A biomass gas equipment fault intelligent diagnosis method and system

Abstract

The present application relates to the technical field of biomass gas equipment, and discloses a kind of biomass gas equipment fault intelligent diagnosis method and system, the method comprises: obtaining operation monitoring data;Based on equipment process flow, the operation monitoring data of adjacent operation unit is checked for corresponding relationship, when there is abnormal corresponding relationship, generate exception state identifier;According to exception state identifier and equipment process flow, determine fault area;The change characteristics of each operation monitoring data in fault area are matched with the fault characteristics in fault characteristic library;When change characteristics and fault characteristics match successfully, generate fault type;According to fault type, check associated operation data;When associated operation data meets fault propagation relationship, confirm fault, generate fault result;When associated operation data does not meet fault propagation relationship, re-matching;According to fault result, generate fault diagnosis information.The present application improves the reliability of fault diagnosis result.

Description

Technical Field

[0001] This invention relates to the field of biomass gasification equipment technology, and more specifically, to an intelligent fault diagnosis method and system for biomass gasification equipment. Background Technology

[0002] Biomass gasification equipment can produce biomass gas by processing biomass raw materials through processes such as feeding and conveying, pyrolysis, catalytic reforming, gas purification and conveying. It is widely used in distributed energy supply and comprehensive utilization of biomass resources. It typically includes a feeding and conveying system, a pyrolysis furnace, a catalytic reforming reactor, a gas purification device, a gas conveying device and a control system.

[0003] Existing patent CN105482835A discloses an integrated biomass gas catalytic synthesis and purification device, which uses a PLC or MCGS control system to monitor operating parameters such as temperature, pressure, and flow rate, and triggers alarms based on whether the parameters exceed preset thresholds. However, during equipment operation, faults such as abnormal feeding, abnormal pyrolysis temperature, abnormal catalytic reforming, blockage of the purification device, abnormal gas delivery, and abnormal control system operation can all cause changes in operating monitoring data such as temperature, pressure, flow rate, current, and actuator status, and may propagate along the equipment's process flow to other operating units, causing multiple operating units to experience abnormalities simultaneously. Existing methods can only reflect the abnormal state of a single operating parameter and are insufficient to determine the source of the abnormality by combining the process correlation between various operating units. When different faults produce similar changes in operating parameters, or when the same fault causes multiple operating units to malfunction simultaneously, problems such as inaccurate fault location, difficulty in fault type identification, and false alarms are likely to occur, affecting the stability of equipment operation and maintenance efficiency.

[0004] Therefore, it is necessary to design an intelligent fault diagnosis method and system for biomass gasification equipment to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes an intelligent fault diagnosis method and system for biomass gasification equipment, which aims to solve the problems of inaccurate fault location, difficulty in fault type identification, and false alarms when different faults cause similar changes in operating parameters, or when the same fault causes multiple operating units to malfunction simultaneously.

[0006] In one aspect, the present invention proposes an intelligent fault diagnosis method for biomass gasification equipment, comprising: Acquire operational monitoring data for each operating unit of the biomass gasification equipment; the operating units include a feeding and conveying unit, a pyrolysis furnace operating unit, a catalytic reforming operating unit, a gas purification operating unit, a gas transmission operating unit, and a control system operating unit; Based on the equipment process flow, the operation monitoring data of adjacent operating units are checked for correspondence. When an abnormal correspondence exists, an abnormal status identifier is generated. Based on the abnormal status identifier and the equipment process flow, the fault area is determined. The change characteristics of each of the operational monitoring data within the fault area are matched with the fault characteristics in the fault characteristic library; when the change characteristics successfully match the fault characteristics, a fault type is generated; Based on the fault type, the associated operational data is checked; when the associated operational data matches the fault propagation relationship, the fault is confirmed and a fault result is generated; when the associated operational data does not match the fault propagation relationship, the matching is performed again. Fault diagnosis information is generated based on the fault results.

[0007] Furthermore, the operational monitoring data includes: The feeding unit's screw conveying status, feeding continuity, and conveying current; the pyrolysis furnace operation unit's zone temperature, furnace pressure, and residence time; the catalytic reforming operation unit's two-stage catalytic temperature, reaction pressure, reflux valve opening, and reflux flow rate; the gas purification operation unit's decoking liquid circulation status, dehydration status, desulfurization status, and purification pressure drop; the gas transmission operation unit's gas storage pressure, outlet flow rate, and pipeline pressure; and the control system operation unit's actuator operation status and alarm records.

[0008] Furthermore, when checking the correspondence between the operational monitoring data of adjacent operating units based on the equipment process flow, this includes: Check whether the operation monitoring data of adjacent operating units show continuous changes, sudden changes, long-term stability, or inconsistencies in changes between preceding and following processes; if so, determine that there is an abnormal correspondence; and mark the abnormal status identifier for adjacent operating units with the abnormal correspondence.

[0009] Furthermore, when identifying the fault area, this includes: When the operating unit marked with the abnormal status identifier includes the control system operating unit, it is determined whether the action status of the actuator corresponds to the alarm record; when the action status of the actuator does not correspond to the alarm record, the control system operating unit is identified as a fault area; otherwise, according to the equipment process flow, the operating units with the abnormal correspondence are investigated step by step, and the operating unit that first shows an abnormality is identified as the fault area. When the operating unit marked with the abnormal status identifier does not include the operating unit of the control system, the operating units with the abnormal correspondence are investigated step by step according to the equipment process flow, and the upstream operating unit where the abnormal correspondence first appears is identified as the fault area.

[0010] Furthermore, when matching the change characteristics of each of the operational monitoring data within the fault area with the fault characteristics in the fault characteristic library, the process includes: The system acquires time-series data of the operational monitoring data within a preset time window; extracts features from the time-series data to obtain an operational feature set; the operational feature set includes data change rate, fluctuation amplitude, and steady-state deviation; calculates the matching degree between the operational feature set and the standard fault feature set of each fault type in the fault feature library; when the matching degree calculation result meets the matching condition, the matching is determined to be successful, and the fault type is output.

[0011] Furthermore, when calculating the matching degree between the set of operational features and the standard fault feature sets of each fault type in the fault feature library, the following steps are included: Calculate the sub-feature similarity between the data change rate, fluctuation amplitude, and steady-state deviation and the standard data change rate, standard fluctuation amplitude, and standard steady-state deviation in the standard fault feature set; perform weighted fusion on each sub-feature similarity to obtain the total feature similarity between the operating feature set and the standard fault feature set; when the total feature similarity is greater than or equal to the similarity threshold, it is determined that the matching condition is met.

[0012] Furthermore, when matching the change characteristics of each of the operational monitoring data within the fault area with the fault characteristics in the fault characteristic library, the method further includes: When the change feature fails to match the fault feature, a matching failure alarm message is generated.

[0013] Furthermore, when verifying the associated operational data according to the aforementioned fault type, the following steps are included: Based on the fault type, obtain associated operational data; determine whether the change status of the associated operational data conforms to the fault propagation relationship; when the change status of the associated operational data conforms to the fault propagation relationship, confirm the fault; when the change status of the associated operational data does not conform to the fault propagation relationship, recalculate the matching degree.

[0014] Furthermore, the fault diagnosis information includes the faulty device, fault location, fault type, fault level, and recommended handling method.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By collecting operational monitoring data from each operating unit of the biomass gasification equipment and checking the correspondence between the operational monitoring data of adjacent operating units in conjunction with the equipment's process flow, it is possible to identify correlation anomalies between operating parameters, thereby improving the ability to detect equipment operational anomalies. Based on the anomaly status identifier and combined with the equipment's process flow, the fault area can be determined, distinguishing between the fault source and the fault propagation phenomenon, avoiding misjudging the chain reaction caused by upstream anomalies in downstream operating units as the fault source, thus improving the accuracy of fault location. Matching the change characteristics of the operational monitoring data within the fault area with fault features in the fault feature database allows for the identification of operational characteristics corresponding to different faults, enabling accurate identification of different fault types such as feed conveying anomalies, pyrolysis anomalies, catalytic reforming anomalies, gas purification anomalies, gas transmission anomalies, and control system anomalies, thus improving the specificity of fault diagnosis. After generating the fault type, the diagnostic results are checked against the associated operational data and verified based on the fault propagation relationship. The fault is only confirmed when the associated operational data matches the fault propagation relationship, reducing misdiagnosis and missed diagnosis caused by local parameter fluctuations, instantaneous anomalies, or false alarms, thereby improving the reliability of fault diagnosis results.

[0016] On the other hand, this application also provides an intelligent fault diagnosis system for biomass gasification equipment, used to apply the above-mentioned intelligent fault diagnosis method for biomass gasification equipment, including: The data acquisition module is configured to acquire operational monitoring data of each operating unit of the biomass gasification equipment; the operating units include a feeding and conveying unit, a pyrolysis furnace operating unit, a catalytic reforming operating unit, a gas purification operating unit, a gas transmission operating unit, and a control system operating unit; The inspection module is configured to check the correspondence between the operation monitoring data of adjacent operating units based on the equipment process flow. When an abnormal correspondence exists, an abnormal status identifier is generated. The fault area is determined based on the abnormal status identifier and the equipment process flow. The matching module is configured to match the change characteristics of each of the operation monitoring data in the fault area with the fault characteristics in the fault characteristic library; when the change characteristics match the fault characteristics successfully, a fault type is generated; The verification module is configured to verify the associated operational data according to the fault type; when the associated operational data matches the fault propagation relationship, the fault is confirmed and a fault result is generated; when the associated operational data does not match the fault propagation relationship, the matching is performed again. The diagnostic module is configured to generate fault diagnosis information based on the fault results.

[0017] It is understandable that the above-mentioned intelligent fault diagnosis methods and systems for biomass gasification equipment have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart of the intelligent fault diagnosis method for biomass gasification equipment provided in an embodiment of the present invention; Figure 2 A logic flowchart for intelligent fault diagnosis of biomass gasification equipment provided in an embodiment of the present invention; Figure 3 This is a functional block diagram of the intelligent fault diagnosis system for biomass gasification equipment provided in an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] In some embodiments of this application, see Figure 1-2 As shown, a method for intelligent fault diagnosis of biomass gasification equipment is proposed, including: S100: Acquire operational monitoring data for each operating unit of the biomass gasification equipment; the operating units include the feeding and conveying unit, the pyrolysis furnace operating unit, the catalytic reforming operating unit, the gas purification operating unit, the gas transmission operating unit, and the control system operating unit; S200: Based on the equipment process flow, check the correspondence between the operation monitoring data of adjacent operating units. When an abnormal correspondence exists, generate an abnormal status identifier; determine the fault area based on the abnormal status identifier and the equipment process flow. S300: Matches the change characteristics of each operational monitoring data within the fault area with the fault characteristics in the fault characteristic database; when the change characteristics and fault characteristics match successfully, a fault type is generated; S400: Based on the fault type, verify the associated operational data; when the associated operational data matches the fault propagation relationship, confirm the fault and generate a fault result; when the associated operational data does not match the fault propagation relationship, re-match; S500: Generates fault diagnosis information based on the fault results.

[0021] In this embodiment, the intelligent fault diagnosis method for biomass gasification equipment is applicable to biomass gasification equipment. The equipment includes a feed conveying unit, a pyrolysis furnace operating unit, a catalytic reforming operating unit, a gas purification operating unit, a gas transmission operating unit, and a control system operating unit. The operating units are connected sequentially according to the process sequence of biomass feedstock conveying, pyrolysis gas production, catalytic reforming, gas purification, gas transmission, and automatic control. The operating status of the preceding operating unit directly affects the operating status of the following operating unit; therefore, there is a clear process correlation and fault propagation among the operating units.

[0022] In step S100, during equipment operation, operational monitoring data of each operating unit is first collected. Specifically, through sensors, actuator status acquisition modules, and control terminals installed on each operating unit, the operating information of the equipment within a preset time interval is obtained, and the collected data is sent to the control system for unified storage and management.

[0023] The operating unit includes at least a feeding and conveying unit, a pyrolysis furnace operating unit, a catalytic reforming operating unit, a gas purification operating unit, a gas transmission operating unit, and a control system operating unit. The corresponding operating monitoring data for each operating unit are as follows: For the feeding and conveying unit, the screw conveying status, feeding continuity, and conveying current are collected. Among them, the screw conveying status is used to characterize whether the conveying mechanism is operating continuously and smoothly, the feeding continuity is used to characterize whether the raw material is intermittent, accumulated, or blocked, and the conveying current is used to characterize the changes in conveying load and mechanical resistance.

[0024] For the pyrolysis furnace operating unit, the temperature, furnace pressure, and residence time of each zone are collected. Among them, the temperature of each zone is used to reflect the heating state of different locations in the pyrolysis furnace, the furnace pressure is used to reflect the generation and discharge of pyrolysis gases, and the residence time is used to characterize the duration of the heating reaction of biomass feedstock in the furnace.

[0025] For the catalytic reforming operation unit, the catalytic temperature, reaction pressure, reflux valve opening, and reflux flow rate of the two catalytic stages are collected. The two catalytic temperatures are used to characterize the working state of different catalytic reaction stages, the reaction pressure is used to reflect the resistance of gas passing through the catalytic reforming reactor and the reaction stability, and the reflux valve opening and reflux flow rate are used to characterize the secondary fuel gas reflux ratio and gas distribution.

[0026] For the gas purification operation unit, the circulation status of the decoking liquid, the dehydration status, the desulfurization status, and the purification pressure drop are collected. Among them, the circulation status of the decoking liquid is used to reflect whether the circulation of the decoking liquid is normal, the dehydration status and the desulfurization status are used to characterize the working status of the purification material, and the purification pressure drop is used to reflect whether there is blockage, obstruction or saturation of the adsorbent material inside the purification device.

[0027] For the gas transmission operation unit, the gas storage pressure, outlet flow rate and pipeline pressure are collected; among them, the gas storage pressure is used to reflect the gas storage status in the syngas storage tank, the outlet flow rate is used to reflect the gas output status, and the pipeline pressure is used to reflect whether there is leakage, blockage or pressure abnormality in the transmission pipeline.

[0028] For the control system operating unit, the action status and alarm records of each actuator are collected; the action status of each actuator is used to reflect the start-up, shutdown and action feedback of equipment such as fans, pumps, valves, conveying motors, and heating units; the alarm records are used to record the alarm status of the system for abnormal events such as over-temperature, over-pressure, material shortage, blockage, and interlocking actions.

[0029] The aforementioned operational monitoring data can be acquired through continuous sampling or uploaded to the control system after sampling at fixed intervals. The collected data can be categorized and stored according to the type of operating unit.

[0030] In step S200, after acquiring the operational monitoring data of each operating unit, the data between adjacent operating units is checked according to the equipment process flow to determine whether there are any abnormal correspondences. Adjacent operating units refer to operating units that have a direct upstream-downstream connection in the process flow, such as the feeding and conveying unit and the pyrolysis furnace operating unit, the pyrolysis furnace operating unit and the catalytic reforming operating unit, the catalytic reforming operating unit and the gas purification operating unit, and the gas purification operating unit and the gas transmission operating unit, etc.

[0031] In this embodiment, when checking the correspondence between the operation monitoring data of adjacent operating units, the following types of abnormal situations are mainly checked: First, check whether there are any inconsistencies in the data of adjacent operating units. That is, when the operating status of an upstream operating unit has changed, but the corresponding parameters of the downstream operating unit have not changed reasonably, it is determined that there is an abnormal correspondence. For example, if the feed continuity of the feeding conveyor unit decreases, but the furnace pressure and temperature of each zone of the pyrolysis furnace operating unit do not fluctuate accordingly, it indicates that the feed change does not match the pyrolysis state.

[0032] Secondly, check for sudden changes in data from adjacent operating units. That is, if the monitoring data of a certain operating unit suddenly changes significantly within a short period of time, while adjacent operating units do not show reasonable correlation, it is considered an abnormal relationship. For example, if the reflux flow rate of the catalytic reforming operating unit suddenly decreases, but the gas production and pressure states of the upstream pyrolysis furnace operating unit do not change accordingly, then an anomaly can be considered to exist in the catalytic process.

[0033] Third, check whether the data of adjacent operating units remain unchanged for a long time. That is, if, under normal operating conditions, parameters that should change with the operating conditions remain in a fixed state for a long time, and the state changes are inconsistent with those of other adjacent units, it is determined that there is an abnormal correspondence. For example, if the purification pressure drop of the gas purification operating unit remains stable for a long time, but the outlet flow rate continues to decrease, it indicates that the purification unit may be blocked, malfunctioning, or have a monitoring abnormality.

[0034] Fourth, check whether there are inconsistencies in the data of adjacent operating units regarding changes in upstream and downstream processes. That is, if the upstream process has shown significant changes, but the downstream process does not respond according to the process rules, or if the changes in the downstream process are in the opposite direction to those in the upstream process, it is considered an abnormal correspondence. For example, if the temperature in each area of ​​the pyrolysis furnace operating unit increases, but the reaction pressure and the two catalytic temperatures in the catalytic reforming operating unit do not adjust accordingly, it indicates an abnormality in the process transfer between the two units.

[0035] When any of the above-mentioned abnormal situations occur in adjacent operating units, the adjacent operating units with the corresponding abnormality are marked as abnormal status identifiers. The abnormal status identifier is used to indicate that there is a process-related abnormality between the adjacent units, but it is not directly equivalent to a specific fault source. Instead, it serves as a basis for subsequent fault area determination.

[0036] After marking the abnormal status, the fault area is further determined based on the abnormal status marking and the equipment process flow. The fault area is used to characterize the range of operating units most likely to fail, avoiding direct judgment of the fault location based solely on a single alarm message.

[0037] In one implementation, when the operating unit marked with an abnormal status identifier includes a control system operating unit, the first step is to determine whether the action status of the actuator in the control system operating unit corresponds to the alarm record. Specifically, if the action status of the actuator does not correspond to the alarm record, it indicates that there is an inconsistency between the action feedback output by the control system and the alarm information. In this case, the control system operating unit can be identified as a fault area. If the action status of the actuator corresponds to the alarm record, it indicates that the control system alarm may only be an interlocking prompt caused by an abnormality in upstream equipment, which is insufficient to directly identify a fault in the control system itself. It is necessary to continue to investigate the operating units with abnormal correspondences step by step according to the equipment process flow, and identify the operating unit that first shows an abnormality as the fault area.

[0038] In another implementation, when the operating units marked with abnormal status indicators do not include the control system operating units, the operating units with abnormal correspondences are checked step by step according to the equipment process flow, and the upstream operating unit where the abnormal correspondence first appears is identified as the fault area. That is, the process is checked sequentially from upstream to downstream, and the operating unit where the abnormal data change first is identified as the fault area, thereby avoiding misjudging subsequent linkage abnormalities as the source of the fault.

[0039] Through the above methods, this implementation method can achieve gradual narrowing and accurate positioning of the fault area based on the process relationship and monitoring data linkage between adjacent operating units, providing a reliable foundation for subsequent fault type identification and fault result confirmation.

[0040] In step S300, after determining the fault area, the operation monitoring data within the fault area is further analyzed to identify the fault type. Specifically, for one or more operating units within the fault area, their operation monitoring data within a preset time window is extracted, and the data changes within that time window are organized to obtain the corresponding change characteristics.

[0041] In this embodiment, the preset time window can be set according to the equipment's operating rhythm, sampling frequency, and fault evolution speed. It can be a fixed-duration window or a rolling update window. By setting a time window, judgments can be avoided based solely on instantaneous data at a single moment, thereby improving the stability and accuracy of fault identification.

[0042] When extracting features from operational monitoring data, the focus is primarily on key features that reflect changes in equipment operating status, including data change rate, fluctuation amplitude, and steady-state deviation. Specifically, the data change rate characterizes the rise, fall, or trend of the corresponding monitoring data within a time window, reflecting whether the fault is continuously escalating or gradually recovering; the fluctuation amplitude characterizes the degree of fluctuation in the monitoring data within a time window, reflecting whether the equipment operation is unstable, intermittently abnormal, or affected by disturbances; and the steady-state deviation characterizes the degree of deviation of the monitoring data from the normal operating state, reflecting the difference between the current state and normal operating conditions.

[0043] In this embodiment, the fault feature library is categorized and stored according to fault type, with each fault type corresponding to a set of standard fault features. The standard fault feature set can be formed from historical operation records, confirmed typical fault samples, and maintenance personnel experience, and can be continuously supplemented and updated as equipment operation data accumulates. For example, the standard fault feature set corresponding to a fault in the feeding and conveying unit may include decreased feeding continuity, increased conveying current, and intermittent interruption of the screw conveyor state; the standard fault feature set corresponding to a fault in the pyrolysis furnace operating unit may include furnace pressure fluctuations, localized temperature anomalies, and excessively short or long residence times; the standard fault feature set corresponding to a fault in the catalytic reforming operating unit may include deviations in the catalytic temperature of the two stages, abnormal reflux flow, and fluctuations in reaction pressure; the standard fault feature set corresponding to a fault in the gas purification operating unit may include increased purification pressure drop, abnormal decoking liquid circulation, and abnormal dehydration or desulfurization status; the standard fault feature set corresponding to a fault in the gas transmission operating unit may include abnormal gas storage pressure, decreased outlet flow, or fluctuations in pipeline pressure; and the standard fault feature set corresponding to a fault in the control system operating unit may include inconsistencies between the actuator's action status and alarm records, sluggish action, and frequent triggering of interlocks.

[0044] When matching the change characteristics of each operational monitoring data within the fault area with the fault characteristics in the fault characteristic library, the time series data of the corresponding operating unit within the fault area are first compared item by item with the standard fault characteristics in the fault characteristic library to determine whether there is consistency or similarity between the two in terms of change trend, fluctuation state and degree of deviation.

[0045] When the change feature and the fault feature are successfully matched, the corresponding fault type is generated; when the match is unsuccessful, it means that the change feature in the current fault area is not completely consistent with any fault type in the existing fault feature library, and an alarm prompt is required.

[0046] In one implementation, to improve recognition stability, the collected time series data can be preprocessed before feature matching, such as removing abnormal interference points, smoothing short-term fluctuations, and unifying the sampling interval, so that the subsequently extracted change features can better reflect the actual operating status of the equipment, rather than short-term noise.

[0047] Furthermore, in this embodiment, the matching judgment between the running feature set and the standard fault feature set does not directly rely on a single parameter, but rather integrates multiple sub-features for corresponding judgment. Specifically, the data change rate, fluctuation amplitude, and steady-state deviation are compared item by item with the standard data change rate, standard fluctuation amplitude, and standard steady-state deviation to obtain the similarity of multiple sub-features.

[0048] Sub-feature similarity is used to reflect the degree of similarity between individual feature items. For example, if the change trend of a certain operating unit is consistent with the typical change trend of a certain type of fault, and its fluctuation amplitude and steady-state deviation are also close to the characteristic performance of that type of fault, then the corresponding sub-feature similarity is high; conversely, if one of them is significantly different, then the corresponding sub-feature similarity is low.

[0049] After obtaining the similarity scores of each sub-feature, a comprehensive evaluation is performed to form the overall feature similarity score. The overall feature similarity score reflects the degree of fit between the current operating feature set and a certain standard fault feature set. When the overall feature similarity score meets the preset matching condition, it is determined that the current operating feature set and the corresponding fault feature set have successfully matched, and the fault type is output. When the overall feature similarity score does not meet the preset matching condition, it is determined that the match has failed, and no fault type is output or the comparison of other candidate fault types is initiated.

[0050] In one implementation, the matching conditions can be set based on equipment type, fault level, historical sample distribution, and actual operating experience, thereby balancing identification sensitivity and false positive suppression capability. For example, for faults with obvious process characteristics such as feed blockage, abnormal purification pressure drop, and abnormal reflux flow, stricter matching conditions can be set; for early minor abnormalities or multiple fault superposition states, a more flexible candidate matching mechanism can be set to avoid prematurely eliminating potential fault types.

[0051] Further, when the variation feature in the fault area cannot meet the matching condition with any fault feature in the fault feature library, a matching failure alarm information is generated. The matching failure alarm information is used to prompt that there is an unrecognized abnormal state in the current fault area, which may correspond to the following situations: first, the type of fault has not been covered in the fault feature library; second, the current abnormality is a composite fault; third, the abnormality is still in the initial stage and does not fully present typical fault features; fourth, the collected data is affected by external disturbances and cannot stably reflect the real fault type. By generating the matching failure alarm information, operation and maintenance personnel can be reminded to further check the equipment status, and continue the analysis by combining manual inspection or subsequent operation data when necessary, so as to avoid misjudgment caused by forcibly outputting the wrong fault type.

[0052] In step S400, after the fault type is generated, it is not directly taken as the final result, but further checked in combination with the associated operation data to verify whether the identified fault type conforms to the propagation law on the equipment process.

[0053] In this embodiment, the associated operation data refers to the data of upstream or downstream operation units associated with the current fault type, as well as the data that can reflect the fault propagation process. For example, when the identification result is a fault of the feeding conveying unit, it can be further checked whether there are linkage abnormalities in the temperature, pressure and residence time of the pyrolysis furnace operation unit; when the identification result is a fault of the catalytic reforming operation unit, it can be further checked whether the output state of the pyrolysis furnace, the inlet pressure of the purification unit and the reflux state conform to the propagation law of catalytic abnormality; when the identification result is a fault of the purification unit, it can be further checked whether the gas storage pressure and the outlet flow rate of the gas conveying unit have synchronous abnormalities.

[0054] When checking the associated operation data, it is mainly judged whether its change state is consistent with the fault propagation relationship corresponding to the identified fault type. The so-called fault propagation relationship refers to the typical transmission path and linkage performance of a certain type of fault in the equipment process flow. For example, upstream material blockage usually causes subsequent feeding changes of the pyrolysis furnace, pressure fluctuations in the furnace and gas production fluctuations; catalytic reforming abnormality usually causes subsequent purification pressure drop changes and gas quality changes; blockage of the purification unit usually causes outlet pressure changes and gas transmission flow abnormalities. When the change state of the associated operation data conforms to the above propagation law, it can be considered that the current fault type is supported by process logic, the fault is confirmed to be established, and a fault result is generated. When the change state of the associated operation data does not conform to the fault propagation relationship, it indicates that the current identification result may have a deviation, and it is necessary to recalculate the matching degree, and re-screen the fault type that is more consistent with the actual variation feature from the fault feature library.

[0055] In one implementation, when recalculating the matching degree, other candidate fault features within the fault area can be compared again, or the changing features can be re-extracted based on new time window data to improve the accuracy and robustness of fault identification. By verifying the associated operational data, misjudging the fault type based solely on local features can be avoided, which is particularly suitable for biomass gasification equipment with multi-unit coupling and obvious fault propagation. Preferably, the number of times the matching degree is recalculated is set to 3. When the number of times the matching degree is recalculated is greater than or equal to 3, a matching failure alarm is triggered.

[0056] In step S500, the fault diagnosis information includes the faulty device, fault location, fault type, fault level, and recommended handling method. Fault diagnosis information is generated after the fault result is confirmed. The fault diagnosis information includes at least the faulty device, fault location, fault type, fault level, and recommended handling method.

[0057] The faulty equipment section indicates the specific operating unit where the abnormality occurred, such as the feeding and conveying unit, pyrolysis furnace operating unit, catalytic reforming operating unit, gas purification operating unit, gas transmission operating unit, or control system operating unit. The fault location section further indicates the specific area or component within that operating unit. The fault type section indicates the category of the fault, such as material blockage, pyrolysis abnormality, catalytic abnormality, purification failure, gas transmission abnormality, or control abnormality. The fault level section indicates the degree of impact of the fault on equipment operation. The suggested handling method section provides maintenance personnel with corresponding handling suggestions, such as clearing blockages, adjusting conveying status, checking temperature control, replenishing or replacing catalytic materials, checking purification materials, overhauling actuators, and resetting the alarm system. By outputting fault diagnosis information containing the above information, maintenance personnel can quickly understand the location, nature, and direction of the fault, thereby shortening troubleshooting time and improving the continuity and stability of equipment operation.

[0058] Based on another preferred embodiment described above, see [link to preferred embodiment]. Figure 3 As shown, this embodiment provides an intelligent fault diagnosis system for biomass gasification equipment, used to apply the above-mentioned intelligent fault diagnosis method for biomass gasification equipment, including: The data acquisition module is configured to acquire operational monitoring data from each operating unit of the biomass gasification equipment; the operating units include the feeding and conveying unit, the pyrolysis furnace operating unit, the catalytic reforming operating unit, the gas purification operating unit, the gas transmission operating unit, and the control system operating unit; The inspection module is configured to check the correspondence between the operation monitoring data of adjacent operating units based on the equipment process flow. When an abnormal correspondence exists, an abnormal status identifier is generated. Based on the abnormal status identifier and the equipment process flow, the fault area is determined. The matching module is configured to match the change characteristics of each operational monitoring data within the fault area with the fault characteristics in the fault characteristic library; when the change characteristics and fault characteristics match successfully, a fault type is generated. The verification module is configured to verify the associated operational data based on the fault type; when the associated operational data matches the fault propagation relationship, the fault is confirmed and a fault result is generated; when the associated operational data does not match the fault propagation relationship, the matching is performed again. The diagnostic module is configured to generate fault diagnosis information based on the fault results.

[0059] In summary, by collecting operational monitoring data from each operating unit of the biomass gasification equipment and checking the correspondence between the operational monitoring data of adjacent operating units in conjunction with the equipment's process flow, it is possible to identify correlational anomalies between operating parameters, thereby improving the ability to detect equipment operational anomalies. Based on anomaly status indicators and combined with the equipment's process flow, the fault area can be determined, distinguishing between the fault source and fault propagation phenomena. This avoids misjudging cascading changes in downstream operating units caused by upstream anomalies as fault sources, improving the accuracy of fault location. Matching the change characteristics of operational monitoring data within the fault area with fault features in the fault feature database allows for the identification of operational characteristics corresponding to different faults. This enables accurate identification of different fault types, such as feed conveying anomalies, pyrolysis anomalies, catalytic reforming anomalies, gas purification anomalies, gas transmission anomalies, and control system anomalies, improving the specificity of fault diagnosis. After generating the fault type, the diagnostic results are checked against the correlated operational data and verified based on the fault propagation relationship. The fault is only confirmed when the correlated operational data matches the fault propagation relationship, reducing misdiagnosis and missed diagnosis caused by local parameter fluctuations, instantaneous anomalies, or false alarms, thus improving the reliability of fault diagnosis results.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for intelligent fault diagnosis of biomass gasification equipment, characterized in that, include: Acquire operational monitoring data for each operating unit of the biomass gasification equipment; the operating units include a feeding and conveying unit, a pyrolysis furnace operating unit, a catalytic reforming operating unit, a gas purification operating unit, a gas transmission operating unit, and a control system operating unit; Based on the equipment process flow, the corresponding relationship of the operation monitoring data of adjacent operating units is checked, and when an abnormal correspondence exists, an abnormal status identifier is generated. The fault area is determined based on the abnormal status identifier and equipment process flow; The change characteristics of each of the operational monitoring data within the fault area are matched with the fault characteristics in the fault characteristic library; when the change characteristics successfully match the fault characteristics, a fault type is generated; Based on the fault type, the associated operational data is checked; When the associated operational data matches the fault propagation relationship, the fault is confirmed and a fault result is generated. If the associated runtime data does not match the fault propagation relationship, re-matching is performed. Fault diagnosis information is generated based on the fault results.

2. The intelligent fault diagnosis method for biomass gasification equipment according to claim 1, characterized in that, The operational monitoring data includes: The feeding unit's screw conveying status, feeding continuity, and conveying current; the pyrolysis furnace operation unit's zone temperature, furnace pressure, and residence time; the catalytic reforming operation unit's two-stage catalytic temperature, reaction pressure, reflux valve opening, and reflux flow rate; the gas purification operation unit's decoking liquid circulation status, dehydration status, desulfurization status, and purification pressure drop; the gas transmission operation unit's gas storage pressure, outlet flow rate, and pipeline pressure; and the control system operation unit's actuator operation status and alarm records.

3. The intelligent fault diagnosis method for biomass gasification equipment according to claim 2, characterized in that, When checking the correspondence between operational monitoring data of adjacent operating units based on the equipment's process flow, the following should be included: Check whether the operation monitoring data of adjacent operating units show continuous changes, sudden changes, long-term stability, or inconsistencies in changes between preceding and following processes; if so, determine that there is an abnormal correspondence; and mark the abnormal status identifier for adjacent operating units with the abnormal correspondence.

4. The intelligent fault diagnosis method for biomass gasification equipment according to claim 3, characterized in that, When identifying the fault area, the following should be included: When the operating unit marked with the abnormal status identifier includes the control system operating unit, it is determined whether the action status of the actuator corresponds to the alarm record; when the action status of the actuator does not correspond to the alarm record, the control system operating unit is identified as a fault area; otherwise, according to the equipment process flow, the operating units with the abnormal correspondence are investigated step by step, and the operating unit that first shows an abnormality is identified as the fault area. When the operating unit marked with the abnormal status identifier does not include the operating unit of the control system, the operating units with the abnormal correspondence are investigated step by step according to the equipment process flow, and the upstream operating unit where the abnormal correspondence first appears is identified as the fault area.

5. The intelligent fault diagnosis method for biomass gasification equipment according to claim 1, characterized in that, When matching the change characteristics of each of the operational monitoring data within the fault area with the fault characteristics in the fault characteristic library, the following is included: The system acquires time-series data of the operational monitoring data within a preset time window; extracts features from the time-series data to obtain an operational feature set; the operational feature set includes data change rate, fluctuation amplitude, and steady-state deviation; calculates the matching degree between the operational feature set and the standard fault feature set of each fault type in the fault feature library; when the matching degree calculation result meets the matching condition, the matching is determined to be successful, and the fault type is output.

6. The intelligent fault diagnosis method for biomass gasification equipment according to claim 5, characterized in that, When calculating the matching degree between the set of operational features and the standard fault feature sets of each fault type in the fault feature library, the following steps are included: Calculate the sub-feature similarity between the data change rate, fluctuation amplitude, and steady-state deviation and the standard data change rate, standard fluctuation amplitude, and standard steady-state deviation in the standard fault feature set; perform weighted fusion on each sub-feature similarity to obtain the total feature similarity between the operating feature set and the standard fault feature set; when the total feature similarity is greater than or equal to the similarity threshold, it is determined that the matching condition is met.

7. The intelligent fault diagnosis method for biomass gasification equipment according to claim 1, characterized in that, When matching the change characteristics of each of the operational monitoring data within the fault area with the fault characteristics in the fault characteristic library, the method further includes: When the change feature fails to match the fault feature, a matching failure alarm message is generated.

8. The intelligent fault diagnosis method for biomass gasification equipment according to claim 1, characterized in that, When verifying the associated operational data according to the fault type, the following are included: Based on the fault type, obtain associated operational data; determine whether the change status of the associated operational data conforms to the fault propagation relationship; when the change status of the associated operational data conforms to the fault propagation relationship, confirm the fault; when the change status of the associated operational data does not conform to the fault propagation relationship, recalculate the matching degree.

9. The intelligent fault diagnosis method for biomass gasification equipment according to claim 1, characterized in that, The fault diagnosis information includes the faulty device, fault location, fault type, fault level, and recommended handling method.

10. A biomass gasification equipment fault intelligent diagnosis system, used to apply the biomass gasification equipment fault intelligent diagnosis method as described in any one of claims 1-9, characterized in that, include: The data acquisition module is configured to acquire operational monitoring data of each operating unit of the biomass gasification equipment; the operating units include a feeding and conveying unit, a pyrolysis furnace operating unit, a catalytic reforming operating unit, a gas purification operating unit, a gas transmission operating unit, and a control system operating unit; The inspection module is configured to check the correspondence between the operation monitoring data of adjacent operating units based on the equipment process flow, and generate an abnormal status identifier when an abnormal correspondence exists. The fault area is determined based on the abnormal status identifier and equipment process flow; The matching module is configured to match the change characteristics of each of the operation monitoring data in the fault area with the fault characteristics in the fault characteristic library; when the change characteristics match the fault characteristics successfully, a fault type is generated; The verification module is configured to verify the associated operational data according to the fault type; When the associated operational data matches the fault propagation relationship, the fault is confirmed and a fault result is generated. If the associated runtime data does not match the fault propagation relationship, re-matching is performed. The diagnostic module is configured to generate fault diagnosis information based on the fault results.

Images