Lampblack purifier lampblack emission real-time monitoring system based on Internet of Things technology

Abstract

The invention relates to the technical field of the Internet of Things, in particular to a real-time monitoring system for oil fume emission of an oil fume purifier based on the Internet of Things technology, and the real-time monitoring system is characterized in that oil fume concentration original data, environmental parameter data and historical oil fume data are acquired through a data acquisition module; the weighted filtering module dynamically distributes filtering weights based on the environmental parameter data, performs weighted fusion on the original data of the oil smoke concentration, and effectively inhibits environmental interference; the weighting correction module calls historical oil smoke data to establish a reference mode and distributes correction weights, secondary optimization is carried out on filtered data, and the data rationality is improved; the state evaluation and decision-making module calculates a comprehensive operation efficiency index based on the timeliness of the correction data, and generates graded early warning in combination with a concentration threshold value; according to the process, the data accuracy is optimized step by step, the problem of misjudgment caused by unreliability of single data is solved, and accurate early warning is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things, and particularly to an oil fume purifier oil fume emission real-time monitoring system based on Internet of Things technology. BACKGROUND

[0002] In the field of emission monitoring of oil fume purification equipment, real-time performance and accuracy are the core indicators for measuring system performance. The existing technology generally adopts a basic framework composed of sensors, data transmission units and monitoring platforms, directly obtains the instantaneous concentration value of the exhaust port by installing an oil fume concentration sensor, and compares it with the preset fixed limit value to trigger a warning. However, this simple mode of direct comparison exposes many inherent defects in actual industrial applications. First, the readings of the oil fume concentration sensor are easily affected by a variety of environmental variables other than cooking conditions, such as the drastic changes in temperature and humidity inside the kitchen, the instantaneous fluctuations in the pressure of the exhaust duct, and the airflow speed differences caused by the operation of the fan, which will interfere with the physical or chemical properties of the sensor element, making the "raw oil fume concentration data" itself contain a lot of noise introduced by non-oil fume factors. Directly using such unprocessed raw data for judgment will lead to frequent false alarms or missed alarms, seriously damaging the reliability of the monitoring results.

[0003] Secondly, this technical path is static and isolated, it only focuses on the single-point data at the current moment, neither considers the specific influence of environmental parameters on the current monitoring value for dynamic correction, nor learns from the "historical oil fume data" accumulated by the long-term operation of the equipment to identify the normal concentration fluctuation rules, resulting in its inability to effectively distinguish between real emission exceedance events and pseudo-signals caused by short-term environmental disturbances or data collection abnormalities.

[0004] In addition, this simple threshold method lacking data fusion and deep analysis cannot make prospective assessment on the running health status of the purification equipment, can only passively alarm after the emission concentration has exceeded the standard, lacks the level and initiative of early warning, and is difficult to meet the management needs of modern intelligent environmental protection for precise governance and prior intervention. SUMMARY

[0005] The present application aims to provide an oil fume purifier oil fume emission real-time monitoring system based on Internet of Things technology to solve the problems raised in the background technology, and the specific technical problems include how to establish a multi-level data optimization and decision-making mechanism with autonomous evolution capability to solve the fundamental defects of the traditional system based on single-point and static monitoring in terms of the credibility of core data and the accuracy of decision-making when facing complex working conditions.

[0006] To achieve the above object, the present application provides the following technical solutions, by constructing a series data processing flow including weighted filtering, weighted correction and state evaluation, by using environment parameters, historical fluctuation patterns and data timeliness to dynamically allocate and apply weights in turn, to realize multiple calibration and comprehensive analysis of original monitoring data, thereby solving the technical problems of unreliable single real-time monitoring data and insufficient early warning decision basis caused by environmental interference, random fluctuation and static judgment.

[0007] The oil fume purification device oil fume emission real-time monitoring system based on the Internet of Things technology comprises a data acquisition module, a weighted filtering module, a weighted correction module and a state evaluation and decision module, wherein: The data acquisition module is used for acquiring oil fume concentration original data, environment parameter data and historical oil fume data, wherein the oil fume concentration original data is acquired by an oil fume concentration sensor; the environment parameter data comprises temperature, humidity, pipeline pressure and exhaust flow rate; and the historical oil fume data is acquired from a time series database. The data acquisition module constructs the perception layer of the system, not only acquires oil fume concentration original data reflecting the essence of emission, but also synchronously acquires environment parameter data that may affect the measurement accuracy, and provides a data basis for subsequent analysis and correction, and the historical oil fume data is called from the time series database; this realizes the comprehensiveness of data acquisition, provides rich raw materials for subsequent intelligent processing, and avoids misjudgment caused by single data source.

[0008] The weighted filtering module dynamically allocates a filtering weight to the oil fume concentration original data based on the reference change of the environment parameter data, and performs weighted fusion to output filtered oil fume concentration data, wherein the generation process of the filtering weight specifically comprises: A reference change interval is preset for each parameter data in the environment parameter data; the environmental deviation of the measured value of each parameter data from the median value in the reference change interval is calculated; the environmental deviation is mapped to a basic weight influence factor by a mathematical function; and the basic weight influence factors of all parameter data are aggregated as the filtering weight; wherein the mapping process of the mathematical function specifically comprises: When the measured value of the parameter data is at the median value of the reference change interval, the basic weight influence factor output is one; when the measured value of the parameter data deviates from the median value of the reference change interval and exceeds the boundary of the reference change interval, the basic weight influence factor decreases to zero according to the functional relationship; wherein the calculation of the filtered oil fume concentration data specifically comprises: The product of each data point value in the oil fume concentration original data and the corresponding filtering weight is calculated to obtain a weighted value; the weighted values are summed to obtain a weighted sum; the filtering weights are summed to obtain a weight sum; and the weighted sum is divided by the weight sum to obtain the filtered oil fume concentration data.

[0009] The core effect of the weighting filtering module is to realize adaptive filtering, which dynamically allocates appropriate filtering weights for each oil fume concentration data point by analyzing the deviation degree of the current measured value of the environmental parameter from the ideal benchmark interval; when the environmental condition is ideal (in the middle of the benchmark interval), the data weight is high and is fully adopted; when the environment is severe (deviates from or even exceeds the benchmark interval), the data weight collected in this environment is reduced and its influence is weakened; through weighted fusion, the finally output filtered oil fume concentration data significantly weakens the interference caused by environmental instantaneous fluctuations, so that the data more truly reflects the changes of oil fume emission itself, and improves the anti-interference ability and preliminary reliability of the data.

[0010] The weighting correction module dynamically allocates correction weights to the filtered oil fume concentration data based on the benchmark fluctuation mode by calling historical oil fume data, and performs weighted fusion to output corrected oil fume concentration data, wherein the generation process of the correction weight specifically includes: The historical oil fume data is called, the benchmark fluctuation mode is extracted through time series analysis; the mode deviation of the filtered oil fume concentration data from the benchmark fluctuation mode is calculated; the mode deviation is mapped to a correction weight between zero and one through a normalization function, wherein when the mode deviation is zero, the correction weight is one, and when the mode deviation increases, the correction weight decreases from one to zero.

[0011] The calculation process of the corrected oil fume concentration data specifically includes: The product of each data point value in the filtered oil fume concentration data and the corresponding correction weight is calculated to obtain a weighted value; the weighted values are summed to obtain a weighted sum; the correction weights are summed to obtain a weight sum; the weighted sum is divided by the weight sum to obtain the corrected oil fume concentration data.

[0012] The weighting correction module introduces intelligent correction based on historical experience, which analyzes historical data to summarize the benchmark fluctuation mode of oil fume concentration (such as normal fluctuation law under different time periods and different workloads), and compares the current filtered data with this mode; if the current data conforms to the historical law (small mode deviation), it is given a high correction weight and is trusted; if the current data is obviously abnormal to the historical mode (large mode deviation), it is given a low weight and is suppressed; the corrected oil fume concentration data output after this step of weighted fusion not only filters out environmental noise, but also corrects abnormal values that do not conform to long-term operation rules although there is no obvious environmental interference, so that the trend and rationality of the data sequence are further enhanced; its effect is to optimize and smooth the data twice, and the finally output corrected oil fume concentration data is high-reliability data verified by environment and history, and this set of high-quality data is the direct basis for the state evaluation and decision module to make accurate evaluation and decision.

[0013] The state evaluation and decision module comprises an evaluation unit which dynamically allocates evaluation weights based on the timeliness of the corrected flue gas concentration data to calculate a comprehensive operation efficiency index of the flue gas purifier, wherein the process of calculating the evaluation weights specifically comprises: The evaluation weight of each corrected flue gas concentration data point is determined by the time difference between its time stamp and the current time, and is calculated by a preset monotonically decreasing function according to the time decay principle.

[0014] The calculation process of the comprehensive operation efficiency index comprises: The value of each corrected flue gas concentration data point is multiplied by its evaluation weight to obtain a weighted value; the weighted values are summed to obtain a weighted sum; the evaluation weights are summed to obtain a weight sum; and the weighted sum is divided by the weight sum to obtain the comprehensive operation efficiency index, which is a normalized value between zero and one.

[0015] The evaluation unit calculates the comprehensive operation efficiency index by introducing the time decay principle (recent data has a large weight and far future data has a small weight), realizes the continuous and weighted evaluation of the health state of the equipment, which makes the index output by the system able to reflect the average performance of the equipment in a period of time and its change trend, rather than being affected by only the latest measurement value, and provides a more comprehensive and stable state basis for decision making; the processing of the evaluation unit completely depends on the output of the weighted correction module, it does not simply use the latest corrected data, but assigns different evaluation weights (recent data has a high weight) to all recent corrected flue gas concentration data based on their time stamps (timeliness), and calculates a comprehensive operation efficiency index reflecting the recent comprehensive performance of the equipment, which makes the evaluation result not a momentary snapshot, but a weighted average health trend judgment.

[0016] The state evaluation and decision module comprises a decision unit, and the process of generating the graded early warning information by the decision unit specifically comprises: The corrected flue gas concentration data is compared with the graded emission standard threshold values to determine an emission level, the graded emission standard threshold values comprising a first threshold value, a second threshold value and a third threshold value; the comprehensive operation efficiency index is compared with efficiency threshold values to determine an efficiency level, the efficiency threshold values comprising a high efficiency threshold value and a low efficiency threshold value; and the graded early warning information is generated based on the emission level and the efficiency level, wherein a primary early warning is generated when the corrected flue gas concentration data exceeds the first threshold value but does not exceed the second threshold value and the comprehensive operation efficiency index is higher than the high efficiency threshold value, a middle early warning is generated when the corrected flue gas concentration data exceeds the second threshold value but does not exceed the third threshold value or the corrected flue gas concentration data exceeds the first threshold value and the comprehensive operation efficiency index is lower than the low efficiency threshold value, and a high early warning is generated when the corrected flue gas concentration data exceeds the third threshold value or the corrected flue gas concentration data exceeds the second threshold value and the comprehensive operation efficiency index is lower than the low efficiency threshold value.

[0017] The decision-making process of the decision-making unit is the final integration of the entire data flow; it will couple the final results (corrected oil smoke concentration data) output by the weighted correction module with the conclusions (comprehensive operation efficiency index) produced by the evaluation unit; the effect is to generate accurate and predictable hierarchical warning information; for example, when the corrected concentration data is slightly over-standard, the decision is not to mechanically trigger a low-level warning, but to combine the comprehensive operation efficiency index (reflecting the device health trend): if the index also shows that the device performance is low, it means that the slight over-standard may be a precursor to a serious failure, thereby triggering a higher level of warning. This reflects the deep association between end decision-making and all data processing links at the front end.

[0018] Compared with the prior art, the beneficial effects of the present application are: By constructing a ring-by-ring data processing chain, first, the original concentration data is adaptively weighted filtered according to real-time environmental parameters, effectively suppressing the noise introduced by environmental interference; then, the filtered data is further weighted corrected using the fluctuation pattern mined from historical data, further correcting abnormal values that do not conform to long-term rules, thereby outputting highly reliable purified concentration data; finally, the system performs multi-factor coupling intelligent decision-making on the operation efficiency index calculated from the double-optimized real-time concentration data and its time-decaying weight, generating accurate and predictable hierarchical warning; this not only greatly improves the anti-interference ability and reliability of the monitoring results, but also realizes the leap from passive alarm to active assessment of device health status, preventing trouble from happening, enhancing the intelligent level and management efficiency of oil smoke emission monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The figure is a schematic diagram of the overall module of the present application; Figure 2 The figure is a schematic diagram of the state evaluation and decision-making module unit of the present application; Figure 3 The figure is a schematic diagram of the core process of the hierarchical warning information of the present application.

[0020] In the figure: 100, data acquisition module; 200, weighted filtering module; 300, weighted correction module; 400, state evaluation and decision-making module; 401, evaluation unit; 402, decision-making unit. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0022] Next, please refer to Figure 1 The present application provides a technical solution: an oil fume purification device oil fume emission real-time monitoring system based on Internet of Things technology, comprising a data acquisition module 100, a weighted filtering module 200, a weighted correction module 300, and a state evaluation and decision module 400.

[0023] The data acquisition module 100 as the perception front end of the entire oil fume monitoring and purification system undertakes the basic acquisition task of multi-source and heterogeneous data. Its core function is to comprehensively, accurately and continuously collect three types of key data, i.e. oil fume concentration original data, environmental parameter data and historical oil fume data, which specifically include: Firstly, the oil fume concentration original data is directly collected through the oil fume concentration sensor deployed in the exhaust pipe. The oil fume concentration sensor is based on optical scattering (such as laser or infrared light source) to capture the instantaneous concentration value of particulate matter and non-methane total hydrocarbon gaseous pollutants in flue gas in real time, generating high-frequency original voltage or digital signals to provide the most fundamental data source for subsequent analysis; Secondly, environmental parameter data is collected synchronously, which includes but is not limited to flue gas temperature and humidity in the pipe obtained through temperature and humidity sensors, pipe static pressure or dynamic pressure obtained through pressure sensors, and exhaust volume or flow rate data obtained through anemometers; environmental parameter data will affect the accuracy of oil fume sensor readings (for example, high temperature and humidity environment may cause sensor reading drift) and the actual diffusion state of oil fume, and is the basis for subsequent dynamic correction by the weighted filtering module 200; Finally, historical oil fume data is called from the built-in time series database, which covers oil fume concentration trend records, device start and stop logs, and past concentration peaks and fluctuation patterns within a specific period (such as the past few hours, days or even months); the introduction of historical oil fume data enables the system to go beyond the judgment of the current instantaneous state, but to understand real-time data in a larger time sequence background, providing the possibility of identifying periodic emission patterns, abnormal events and device performance degradation trends; In summary, the data acquisition module 100 is not only a passive data receiving end, but also a hub for actively building a multi-dimensional data system. It integrates real-time, environmental and historical data to lay a solid and comprehensive information foundation for subsequent filtering, correction, evaluation and decision-making.

[0024] The weighted filtering module 200 dynamically allocates filtering weights to oil fume concentration original data based on the reference changes of environmental parameter data, which specifically includes: preset a normal reference variation interval for each parameter data (such as temperature and humidity) in the environmental parameter data; calculate the environmental deviation of the measured value of each parameter data in the environmental parameter data from the median value in the respective reference variation interval; the environmental deviation will be mapped to a basic weight influence factor corresponding to the parameter data as an input through a preset, continuous or segmented mathematical function; the mapping process of the mathematical function is as follows: When the measured value of each parameter data in the environmental parameter data is at the median value of the reference interval, the basic weight influence factor output is one, representing that the parameter is in the most ideal state, and the positive contribution to the data reliability is the highest; when the measured value of each parameter data in the environmental parameter data deviates from the median value of the reference interval and exceeds the boundary of the reference interval, the basic weight influence factor decreases to zero according to the function relationship, representing that the parameter is in a serious deviation state, and its data reliability contribution can be ignored.

[0025] The basic weight influence factors of each parameter data in the environmental parameter data are aggregated to calculate a value representing the comprehensive stability of the current environment, which is directly defined as the filtering weight of the oil smoke concentration original data point collected at the current time point; therefore, the assignment of the filtering weight is completely determined by the deviation of the environmental parameter relative to its reference, and the value is positively correlated with the environmental stability.

[0026] The weighted filtering module 200 performs weighted fusion on the oil smoke concentration original data based on the filtering weight to output the filtered oil smoke concentration data, specifically including: The product of each data point value in the oil smoke concentration original data and its corresponding filtering weight is calculated to obtain a set of weighted values; the weighted values are summed to obtain a weighted sum; at the same time, all filtering weights are summed to obtain a weight sum; the result of the weighted sum is divided by the result of the weight sum to obtain the filtered oil smoke concentration data, which is a linear combination of all data points in the oil smoke concentration original data, and the contribution of each data point is accurately controlled by its filtering weight; the data point with a larger weight value has a higher contribution ratio to the final result; the data point with a weight value close to zero can be ignored; through the above operation, abnormal fluctuations caused by environmental interference are effectively suppressed, and a filtered data that can more stably reflect the trend of oil smoke concentration is output.

[0027] The weighted correction module 300 dynamically assigns a correction weight to the filtered oil smoke concentration data by calling historical oil smoke data and based on the reference fluctuation mode, specifically including: Firstly, the historical oil fume data is called, and the baseline fluctuation mode of the oil fume concentration under different working conditions (such as different time periods and different equipment loads) is extracted through time series analysis, including the typical fluctuation range, change trend and periodic law; wherein the time series analysis refers to the specific process of quantitatively extracting the typical emission law of the historical oil fume data called from the time series database under a specific working condition through calculation and processing, which specifically includes: The analysis first divides the historical data into data sets reflecting different running states (such as data sets of meal peak period and standby state) according to the equipment start-stop log, time period information, etc.; then, each data set is analyzed to calculate its statistical characteristics, thereby quantitatively defining the baseline fluctuation mode under the working condition, and these characteristics include but are not limited to: the normal fluctuation range of the oil fume concentration (such as determined by the average value and the standard deviation), the typical change trend (such as determined by the fitting curve whether it is in a certain period of time is in an upward, downward or stable trend), and the inherent periodic law (such as identified by spectrum analysis whether there is a regular peak and valley alternation period in a certain time); finally, these quantified statistical characteristics jointly constitute a standard that can be used for real-time comparison, i.e. the baseline fluctuation mode.

[0028] When the new filtered oil fume concentration data is input, it will be compared with the baseline fluctuation mode under the corresponding working condition in real time, and the deviation degree is calculated; the deviation degree is mapped to a continuous value between zero and one through a pre-defined normalization function, and the value is the correction weight of the current data point; specifically including, If the fluctuation characteristics of the current filtered oil fume concentration data are consistent with the baseline fluctuation mode, the mode deviation is zero, and the correction weight is assigned as one; if the fluctuation characteristics of the current filtered oil fume concentration data are inconsistent with the baseline fluctuation mode, the mode deviation increases, and the correction weight decreases from one to zero according to the mapping function relationship; therefore, the correction weight is a specific scalar value obtained through deterministic mathematical calculation according to the degree of agreement between the current filtered oil fume concentration data and the baseline fluctuation mode, and the size is positively correlated with the historical agreement degree of the data.

[0029] The weighted correction module 300 performs weighted fusion on the filtered oil fume concentration data based on the calculated correction weight, and outputs the corrected oil fume concentration data, specifically including: Assuming that there are a series of filtered oil fume concentration data points to be corrected in a processing cycle, each data point has obtained a certain correction weight; the weighted fusion process has the same logic as the weighted fusion process in the weighted filtering module 200, and each data point in the filtered oil fume concentration data is multiplied by its corresponding correction weight to obtain a set of weighted values; the weighted values are arithmetically summed to obtain a weighted sum; at the same time, all correction weights are arithmetically summed to obtain a weight sum; the final corrected oil fume concentration data is determined by the quotient of the weighted sum divided by the weight sum.

[0030] Referring to Figure 2 , the evaluation unit 401 in the state evaluation and decision module 400 dynamically allocates evaluation weights based on the timeliness of the corrected oil fume concentration data to calculate the comprehensive operation efficiency index of the oil fume purifier, which specifically includes: The process of assigning evaluation weights to each corrected oil fume concentration data point is a calculation based on the time decay principle, which obtains the timestamp of each corrected oil fume concentration data point and calculates the time difference from the current time; the time difference is used as an input parameter and sent to a pre-set monotonically decreasing function for calculation; the characteristics of the monotonically decreasing function determine that its output value (i.e. the evaluation weight) strictly decreases with the increase of the input value (the time difference); therefore, the data points closer to the current time (i.e. the smaller the time difference), the evaluation weight calculated by the function is larger; on the contrary, the data points farther from the current time (i.e. the larger the time difference), the evaluation weight is smaller; through this mechanism, it is ensured that the final comprehensive operation efficiency index can more significantly reflect the recent device operation state; After completing the calculation of the evaluation weights of all data points, the calculation of the comprehensive operation efficiency index uses a weighted fusion algorithm, which multiplies the value of each corrected oil fume concentration data point by its corresponding evaluation weight to obtain the weighted value of each data point; the sum of all weighted values is obtained to obtain a weighted sum; at the same time, the sum of all evaluation weights is obtained to obtain a weight sum; the comprehensive operation efficiency index is finally determined by the quotient of the weighted sum divided by the weight sum; the index is a normalized value between zero and one, and the closer the value is to one, the higher the comprehensive operation efficiency of the oil fume purifier in the near future.

[0031] The decision unit 402 in the state evaluation and decision module 400 compares the corrected oil fume concentration data with the pre-set emission standard threshold and generates corresponding graded warning information in combination with the comprehensive operation efficiency index, which specifically includes: First, compare the latest corrected oil smoke concentration data with a set of predefined emission standard threshold values (e.g., first level threshold, second level threshold, third level threshold) to determine the grade to which the instantaneous emission level belongs, i.e., the emission grade; at the same time, compare the calculated comprehensive operation efficiency index with another set of predefined efficiency threshold values (e.g., high efficiency threshold, low efficiency threshold) to determine the current operation efficiency state of the equipment, i.e., the efficiency grade; The final hierarchical early warning information is determined by the emission grade and the efficiency grade, and the generation rule is as follows: When the corrected oil smoke concentration data exceeds the first level threshold but does not exceed the second level threshold, and the comprehensive operation efficiency index is higher than the high efficiency threshold, the system generates a primary early warning; when the corrected oil smoke concentration data exceeds the second level threshold but does not exceed the third level threshold, or the corrected oil smoke concentration data exceeds the first level threshold and the comprehensive operation efficiency index is lower than the low efficiency threshold, the system generates a middle-level early warning; when the corrected oil smoke concentration data exceeds the third level threshold, or the corrected oil smoke concentration data exceeds the second level threshold and the comprehensive operation efficiency index is lower than the low efficiency threshold, the system generates a high-level early warning; Please refer to Figure 3 , the core process of the hierarchical early warning information is as follows: Start with the input corrected oil smoke concentration data C and the comprehensive operation efficiency index E; these data come from the front-end weighted correction module and the evaluation unit, ensuring high reliability of the input; First, check if C exceeds the third level threshold T3; if so, directly trigger a high-level early warning, as this indicates a serious over-standard; If not exceeding T3, check if C exceeds the second level threshold T2. If exceeding T2, judge in combination with E: if E is lower than the low efficiency threshold L, indicating that the equipment efficiency is low, trigger a high-level early warning; otherwise, trigger a middle-level early warning; If not exceeding T2, check if C exceeds the first level threshold T1; if exceeding T1, judge in combination with E: if E is higher than the high efficiency threshold H, trigger a primary early warning; otherwise, trigger a middle-level early warning (as the equipment efficiency is low, the risk is amplified); If C does not exceed T1, output no early warning or normal state; All paths eventually converge to the early warning information output, ensuring a closed loop process.

[0032] The threshold values are defined as follows: Emission threshold values: first level threshold T1, second level threshold T2, and third level threshold T3, satisfying T1 < T2 < T3.

[0033] Efficiency threshold values: high efficiency threshold H and low efficiency threshold L, satisfying H > L.

[0034] This linkage judgment mechanism ensures that the early warning information can reflect the severity of emission over-standard and also reflect the operation health status of the equipment itself, thereby providing accurate decision support.

[0035] The foregoing merely illustrates the principles of the application and application of its more prominent features. Those skilled in the art will appreciate that the application is not limited to the embodiments described above, but rather that various changes and modifications can be made thereto without departing from the spirit and scope of the application. The scope of the application is defined by the appended claims and their equivalents.

Claims

1. The oil fume emission real-time monitoring system of the oil fume purifier based on the Internet of Things technology, characterized in that, It comprises a data acquisition module (100), a weighted filtering module (200), a weighted correction module (300) and a state evaluation and decision module (400), wherein: The data acquisition module (100) is used for collecting original data of oil fume concentration, environmental parameter data and historical oil fume data. The weighted filtering module (200) dynamically allocates filtering weights to the original data of oil fume concentration based on the reference change of the environmental parameter data, and performs weighted fusion to output filtered oil fume concentration data. The weighted correction module (300) dynamically allocates correction weights to the filtered oil fume concentration data based on the reference fluctuation mode by calling the historical oil fume data, and performs weighted fusion to output corrected oil fume concentration data. The state evaluation and decision module (400) dynamically allocates evaluation weights based on the timeliness of the corrected oil fume concentration data to calculate the comprehensive operation efficiency index of the oil fume purifier; compares the corrected oil fume concentration data with the preset emission standard threshold, and generates corresponding graded early warning information in combination with the comprehensive operation efficiency index. 2.The oil fume emission real-time monitoring system of the oil fume purifier based on the Internet of Things technology according to claim 1, characterized in that, The original data of oil fume concentration in the data acquisition module (100) is collected by an oil fume concentration sensor; the environmental parameter data includes temperature, humidity, pipeline pressure and exhaust flow rate; the historical oil fume data is obtained from a time series database. 3.The oil fume emission real-time monitoring system of the oil fume purifier based on the Internet of Things technology according to claim 1, characterized in that, The generation process of the filtering weight specifically includes: A reference change interval is preset for each parameter data in the environmental parameter data; the environmental deviation of the measured value of each parameter data from the median value in the reference change interval is calculated; the environmental deviation is mapped to a basic weight influence factor by a mathematical function; and the basic weight influence factors of all parameter data are aggregated to obtain the filtering weight. 4.The oil fume emission real-time monitoring system of the oil fume purifier based on the Internet of Things technology according to claim 3, characterized in that, The mapping process of the mathematical function specifically includes: When the measured value of the parameter data is at the median value of the reference change interval, the basic weight influence factor output is one; when the measured value of the parameter data deviates from the median value of the reference change interval and exceeds the boundary of the reference change interval, the basic weight influence factor decreases to zero according to the function relationship. 5.The oil fume emission real-time monitoring system of the oil fume purifier based on the Internet of Things technology according to claim 1, characterized in that, The calculation of the filtered oil fume concentration data specifically includes: The product of each data point value in the original data of oil fume concentration and the corresponding filtering weight is calculated to obtain a weighted value; the weighted values are summed to obtain a weighted sum; the filtering weights are summed to obtain a weight sum; and the filtered oil fume concentration data is obtained by dividing the weighted sum by the weight sum. 6.The oil fume emission real-time monitoring system of the oil fume purifier based on the Internet of Things technology according to claim 1, characterized in that, The generation process of the correction weight specifically includes: The historical oil fume data is called to extract the reference fluctuation mode by time series analysis; the mode deviation of the filtered oil fume concentration data from the reference fluctuation mode is calculated; and the mode deviation is mapped to a correction weight between zero and one by a normalization function, wherein the correction weight is one when the mode deviation is zero, and the correction weight decreases from one to zero as the mode deviation increases. 7.The oil fume emission real-time monitoring system based on the Internet of Things according to claim 1, characterized in that, The calculation process of the corrected oil fume concentration data specifically includes: The product of each data point value in the filtered oil fume concentration data and the corresponding correction weight is calculated to obtain a weighted value, the weighted values are summed to obtain a weighted sum, the correction weights are summed to obtain a weight sum, and the weighted sum is divided by the weight sum to obtain corrected oil fume concentration data. 8.The oil fume emission real-time monitoring system based on the Internet of Things according to claim 1, characterized in that, The state evaluation and decision module (400) comprises an evaluation unit (401), and the process of calculating the evaluation weight specifically comprises: The evaluation weight of each corrected oil fume concentration data point is determined by the time difference between the time stamp and the current time, and is calculated by a preset monotonically decreasing function according to the time decay principle. 9.The oil fume emission real-time monitoring system based on the Internet of Things according to claim 1, characterized in that, The calculation process of the comprehensive operation efficiency index comprises: Each corrected oil fume concentration data point value is multiplied by the evaluation weight to obtain a weighted value, the weighted values are summed to obtain a weighted sum, the evaluation weights are summed to obtain a weight sum, and the weighted sum is divided by the weight sum to obtain the comprehensive operation efficiency index, which is a normalized value between zero and one. 10.The oil fume emission real-time monitoring system based on the Internet of Things according to claim 1, characterized in that, The state evaluation and decision module (400) comprises a decision unit (402), and the process of generating the hierarchical early warning information specifically comprises: The corrected oil fume concentration data is compared with the hierarchical emission standard threshold to determine the emission level, the hierarchical emission standard threshold comprises a first threshold, a second threshold and a third threshold, the comprehensive operation efficiency index is compared with the efficiency threshold to determine the efficiency level, the efficiency threshold comprises a high efficiency threshold and a low efficiency threshold, and the hierarchical early warning information is generated based on the emission level and the efficiency level, wherein the primary early warning is generated when the corrected oil fume concentration data exceeds the first threshold but does not exceed the second threshold and the comprehensive operation efficiency index is higher than the high efficiency threshold, the intermediate early warning is generated when the corrected oil fume concentration data exceeds the second threshold but does not exceed the third threshold or the corrected oil fume concentration data exceeds the first threshold and the comprehensive operation efficiency index is lower than the low efficiency threshold, and the high-level early warning is generated when the corrected oil fume concentration data exceeds the third threshold or the corrected oil fume concentration data exceeds the second threshold and the comprehensive operation efficiency index is lower than the low efficiency threshold.

Images