power at time t, S(t) represents the attenuated signal; the attenuated signal S(t) is subjected to empirical mode decomposition, and S(t) is decomposed into multiple IMF components and a residual term, expressed as: Where n is the number of IMFs obtained by decomposition, rn(t) is the residual trend term, and the energy of each IMF is calculated as: E i =∑ t IMFi 2 (t); where: E i is the energy of the ith IMF component, IMFi(t) is the ith IMF signal; the energy ratio R of the high-frequency IMF component is defined as HF , the expression is: Where: k is the number of high-frequency IMF components, and the echo signal attenuation anomaly index EAI is calculated, and the expression is: EAI = -log(1-R HF ).

4. The environmental pollution intelligent assessment and monitoring platform based on industrial big data according to claim 3 is characterized by: The angle offset feature acquisition module includes: installing multiple temperature sensors around the lidar site, setting them at different heights, regularly recording the temperature data at each height, setting the temperature at height hi as Ti, and forming a data set; calculating the temperature gradient between adjacent heights: Where: Thigh and Tlow are the temperatures at high and low positions, and hhigh and hlow are the corresponding heights; calculate the average temperature gradient of multiple measurement points, record the calculated temperature gradient value, and calculate the refractive index gradient based on the atmospheric refraction formula The expression is: Where: r is the refractive sensitivity coefficient, is the temperature gradient.

5. The environmental pollution intelligent assessment and monitoring platform based on industrial big data according to claim 4 is characterized by: The laser propagation angle deviation index is generated by analyzing the angle deviation characteristics of the laser propagation path. The laser propagation angle deviation index is obtained by calculating the temperature gradient The atmospheric refraction formula is used to calculate the laser propagation angle deviation θ(t), which is expressed as: Where: r is the refractive sensitivity coefficient, d is the length of the laser propagation path, and the transient changes of the angle offset θ(t) at different scales are analyzed by wavelet transform to extract high-frequency abnormal features. The wavelet transform formula is: WT(θ)(a,b)=∫θ(t)ψ a,b (t)dt; where: ψ a,b (t) is the wavelet basis function, a is the control frequency scale, b is the control time offset, and WT(θ)(a,b) represents the wavelet transform result of the angle offset signal θ(t) under different control frequency scales a and control time offsets b; In order to quantify the transient deviation of the laser propagation angle, the wavelet energy at different frequency scales is calculated, and the wavelet energy E(a) of each control frequency scale a is calculated. The expression is: E(a) = Σ b |WT(θ)(a,b)| 2 ; Where: E(a) represents the wavelet energy corresponding to the control frequency scale a, WT(θ)(a,b) is the wavelet transform coefficient of the angle offset signal, and the total energy under the control frequency scale a is obtained by summing all control time offsets b, and the total energy proportion of the high-frequency part is calculated, that is, the laser propagation angle offset index DFC is calculated, and the expression is: Where: A is the set of all scales, H is the high frequency scale range.

6. The environmental pollution intelligent assessment and monitoring platform based on industrial big data according to claim 5 is characterized by: The measurement accuracy calculation module specifically includes: normalizing the echo signal attenuation anomaly index and the laser propagation angle deviation index so that they are both between [0,1], and calculating the measurement accuracy value of the pollutant based on the normalized echo signal attenuation anomaly index and the laser propagation angle deviation index.

7. The environmental pollution intelligent assessment and monitoring platform based on industrial big data according to claim 6 is characterized by: The measurement error correction module compares the acquired pollutant measurement accuracy value with the measurement accuracy standard threshold value pre-set according to historical data. If the pollutant measurement accuracy value is greater than or equal to the pre-set measurement accuracy standard threshold value, it means that the pollutant measurement accuracy is high, and the pollutant measurement result is classified as an accurate measurement result; if the pollutant measurement accuracy value is less than the pre-set measurement accuracy standard threshold value, it means that the pollutant measurement accuracy is low, and the pollutant measurement result is classified as an inaccurate measurement result.

8. The environmental pollution intelligent assessment and monitoring platform based on industrial big data according to claim 7 is characterized by: Transmit power P emit With the received power P receive The relationship is expressed as: receive =P emit ·e -zd ; Where: z is the attenuation coefficient, d is the length of the laser propagation path; In order to reduce the echo signal attenuation abnormality index EAI, adjust the new transmission power P emit,new :P emit,new =P emit ×(1+β1·EAI norm ), where: β1 is the power adjustment factor, EAI norm is the normalized echo signal attenuation anomaly index; if EAI norm If it is greater than 0.5, it means that the signal attenuation is abnormally serious, then increase the transmission power P emit To enhance the echo signal; if EAI norm If it is less than or equal to 0.5, the original power is maintained; To compensate for the angle deviation, adjust the new emission angle θ emit,new , the expression is: θ emit,new =θ emit -β2·DFC norm ·θ(t); where: β2 is the angle adjustment coefficient, DFC norm is the normalized laser propagation angle deviation index; if DFC norm If it is greater than 0.5, it means that the angle deviation is serious, so adjust the emission angle θ emit Compensate for refraction effects; if DFC norm If it is less than or equal to 0.5, the original emission angle is maintained; After adjusting the laser power and emission angle, recalculate the measurement accuracy value of the pollutant. If it is greater than or equal to the preset measurement accuracy standard threshold, it means that the adjustment is effective and the measurement accuracy is improved; otherwise, continue to optimize the parameters.

Images