Pollution source monitoring method and system based on multi-source data acquisition and analysis

Abstract

The invention relates to the technical field of pollution monitoring, in particular to a pollution source monitoring method and system based on multi-source data acquisition and analysis. The method comprises the following steps: firstly, analyzing and acquiring fluctuation characteristic parameters of each pollution index, and acquiring a working condition correlation coefficient of each pollution index in combination with historical time sequence change correlation between emission data under each pollution index and working condition data under each working condition index; and then, in combination with short-term change characteristics of the emission data under the pollution indexes and the working condition data under the working condition indexes at the current monitoring moment, evaluating an abnormal state coefficient under each pollution index so as to carry out abnormal emission early warning. According to the method, the response sensitivity of the pollution index to the working condition change is quantified, the working condition association condition of the pollution index is further quantified, the working condition association condition of the pollution index at the current monitoring moment is further analyzed, the abnormal emission behavior of the pollution source which does not conform to the working condition is accurately evaluated, and the monitoring accuracy of the pollution source is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pollution monitoring, in particular to a pollution source monitoring method and system based on multi-source data collection and analysis. BACKGROUND

[0002] Precise monitoring of pollution sources and emission early warning management have become an important goal of environmental supervision. With the development of technologies such as the Internet of Things and big data, pollution source monitoring is gradually evolving towards automation, intelligence and networking. Through the collection and fusion analysis of multi-source data, various types of monitoring data can be supplemented and verified, thereby reducing the errors and uncertainties that may exist in single monitoring data and improving the accuracy and reliability of pollution source monitoring.

[0003] Traditional monitoring mainly focuses on whether the concentration or content of pollutants at the pollution source discharge port exceeds the standard, without considering the dynamic correlation between the working condition parameters and pollutant generation in the actual production process. For example, when monitoring the discharge of a factory boiler, the changes of working condition parameters such as boiler load, coal quality and treatment facility operation state are not considered to affect the emission of pollutants. When the production conditions change, it is difficult to accurately determine whether the change of pollutant concentration is normal transient fluctuation or abnormal emission, which may easily lead to false positives or false negatives, thereby affecting the monitoring effect of pollution sources. SUMMARY

[0004] In order to solve the problem of low accuracy of pollution source monitoring, the purpose of the present application is to provide a pollution source monitoring method and system based on multi-source data collection and analysis, and the technical solution is as follows: The pollution source monitoring method based on multi-source data collection and analysis comprises: Obtaining multi-source monitoring data at each monitoring time in a historical monitoring period, the multi-source monitoring data at least including emission data of each pollution index and working condition data of each working condition index at the pollution source discharge port; In the historical monitoring period, obtaining the fluctuation characteristic parameters of each pollution index according to the time sequence fluctuation change of the emission data of each pollution index, and obtaining the working condition correlation coefficient of each pollution index by combining the time sequence change correlation between the emission data of each pollution index and the working condition data of each working condition index; At the current monitoring time, obtaining the current working condition correlation coefficient of each pollution index in a preset historical period at the current monitoring time according to the change characteristics of the emission data of each pollution index and the working condition data of each working condition index, obtaining the abnormal state coefficient of each pollution index according to the difference between the current working condition correlation coefficient and the working condition correlation coefficient of each pollution index, and performing abnormal emission early warning based on the abnormal state coefficient.

[0005] Further, the method for obtaining the fluctuation characteristic parameter comprises: dividing the historical monitoring period to obtain all time windows; in each time window, obtaining the fluctuation parameter of each pollution index according to the fluctuation change of the emission data under each pollution index; obtaining the stability characteristic parameter of each pollution index in the historical monitoring period according to the discrete characteristics of the fluctuation parameters of each pollution index in different time windows; obtaining the fluctuation characteristic parameter of each pollution index according to the fluctuation parameter and the stability characteristic parameter in each time window.

[0006] Further, the method for obtaining the time window comprises: dividing the historical monitoring period into all time windows with a preset length.

[0007] Further, the method for obtaining the working condition correlation coefficient comprises: in each time window, obtaining the collaborative change sub-parameter between each pollution index and each working condition index according to the change correlation between the emission data under each pollution index and the working condition data under each working condition index; obtaining the collaborative change parameter between each pollution index and each working condition index by comprehensively considering the differences between the collaborative change sub-parameters in all time windows in the historical monitoring period; obtaining the working condition correlation coefficient of each pollution index according to the collaborative change parameter between each pollution index and each working condition index and the fluctuation characteristic parameter of each pollution index.

[0008] Further, the method for obtaining the collaborative change sub-parameter comprises: at each monitoring time in each time window, determining the correlation characteristic value between each pollution index and each working condition index according to the emission data under each pollution index and the working condition data under each working condition index; obtaining the collaborative change sub-parameter between each pollution index and each working condition index by comprehensively considering the correlation characteristic values at adjacent monitoring times in the time window.

[0009] Further, obtaining the abnormal state coefficient under each pollution index comprises: under each pollution index, taking the negative correlation normalization result of the ratio between the current working condition correlation coefficient and the working condition correlation coefficient as the abnormal state coefficient.

[0010] Further, the abnormal emission early warning based on the abnormal state coefficient comprises: when the abnormal state coefficient under any pollution index is greater than a preset threshold, performing abnormal emission early warning.

[0011] Further, the acquisition method of the fluctuation parameter comprises: For each pollution index, in each time window, an emission reference value is acquired according to the centralized features of the emission data at all monitoring moments, and a fluctuation parameter is acquired based on the deviation of the emission data at each monitoring moment relative to the emission reference value.

[0012] The pollution source monitoring system based on multi-source data acquisition and analysis comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the pollution source monitoring method based on multi-source data acquisition and analysis when executing the computer program.

[0013] The present application has the following advantages: The present application firstly acquires the emission data of each pollution index and the working condition data of each working condition index at each monitoring moment in a historical monitoring period; then in the historical monitoring period, the fluctuation characteristic parameter reflecting the sensitive response of each pollution index to working condition change is preliminarily acquired according to the time sequence fluctuation change of the emission data of each pollution index, and the working condition correlation coefficient of each pollution index is acquired in combination with the time sequence change correlation between the emission data of each pollution index and the working condition data of each working condition index, which quantifies the influence degree of each pollution index on the working condition index change and prepares for subsequent abnormal emission early warning; further, at the current monitoring moment, the working condition correlation of each pollution index at the current monitoring moment is preliminarily evaluated according to the change features of the emission data of each pollution index and the working condition data of each working condition index, and the abnormal state coefficient of each pollution index is accurately evaluated and acquired in combination with the historical reference benchmark provided by the working condition correlation coefficient of each pollution index in the historical monitoring period; finally, the abnormal emission early warning is performed based on the abnormal state coefficient. The present application firstly analyzes the fluctuation change of the emission data in the historical period, quantifies the response sensitivity of the pollution index to the working condition change, further evaluates the working condition correlation of the pollution index in combination with the time sequence change correlation between the pollution index and the working condition index, analyzes the working condition correlation of the pollution index at the current monitoring moment, evaluates the abnormal emission behavior of the pollution source not meeting the working condition, and thus improves the accuracy of the pollution source monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0015] Figure 1A flow chart of a pollution source monitoring method based on multi-source data collection and analysis is provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined object of the application, the specific embodiments, structure, features and effects of the pollution source monitoring method and system based on multi-source data collection and analysis according to the present application are described in detail as follows in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0018] The specific scheme of the pollution source monitoring method and system based on multi-source data collection and analysis provided by the present application is specifically described below in combination with the accompanying drawings.

[0019] Please refer to Figure 1 which shows a flow chart of a pollution source monitoring method based on multi-source data collection and analysis provided in an embodiment of the present application, specifically including: Step S1, obtaining multi-source monitoring data at each monitoring time in a historical monitoring period, the multi-source monitoring data at least including emission data of each pollution index and working condition data of each working condition index at the emission port of the pollution source.

[0020] It should be noted that the purpose of the embodiment of the present application is to distinguish the normal production fluctuation caused by working condition adjustment and the real abnormal emission behavior, so as to perform abnormal early warning, avoid the interference of working condition factors such as boiler load adjustment and coal quality fluctuation, and make the monitoring result reliable. The abnormal emission behavior refers to the abnormal behavior that the emission does not change with the working condition, that is, even if the emission data does not exceed the pollution threshold, there is a high possibility of abnormal conditions such as equipment failure, illegal discharge or decreased efficiency of treatment facilities.

[0021] The embodiment of the present application takes the flue gas of the coal-fired power plant boiler as the pollution source for example analysis and description; in other embodiments, the implementer can also monitor and analyze the industrial wastewater as the pollution source.

[0022] In one embodiment of the present application, the real-time emission behavior is monitored by using the fixed continuous emission monitoring system (CEMS) installed at the emission port of the boiler flue gas (after treatment by the flue gas treatment equipment), and the operation state of the flue gas treatment equipment and the production state of the boiler are monitored in real time by using the industrial control system (DCS) of the coal-fired power plant, so as to obtain multi-source monitoring data at each monitoring time within a preset historical monitoring period.

[0023] The historical monitoring period refers to the historical continuous operation period of the current monitoring time, and in this embodiment, the historical 24 hours of continuous operation are taken as an example, and the implementer can also adjust it according to the specific operation situation; the multi-source monitoring data at least includes emission data of each pollution index at the emission port of the pollution source and working condition data of each working condition index; the emission data reflects the emission state, and the working condition data reflects the operation state of the flue gas treatment equipment and the production state of the boiler.

[0024] Among them, the pollution index at least includes sulfur dioxide concentration, nitrogen oxide concentration, particulate matter concentration, flue gas temperature, flue gas pressure, flue gas flow rate, flue gas humidity, flue gas oxygen content, etc.; the working condition index at least includes the operation load of the flue gas treatment equipment, the fan current, the desulfurization tower pH value, the circulating pump current, the ammonia injection flow rate, and the production load of the boiler; the above index parameters are monitored and collected by the existing industrial control system, and the specific acquisition process will not be described again; in other embodiments, the implementer can also adjust it according to the actual situation, such as increasing or decreasing the pollution index or the working condition index.

[0025] It should be noted that the monitoring frequency of each pollution index and each working condition index is the same and is monitored synchronously, and in this embodiment, it is set to 1Hz; in other embodiments, the implementer can also adjust the monitoring frequency according to the actual situation, but the index data under different monitoring frequencies need to be resampled to ensure time sequence alignment, so as to facilitate subsequent analysis.

[0026] After a certain data cleaning and standardization processing of the data under each (pollution or working condition) index, such as removing duplicate values or filling missing values, the data under each (pollution index or working condition index) index is dimensionless and mapped to a preset range such as 0-10; the data cleaning and standardization processing are all known technical means, and will not be described again.

[0027] Step S2, within the historical monitoring period, according to the time sequence fluctuation change of the emission data under each pollution index, the fluctuation characteristic parameters of each pollution index are obtained, and the time sequence change correlation between the emission data under each pollution index and the working condition data under each working condition index is obtained. The working condition correlation coefficient of each pollution index is obtained.

[0028] During coal-fired power generation, emission data under various pollution indicators are not static but fluctuate with changes in normal operating conditions. Furthermore, the fluctuations in emission data under different pollution indicators are not entirely the same. For example, under the same operating conditions, some pollution indicators respond significantly to changes in operating conditions and can provide effective signals of changes in pollution emission status; while the responses of other pollution indicators to changes in operating conditions may have a certain lag or coupling relationship, and their values ​​change slowly or are unpredictable due to random noise interference, providing weak and unstable signals of changes in pollution emission status.

[0029] Considering that the more drastic or significant the fluctuations in emission data under pollution indicators, and the more regular or repeatable the fluctuation pattern in time series, the more sensitive the data is to changes in operating conditions, and the higher its reference value for assessing subsequent abnormal emissions; on the other hand, the weaker the fluctuations in emission data under pollution indicators, or the more random and irregular the fluctuations in time series, the more sluggish the data is to changes in operating conditions or the presence of noise interference, and the lower its reference value for assessing subsequent abnormal emissions. Based on this, the embodiments of the present invention will first obtain the fluctuation characteristic parameters of each pollution index according to the temporal fluctuation changes of emission data under each pollution index during the historical monitoring period. The fluctuation characteristic parameters characterize the sensitivity level of the pollution index to changes in operating conditions during the historical monitoring period. The larger the fluctuation characteristic parameters, the more sensitive the corresponding pollution index is to changes in operating conditions, the higher the reference value for the assessment of subsequent abnormal emissions, and the greater the reference value for the analysis and correlation between subsequent pollution indexes and operating condition indicators.

[0030] Preferably, in one embodiment of the present invention, considering that long-term time-series analysis may smooth out the fluctuation characteristics of emission data, thereby affecting subsequent analysis of fluctuation information, the historical monitoring period is first divided to obtain all time windows. Then, the local fluctuation characteristics of emission data are analyzed within each time window to help assess its sensitivity or responsiveness to changes in operating conditions. Furthermore, the repeatability or consistency of the local fluctuation characteristics of emission data within different time windows is analyzed to obtain the fluctuation characteristic parameters for each pollution indicator. Therefore, the method for obtaining the fluctuation characteristic parameters includes: Divide historical monitoring periods and obtain all time windows; within each time window, obtain the fluctuation parameters of each pollution indicator based on the fluctuation changes of emission data under each pollution indicator. Based on the discrete characteristics of the fluctuation parameters of each pollution index within different time windows, the stable characteristic parameters of each pollution index within the historical monitoring period are obtained. Based on the fluctuation parameters and stable characteristic parameters within each time window, the fluctuation characteristic parameters of each pollution index are obtained.

[0031] It should be noted that the analysis method for the fluctuation characteristic parameters of each pollution indicator is the same. Here, we will only take one pollution indicator as an example for analysis and description, and will not go into detail about each one.

[0032] In a preferred embodiment of the present invention, the method for obtaining the time window includes: uniformly dividing the historical monitoring period into a preset length to determine all time windows.

[0033] The preset length is at least 10, which is set to 10 in this embodiment. The implementer can also adjust it according to their own needs, for example, based on the stage of change in working conditions. Starting from the beginning of the historical monitoring period, every 10 monitoring moments are a time window, and the division is carried out sequentially until the entire historical monitoring period is covered, ensuring that the length of each time window is consistent and there is no overlap. This division method is conducive to reducing the interference caused by the boundary effect and improving the stability and comparability of fluctuation feature extraction.

[0034] Furthermore, within each time window, fluctuation parameters are calculated based on the fluctuations in emission data under the pollution indicators. These fluctuation parameters reflect the fluctuations in emission data and may be the instantaneous sensitive response of pollution indicators to changes in operating conditions within a local time window, or they may be occasional instantaneous values ​​such as noise. This prepares the parameters for the subsequent comprehensive evaluation of the fluctuation characteristics of pollution indicators.

[0035] In a preferred embodiment of the present invention, considering that the concentrated characteristics of emission data under pollution indicators within a time window, such as the mean, can characterize average emission behavior, an emission reference value reflecting a continuous and stable emission level can first be obtained; then, based on the deviation of the emission data relative to the emission reference value at each monitoring time, a fluctuation parameter reflecting the instantaneous fluctuation characteristics within the time window is obtained; therefore, the method for obtaining the fluctuation parameter includes: For each pollution indicator, within each time window, an emission reference value is obtained based on the concentrated characteristics of emission data at all monitoring times, and a fluctuation parameter is obtained based on the deviation of emission data at each monitoring time from the emission reference value.

[0036] Specifically, within each time window, the mean of emission data at all monitoring times is used as the emission reference value. The sum of the absolute values ​​of the differences between the emission data at each monitoring time and the emission reference value is normalized, for example, by multiplying by a preset scaling factor of 0.1 and then mapping it to the sigmoid function to prevent the output from saturating (approaching 1) due to excessively large input values ​​and losing discriminability. The normalized mapping result is used as the fluctuation parameter.

[0037] In other embodiments, the implementer may also use the normalized result of the fitting error of the least squares linear fitting of the emission data within the time window as the fluctuation parameter, or other normalization methods such as linear normalization may be used. These are all well-known techniques and will not be described in detail here.

[0038] Furthermore, based on the discrete characteristics of the fluctuation parameters of each pollution indicator within different time windows, stable characteristic parameters of each pollution indicator within historical monitoring periods are obtained. Stable characteristic parameters reflect the repeatability or regularity of the fluctuation changes of pollution indicators within different time windows. The higher the repeatability, the more consistent the instantaneous sensitivity response of pollution indicators to changes in operating conditions within different local time windows is, and the greater the possibility of reflecting the true fluctuation of emission data. Conversely, the lower the repeatability, the greater the possibility of occasional instantaneous values ​​such as noise, thus preparing for subsequent comprehensive evaluation of the fluctuation characteristic parameters of pollution indicators.

[0039] Specifically, the absolute value of the difference between the fluctuation parameter and the minimum fluctuation parameter of the pollution index in each time window is calculated. The absolute value of the difference is averaged and then negatively correlated and normalized. For example, the mean of the absolute values ​​of the difference is added to a constant 1 and then the reciprocal is used for negative correlation normalization to obtain stable characteristic parameters.

[0040] In other embodiments, implementers may also use other negative correlation normalization methods, such as mapping to an exponential function exp(-x) with the natural constant e as the base; or they may use the maximum fluctuation parameter, the mean of the fluctuation parameters, etc., instead of the minimum fluctuation parameter.

[0041] Then, based on the fluctuation parameters and stable characteristic parameters within each time window, the fluctuation characteristic parameters of each pollution index are obtained. Specifically, the fluctuation parameters within all time windows are averaged, and then the average fluctuation parameter is multiplied and fused with the stable characteristic parameter. The stable characteristic parameter integrates the repetitive patterns of the fluctuation characteristics of emission data within all time windows, providing a confidence reference for the long-term fluctuation of emission data within historical monitoring periods. The average fluctuation parameter can provide a relevant reference for the instantaneous sensitive response of pollution index to changes in operating conditions within a preset historical monitoring period. Therefore, multiplying and fusing the two yields the fluctuation characteristic parameters of the pollution index under the pollution index.

[0042] After obtaining the fluctuation characteristic parameters of each pollution indicator, this embodiment of the invention further analyzes the temporal change correlation between emission data under each pollution indicator and operating data under each operating condition indicator, and obtains the operating condition correlation coefficient of each pollution indicator. The operating condition correlation coefficient, combined with the fluctuation characteristic parameters that reflect the sensitivity of the pollution indicator to changes in operating conditions, analyzes the actual temporal change correlation between the pollution indicator and the operating condition indicator, quantifies the degree of influence of each pollution indicator on changes in operating conditions, and prepares for subsequent abnormal emission early warning.

[0043] Preferably, in one embodiment of the present invention, considering the correlation between emission data under each pollution index and operating condition data under each operating condition index, comprehensively assessing the synergistic changes of the two across all time windows can help evaluate their coordinated changes; further, the overall coordinated changes of pollution indices with operating conditions can be evaluated by comprehensively considering the synergistic change parameters between each pollution index and all operating condition indices, while combining the fluctuation characteristic parameters reflecting the sensitivity of pollution indices to changes in operating conditions, the degree of influence of changes in operating condition indices on pollution indices is comprehensively evaluated, and the operating condition correlation coefficient is determined; the method for obtaining the operating condition correlation coefficient includes: Within each time window, based on the correlation between the changes in emission data under each pollution indicator and the operating condition data under each operating condition indicator, the co-change sub-parameters between each pollution indicator and each operating condition indicator are obtained. By integrating the co-variation sub-parameters across all time windows during the historical monitoring period, the co-variation parameters between each pollution index and each operating condition index are obtained. Based on the co-variation parameters between each pollution index and each operating condition index, as well as the fluctuation characteristic parameters of each pollution index, the operating condition correlation coefficient of each pollution index is obtained.

[0044] In a preferred embodiment of the present invention, the method for obtaining the cooperative variation sub-parameter includes: At each monitoring moment within each time window, based on the emission data for each pollution indicator and the operating data for each operating condition indicator, the correlation characteristic value between each pollution indicator and each operating condition indicator is determined; By considering the differences between associated feature values ​​at adjacent monitoring times within a comprehensive time window, we can obtain the co-variation sub-parameters between each pollution index and each operating condition index.

[0045] As an example, we will analyze and describe the situation using any pollution index and any operating condition index. Specifically, at each monitoring moment within each time window, the emission data under the pollution index is used as the numerator, the operating condition data under the operating condition index is used as the denominator, and the ratio of the fractions is used as the correlation characteristic value between the pollution index and the operating condition index. To avoid the denominator being 0, a very small positive parameter, such as 0.01, can be added to the operating condition data before using it as the denominator. The correlation characteristic value represents the instantaneous proportional relationship between the pollution index and the operating condition index, preparing for subsequent measurement of changes and correlation. Then, the absolute value of the difference between the associated characteristic values ​​at adjacent monitoring times within the time window is calculated. The smaller the absolute value of the difference, the more stable the instantaneous proportional relationship between the pollution index and the operating condition index, and the greater the possibility of coordinated change between the pollution index and the operating condition index. Therefore, the absolute values ​​of the difference between the associated characteristic values ​​at all adjacent monitoring times are averaged and negatively correlated and normalized. For example, the average absolute value of the difference is added to a constant 1 and then the reciprocal is used for negative correlation normalization to obtain the co-change sub-parameter.

[0046] In other embodiments, implementers may also employ other negative correlation normalization methods, such as mapping to an exponential function exp(-x) with the natural constant e as the base.

[0047] Further, by comprehensively analyzing the co-variation sub-parameters within all time windows during the historical monitoring period, the co-variation parameters between each pollution index and each operating condition index are obtained. Specifically, the absolute value of the difference between the co-variation sub-parameter within each time window and the mean of the co-variation sub-parameters across all time windows is calculated. The smaller the absolute value of the difference, the more consistent the co-variation sub-parameters are across different time windows. Then, the absolute values ​​of the differences corresponding to all time windows are summed and negatively correlated and normalized. For example, the sum of the absolute values ​​of the differences is increased by a constant 1 and then the reciprocal is used for negative correlation normalization to obtain the co-variation parameters.

[0048] Then, the co-variance parameter between each pollution index and each operating condition index is multiplied by the fluctuation characteristic parameter of each pollution index, and the product is used as the operating condition correlation coefficient of each pollution index.

[0049] Step S3: At the current monitoring time, based on the change characteristics of emission data under each pollution indicator and operating data under each operating condition indicator, obtain the current operating condition correlation coefficient of each pollution indicator within the preset historical period at the current monitoring time; based on the difference between the current operating condition correlation coefficient and the operating condition correlation coefficient under each pollution indicator, obtain the abnormal state coefficient under each pollution indicator; and conduct abnormal emission early warning based on the abnormal state coefficient.

[0050] Considering that the operating condition correlation coefficient reflects the sensitivity of pollution indicators to changes in operating conditions during historical monitoring periods, it can provide a certain historical benchmark. If the sensitivity of pollution indicators to changes in operating conditions at the current monitoring time is closer to the historical benchmark, it indicates that the operating condition correlation at the current monitoring time has not been disrupted, and the pollution indicators change with the operating conditions, which is consistent with normal emissions. Conversely, it indicates that the operating condition correlation has been disrupted, for example, the operating conditions have not changed but the emission data fluctuates drastically, and the possibility of abnormal emissions is greater. Based on this, in a preset historical period within the current monitoring time, this embodiment of the invention preliminarily assesses the correlation between the operating conditions of each pollution indicator and the emission data and operating condition data of each operating condition indicator within the current monitoring time, and obtains the current operating condition correlation coefficient. Furthermore, by combining the operating condition correlation coefficients of each pollution indicator within the historical monitoring period, an abnormal state coefficient is obtained for each pollution indicator. The abnormal state coefficient reflects the change in the operating condition correlation of the corresponding pollution indicator, and further reflects the possibility of abnormal emissions provided under the corresponding pollution indicator, providing a basis for subsequent abnormal emission early warning.

[0051] In one embodiment of the present invention, the current operating condition correlation coefficient of each pollution indicator can be calculated first based on the analysis steps of the operating condition correlation coefficient of each pollution indicator during the historical monitoring period; then the difference between the current operating condition correlation coefficient and the operating condition correlation coefficient is compared to evaluate the consistency of the operating condition correlation, and then the abnormal state coefficient under the pollution indicator is obtained.

[0052] As an example, taking any pollution indicator as an example, firstly, within a preset historical time period of the current monitoring time, obtain the current operating condition correlation coefficient of the pollution indicator; wherein, the preset historical time period is 10-15 minutes of the history of the current monitoring time, and 10 minutes is taken in this embodiment, but the implementer can also adjust it himself; divide the preset historical time period into several time windows for analysis and evaluation. The analysis method of the current operating condition correlation coefficient is the same as that of the operating condition correlation coefficient in the historical monitoring time period, but the analysis and calculation are performed within the preset historical time period of the current monitoring time, and the acquisition process will not be described again.

[0053] In a preferred embodiment of the present invention, under each pollution index, the negative correlation normalization result of the ratio between the current working condition correlation coefficient and the working condition correlation coefficient is used as the abnormal state coefficient; the current working condition correlation coefficient is used as the numerator, the working condition correlation coefficient is used as the denominator, and the ratio of the fraction is negatively normalized, for example, mapped to the exponential function exp(-x) with the natural constant e as the base, and the negative correlation normalization mapping result is used as the abnormal state coefficient.

[0054] When the correlation coefficient of the current operating condition is greater than or equal to the correlation coefficient of the operating condition, the ratio is greater than or equal to 1, indicating that the change of the pollution index with the operating condition in the preset historical period at the current monitoring time is greater than the historical benchmark, the probability of normal emission of the pollution source under the pollution index is relatively higher, and the abnormal state coefficient is smaller; conversely, if the ratio is less than 1, it indicates that the correlation between the pollution index and the operating condition in the preset historical period at the current monitoring time is not strong, there may be abnormal emission behavior, and the abnormal state coefficient is larger.

[0055] Once the abnormal state coefficient for each pollution indicator is determined, abnormal emission warnings can be issued based on the abnormal state coefficient.

[0056] Preferably, in one embodiment of the present invention, considering that when any pollution indicator has a high probability of abnormal emission, the pollution source has the possibility of abnormal emission and it is necessary to issue an abnormal emission warning; therefore, issuing an abnormal emission warning based on the abnormal state coefficient includes: issuing an abnormal emission warning when the abnormal state coefficient of any pollution indicator is greater than a preset threshold.

[0057] The preset threshold is set to 0.7. The preset threshold is an example value calculated under a typical working condition (the equipment is in a healthy / normal operating state, such as 24 hours after the system has just been overhauled). In actual applications, it can be adjusted according to the tolerance for false alarm rate. When the abnormal state coefficient under any pollution indicator is greater than 0.7, an abnormal emission warning is issued. When the abnormal state coefficient under all pollution indicators is less than or equal to 0.7, monitoring continues.

[0058] Based on the same inventive concept, one embodiment of the present invention also proposes a pollution source monitoring system based on multi-source data acquisition and analysis. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the pollution source monitoring method based on multi-source data acquisition and analysis described in steps S1-S3 above.

[0059] In summary, this invention first acquires multi-source monitoring data at each monitoring moment within a historical monitoring period. Then, based on the temporal fluctuations of emission data for each pollution indicator, it obtains the fluctuation characteristic parameters of each pollution indicator. Furthermore, by combining the temporal correlation between emission data for each pollution indicator and operating condition data for each operating condition indicator, it obtains the operating condition correlation coefficient for each pollution indicator. At the current monitoring moment, based on the change characteristics of emission data for each pollution indicator and operating condition data for each operating condition indicator, it obtains the current operating condition correlation coefficient for each pollution indicator within a preset historical period at the current monitoring moment. Further, by comparing the current operating condition correlation coefficient with the operating condition correlation coefficient for each pollution indicator, it obtains the abnormal state coefficient for each pollution indicator. Based on the abnormal state coefficient, it provides early warning of abnormal emissions. This invention quantifies the response sensitivity of pollution indicators to changes in operating conditions by analyzing the fluctuations of emission data within a historical period. It further assesses the operating condition correlation of pollution indicators by combining the temporal correlation between pollution indicators and operating condition indicators, and then analyzes the operating condition correlation of pollution indicators at the current monitoring moment. This accurately assesses abnormal emission behavior of pollution sources that does not conform to operating conditions, thereby improving the accuracy of pollution source monitoring.

[0060] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A pollution source monitoring method based on multi-source data acquisition and analysis, characterized in that, The method includes: Acquire multi-source monitoring data at each monitoring moment within a historical monitoring period. The multi-source monitoring data includes at least emission data for each pollution index and operating condition data for each operating condition index at the pollution source emission outlet. During the historical monitoring period, based on the temporal fluctuations of the emission data under each pollution indicator, the fluctuation characteristic parameters of each pollution indicator are obtained, and the working condition correlation coefficient of each pollution indicator is obtained by combining the temporal variation correlation between the emission data under each pollution indicator and the working condition data under each working condition indicator. At the current monitoring time, based on the change characteristics of the emission data under each pollution indicator and the operating condition data under each operating condition indicator, the current operating condition correlation coefficient of each pollution indicator within a preset historical period at the current monitoring time is obtained; based on the difference between the current operating condition correlation coefficient and the operating condition correlation coefficient under each pollution indicator, the abnormal state coefficient under each pollution indicator is obtained; and abnormal emission early warning is issued based on the abnormal state coefficient.

2. The pollution source monitoring method based on multi-source data acquisition and analysis according to claim 1, characterized in that, The method for obtaining the fluctuation characteristic parameters includes: Divide historical monitoring periods and obtain all time windows; within each time window, obtain the fluctuation parameters of each pollution index based on the fluctuation changes of the emission data for each pollution index. Based on the discrete characteristics of the fluctuation parameters of each pollution index within different time windows, the stable characteristic parameters of each pollution index within the historical monitoring period are obtained. Based on the fluctuation parameters and the stable characteristic parameters within each time window, the fluctuation characteristic parameters of each pollution index are obtained.

3. The pollution source monitoring method based on multi-source data acquisition and analysis according to claim 2, characterized in that, The method for obtaining the time window includes: Historical monitoring periods are evenly divided into preset lengths to determine all time windows.

4. The pollution source monitoring method based on multi-source data acquisition and analysis according to claim 2 or 3, characterized in that, The method for obtaining the working condition correlation coefficient includes: Within each time window, based on the correlation between the emission data under each pollution index and the operating condition data under each operating condition index, the co-change sub-parameters between each pollution index and each operating condition index are obtained; By comprehensively considering the differences among the co-variation sub-parameters within all time windows during the historical monitoring period, the co-variation parameters between each pollution index and each operating condition index are obtained. Based on the co-variation parameters between each pollution index and each operating condition index, and the fluctuation characteristic parameters of each pollution index, the operating condition correlation coefficient of each pollution index is obtained.

5. The pollution source monitoring method based on multi-source data acquisition and analysis according to claim 4, characterized in that, The method for obtaining the cooperative change sub-parameters includes: At each monitoring moment within each time window, based on the emission data for each pollution indicator and the operating condition data for each operating condition indicator, the correlation characteristic value between each pollution indicator and each operating condition indicator is determined; By combining the associated feature values ​​at adjacent monitoring times within the combined time window, the co-variation sub-parameters between each pollution index and each operating condition index are obtained.

6. The pollution source monitoring method based on multi-source data acquisition and analysis according to claim 1, characterized in that, The abnormal state coefficients for each pollution index include: For each pollution index, the negative correlation normalization result of the ratio between the current operating condition correlation coefficient and the operating condition correlation coefficient is used as the abnormal state coefficient.

7. The pollution source monitoring method based on multi-source data acquisition and analysis according to claim 1, characterized in that, Abnormal emission early warning based on the aforementioned abnormal state coefficient includes: An abnormal emission warning is issued when the abnormal state coefficient under any pollution indicator is greater than a preset threshold.

8. The pollution source monitoring method based on multi-source data acquisition and analysis according to claim 2, characterized in that, The method for obtaining the fluctuation parameters includes: For each pollution indicator, within each time window, an emission reference value is obtained based on the concentrated characteristics of the emission data at all monitoring times, and a fluctuation parameter is obtained based on the deviation of the emission data at each monitoring time from the emission reference value.

9. A pollution source monitoring system based on multi-source data acquisition and analysis, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the pollution source monitoring method based on multi-source data acquisition and analysis as described in any one of claims 1 to 8.

Images