A method and apparatus for online analysis of components in chemical production tail gas

Abstract

This invention discloses an online analysis method and apparatus for chemical production exhaust gas components, relating to the field of industrial exhaust gas online monitoring technology. The method includes associating known component diagnostic results from an exhaust gas component causal diagnostic spectrum with unknown abnormal features based on their occurrence time and process path overlap; matching response features for unknown byproducts whose names cannot be directly identified and writing them into a surrogate registry; and then estimating temporary concentrations based on the response intensity corresponding to the surrogate registry entries to generate reliable conclusions about the exhaust gas components. This invention, by generating source tracing matching information for unknown byproducts and establishing surrogate registry entries, enables unknown byproducts whose names cannot be identified to participate in temporary quantification and reliable conclusion generation, ultimately improving the completeness and diagnostic reliability of reliable conclusions about exhaust gas components.

Description

Technical Field

[0001] This invention relates to the field of online monitoring technology for industrial exhaust gas, and in particular to a method and apparatus for online analysis of components in chemical production exhaust gas. Background Technology

[0002] Online analysis of chemical production tail gas components belongs to the field of online gas analysis technology. Conventional methods usually involve obtaining the emission gas through a tail gas sampling unit, then using FTIR for multi-component spectral response analysis, using TDLAS for selective detection of specific gases, using GC or MS to identify complex volatile components, and combining production control data to continuously monitor and record the tail gas emission status.

[0003] The conventional methods described above have two main areas for improvement in complex chemical engineering scenarios: First, online concentration changes are usually difficult to further correlate with reaction generation pathways and tail gas treatment pathways, and the location of anomalies depends on subsequent analysis; second, when unregistered spectral segments, retention time peaks, or mass-to-charge ratio peaks appear, it is difficult to incorporate them into the reliable analysis process in a timely manner. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an online analysis method for chemical production tail gas components to solve the problems of difficulty in associating online concentration changes with process paths for anomaly tracing and difficulty in including unknown byproducts in reliable analysis.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an online analysis method for chemical production tail gas components, which includes: collecting chemical production tail gas data and process operation data, performing time-series correlation and path attribution analysis on component changes and production stages, constructing a tail gas process component correlation benchmark map, matching real-time concentration curves window by window and locating the deviation start point and transmission direction, and generating a tail gas component causal deviation chain. Along the causal deviation chain of exhaust gas components, the reaction generation path and exhaust gas treatment path of abnormal components are traced. The source of the abnormality and the expansion link are determined by combining the deviation sequence. The correspondence between component deviation and process path is transformed into a traceable diagnostic correlation structure, and an exhaust gas component causal diagnostic map is constructed. The diagnostic results of known components in the causal diagnostic spectrum of exhaust gas components are correlated with unknown abnormal features according to the overlap of occurrence time and process path. For unknown byproducts whose names cannot be directly confirmed, response features are matched and written into the substitute registration item. Then, the temporary concentration is estimated based on the response intensity corresponding to the substitute registration item to generate a reliable conclusion on exhaust gas components.

[0007] As a preferred embodiment of the online analysis method for chemical production tail gas components described in this invention, the steps of performing time-series correlation and path attribution analysis on component changes and production stages to construct a baseline map of tail gas process component correlation are as follows: The data on chemical production exhaust gas and process operation are unified in terms of timestamps and aligned in terms of sampling period. Continuous time windows are divided according to the production stage. The component concentration trends and changes before and after treatment within each time window are extracted to form a table of component stage change relationships. Based on the component stage change relationship table, the source of component change is determined. When the component peak first appears in the time window corresponding to the feeding stage to the distillation separation and the concentration increases synchronously before and after treatment, it is determined to belong to the production generation end. When the component peak mainly appears before and after the treatment unit and the concentration after treatment does not reach the normal decay range, it is determined to belong to the tail gas treatment end. The transmission path of component change characteristics is determined and a component path attribution relationship table is formed. Write the transmission paths in the component path attribution table into the production stage window corresponding to the component stage change relationship table, and use the normal component change range and concentration decay relationship before and after treatment in each stage as the benchmark judgment conditions to construct the tail gas process component association benchmark diagram.

[0008] As a preferred embodiment of the online analysis method for chemical production tail gas components according to the present invention, the steps for constructing the tail gas process component correlation benchmark map are as follows: The normal component variation range of the target component within each production stage is set as the stage benchmark; the normal decrease in concentration from the inlet to the outlet of the treatment unit is set as the concentration decay relationship before and after treatment. The normal component variation range and the concentration decay relationship before and after treatment are bound together according to the transmission path of component change characteristics. When a component meets both the normal component variation range and the concentration decay relationship before and after treatment, a normal correlation node is established between the current component and the current production stage and the transmission path, and connected in the order of production stages to form a baseline diagram of tail gas process component correlation.

[0009] As a preferred embodiment of the online analysis method for chemical production tail gas components described in this invention, the step of matching and locating the deviation start point and propagation direction of the real-time concentration curve window by window to generate a causal deviation chain for tail gas components is as follows: Collect the concentration values ​​of the target components within a continuous sampling period, and arrange the concentration values ​​of each target component into a real-time concentration curve according to the sampling time; The real-time concentration curve is divided into continuous time windows, and window-by-window matching is performed with the normal component change range of the corresponding production stage in the tail gas process component correlation benchmark diagram. Time windows that do not conform to the normal component change range are marked as deviation windows, and the earliest abnormal change of the component in the deviation window and the corresponding production stage are located to obtain the component deviation start information. Based on the information on the deviation of components from the starting point, the concentration changes of relevant components and the relationship between concentration decay before and after treatment are read in the subsequent time window. It is then determined whether the deviation change is transmitted from the production end to the treatment detection point or from the exhaust gas treatment end to the outlet side, thus obtaining the information on the direction of component deviation transmission. By linking the information on the starting point of component deviation and the information on the direction of component deviation according to the chronological order and the corresponding process path, a continuous causal relationship is formed from the first deviation of the abnormal component to its subsequent expansion, thus generating a causal deviation chain of exhaust gas components.

[0010] As a preferred embodiment of the online analysis method for chemical production tail gas components according to the present invention, the steps of tracing the reaction generation path and tail gas treatment path of abnormal components along the causal deviation chain of tail gas components are as follows: Based on the causal deviation chain of exhaust gas components, the time window and corresponding production stage of the first deviation of the abnormal component are read, and the concentration transfer direction of the abnormal component in the subsequent time window is determined to obtain the abnormal component transfer location information. Based on the location information of abnormal component transmission, the transmission path of component change characteristics corresponding to the current production stage and the current abnormal component is found in the tail gas process component association benchmark diagram, and the reaction generation path and tail gas treatment path of the abnormal component are determined along the transmission path.

[0011] As a preferred embodiment of the online analysis method for chemical production tail gas components described in this invention, the steps of determining the source and extension of the anomaly by combining the sequence of deviations, transforming the correspondence between component deviations and process paths into a traceable diagnostic correlation structure, and constructing a causal diagnostic map of tail gas components are as follows: Based on the reaction generation path and the exhaust gas treatment path, the abnormal transmission status is determined; when the component deviation first appears in the reaction generation path and is transmitted to the exhaust gas treatment path with the exhaust gas flow direction, the abnormality is determined to be caused by the production generation end; when the component deviation first appears in the exhaust gas treatment path and the concentration decay fails after treatment, the abnormality is determined to be extended from the exhaust gas treatment end, and deviation sequence determination information is generated. Based on the deviation sequence information, diagnostic associations are established between abnormal components and corresponding path nodes to obtain information on the source and extension links of the abnormality. The source judgment and extension location of the abnormal components are written back along the reaction generation path and the exhaust gas treatment path. Component deviations that occur consecutively under the same abnormal source are merged into a diagnostic association structure. Based on the sequential relationship of the process path, establish the transmission connection between nodes in the diagnostic association structure, and mark the corresponding abnormal transmission type for each transmission connection according to the deviation sequence judgment information, and construct the causal diagnostic map of exhaust gas components.

[0012] As a preferred embodiment of the online analysis method for chemical production tail gas components according to the present invention, the step of associating the diagnostic results of known components in the tail gas component causal diagnostic spectrum with unknown abnormal features according to the overlap of occurrence time and process path is as follows: Extracting unknown anomalies from chemical production exhaust gas data that fail to match existing component libraries; When an unknown abnormal feature cannot be matched with a corresponding component name in the existing component library and has not yet been verified offline, it is marked as an unknown byproduct whose name cannot be confirmed. Read the occurrence time and process path location corresponding to the known component diagnosis results from the exhaust gas component cause-effect diagnosis map, and mark the current occurrence time and process path location as the known component association reference point; The occurrence time and process path of the unknown abnormal feature are matched with the reference point associated with the known components. When the two fall into the same abnormal process and correspond to the same path transmission range, a time path overlap record is generated. The corresponding diagnostic results of the known components are associated with the unknown abnormal feature to generate unknown byproduct traceability matching information.

[0013] As a preferred embodiment of the online analysis method for chemical production tail gas components according to the present invention, the steps for generating reliable conclusions about the tail gas components are as follows: Based on the source matching information of unknown byproducts, select the time period when unknown abnormal features and known components of the same path appear synchronously, and determine the matching response features of unknown byproducts based on the synchronous change relationship of responses within unknown abnormal features and known components of the same path; based on the matching response features of unknown byproducts, establish a substitute correspondence between unknown byproducts and corresponding unknown abnormal features, and write it into the substitute registration item; Based on the response conversion relationship between unknown byproducts in the surrogate registration entry and known components in the same path, the temporary concentration of unknown byproducts is estimated and added to the surrogate registration entry. Combined with the diagnostic results of known components, a reliable conclusion on the exhaust gas components is generated.

[0014] As a preferred embodiment of the online analysis method for chemical production tail gas components described in this invention, the step of establishing a substitute correspondence between unknown byproducts and corresponding unknown abnormal features is as follows: The unknown byproducts are assigned temporary byproduct numbers, and the unknown anomalous feature that triggers the current number is used as the identification entry point for the temporary byproduct number; the response conversion relationship between the unknown anomalous feature and the known components in the same path is determined based on the occurrence time and response intensity change ratio of the unknown anomalous feature in the same transmission path. By binding temporary byproduct numbers with unknown anomaly features and using response conversion relationships as quantitative rules between the two, a substitute correspondence relationship between unknown byproducts and corresponding unknown anomaly features is established.

[0015] Secondly, the present invention provides an online analysis device for chemical production tail gas components, including: a tail gas process benchmark construction module, which collects chemical production tail gas data and process operation data, performs time-series correlation and path attribution analysis on component changes and production stages, constructs a tail gas process component correlation benchmark map, matches real-time concentration curves window by window and locates the deviation starting point and transmission direction, and generates a tail gas component causal deviation chain. The exhaust gas causal diagnosis construction module traces the reaction generation path and exhaust gas treatment path of abnormal components along the causal deviation chain of exhaust gas components. It combines the deviation sequence to determine the source of abnormality and the extension link, and transforms the correspondence between component deviation and process path into a traceable diagnostic association structure to construct an exhaust gas component causal diagnosis map. The unknown byproduct credibility determination module associates the diagnostic results of known components in the causal diagnostic spectrum of exhaust gas components with unknown abnormal features according to the overlap of occurrence time and process path. For unknown byproducts whose names cannot be directly confirmed, the module matches response features and writes them into the substitute registration item. Then, it estimates the temporary concentration based on the response intensity corresponding to the substitute registration item and generates a credibility conclusion for exhaust gas components.

[0016] The beneficial effects of this invention are as follows: by constructing a causal diagnostic map of exhaust gas components, the deviation position of components, the type of abnormal transmission, and the process path nodes are associated, so as to realize the traceability judgment of the source and extension process of abnormalities; by generating source matching information of unknown by-products and establishing a substitute registration item, unknown by-products whose names cannot be confirmed can participate in the generation of temporary quantitative and credible conclusions, thereby improving the completeness and diagnostic reliability of credible conclusions of exhaust gas components. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an online analysis method for components in chemical production tail gas.

[0019] Figure 2 This is a schematic diagram illustrating the generation of a causal deviation chain in exhaust gas components.

[0020] Figure 3 This is a schematic diagram for determining the reliability of unknown byproducts.

[0021] Figure 4 This is a comparison chart of the concentrations of unknown byproducts.

[0022] Figure 5 This is a time-series diagram showing the concentration changes of the target component in the reaction generation path and the exhaust gas treatment path. Detailed Implementation

[0023] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0024] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0025] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0026] Reference Figures 1-5 This is one embodiment of the present invention, which provides an online analysis method for components in chemical production tail gas, comprising the following steps: S1. Collect chemical production tail gas data and process operation data, perform time-series correlation and path attribution analysis on component changes and production stages, construct a tail gas process component correlation benchmark map, match real-time concentration curves window by window and locate the deviation starting point and transmission direction, and generate tail gas component causal deviation chain.

[0027] The data on chemical production exhaust gas includes sampling point, sampling time, exhaust gas flow rate, exhaust gas temperature, exhaust gas pressure, exhaust gas humidity, target component concentration, FTIR spectrum response, GC retention time peak, MS mass-to-charge ratio peak, and component concentration at the inlet and outlet of the treatment unit.

[0028] Process operation data includes the current production stage, reaction temperature, reaction pressure, feed flow rate, catalyst usage status, condenser outlet temperature, absorber pH value, absorber circulating liquid flow rate, combustion furnace temperature, and vent valve opening.

[0029] It should be noted that the chemical production tail gas data is collected through tail gas sampling probes, flow meters, temperature, humidity and pressure sensors, FTIR analyzers, GC analyzers, MS analyzers and online monitoring points at the inlet and outlet of the treatment unit. The process operation data is collected through the DCS system, reaction equipment sensors, condenser measuring points, absorption tower measuring points, combustion furnace measuring points and vent valve opening feedback signals.

[0030] The data on chemical production exhaust gas and process operation are unified in terms of timestamps and sampling periods. Continuous time windows are divided according to production stages. The compositional concentration trends and changes before and after treatment within each time window are extracted to form a compositional stage change relationship table. The production stages include the feeding stage, the heating reaction stage, the isothermal reaction stage, the distillation separation stage, the condensation recovery stage, and the venting and replacement stage.

[0031] Furthermore, for each production stage corresponding to a time window, the change process of the target component concentration curve from the window start point to the peak point and then to the end point is read to determine the concentration rise and fall status, the peak occurrence period and the abnormal deviation status (for example, if the concentration rises continuously from the window start point to the peak and then falls back, it is determined as a rise followed by a fall; the sampling time where the peak is located is determined as the peak occurrence period; and the concentration continuously exceeds the normal component change range is determined as an abnormal deviation status). Based on the decrease in concentration from the inlet concentration to the outlet concentration of the treatment unit, the concentration decay status before and after treatment is extracted. The above component change characteristics, the current production stage and the target component are written into the same stage record, and summarized in the order of production stages to form a component stage change relationship table.

[0032] Based on the component stage change relationship table, the source of component change is determined. When the component peak first appears in the time window corresponding to the feeding stage to the distillation separation and the concentration increases synchronously before and after treatment, it is determined to belong to the production generation end. When the component peak mainly appears before and after the treatment unit and the concentration after treatment does not reach the normal decay range, it is determined to belong to the tail gas treatment end. The transmission path of component change characteristics is determined and a component path attribution relationship table is formed.

[0033] Furthermore, after determining whether the component change belongs to the production generation end or the exhaust gas treatment end, the path from the corresponding production stage to the treatment unit inlet and then to the treatment unit outlet is determined according to the time window in which the component peak first appears, the detection location where the subsequent concentration change reaches, and the concentration decay state before and after treatment. Then, the target component, component change characteristics, end-side determination, transmission path, and corresponding production stage are written into the same path attribution record, and a component path attribution relationship table is formed by summarizing the target component and production stage.

[0034] Write the transmission paths in the component path attribution table into the production stage window corresponding to the component stage change relationship table, and use the normal component change range and concentration decay relationship before and after treatment in each stage as the benchmark judgment conditions to construct the tail gas process component association benchmark diagram.

[0035] Furthermore, the transfer path of the target component corresponding to the component path attribution table is written into the production stage window corresponding to the first appearance of the target component peak in the component stage change table. For example, if the peak of a certain VOCs component first appears in the distillation separation stage, then the transfer path from the distillation separation stage to the treatment unit inlet and then to the treatment unit outlet is written into the distillation separation stage window. The construction of the exhaust gas process component correlation benchmark diagram is as follows: The normal component variation range of the target component within each production stage is set as the stage benchmark; the normal decrease in concentration from the inlet to the outlet of the treatment unit is set as the concentration decay relationship before and after treatment; the normal component variation range and the concentration decay relationship before and after treatment are bound according to the transmission path of component change characteristics; when a component simultaneously meets both the normal component variation range and the concentration decay relationship before and after treatment, a normal correlation node is established between the current component and the current production stage and the transmission path, and connected in chronological order of production stages to form the exhaust gas process component correlation benchmark diagram. The exhaust gas process component correlation benchmark diagram is used to uniformly correlate production stages, normal component variation ranges, concentration decay relationships before and after treatment, and the transmission path of component change characteristics, providing a benchmark basis for subsequent identification of component deviations from the starting point and tracking of abnormal transmission directions.

[0036] It should be noted that the normal component variation range refers to the upper limit, lower limit and fluctuation trend of the target component concentration allowed to occur in the corresponding production stage, which is set by the target component concentration variation record in the same production stage from historical stable operation data.

[0037] Collect the concentration values ​​of the target components within a continuous sampling period, and arrange the concentration values ​​of each target component into a real-time concentration curve according to the sampling time.

[0038] The real-time concentration curve is divided into continuous time windows, and window-by-window matching is performed with the normal component change range of the corresponding production stage in the tail gas process component correlation benchmark diagram. Time windows that do not conform to the normal component change range are marked as deviation windows. The earliest abnormal change of the component in the deviation window and the corresponding production stage are located to obtain the component deviation starting point information.

[0039] Furthermore, the real-time concentration curve is divided into continuous time windows consistent with the production stage windows according to the sampling time sequence. The target component concentration change within each continuous time window is then compared with the normal component change range of the same production stage in the tail gas process component correlation benchmark diagram. When the upper limit, lower limit, and fluctuation trend of the concentration all meet the corresponding benchmark, the continuous time window is determined to be matched. The target component concentration within the current continuous time window is compared point by point with the normal upper limit and lower limit of concentration, and the actual rise and fall trend of the target component is compared with the allowable fluctuation trend. When the concentration at any sampling point exceeds the normal upper limit or falls below the normal lower limit, or the actual rise and fall trend is inconsistent with the allowable fluctuation trend and continues to reach the set number of samplings, the continuous time window is marked as a deviation window. Then, the target component that triggered the exceedance or inconsistency trend earliest and its corresponding production stage are read from the deviation window to obtain the component deviation starting point information.

[0040] Based on the information on the deviation of components from the starting point, the concentration changes of relevant components and the concentration decay relationship before and after treatment are read in the subsequent time window. It is then determined whether the deviation change is transmitted from the production end to the treatment detection point or from the exhaust gas treatment end to the outlet side, thus obtaining the information on the direction of component deviation transmission.

[0041] Furthermore, based on the component deviation from the starting point information, the subsequent concentration changes at the production end detection point, the treatment inlet detection point, and the treatment outlet detection point are read. When the concentration at a detection point exceeds the normal concentration range or the trend deviates from the allowable fluctuation trend and continues to reach the set number of samplings, the detection point is determined to have met the deviation judgment condition. If the production end detection point meets the deviation judgment condition first and the subsequent treatment detection points deviate in sequence, it is determined that the deviation is transmitted from the production end to the treatment detection point. If the treatment inlet detection point or the treatment outlet detection point meets the deviation judgment condition first and the concentration after treatment does not reach the normal attenuation range, it is determined that the deviation is extended from the exhaust gas treatment end to the outlet side, thus obtaining the component deviation transmission direction information.

[0042] By linking the information on the starting point of component deviation and the information on the direction of component deviation according to the chronological order and the corresponding process path, a continuous causal relationship is formed from the first deviation of the abnormal component to its subsequent expansion, thus generating a causal deviation chain of exhaust gas components.

[0043] Furthermore, the starting point information of component deviation is used as the first node of the chain. Subsequent nodes are established sequentially according to the order in which each detection point in the component deviation transmission direction information meets the deviation judgment condition. Adjacent nodes are connected according to the transmission path in the tail gas process component association benchmark diagram. When the deviation of a subsequent node occurs after the previous node and belongs to the same transmission path of the same target component, a continuous causal relationship is determined between the two nodes. Finally, the first node, subsequent nodes, and node connection relationships are organized into a tail gas component causal deviation chain. The tail gas component causal deviation chain is used to connect the first deviation location, deviation transmission direction, and process path correspondence of abnormal components, transforming tail gas anomalies from single-point concentration alarms into a continuous causal judgment basis that can trace the source and expansion process.

[0044] It should be noted that the process path refers to the continuous path of the target component after it is generated in the corresponding production stage, passing through the reaction generation stage, the transport stage, and the detection point of the treatment unit along with the exhaust gas flow. Abnormal components refer to target components that do not conform to the normal component variation range within the corresponding production stage, or that do not conform to the normal concentration decay relationship before and after treatment.

[0045] S2. Along the causal deviation chain of exhaust gas components, trace the reaction generation path and exhaust gas treatment path of abnormal components, and determine the source and extension links of the anomaly by combining the deviation sequence. Transform the correspondence between component deviation and process path into a traceable diagnostic correlation structure and construct a causal diagnostic map of exhaust gas components.

[0046] Based on the causal deviation chain of exhaust gas components, the time window and corresponding production stage of the first deviation of the abnormal component are read, and the concentration transfer direction of the abnormal component in the subsequent time window is determined to obtain the abnormal component transfer location information. Furthermore, continuous detection data of abnormal components after the first deviation time window are read, and the order in which the deviation judgment conditions are met at the production end detection point, the treatment inlet detection point, and the treatment outlet detection point are compared. When the concentration deviation occurs in the order of production end detection point to treatment inlet detection point and then to treatment outlet detection point, it is determined that the abnormal component is transmitted from the production end to the exhaust gas treatment end. When the concentration deviation first appears at the treatment inlet detection point or the treatment outlet detection point and the concentration after treatment does not reach the normal attenuation range, it is determined that the abnormal component is spreading from the exhaust gas treatment end to the outlet side. Then, the first deviation time window, the corresponding production stage, the starting detection point, and the concentration transmission direction are written into the same record to obtain the abnormal component transmission location information.

[0047] Based on the location information of abnormal component transmission, the transmission path of component change characteristics corresponding to the current production stage and the current abnormal component is found in the tail gas process component association benchmark diagram, and the reaction generation path and tail gas treatment path of the abnormal component are determined along the transmission path.

[0048] Furthermore, based on the abnormal component transmission location information, the current production stage and the current abnormal component are used as search conditions in the exhaust gas process component association benchmark diagram to find normal association nodes that simultaneously contain the production stage and the abnormal component, and the transmission path of the component change characteristics recorded in the normal association node is read; according to the connection order of the transmission path from the production stage window to the treatment unit inlet detection point and to the treatment unit outlet detection point, the path to the production stage window is determined as the reaction generation path, and the path to the treatment unit inlet detection point to the treatment unit outlet detection point is determined as the exhaust gas treatment path.

[0049] Based on the reaction generation path and the exhaust gas treatment path, the abnormal transmission status is determined; when the component deviation first appears in the reaction generation path and is transmitted to the exhaust gas treatment path with the exhaust gas flow direction, the abnormality is determined to be caused by the production generation end; when the component deviation first appears in the exhaust gas treatment path and the concentration decay fails after treatment, the abnormality is determined to be extended from the exhaust gas treatment end, and deviation sequence determination information is generated. Furthermore, the process for determining the failure of concentration decay after treatment is as follows: within the same continuous time window, read the inlet concentration and outlet concentration of the abnormal component in the treatment unit, calculate the actual decrease in concentration from the inlet to the outlet, and compare the actual decrease with the normal decay amplitude corresponding to the production stage in the tail gas process component correlation benchmark diagram; when the actual decrease is lower than the normal decay amplitude, and this state continues to reach the set number of samplings, the concentration decay after treatment is determined to be failed.

[0050] It should be noted that the deviation sequence determination information is used to record the first location of the component deviation and the subsequent transmission order, thereby distinguishing whether the anomaly is caused by the production end or extended by the exhaust gas treatment end, and providing a basis for judgment to construct the exhaust gas component causal diagnostic map.

[0051] Based on the deviation sequence information, diagnostic associations are established between abnormal components and corresponding path nodes to obtain information on the source and extension links of the abnormality. The source judgment and extension location of the abnormal components are written back along the reaction generation path and the exhaust gas treatment path. Component deviations that occur consecutively under the same abnormal source are merged into a diagnostic association structure. Furthermore, the abnormal components are written into the path nodes where the component deviation occurs, and the source type of the path node is marked as either initiated at the production end or extended at the exhaust gas treatment end based on the deviation sequence information. Then, the path nodes where the component deviation occurs subsequently are marked as extension links to obtain information on the abnormal source and extension links. Along the reaction generation path and the exhaust gas treatment path, the path node where the component deviation occurs for the first time is taken as the diagnostic starting point, and the path nodes where the component deviation occurs subsequently are taken as extension nodes. The connection relationship between the starting point and the extension nodes is established according to the adjacent time windows and the path sequence relationship to obtain the diagnostic association structure.

[0052] Based on the sequential relationship of the process path, establish the transmission connection between nodes in the diagnostic association structure, and mark the corresponding abnormal transmission type for each transmission connection according to the deviation sequence judgment information, and construct the causal diagnostic map of exhaust gas components.

[0053] Furthermore, according to the sequence of the process paths, nodes in the upstream path of the diagnostic association structure are designated as preceding nodes, and nodes in the adjacent downstream paths where component deviation subsequently occurs are designated as subsequent nodes. A connection representing the direction of component deviation transmission is established between the preceding and subsequent nodes. Based on the deviation sequence determination information, the positions of the preceding and subsequent nodes of each transmission connection are read. When the preceding node is located in the reaction generation path and the subsequent node is located in the exhaust gas treatment path, the transmission connection is marked as a production generation end initiation type. When the preceding node is located in the exhaust gas treatment path and the subsequent node continues to point to the treatment outlet side, the transmission connection is marked as an exhaust gas treatment end extension type. Each path node in the diagnostic correlation structure is treated as a graph node, and the connections between nodes representing component deviations from the transmission direction are treated as graph edges. Each graph edge is labeled as either a production-end initiation type or an exhaust gas treatment-end extension type. Simultaneously, the abnormal component, source type, and extension location are written into the corresponding graph node, ultimately forming an exhaust gas component causal diagnostic graph capable of tracing the source and extension process of anomalies. The exhaust gas component causal diagnostic graph is used to associate the component deviation location, anomaly transmission type, and process path nodes of abnormal components, transforming exhaust gas component analysis from concentration anomaly identification to anomaly source localization and extension process tracing.

[0054] Figure 5 The solid line, dashed line, and dotted line represent the target component concentration at the reaction generation path, the target component concentration at the inlet of the treatment unit, and the target component concentration at the outlet of the treatment unit, respectively. The vertical dashed line corresponds to the time boundary of different production stages. It can be seen that the target component shows a significant peak value before the reaction generation path in the high-temperature intensification stage, and then forms corresponding changes at the inlet and outlet of the treatment unit. This can support the judgment of the deviation sequence and the anomaly transmission type, indicating that the present invention can link the component deviation position, process path node, and anomaly expansion process, and improve the traceability judgment ability of anomaly source and expansion link.

[0055] S3. Associate the diagnostic results of known components in the causal diagnostic spectrum of exhaust gas components with unknown abnormal features according to the overlap of occurrence time and process path. Match response features for unknown byproducts whose names cannot be directly confirmed and write them into the substitute registration item. Then estimate the temporary concentration based on the response intensity corresponding to the substitute registration item to generate a reliable conclusion on exhaust gas components.

[0056] Extracting unknown anomalies from chemical production exhaust gas data that fail to match existing component libraries; Furthermore, the FTIR spectrum response, GC retention time peak, and MS mass-to-charge ratio peak were read from the chemical production tail gas data. Each online response feature was matched with the standard spectrum, standard retention time, and standard mass-to-charge ratio of known components in the existing component library. The matched responses of known components were removed, and the online response features that did not reach the matching threshold and recurred repeatedly within the continuous time window were retained to obtain unknown abnormal features.

[0057] It should be noted that the existing component library is a component identification database built upon historical exhaust gas detection records, standard gas calibration records, and offline verification records, used for online response feature matching. Unknown anomaly features refer to online response features extracted from chemical production exhaust gas data that fail to match the response features of known components in the existing component library and repeatedly appear within a continuous time window. Examples include unregistered absorption peaks in FTIR spectra, unregistered retention time peaks in GC analysis, or unregistered mass-to-charge ratio peaks in MS analysis.

[0058] The known component diagnostic results include the determination of the source of the abnormal component, the location of its spread, and the type of abnormal transmission. When an unknown abnormal feature cannot be matched with a corresponding component name in the existing component library and has not yet been verified offline, it is marked as an unknown byproduct whose name cannot be confirmed. Furthermore, when the spectral position, retention time, or mass-to-charge ratio of an unknown abnormal feature does not fall within the response feature matching threshold range of any known component in the existing component library, or although there are some similarities, they cannot uniquely correspond to a specific component name, it is determined that the corresponding component name cannot be matched; offline verification refers to sending exhaust gas samples from the same sampling period into the laboratory testing process, and verifying the true component name of the unknown abnormal feature through standard sample comparison, chromatographic mass spectrometry retesting, or manual spectral confirmation; before the offline verification is completed or a confirmed name is given, the unknown abnormal feature is marked as an unknown byproduct whose name cannot be confirmed.

[0059] It should be noted that the matching threshold is set based on historical exhaust gas detection records, standard gas calibration records, and offline verification records. It is determined by statistically analyzing the allowable deviation range of known component standard spectral segments, standard retention times, and standard mass-to-charge ratios under stable detection conditions. An exemplary range is that the standard spectral segment deviation does not exceed ±2 cm. - ¹, The retention time offset shall not exceed ±0.05 min, and the mass-to-charge ratio offset shall not exceed ±0.2 m / z.

[0060] Read the occurrence time and process path location corresponding to the known component diagnosis results from the exhaust gas component cause-effect diagnosis map, and mark the current occurrence time and process path location as the known component association reference point; The occurrence time and process path of the unknown abnormal feature are matched with the reference point associated with the known components. When the two fall into the same abnormal process and correspond to the same path transmission range, a time path overlap record is generated. The corresponding diagnostic results of the known components are associated with the unknown abnormal feature to generate unknown byproduct traceability matching information.

[0061] Furthermore, the occurrence time of the unknown abnormal feature is compared with the occurrence time of the known component-related benchmark point, and the process path where the unknown abnormal feature is located is compared with the process path position of the known component-related benchmark point. When the occurrence time of the unknown abnormal feature overlaps with the occurrence time of the known component diagnosis result, and the path node where the unknown abnormal feature is located is located in the same component change feature transmission path, it is determined that the two belong to the same abnormal process, and the overlapping time period, overlapping path node and corresponding known component diagnosis result are written into the same record to generate a time period path overlap record.

[0062] Based on the overlapping time-period path records, the overlapping time periods, overlapping path nodes, and corresponding known component diagnostic results are read. Unknown anomalies appearing within the same overlapping time period and located within the same overlapping path node range are linked to the anomaly transmission type corresponding to the known component diagnostic result. The source path of the unknown anomaly feature appearing together with this anomaly transmission type is recorded. This establishes a traceability association between the known component diagnostic results and the unknown anomaly features, generating unknown byproduct traceability matching information. The unknown byproduct traceability matching information is used to link unknown anomalies whose names cannot be confirmed to the anomaly transmission type and source path corresponding to the known component diagnostic results, so that unknown byproducts can still obtain traceable process source judgment basis before offline confirmation is completed.

[0063] Based on the source matching information of unknown byproducts, select the time period when unknown abnormal features and known components of the same path appear synchronously, and determine the matching response features of unknown byproducts based on the synchronous change relationship of responses within unknown abnormal features and known components of the same path; based on the matching response features of unknown byproducts, establish a substitute correspondence between unknown byproducts and corresponding unknown abnormal features, and write it into the substitute registration item; Furthermore, based on the source matching information of unknown byproducts, the online response curves of unknown abnormal features and known components of the same path within the same time period are read, and the peak response time, response rise and fall direction, and response duration of the unknown abnormal feature are compared with those of the known components of the same path. When the peak response time coincides and the response rise and fall direction is consistent, and the response duration falls within the same abnormal time period, the unknown abnormal feature is identified as the unknown byproduct matching response feature.

[0064] Based on the matching response characteristics of unknown byproducts, a temporary byproduct number is generated for the unknown byproduct, and the unknown anomaly feature that generates the matching response characteristic is used as the identification marker for the temporary byproduct number. The temporary byproduct number, the unknown anomaly feature, and the unknown byproduct matching response characteristic are then written into the same surrogate registration item, enabling the unknown byproduct to be identified through the corresponding unknown anomaly feature before its name is confirmed, thereby establishing a surrogate correspondence between the unknown byproduct and its corresponding unknown anomaly feature. This surrogate correspondence is used to bind the temporary byproduct number, the unknown anomaly feature, and the unknown byproduct matching response characteristic into the same computable object before the unknown byproduct's name is confirmed. This prevents the unknown byproduct from being eliminated as unidentified interference, but allows it to participate in source tracing, temporary quantification, and the generation of reliable conclusions regarding exhaust gas components along the same path response relationship as known components.

[0065] Based on the response conversion relationship between unknown byproducts in the surrogate registration entry and known components in the same path, the temporary concentration of unknown byproducts is estimated and added to the surrogate registration entry. Combined with the diagnostic results of known components, a reliable conclusion on the exhaust gas components is generated.

[0066] Furthermore, the response conversion relationship in the surrogate registration item refers to the conversion rule that uses known components along the same path as a reference to convert the matching response characteristics of unknown byproducts into temporary concentrations. Specifically, it first compares the response intensity of the matching response characteristics of unknown byproducts with that of known components along the same path within the same time period to determine the relative strength of the response of unknown byproducts to known components along the same path. Then, it compares the decrease in response intensity of the matching response characteristics of unknown byproducts from the inlet to the outlet of the treatment unit with the decrease in concentration intensity of known components along the same path from the inlet to the outlet of the treatment unit to determine whether unknown byproducts are more difficult to remove than known components along the same path during the treatment process.

[0067] When estimating the temporary concentration of unknown byproducts, the concentration of known components along the same path at the treatment outlet is used as a reference concentration. A preliminary conversion is performed based on the relative strength of the response of unknown byproducts to known components along the same path. The concentration is then corrected based on the difference in the decrease of unknown byproducts and known components along the same path before and after treatment. The temporary concentration of unknown byproducts is then added to the surrogate registration item.

[0068] Confidential conclusions regarding exhaust gas components include diagnostic results for known components, matching response characteristics of unknown byproducts, surrogate registration items, temporary concentrations of unknown byproducts, and the correspondence between unknown byproducts and known components along the same pathway.

[0069] The temporary concentration of the unknown byproduct is estimated using the following expression: ; in, Indicates at time Unknown byproducts Provisional estimated concentration at the outlet side of the treatment unit. Indicates at time Known components along the same path The measured concentration at the outlet side of the treatment unit, Indicates at time The response intensity of matching response characteristics to unknown byproducts. Indicates at time The response intensity of the known components along the same path. Indicates at time The concentration decay rate of known components before and after treatment along the same path. Indicates at time The response decay rate before and after treatment of unknown byproducts matching response characteristics. Indicates an unknown byproduct. This refers to a known component whose name has been confirmed and which is on the same transport path as the unknown byproduct. This represents the preset minimum attenuation correction constant; To establish a substitute correspondence between unknown byproducts and their corresponding unknown anomalies, the following steps are taken: First, a temporary byproduct number is assigned to the unknown byproduct, and the unknown anomaly that triggers this number is used as the entry point for identifying the temporary byproduct number. Second, based on the occurrence time and response intensity change ratio of the unknown anomaly in the same transmission path, the response conversion relationship between the unknown anomaly and the known components in the same path is determined. Third, the temporary byproduct number is bound to the unknown anomaly, and the response conversion relationship is used as a quantitative rule between the two to establish a substitute correspondence between the unknown byproduct and its corresponding unknown anomaly.

[0070] It should be noted that the response decay rate of the unknown byproduct matching response characteristics before and after treatment is obtained by reading the response intensities of the unknown byproduct matching response characteristics at the inlet and outlet sides of the treatment unit within the same time window, and calculating the percentage decrease in response intensity from the inlet side to the outlet side. The minimum decay correction constant is set based on the lowest effective response decay rate of the unknown byproduct matching response characteristics in historical stable operating data, and is a fixed small value greater than the lower limit of the instrument response noise, to avoid abnormal amplification of temporary concentration estimation when the response decay rate before and after treatment is close to zero.

[0071] Figure 4 The solid line represents the actual concentration of the unknown byproduct, and the dashed line represents the temporary concentration of the unknown byproduct. It can be seen that during the abnormally active period, the variation range and peak position of the two remain highly consistent. This indicates that the present invention does not simply record the unknown abnormal features as interference signals, but can transform them into analytical objects that can participate in the calculation by means of substitute registration items, thereby supporting the temporary quantification of unknown byproducts and improving the completeness and diagnostic reliability of the credible conclusions of exhaust gas components.

[0072] This embodiment also provides an online analysis device for chemical production tail gas components, including: a tail gas process benchmark construction module, which collects chemical production tail gas data and process operation data, performs time-series correlation and path attribution analysis on component changes and production stages, constructs a tail gas process component correlation benchmark map, matches real-time concentration curves window by window and locates the deviation start point and transmission direction, and generates a tail gas component causal deviation chain. The exhaust gas causal diagnosis construction module traces the reaction generation path and exhaust gas treatment path of abnormal components along the causal deviation chain of exhaust gas components. It combines the deviation sequence to determine the source of abnormality and the extension link, and transforms the correspondence between component deviation and process path into a traceable diagnostic association structure to construct an exhaust gas component causal diagnosis map. The unknown byproduct credibility determination module associates the diagnostic results of known components in the causal diagnostic spectrum of exhaust gas components with unknown abnormal features according to the overlap of occurrence time and process path. For unknown byproducts whose names cannot be directly confirmed, the module matches response features and writes them into the substitute registration item. Then, it estimates the temporary concentration based on the response intensity corresponding to the substitute registration item and generates a credibility conclusion for exhaust gas components.

[0073] In summary, this invention achieves traceable judgment of the source and extension process of anomalies by constructing a causal diagnostic map of exhaust gas components, establishing associations between component deviation locations, abnormal transmission types, and process path nodes; and by generating source matching information for unknown byproducts and establishing surrogate registration items, enabling unknown byproducts whose names cannot be confirmed to participate in the generation of temporary quantitative and credible conclusions, ultimately improving the completeness and diagnostic reliability of credible conclusions for exhaust gas components.

[0074] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for online analysis of components in chemical production tail gas, characterized in that, include: Collect chemical production tail gas data and process operation data, perform time-series correlation and path attribution analysis on component changes and production stages, construct a tail gas process component correlation benchmark map, match real-time concentration curves window by window and locate the deviation start point and transmission direction, and generate tail gas component causal deviation chain. Along the causal deviation chain of exhaust gas components, the reaction generation path and exhaust gas treatment path of abnormal components are traced. The source of the abnormality and the expansion link are determined by combining the deviation sequence. The correspondence between component deviation and process path is transformed into a traceable diagnostic correlation structure, and an exhaust gas component causal diagnostic map is constructed. The diagnostic results of known components in the causal diagnostic spectrum of exhaust gas components are correlated with unknown abnormal features according to the overlap of occurrence time and process path. For unknown by-products whose names cannot be directly confirmed, response features are matched and written into the substitute registration item. Then, the temporary concentration is estimated based on the response intensity corresponding to the substitute registration item to generate a reliable conclusion on exhaust gas components. The steps to arrive at the credible conclusion regarding the generated exhaust gas components are as follows: Based on the source matching information of unknown byproducts, select the time period when unknown abnormal features and known components of the same path appear synchronously, and determine the matching response features of unknown byproducts based on the synchronous change relationship of responses within unknown abnormal features and known components of the same path; based on the matching response features of unknown byproducts, establish a substitute correspondence between unknown byproducts and corresponding unknown abnormal features, and write it into the substitute registration item; Based on the response conversion relationship between unknown byproducts in the surrogate registration and known components in the same pathway, the temporary concentration of unknown byproducts is estimated and added to the surrogate registration. Combined with the diagnostic results of known components, a reliable conclusion on exhaust gas components is generated. The steps for establishing a substitute correspondence between unknown byproducts and corresponding unknown abnormal features are as follows: The unknown byproducts are assigned temporary byproduct numbers, and the unknown anomalous feature that triggers the current number is used as the identification entry point for the temporary byproduct number; the response conversion relationship between the unknown anomalous feature and the known components in the same path is determined based on the occurrence time and response intensity change ratio of the unknown anomalous feature in the same transmission path. By binding temporary byproduct numbers with unknown anomaly features and using response conversion relationships as quantitative rules between the two, a substitute correspondence relationship between unknown byproducts and corresponding unknown anomaly features is established.

2. The online analysis method for chemical production tail gas components as described in claim 1, characterized in that, The steps for performing time-series correlation and path attribution analysis on component changes and production stages, and constructing a baseline diagram of component correlation in the exhaust gas process, are as follows: The data on chemical production exhaust gas and process operation are unified in terms of timestamps and aligned in terms of sampling period. Continuous time windows are divided according to the production stage. The component concentration trends and changes before and after treatment within each time window are extracted to form a table of component stage change relationships. Based on the component stage change relationship table, the source of component change is determined. When the component peak first appears in the time window corresponding to the feeding stage to the distillation separation and the concentration increases synchronously before and after treatment, it is determined to belong to the production generation end. When the component peak mainly appears before and after the treatment unit and the concentration after treatment does not reach the normal decay range, it is determined to belong to the tail gas treatment end. The transmission path of component change characteristics is determined and a component path attribution relationship table is formed. Write the transmission paths in the component path attribution table into the production stage window corresponding to the component stage change relationship table, and use the normal component change range and concentration decay relationship before and after treatment in each stage as the benchmark judgment conditions to construct the tail gas process component association benchmark diagram.

3. The online analysis method for chemical production tail gas components as described in claim 2, characterized in that, The steps for constructing the baseline diagram relating exhaust gas process components are as follows: The normal component variation range of the target component within each production stage is set as the stage benchmark; the normal decrease in concentration from the inlet to the outlet of the treatment unit is set as the concentration decay relationship before and after treatment. The relationship between the range of normal component changes and the concentration decay before and after treatment is linked according to the transmission path of component change characteristics; When a component simultaneously meets the normal component variation range and the concentration decay relationship before and after treatment, a normal correlation node is established between the current component and the current production stage and the transmission path, and connected in the order of production stages to form a baseline diagram of the correlation of exhaust gas process components.

4. The online analysis method for chemical production tail gas components as described in claim 3, characterized in that, The steps for matching and locating the deviation start point and propagation direction of the real-time concentration curve window by window to generate the causal deviation chain of exhaust gas components are as follows: Collect the concentration values ​​of the target components within a continuous sampling period, and arrange the concentration values ​​of each target component into a real-time concentration curve according to the sampling time; The real-time concentration curve is divided into continuous time windows, and window-by-window matching is performed with the normal component change range of the corresponding production stage in the tail gas process component correlation benchmark diagram. Time windows that do not conform to the normal component change range are marked as deviation windows, and the earliest abnormal change of the component in the deviation window and the corresponding production stage are located to obtain the component deviation start information. Based on the information on the deviation of components from the starting point, the concentration changes of relevant components and the relationship between concentration decay before and after treatment are read in the subsequent time window. It is then determined whether the deviation change is transmitted from the production end to the treatment detection point or from the exhaust gas treatment end to the outlet side, thus obtaining the information on the direction of component deviation transmission. By linking the information on the starting point of component deviation and the information on the direction of component deviation according to the chronological order and the corresponding process path, a continuous causal relationship is formed from the first deviation of the abnormal component to its subsequent expansion, thus generating a causal deviation chain of exhaust gas components.

5. The online analysis method for chemical production tail gas components as described in claim 1, characterized in that, The steps for tracing the reaction generation path and exhaust gas treatment path of abnormal components along the causal deviation chain of exhaust gas components are as follows: Based on the causal deviation chain of exhaust gas components, the time window and corresponding production stage of the first deviation of the abnormal component are read, and the concentration transfer direction of the abnormal component in the subsequent time window is determined to obtain the abnormal component transfer location information. Based on the location information of abnormal component transmission, the transmission path of component change characteristics corresponding to the current production stage and the current abnormal component is found in the tail gas process component association benchmark diagram, and the reaction generation path and tail gas treatment path of the abnormal component are determined along the transmission path.

6. The online analysis method for chemical production tail gas components as described in claim 5, characterized in that, The process of determining the source and extension of anomalies by combining the sequence of deviations transforms the correspondence between component deviations and process paths into a traceable diagnostic correlation structure, constructing a causal diagnostic map of exhaust gas components. The steps are as follows: Based on the reaction generation path and the exhaust gas treatment path, the abnormal transmission trend is determined; when the deviation of the component first appears in the reaction generation path and is transmitted to the exhaust gas treatment path with the exhaust gas flow direction, it is determined that the abnormality was caused by the production generation end. When the deviation of a component first appears in the exhaust gas treatment path and the concentration decay fails after treatment, the anomaly is determined by extending from the exhaust gas treatment end, generating deviation sequence determination information. Based on the deviation sequence information, diagnostic associations are established between abnormal components and corresponding path nodes to obtain information on the source and extension links of the abnormality. The source judgment and extension location of the abnormal components are written back along the reaction generation path and the exhaust gas treatment path. Component deviations that occur consecutively under the same abnormal source are merged into a diagnostic association structure. Based on the sequential relationship of the process path, establish the transmission connection between nodes in the diagnostic association structure, and mark the corresponding abnormal transmission type for each transmission connection according to the deviation sequence judgment information, and construct the causal diagnostic map of exhaust gas components.

7. The online analysis method for chemical production tail gas components as described in claim 1, characterized in that, The steps for associating the diagnostic results of known components in the exhaust gas component causal diagnostic map with unknown abnormal features according to the overlap of occurrence time and process path are as follows: Extracting unknown anomalies from chemical production exhaust gas data that fail to match existing component libraries; When an unknown abnormal feature cannot be matched with a corresponding component name in the existing component library and has not yet been verified offline, it is marked as an unknown byproduct whose name cannot be confirmed. Read the occurrence time and process path location corresponding to the known component diagnosis results from the exhaust gas component cause-effect diagnosis map, and mark the current occurrence time and process path location as the known component association reference point; The occurrence time and process path of the unknown abnormal feature are matched with the reference point associated with the known components. When the two fall into the same abnormal process and correspond to the same path transmission range, a time path overlap record is generated. The corresponding diagnostic results of the known components are associated with the unknown abnormal feature to generate unknown byproduct traceability matching information.

8. An online analysis device for chemical production tail gas components, based on the online analysis method for chemical production tail gas components according to any one of claims 1 to 7, characterized in that, include: The tail gas process baseline construction module collects tail gas data and process operation data from chemical production, performs time-series correlation and path attribution analysis on component changes and production stages, constructs a tail gas process component correlation baseline map, matches real-time concentration curves window by window and locates the deviation start point and transmission direction, and generates a tail gas component causal deviation chain. The exhaust gas causal diagnosis construction module traces the reaction generation path and exhaust gas treatment path of abnormal components along the causal deviation chain of exhaust gas components. It combines the deviation sequence to determine the source of abnormality and the extension link, and transforms the correspondence between component deviation and process path into a traceable diagnostic association structure to construct an exhaust gas component causal diagnosis map. The unknown byproduct credibility determination module associates the diagnostic results of known components in the causal diagnostic spectrum of exhaust gas components with unknown abnormal features according to the overlap of occurrence time and process path. For unknown byproducts whose names cannot be directly confirmed, the module matches response features and writes them into the substitute registration item. Then, it estimates the temporary concentration based on the response intensity corresponding to the substitute registration item and generates a credibility conclusion for exhaust gas components.

Images