Method for screening abnormal data in atmospheric particulate matter component monitoring data and computing equipment

Abstract

The embodiment of the invention provides a method for screening abnormal data in atmospheric particulate matter component monitoring data and computing equipment. The abnormal data screening method comprises the following steps: determining the current mass concentration monitoring data of various particulate matter components and the actual mass concentration of particulate matters; performing weighted summation on the mass concentration monitoring data of each component and a corresponding preset weight coefficient to obtain a conversion total mass concentration; the mass concentration ratio of the total mass concentration to the actual mass concentration of the particulate matter is calculated and converted; and under the condition that the mass concentration ratio is out of the first preset ratio interval, determining the current mass concentration monitoring data of each component as overall screening abnormal data. By adopting the scheme, the overall screening efficiency and effect of the current concentration quality monitoring data of the atmospheric particulate matter components can be better.

Description

Technical Field

[0001] The present disclosure relates to the technical field of environmental monitoring, and in particular to a method and computing device for screening abnormal data in atmospheric particulate matter component monitoring data. Background Art

[0002] Fine particulate matter in the atmosphere (PM 2.5 ) has complex sources, including primary particulate matter emitted directly from pollution sources and secondary particulate matter generated through photochemical reactions. Determining the source of fine particulate matter is crucial for solving air pollution problems in a targeted manner.

[0003] In order to determine the source of air pollution, the relevant technology proposes to use monitoring equipment to monitor the components in atmospheric particulate matter, and then determine the type and source of the particulate matter by analyzing the components of the particulate matter. When using monitoring equipment to monitor the components of atmospheric particulate matter, due to failures of the monitoring equipment, random errors, environmental factors, and human intentions, there are abnormal data in the monitoring data that cannot truly reflect the concentration of pollutants in the ambient atmosphere. Before performing data analysis, the aforementioned abnormal data needs to be identified and eliminated. However, the existing methods for screening abnormal data of particulate matter components are mostly manual screening methods. Due to the time limit requirements and the problem that manual identification requires prior experience knowledge, the use of manual screening methods is inefficient and ineffective in practical applications. Summary of the invention

[0004] In order to realize automatic identification of abnormal data in atmospheric particulate matter component monitoring data, the embodiments of the present disclosure provide a method and a computing device for screening abnormal data in atmospheric particulate matter component monitoring data.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for screening abnormal data in atmospheric particulate matter component monitoring data, comprising:

[0006] Determine the current mass concentration monitoring data of various components obtained by measuring the components of atmospheric particulate matter taken during the current period; and determine the actual mass concentration of particulate matter obtained by monitoring atmospheric particulate matter during the target period;

[0007] The mass concentration monitoring data of each component and the corresponding preset weight coefficient are weighted and summed to obtain the converted overall mass concentration;

[0008] Calculating a mass concentration ratio of the converted overall mass concentration to the actual mass concentration of the particulate matter, and determining whether the mass concentration ratio is within a first preset ratio interval;

[0009] In the case where the mass concentration ratio is outside the first preset ratio interval, the current mass concentration monitoring data of the various components are determined as overall screening abnormal data.

[0010] Optionally, the method further comprises: determining a target ion component having an associated correlation among the various components, and an atomic weight and an ion valence of the target ion component;

[0011] Calculating the unit cation molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the positively charged ions in the target ion component; and calculating the unit anion molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the negatively charged ions in the target ion component;

[0012] Calculating the ion molar concentration ratio of the unit cation molar concentration to the unit anion molar concentration, and determining whether the ion molar concentration ratio is within a second preset ratio interval;

[0013] When the ion molar concentration ratio is outside the second preset ratio interval, the current mass concentration monitoring data of the target ion component is determined to be logic screening abnormal data.

[0014] Optionally, determining that the target ion components with associated positive correlation among the various components are sulfate ions, nitrate ions and ammonium ions;

[0015] The calculating unit cation molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the positively charged ions in the target ion component comprises: calculating the unit cation molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the ammonium ions;

[0016] The method of calculating the unit anion molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the negatively charged ions in the target ion component includes calculating the unit anion molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the nitrate ions and sulfate ions.

[0017] Optionally, the method further includes: determining current mass concentration monitoring data of organic carbon and elemental carbon in the various components, and calculating a current carbon mass concentration ratio between the current mass concentration monitoring data of organic carbon and the current mass concentration monitoring data of elemental carbon;

[0018] Determine whether the current carbon mass concentration ratio is within a first set confidence interval, wherein the first set confidence interval is determined according to a mean and a standard deviation of a historical carbon mass concentration ratio, wherein the historical carbon mass concentration ratio is a ratio of historical mass concentration monitoring data of an organic carbon historical mass concentration monitoring data object in a historical period;

[0019] When the current carbon mass concentration ratio is not within the first set confidence interval, the current mass concentration monitoring data of the organic carbon and the elemental carbon are determined to be logic screening abnormal data.

[0020] Optionally, it also includes: for one of the various components, calculating the mass concentration prediction data corresponding to the current period according to the mass concentration monitoring data of the historical period;

[0021] Calculating a current mass concentration residual according to the current mass concentration monitoring data of the one component and the mass concentration prediction data;

[0022] Determining whether the current mass concentration residual is within a second set confidence interval, where the second set confidence interval is determined based on an average value and a standard deviation of historical mass concentration residuals of the one component;

[0023] When it is determined that the current mass concentration residual is not within the second set confidence interval, the current mass concentration monitoring data of the component is determined to be mutation abnormal data.

[0024] Optionally, for one component of various components, the mass concentration prediction data corresponding to the current period is calculated based on the mass concentration monitoring data of the historical period, including:

[0025] For one of the various components, a mass concentration monitoring data sequence of a first historical period and a noise sequence of the first-order difference of the mass concentration of a second historical period are obtained, wherein the first historical period and the second historical period are both historical periods adjacent to the current period;

[0026] Calculate the data difference of adjacent data in the mass concentration monitoring data sequence, and perform weighted summation according to each of the data differences and the preset weights corresponding to each of the data differences to obtain the autoregressive difference sum; and perform weighted summation according to each of the random noise data of the concentration monitoring data random noise sequence and the corresponding preset weights to obtain the moving average random noise sum;

[0027] Calculating the first-order difference of the current mass concentration of the component according to the autoregressive difference sum, the moving average random noise and a preset constant term;

[0028] The mass concentration prediction data at the current moment is calculated according to the first-order difference of the current mass concentration of the component and the mass concentration monitoring data of the previous time period.

[0029] Optionally, according to one of the various components, it is determined that the constant duration of the adjacent historical mass concentration monitoring data is a preset duration and the current mass concentration monitoring number is the same as the historical mass concentration monitoring data;

[0030] When the duration exceeds a preset duration and the current mass concentration monitoring data is the same as the historical mass concentration monitoring data, it is determined that the current mass concentration monitoring data of the component is abnormal fluctuation data.

[0031] Optionally, it also includes: obtaining a device status identifier of a monitoring device for detecting various components in the current time period;

[0032] In the case where the device status mark is an abnormal status mark, the current mass concentration monitoring data of the corresponding component is marked as abnormal measurement data.

[0033] Optionally, it also includes: determining whether the current mass concentration monitoring data of a component exceeds the monitoring range or subsequent data processing range of the corresponding monitoring device for the component;

[0034] When the current mass concentration data of the one component exceeds the monitoring range or the subsequent data processing range, at least the current mass concentration data of the one component is determined to be out-of-range abnormal data.

[0035] In a second aspect, an embodiment of the present disclosure provides a computing device, including a processor and a memory, wherein the memory is used to store a computer program; when the computer program is loaded by the processor, the processor executes the method for screening abnormal data in atmospheric particulate matter component monitoring data as described above.

[0036] The method for screening abnormal data in atmospheric particulate matter component monitoring data provided by the present disclosure takes the actual mass concentration of particulate matter as a reference, and evaluates the credibility of the converted overall mass concentration data obtained by weighted summation based on the mass concentration monitoring data of various components. In the case where it is determined that the data of the converted overall mass concentration is unreliable, it is reversely determined that the current mass concentration monitoring data of each component is the overall screening abnormal data. In this way, the overall screening of the current concentration quality monitoring data of atmospheric particulate matter components can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work, including:

[0039] Figure 1is a flow chart of a method for screening abnormal data in atmospheric particulate matter component monitoring data provided by an embodiment of the present disclosure;

[0040] Figure 2 is a flow chart of a method for determining whether the mass concentration monitoring data of some components are abnormal in some embodiments of the present disclosure;

[0041] Figure 3 is a flow chart of a method for determining whether the mass concentration monitoring data of some components are abnormal in other embodiments of the present disclosure;

[0042] Figure 4 is a flow chart of a method for determining whether the mass concentration monitoring data of some components are abnormal in some embodiments of the present disclosure;

[0043] Figure 5 is a flow chart of a method for determining whether the mass concentration monitoring data of some components are abnormal in some embodiments of the present disclosure;

[0044] Figure 6 It is a schematic diagram of the structure of a computing device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0045] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0046] The term "including" and its variations used in this document are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions of other terms will be given in the description below. In this document, relational terms such as "first" and "second" are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0047] In order to solve the problem that the existing abnormal data of particulate matter components are not available by manual screening, the embodiment of the present disclosure provides an automated abnormal data screening method. The premise for the implementation of the automated abnormal data screening method is to formulate a reasonable automated screening strategy. The core of the method for screening abnormal data of atmospheric particulate matter component monitoring data provided by the embodiment of the present disclosure is to provide a practical automated screening strategy to achieve available automated data screening.

[0048] Figure 1 1 is a flow chart of a method for screening abnormal data in atmospheric particulate matter component monitoring data provided by an embodiment of the present disclosure. Figure 1 As shown, the method for screening abnormal data in atmospheric particulate matter component monitoring data provided by the embodiment of the present disclosure includes S110-S160. The data processing step in the implementation method of the present disclosure is performed by a computing device.

[0049] S110: Determine to measure the components of the atmospheric particulate matter obtained in the current period to obtain current mass concentration monitoring data of various components; and determine the actual mass concentration of the particulate matter obtained by monitoring the atmospheric particulate matter in the target period.

[0050] In a specific implementation, various existing monitoring equipment can be used to measure the components of atmospheric particulate matter to obtain the current command concentration monitoring data of various components in the current time period. In a specific implementation, monitoring equipment such as online particle chromatograph, carbonaceous component analyzer, atmospheric heavy metal online detector, etc. can be used to measure the component concentration of a specific type of particulate matter components to obtain the corresponding current mass concentration monitoring data. The aforementioned components include ionic components, substances and elemental components, carbonaceous elemental components, etc. in atmospheric particulate matter. For example, in a specific application, the monitored components in atmospheric particulate matter include NO 3— 、SO4 2— NH 4+ , Cl — , K + , Ca 2+ Mg 2+ , OC, EC, K, Ca, V, Cr, Mn, Co, Ni, Cu, Zn, As, Cd, Sn, Sb, Ba, Pb, Al, Si, Cl, Ti, Fe.

[0051] The above-mentioned current time period refers to the time period when the component measurement is required, and is not limited to the time period when the computing device processes the data. In the specific implementation, the current mass concentration monitoring data is such as g / m 3 Etc. Data expressed using mass per unit volume.

[0052] While executing the above determination of the current mass concentration monitoring data of various components acquired in the current period, the computing device can obtain the actual mass concentration of particulate matter obtained by the particulate matter monitoring device monitoring the atmospheric particulate matter, for example, the actual mass concentration of particulate matter can be obtained by measuring the manual weighing method, quartz micro-oscillating balance or particle size spectrometer based on the light scattering principle. It should be noted that the actual mass concentration of particulate matter obtained here may still not be an absolutely accurate concentration, but only a relatively accurate mass concentration.

[0053] S120: performing weighted summation on the mass concentration monitoring data of each component and the corresponding preset weight coefficient to obtain the converted overall mass concentration.

[0054] Since atmospheric particulate matter is composed of a mixture of primary pollutants emitted by specific pollution sources, or the aforementioned primary pollutants are formed through chemical reactions, the relative content of each component in the atmospheric pollutants is relatively certain for atmospheric particulate matter formed by specific types of pollution sources, that is, the content of various components in atmospheric particulate matter is within a specific range. Based on this, according to the mass concentration monitoring data of various components monitored, the converted overall mass concentration of atmospheric particulate matter can be roughly estimated (it should be noted here that even if some components are not monitored, because such components are associated with the monitored components, the corresponding prediction of the overall mass concentration of atmospheric particulate matter components is feasible). The converted overall mass concentration is the overall mass concentration obtained by converting the mass concentrations of various components to specific control conditions, which does not truly represent the actual mass concentration of particulate matter under actual atmospheric conditions.

[0055] Before executing the aforementioned S120, the preset weight coefficients corresponding to various components can be determined by experimental testing and data fitting methods. After obtaining the mass concentration monitoring data of each component, the aforementioned mass concentration monitoring data and the corresponding preset weight coefficients are weighted and summed to obtain the converted overall mass concentration.

[0056] In one specific implementation, the converted total mass concentration calculated by the computing device is the mass concentration converted to PM2.5, and the corresponding converted total mass concentration is Where m(OC) is the mass concentration of organic carbon, is the current mass concentration of nitrate, is the current mass concentration of sulfate, is the current mass concentration of ammonium, m(EC) is the mass concentration of elemental carbon, m(Cl - ) is the current mass concentration of chloride ions, m(CM) is the current mass concentration of crustal materials, m(TE) is the current trace element mass concentration, m(TE)=m(K+ )+m(Mg 2+ )+m(F - )+m(Ba)+m(Cd)+m(Sn)+m(V)+m(Cr)+m(Mn)+m(Co)+m(Ni)+m(Cu)+m(Zn)++m(Sc)+m(P)+m(Na + ), k in the above formulas i is the preset weight coefficient of the i-th component (that is, the conversion coefficient for converting it to the overall mass). i In this case, the default weight coefficient corresponding to this component is 1.

[0057] S130: Calculate the mass concentration ratio between the converted overall mass concentration and the actual mass concentration of the particulate matter.

[0058] After obtaining the above-mentioned converted total mass concentration and actual mass concentration of particulate matter, the mass concentration ratio can be obtained by comparing the two. The mass concentration ratio reflects the relative proportion of the converted total mass concentration and the actual mass concentration of particulate matter.

[0059] S140: Determine whether the mass concentration ratio is within a first preset ratio interval; if not, execute S150.

[0060] S150: Determine the current mass concentration monitoring data of various components as overall screening abnormal data.

[0061] In practical applications, the following assumptions are made: the various weights in the converted overall mass concentration determined in S120 are reasonable, and the actual mass concentration of particulate matter is reliable. Based on the above assumptions, when the current command concentration monitoring data of various components are reasonable, the difference between the converted overall mass concentration and the actual mass concentration of particulate matter will not be too large, and the command concentration ratio divided by the two should be within a reasonable ratio range (that is, the first preset ratio range). If the mass concentration ratio is not within the first preset ratio range, it can be reversed to determine that there is a high probability of abnormal data in the current mass concentration monitoring data of various components. At this time, because it is evaluated from an overall perspective, it is impossible to determine which component of the various components has abnormal data, so the current mass concentration monitoring data of various components is used as the overall screening abnormal data; on the contrary, if the aforementioned mass concentration ratio is normal, the current mass concentration monitoring data of various components is determined to be the overall screening normal data. The overall screening mentioned here is the overall screening of the components monitored in the particulate matter.

[0062] Through the above method, by taking the actual mass concentration of particulate matter as a reference, the credibility of the converted overall mass concentration data obtained by weighted summation based on the mass concentration monitoring data of various components is evaluated (evaluated by using the mass concentration ratio method). In the case where it is determined that the data of the converted overall mass concentration is unreliable, it is reversely determined that the current mass concentration monitoring data of each component is the overall screening abnormal data. In this way, the overall screening of the current concentration quality monitoring data of atmospheric particulate matter components can be achieved.

[0063] The above method only determines whether the current mass concentration monitoring data of atmospheric particulate matter components is abnormal data for overall screening, and cannot determine which component in the particulate matter has abnormal mass concentration monitoring. To this end, the embodiment of the present disclosure also provides the following method for determining whether the current mass concentration monitoring data of certain components is abnormal data.

[0064] Figure 2 1 is a flow chart of a method for determining whether the mass concentration monitoring data of some components are abnormal in some embodiments of the present disclosure. Figure 2 As shown, in some embodiments of the present disclosure, the method for determining whether the mass concentration monitoring data of some components is abnormal includes S210-S240.

[0065] S210: Determine target ion components with associated correlation among various components, as well as atomic weights and ion valences of the target ion components.

[0066] The aforementioned associated correlation means that when the mass concentration of a certain component in the particulate matter is higher, the mass concentration of another component will be positively correlated and higher or negatively correlated and lower. In practical applications, part of the aforementioned components with associated positive correlation are cationic components, and part are anionic components, and the charges of the aforementioned anions and cations are basically balanced with each other. Accordingly, determining the target ion components with positive correlation among various components is to determine the cationic components and anionic components with associated positive correlation among various components.

[0067] Through a large amount of experimental data, it is found that the ion components with associated positive correlation in atmospheric particulate matter include ammonium ions as cations, and sulfate ions and nitrate ions as anions. Correspondingly, the determined target ion components can be ammonium ions, sulfate ions, and nitrate ions. It should be noted here that sulfate ions and nitrate ions, as negatively charged ions, have an associated positive correlation with ammonium ions as positively charged ions as a whole.

[0068] S220: Calculate the unit molar concentration of cations based on the current mass concentration monitoring data, atomic weight and ion valence of the positively charged ions in the target ion component; and calculate the unit molar concentration of anions based on the current mass concentration monitoring data, atomic weight and ion valence of the negatively charged ions in the target ion component.

[0069] In the embodiments of the present disclosure, the unit cation molar concentration and are the molar concentrations calculated based on the cations in the target ion component, and the unit anion molar concentration is the molar concentration calculated based on the anions in the target ion component. Both are not the molar concentrations of all positrons or negative electrons in the atmospheric particulate matter.

[0070] According to chemical knowledge, the unit cation molar concentration or unit anion molar concentration is proportional to the current mass concentration monitoring data and the ion valence, and inversely proportional to the atomic weight of the corresponding ion. Therefore, it can be obtained by multiplying the current command concentration monitoring data and the ion valence and then dividing it by the corresponding atomic weight. Correspondingly, the unit anion molar concentration can also be obtained by the above method.

[0071] According to the above analysis, based on the ammonium ion, sulfate ion and nitrate ion determined in the above text, the ion valence of the above three target ion components is +1, -2, -1, and the atomic weight of the above three target ion components is 96, 62 and 18, the unit cation molar concentration is determined to be Determine the unit anion molar concentration as

[0072] S230: Calculate the ion molar concentration ratio of the unit cation electron molar concentration to the unit anion molar concentration.

[0073] The ion molar concentration ratio of the unit cation molar concentration to the unit anion molar concentration is calculated by directly comparing the two data mentioned above. For example, for the above-mentioned ammonium ion, sulfate ion and nitrate ion, the obtained ion molar mass concentration is

[0074] S240: Determine whether the ion molar concentration ratio is within a second preset ratio range; if not, execute S250.

[0075] S250: At least determine the current mass concentration monitoring data of the target ion component as logic screening abnormal data.

[0076] According to the premise that having an associated positive correlation means that when the mass concentration of a certain component in the particulate matter is high, the mass concentration of another component will also be high, the ion molar concentration ratio of the positive ions and negative ions of the aforementioned components with an associated positive correlation is within a certain reasonable range. Through the previous experiments, the aforementioned reasonable range was determined to be the second preset ratio interval. Therefore, if it is judged that the ion molar concentration ratio is not within the second preset ratio interval, the current mass concentration monitoring data in the corresponding target ion component is the logic screening abnormal data. The aforementioned logic screening abnormal data is based on logic screening (based on the fact that the number of positive and negative valence electrons is basically the same).

[0077] It should be noted that the logic screening of abnormal data determined by the aforementioned S210 - S240 is for the target ion components with associated positive correlation as a whole, not for a single target ion component.

[0078] Figure 3 4 is a flow chart of a method for determining whether the mass concentration monitoring data of some components is abnormal in some other embodiments of the present disclosure. Figure 3 As shown, in some embodiments of the present disclosure, the method for determining whether the mass concentration monitoring data of some components is abnormal includes S310-S330.

[0079] S310: Determine current mass concentration monitoring data of organic carbon and elemental carbon in various components, and calculate a current carbon mass concentration ratio of the current mass concentration monitoring data of organic carbon to the current mass concentration monitoring data of elemental carbon.

[0080] S320: Determine whether the current carbon mass concentration ratio is within a first set confidence interval; if not, execute S330.

[0081] S330: Determine the current mass concentration monitoring data of organic carbon and elemental carbon as logical screening abnormal data.

[0082] Through the analysis of carbon elements in a large amount of particulate matter component data, it is found that the mass concentrations of organic carbon and elemental carbon in atmospheric particulate matter change continuously with the progress of the oxidation reaction (organic carbon is oxidized to elemental carbon). Because the oxidation reaction rate of organic carbon in the atmosphere is slow, the aforementioned rate of change is slow. Accordingly, the mass concentration ratio of organic carbon and elemental carbon will have a specific change trend, and within a period of time, the mass concentration ratio of organic carbon and elemental carbon is stable.

[0083] Based on the foregoing analysis, in an embodiment of the present disclosure, the current mass concentration monitoring data of organic carbon in the current period and the current carbon mass concentration ratio of elemental carbon are calculated, and a first set confidence interval determined by the carbon mass concentration ratio in the most recent reasonable historical time is used to determine whether the current carbon mass concentration ratio is reasonable, and then a determination is made as to whether the current mass concentration monitoring data of organic carbon and elemental carbon are logical screening abnormal data.

[0084] The first set confidence interval is determined by mathematical statistics of the historical carbon mass concentration ratio, specifically, the mean and standard deviation are determined based on the historical carbon concentration ratio within a period (the calculation method of the historical carbon concentration ratio is similar to the method of calculating the current carbon mass concentration ratio in S310), and the first set confidence interval is determined based on the mean and standard deviation. For example, the 3σ method can be used to set the first set confidence interval to (μ K -3σ,μ K +3σ)(μ in the above formula K is the mean, σ is the standard deviation).

[0085] If it is determined that the current carbon mass concentration ratio is not within the aforementioned first set confidence interval, the computing device determines that the current mass concentration monitoring data of organic carbon and elemental carbon are abnormal data. It should be noted that at this time, it is not possible to determine whether the current mass concentration monitoring data of organic carbon and elemental carbon are abnormal data alone.

[0086] Figure 4 4 is a flow chart of a method for determining whether the mass concentration monitoring data of some components is abnormal in some embodiments of the present disclosure. Figure 4 As shown, in some embodiments, the computing device uses the following S410-S440 to determine whether the current mass concentration monitoring data of a component is mutation abnormal data.

[0087] S410: For one of the various components, the mass concentration prediction data corresponding to the current time period is calculated based on the mass concentration monitoring data of the historical time period.

[0088] Through a large number of component historical data, it is found that for a single component, the data sequence obtained after the mass concentration time series is processed by reasonable order difference is stable. Based on the premise that the aforementioned data series can achieve stability, reasonable mathematical tools can be used to calculate the mass concentration prediction data corresponding to the current period based on the mass concentration monitoring data of a component in the historical period. It should be noted that the aforementioned mass concentration prediction data is the command concentration data of a component in the current period, not the real data.

[0089] In some embodiments, through data analysis, it is found that the first-order difference sequence of the mass concentration monitoring data of a component is a stationary time series. Based on this data analysis, the computing device uses a model for processing stationary time series (such as an integrated moving average autoregressive model) to determine the first-order difference of the current mass concentration monitoring data of a component (relative to the first-order difference of the mass concentration monitoring data of the previous time period) according to the following S411-S414, and then calculates the mass concentration prediction data corresponding to the current time period based on the first-order difference. The following takes the integrated moving average autoregressive model as an example to explain how to calculate the first-order difference of the command concentration monitoring data.

[0090] S411: For one component of various components, obtain a mass concentration monitoring data sequence of a first historical period and a noise sequence of a first-order difference of mass concentration in a second historical period.

[0091] In the embodiment of the present disclosure, the first historical period and the second historical period are both historical periods adjacent to the current period. The lengths of the first historical period and the second historical period can be obtained by exhaustive enumeration in the training phase of the autoregressive moving average model to meet the Akaike information amount or the Bayesian information amount, and the two may be the same or different.

[0092] Each value in the aforementioned noise sequence of the first-order difference of mass concentration is calculated by calculating the difference between the predicted value and the actual value of the first-order difference of mass concentration in each historical period.

[0093] S412: Calculate the data difference of adjacent data in the mass concentration monitoring data sequence, and perform weighted summation according to each data difference and the preset weight corresponding to each data difference to obtain the autoregressive difference sum; and perform weighted summation according to each random noise data of the random noise sequence of the concentration monitoring data and the corresponding preset weight to obtain the moving average random noise sum.

[0094] After obtaining the mass concentration monitoring data sequence, the difference between the later data and the previous data in the adjacent data is calculated, that is, the data difference is subtracted from the previous data. After obtaining each data difference, the preset weights obtained in the model training stage are used to perform weighted summation processing with each data difference, and the autoregressive difference sum can be obtained. After obtaining the random sequence of concentration monitoring data, the weighted summation processing of each random noise data and the preset weights obtained in the model training stage can also be performed to obtain the moving average random noise sum.

[0095] S413: Calculate the first-order difference of the current mass concentration of a component based on the autoregressive difference sum, the moving average random noise and the preset constant term.

[0096] After obtaining the aforementioned autoregressive difference sum and moving average random noise sum, the two are added to the preset constant term calculated in the model training phase to obtain the first-order difference of the current mass concentration of a component.

[0097] In the specific implementation, the mass concentration monitoring data sequence of the first historical period obtained according to the above method is [y t-p ,……,y t-2 ,y t-1 ], the noise sequence of the first-order difference of mass concentration in the second historical period is [ε t-q ,ε t-q+1 ,……,ε t-2 ,ε t-1 ], the data difference sequence obtained by calculation is [y′ t-p ,……,y′ t-2 ,y′ t-1 ], then the first-order difference of the current mass concentration y′ t =c+φ1y′ t-1 +…+φ p y′ t-p +θ1ε t-1 +…θ q ε t-q , where c is a preset constant, φ i and θ j are the corresponding preset weights respectively.

[0098] S414: Calculate the mass concentration prediction data at the current moment according to the first-order difference of the current mass concentration of a component and the mass concentration monitoring data of the previous period.

[0099] After obtaining the first-order difference of the current mass concentration of a component, y t =y' t +y t-1 Calculate the current mass concentration prediction data y t , where y t-1 It is the mass concentration monitoring data of the previous period.

[0100] S420: Calculate the current mass concentration residual according to the current mass concentration monitoring data and mass concentration prediction data of a component.

[0101] After obtaining the current mass concentration monitoring data of a component and the current mass concentration prediction data of the mass concentration, the current mass concentration residual can be obtained by subtracting the two.

[0102] S430: Determine whether the current mass concentration residual is within a second set confidence interval; if not, execute S440.

[0103] S440: Determine that the current mass concentration monitoring data of a component is mutation abnormal data.

[0104] According to the principle of statistical analysis, if the current mass concentration is normal data, it should be within the second set confidence interval determined by the mean value and standard deviation of the historical mass concentration residuals; if it is not within the aforementioned confidence interval, it is determined that the difference between the mass concentration monitoring data and the mass concentration prediction data is too large, and the current mass concentration monitoring data can be determined as abnormal data. The aforementioned second set confidence interval can be determined using the 3σ method.

[0105] Figure 5 4 is a flow chart of a method for determining whether the mass concentration monitoring data of some components are abnormal in some embodiments of the present disclosure. Figure 5 As shown, in some embodiments, the method for determining whether the monitoring data of mass concentration of some components is abnormal includes S510-S520.

[0106] S510: According to one of the various components, determine whether the constant duration of the recent historical mass concentration monitoring data is a preset duration and whether the current mass concentration monitoring number is the same as the recent historical mass concentration monitoring data; if both are the same, execute S520.

[0107] S520: Determine that the current mass concentration monitoring data of a component is abnormal fluctuation data.

[0108] In practical applications, considering the error characteristics of actual measurements and the fluctuation characteristics of particulate mass components, the mass concentration monitoring data of a certain component is likely to be in a constant state for a short period of time by chance, and the chance of being in a constant state for a long time is basically zero, that is, an impossible event has occurred. In this case, if the constant duration of the historical mass concentration monitoring data has been the preset duration, and the current mass concentration monitoring data is still the same as the historical mass concentration monitoring data, it is determined that an impossible event has occurred, and there is reason to believe that the mass concentration prediction data of the current quality is abnormal fluctuation data.

[0109] In actual use, it may happen that the mass concentration monitoring data of a certain component is obtained when the corresponding monitoring device is in an abnormal state. Since the monitoring device is in an abnormal state, the corresponding mass concentration monitoring data may also be in an abnormal state. When the computing device obtains the mass concentration monitoring data of various components, it will obtain the abnormal state mark of the monitoring device that monitors the corresponding component in the current time period. If the device state mark is abnormal, it will determine that the corresponding mass concentration monitoring data may be abnormal measurement data, and therefore mark it. In actual applications, the aforementioned equipment state mark may be an abnormal equipment motion state mark, an operation and maintenance mark, a quality control mark, etc.

[0110] In some embodiments, the current mass concentration monitoring data of a component may be data that exceeds the detection range of the corresponding monitoring equipment or data that exceeds the subsequent data processing range. According to the monitoring principle or actual situation of the monitoring equipment, the probability of the above situation is very small. In this case, the current mass concentration data of a component is identified as over-range abnormal data. The aforementioned subsequent data processing range can be a range determined in the monitoring industry standard (such as HJ 799, HJ 800, HJ 829, HJ830, HJ 1327, HJ 1328, HJ 1329).

[0111] The above method realizes the identification of abnormal data in atmospheric particulate matter component monitoring data. After identifying the corresponding type of abnormal data and determining the abnormal type of the abnormal data, the abnormal type can be associated with the corresponding component, and the abnormal type can be written into the database storing the atmospheric particulate matter component mass concentration monitoring data for subsequent data processing.

[0112] In practical applications, after determining the aforementioned abnormal data, when the mass concentration monitoring data of various components are displayed and output through an output device, the data volume, data type, etc. of the abnormal data obtained by the aforementioned screening can also be displayed through various chart presentation methods.

[0113] The embodiment of the present disclosure also provides a computing device for implementing the aforementioned method. Figure 6 Schematic diagram of the structure of the computing device provided by the embodiment of the present disclosure. Figure 6 , which shows a schematic diagram of the structure of a computing device 600 suitable for implementing the method of the embodiment of the present disclosure. Figure 6 The computing device shown is only an example and should not bring any limitation to the functionality and scope of use of the embodiments of the present disclosure.

[0114] like Figure 6 As shown, the computing device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory ROM 602 or a program loaded from a storage device 608 to a random access memory RAM 603. Various programs and data required for the operation of the computing device 600 are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output I / O interface 605 is also connected to the bus 604.

[0115] Typically, the following devices may be connected to the I / O interface 605: input devices 606 including, for example, a touch screen, a touch pad, a camera, a microphone, etc.; output devices 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 608 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 609. The communication devices 609 may allow the computing device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 5 The computing device 600 is shown with various devices, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have instead.

[0116] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0117] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable storage medium, a computer-readable signal medium, or any combination of the above two.

[0118] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0119] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer readable program codes. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than a computer readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0120] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText TraNsferProtocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0121] The computer-readable medium may be included in the computing device, or may exist independently without being incorporated into the computing device.

[0122] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, SMalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the tester computer, partially on the tester computer, as a separate software package, partially on the tester computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the tester computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0123] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated system based on hardware that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0124] The units involved in the embodiments described in the present disclosure may be implemented by software or by hardware. The name of the unit does not, in some cases, constitute a limitation on the unit itself. The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0125] The above are only specific embodiments of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for screening abnormal data in atmospheric particulate matter component monitoring data, characterized in that: include: Determine the component measurement of the atmospheric particulate matter collected during the current period to obtain the current mass concentration monitoring data of various components; and, determining the actual mass concentration of particulate matter obtained by monitoring atmospheric particulate matter during the target period; The mass concentration monitoring data of each component and the corresponding preset weight coefficient are weighted and summed to obtain the converted overall mass concentration; Calculating a mass concentration ratio of the converted overall mass concentration to the actual mass concentration of the particulate matter, and determining whether the mass concentration ratio is within a first preset ratio interval; In the case where the mass concentration ratio is outside the first preset ratio interval, the current mass concentration monitoring data of the various components are determined as overall screening abnormal data.

2. The method according to claim 1, characterized in that: The method further comprises: Determining target ion components having correlation among the various components, as well as atomic weights and ion valences of the target ion components; Calculating the unit cation molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the positively charged ions in the target ion component; and calculating the unit anion molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the negatively charged ions in the target ion component; Calculating the ion molar concentration ratio of the unit cation molar concentration to the unit anion molar concentration, and determining whether the ion molar concentration ratio is within a second preset ratio interval; When the ion molar concentration ratio is outside the second preset ratio interval, the current mass concentration monitoring data of the target ion component is determined to be logic screening abnormal data.

3. The method according to claim 2, characterized in that Determining that the target ion components with associated positive correlation among the various components are sulfate ions, nitrate ions and ammonium ions; The calculating unit cation molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the positively charged ions in the target ion component comprises: calculating the unit cation molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the ammonium ions; The method of calculating the unit anion molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the negatively charged ions in the target ion component includes calculating the unit anion molar concentration based on the current mass concentration monitoring data, atomic weight and ion valence of the nitrate ions and sulfate ions.

4. The method according to claim 1, characterized in that Also includes: Determining current mass concentration monitoring data of organic carbon and elemental carbon in the various components, and calculating a current carbon mass concentration ratio of the current mass concentration monitoring data of organic carbon to the current mass concentration monitoring data of elemental carbon; Determine whether the current carbon mass concentration ratio is within a first set confidence interval, wherein the first set confidence interval is determined according to a mean and a standard deviation of a historical carbon mass concentration ratio, wherein the historical carbon mass concentration ratio is a ratio of historical mass concentration monitoring data of an organic carbon historical mass concentration monitoring data object in a historical period; When the current carbon mass concentration ratio is not within the first set confidence interval, the current mass concentration monitoring data of the organic carbon and the elemental carbon are determined to be logic screening abnormal data.

5. The method according to claim 1, characterized in that Also includes: For one of the various components, the mass concentration prediction data corresponding to the current period is calculated based on the mass concentration monitoring data of the historical period; Calculating a current mass concentration residual according to the current mass concentration monitoring data of the one component and the mass concentration prediction data; Determining whether the current mass concentration residual is within a second set confidence interval, where the second set confidence interval is determined based on an average value and a standard deviation of historical mass concentration residuals of the one component; When it is determined that the current mass concentration residual is not within the second set confidence interval, the current mass concentration monitoring data of the component is determined to be mutation abnormal data.

6. The method according to claim 5, characterized in that For one component of various components, the mass concentration prediction data corresponding to the current period is calculated based on the mass concentration monitoring data of the historical period, including: For one of the various components, a mass concentration monitoring data sequence of a first historical period and a noise sequence of the first-order difference of the mass concentration of a second historical period are obtained, wherein the first historical period and the second historical period are both historical periods adjacent to the current period; Calculate the data difference of adjacent data in the mass concentration monitoring data sequence, and perform weighted summation according to each of the data differences and the preset weights corresponding to each of the data differences to obtain the autoregressive difference sum; and perform weighted summation according to each of the random noise data of the concentration monitoring data random noise sequence and the corresponding preset weights to obtain the moving average random noise sum; Calculating the first-order difference of the current mass concentration of the component according to the autoregressive difference sum, the moving average random noise and a preset constant term; The mass concentration prediction data at the current moment is calculated according to the first-order difference of the current mass concentration of the component and the mass concentration monitoring data of the previous time period.

7. The method according to claim 1, characterized in that: According to one of the various components, it is determined that the constant duration of the adjacent historical mass concentration monitoring data is a preset duration and the current mass concentration monitoring number is the same as the historical mass concentration monitoring data; When the duration exceeds a preset duration and the current mass concentration monitoring data is the same as the historical mass concentration monitoring data, it is determined that the current mass concentration monitoring data of the component is abnormal fluctuation data.

8. The method according to claim 1, characterized in that Also includes: Obtaining device status identifiers of monitoring devices for detecting various components during the current period; In the case where the device status mark is an abnormal status mark, the current mass concentration monitoring data of the corresponding component is marked as abnormal measurement data.

9. The method according to claim 1, characterized in that: Also includes: Determine whether the current mass concentration monitoring data of a component exceeds the monitoring range or subsequent data processing range of the corresponding monitoring equipment for the component; When the current mass concentration data of the one component exceeds the monitoring range or the subsequent data processing range, at least the current mass concentration data of the one component is determined to be out-of-range abnormal data.

10. A computing device, characterized in that It comprises a processor and a memory, wherein the memory is used to store a computer program; when the computer program is loaded by the processor, the processor executes the method for screening abnormal data in atmospheric particulate matter component monitoring data as described in any one of claims 1 to 9.

Images