Early warning model test method and device for supercritical unit under mechanism diagnosis analysis

Abstract

The application relates to a test method and device for a warning model of a supercritical unit under mechanism diagnosis analysis. The specific scheme is as follows: obtaining measurement point data of multiple mechanism models; verifying multiple measurement points based on the measurement point names, measurement point identifiers and monitoring values of the multiple measurement points of the mechanism models; in response to the fact that the multiple measurement points all pass the verification, determining the fixed values of multiple operator blocks in the mechanism models; inputting the detection values of the multiple measurement points into the mechanism models to obtain a warning result output by the mechanism models based on the fixed values of the multiple operator blocks and the detection values of the multiple measurement points; comparing the warning result with a true result to obtain a first comparison result; and in response to the fact that each first comparison result meets preset requirements, performing joint testing on the multiple mechanism models based on a preset total model and the multiple measurement point data of the multiple mechanism models to obtain a test result. The application verifies the accuracy, timeliness and authenticity of the mechanism models and the measurement point data.

Description

Technical Field

[0001] This application relates to the field of big data platform technology for supercritical units, and in particular to a method and apparatus for testing early warning models of supercritical units under mechanism diagnosis and analysis. Background Technology

[0002] In related technologies, thermal power plants encompass multiple systems, equipment, and specialties, resulting in numerous unstable and uncertain factors. Failure to promptly identify and address these issues can severely impact the safety of various systems and equipment, leading to accidents such as equipment damage, unauthorized unit outages, and personal injuries. The root cause of these accidents lies in the unclear and ambiguous definition of the boundary conditions and information for each system and equipment. This results in the failure to provide timely, accurate, and effective fault warnings and guidance to personnel. Summary of the Invention

[0003] Therefore, this application provides a method and apparatus for testing early warning models of supercritical units under mechanistic diagnostic analysis. The technical solution of this application is as follows:

[0004] According to a first aspect of the embodiments of this application, a method for testing an early warning model of a supercritical unit under mechanism diagnostic analysis is provided, the method comprising:

[0005] The measurement point data of each of the multiple mechanistic models are obtained; wherein, the multiple mechanistic models are all pre-built based on fault early warning data of multiple professional fields corresponding to supercritical units; the measurement point data includes the measurement point name, measurement point identifier and monitoring value of each of the multiple measurement points;

[0006] For each mechanism model, based on the measurement point name, measurement point identifier, and monitoring value of each measurement point of the mechanism model, the multiple measurement points are respectively verified.

[0007] In response to the fact that all the multiple measurement points have passed verification, the fixed values ​​of each of the multiple operator blocks in the mechanism model are determined;

[0008] The detection values ​​of each of the multiple measuring points are input into the mechanism model to obtain the early warning result output by the mechanism model based on the fixed values ​​of each of the multiple operator blocks and the detection values ​​of each of the multiple measuring points;

[0009] The warning result is compared with the actual result to obtain the first comparison result;

[0010] In response to the fact that the first comparison results of each mechanism model meet the preset requirements, the multiple mechanism models are jointly tested based on the preset overall model and the multiple measurement point data of each of the multiple mechanism models to obtain the test results; wherein, the overall model is a supercritical unit early warning model generated based on the multiple mechanism models.

[0011] According to one embodiment of this application, the verification process for the multiple measurement points based on the mechanism model, including their respective measurement point names, measurement point identifiers, and measurement point attribute information, includes:

[0012] Based on the measurement point names and measurement point identifiers of multiple measurement points in the aforementioned mechanism model, search for the multiple measurement point identifiers of the aforementioned mechanism model that are pre-stored in the database.

[0013] In response to the fact that the measurement point identifier of at least one of the multiple measurement points of the mechanism model does not correspond to the measurement point identifier of the mechanism model stored in the database, based on the measurement point name of the at least one measurement point, it is determined whether the measurement point identifier of the related measurement point of the at least one measurement point corresponds to the measurement point identifier of the mechanism model stored in the database.

[0014] In response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it is determined that at least one measurement point of the mechanism model has failed the verification. The parameter tuning process is performed on the at least one measurement point respectively, and the step of verifying the multiple measurement points is repeated.

[0015] According to one embodiment of this application, the verification process for the multiple measuring points based on the mechanism model, including their respective measuring point names, measuring point identifiers, and monitored values, further includes:

[0016] Based on the monitoring values ​​of multiple measuring points in the aforementioned mechanism model, the input attributes of each monitoring value are determined.

[0017] For each measuring point, obtain the actual attributes of the monitored value of that measuring point;

[0018] The input attribute is compared with the actual attribute to obtain a second comparison result;

[0019] In response to the second comparison result that the input attribute does not match the actual attribute, based on the measurement point name, the related measurement points of the measurement point are determined, and it is determined whether the measurement point identifier of the related measurement points corresponds to the measurement point identifier of the mechanism model stored in the database;

[0020] In response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it is determined that the measurement point has failed the verification, and parameter tuning is performed on the measurement point. The steps of verifying the multiple measurement points are repeated.

[0021] According to one embodiment of this application, the verification process for the multiple measuring points based on the mechanism model, including their respective measuring point names, measuring point identifiers, and monitored values, further includes:

[0022] For each measuring point, the measuring point name is compared with the actual measuring point name in the distributed control system (DCS).

[0023] In response to the fact that the name of the measuring point does not match the actual name of the DCS measuring point, based on the name of the measuring point, the relevant measuring points of the measuring point are determined, and it is determined whether the name of the relevant measuring points matches the actual name of the DCS measuring point.

[0024] If the name of the relevant measuring point does not match the actual measuring point name of the DCS, it is determined that the measuring point has failed the verification. The parameter adjustment process is performed on the measuring point, and the steps of verifying the multiple measuring points are repeated.

[0025] According to one embodiment of this application, after inputting the detection values ​​of the plurality of measurement points into the mechanism model, the method further includes:

[0026] The trigger time of the mechanism model is obtained, and based on the trigger time, it is determined whether the mechanism model is triggered within a preset time period;

[0027] In response to the fact that the mechanism model is not triggered within a preset time period, it is determined whether there is a delay operator block in the mechanism model;

[0028] In response to the presence of a delay operator block in the mechanism model, the delay time of the delay operator block is determined;

[0029] Based on the delay time and the trigger time, determine whether the mechanism model should postpone triggering according to the delay time;

[0030] In response to the mechanism model not delaying the trigger according to the delay time, the mechanism model is subjected to parameter tuning.

[0031] According to one embodiment of this application, the joint testing of the multiple mechanistic models based on multiple measurement point data of each of the preset overall model and the multiple mechanistic models to obtain test results includes:

[0032] The measurement data from multiple measurement points of each of the multiple mechanistic models are input into the overall model;

[0033] The warning results of each of the multiple mechanistic models and the warning result of the overall model are obtained respectively;

[0034] If the warning results of each of the multiple mechanistic models and the warning result of the overall model are consistent with their respective actual results, the overall model is determined to have passed the test.

[0035] In response to at least one of the warning results of the multiple mechanistic models and the warning result of the overall model not being consistent with their respective actual results, the overall model is subjected to parameter tuning.

[0036] According to one embodiment of this application, the fault warning data in the multiple professional fields includes any one or more of the following: fault warning information data of steam turbine, fault warning information data of boiler, fault warning information data of electrical, fault warning information data of thermal control, fault warning information data of chemistry, fault warning information data of ash removal, fault warning information data of desulfurization, and fault warning information data of coal conveying system.

[0037] According to a second aspect of the embodiments of this application, a test device for an early warning model of a supercritical unit under mechanism diagnostic analysis is provided, the device comprising:

[0038] The acquisition module is used to acquire the measurement point data of each of the multiple mechanism models; wherein, the multiple mechanism models are all pre-built based on fault early warning data of multiple professional fields corresponding to supercritical units; the measurement point data includes the measurement point name, measurement point identifier and monitoring value of each of the multiple measurement points;

[0039] The verification module is used to verify the multiple measurement points for each mechanism model based on their respective measurement point names, measurement point identifiers, and monitoring values.

[0040] A determination module is used to determine the set value of each of the multiple operator blocks in the mechanism model in response to the fact that all the multiple measurement points have passed the verification.

[0041] The input module is used to input the detection values ​​of the multiple measuring points into the mechanism model to obtain the warning result output by the mechanism model based on the fixed values ​​of the multiple operator blocks and the detection values ​​of the multiple measuring points.

[0042] The comparison module is used to compare the warning result with the actual result to obtain a first comparison result;

[0043] The testing module is used to jointly test the multiple mechanistic models based on the preset overall model and the multiple measurement point data of each of the multiple mechanistic models in response to the first comparison result of each of the preset requirements, and to obtain the test results; wherein, the overall model is a supercritical unit early warning model generated based on the multiple mechanistic models.

[0044] According to a third aspect of the embodiments of this application, an electronic device is provided, including: a processor, and a memory communicatively connected to the processor;

[0045] The memory stores computer-executed instructions;

[0046] The processor executes computer execution instructions stored in the memory to implement the method as described in any one of the first aspects.

[0047] According to a fourth aspect of the embodiments of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of the first aspects.

[0048] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects:

[0049] By acquiring measurement point data for multiple mechanistic models; for each mechanistic model, verifying multiple measurement points based on their names, identifiers, and monitoring values; in response to multiple measurement points passing verification, determining the setpoints of multiple operator blocks within the mechanistic model; inputting the detection values ​​of each measurement point into the mechanistic model to obtain early warning results output by the mechanistic model based on the setpoints of the multiple operator blocks and the detection values ​​of the multiple measurement points; comparing the early warning results with the actual results to obtain the first comparison result; in response to the first comparison results of each mechanistic model meeting preset requirements, jointly testing multiple mechanistic models based on the preset overall model and the measurement point data of each mechanistic model to obtain test results. This verifies the accuracy, timeliness, and authenticity of the mechanistic models and measurement point data.

[0050] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0051] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application, and do not constitute an undue limitation of this application.

[0052] Figure 1 This is a flowchart illustrating a method for testing an early warning model of a supercritical unit under mechanistic diagnostic analysis, as described in an embodiment of this application.

[0053] Figure 2 This is a structural block diagram of an early warning model testing device for a supercritical unit under mechanism diagnostic analysis, as described in an embodiment of this application.

[0054] Figure 3This is a block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0055] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0056] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0057] It should be noted that, in related technologies, thermal power plants encompass multiple systems, equipment, and disciplines, resulting in numerous unstable and uncertain factors. Failure to promptly identify and address these issues can severely impact the safety of various systems and equipment, leading to accidents such as equipment damage, unauthorized unit outages, and personal injuries. The root cause of these accidents lies in the unclear and ambiguous definition of the boundary conditions and information for each system and equipment, resulting in the failure to provide timely, accurate, and effective fault warnings and guidance to personnel.

[0058] To address the aforementioned issues, this application proposes a testing method and apparatus for early warning models of supercritical units under mechanistic diagnostic analysis. This method involves acquiring measurement point data from multiple mechanistic models; verifying multiple measurement points for each model based on their names, identifiers, and monitoring values; determining the setpoints of multiple operator blocks within the mechanistic model in response to the verification of all measurement points; inputting the detection values ​​of each measurement point into the mechanistic model to obtain an early warning result output by the model based on the setpoints of the operator blocks and the detection values ​​of the measurement points; comparing the early warning result with the actual result to obtain a first comparison result; and performing a joint test on multiple mechanistic models based on a preset overall model and the measurement point data of each mechanistic model to obtain test results. This verifies the accuracy, timeliness, and authenticity of the mechanistic models and measurement point data. The problem of intelligent operation and smart safety system construction in thermal power plants was solved. A fault early warning information database for multiple key systems across all disciplines was created and formulated. Based on this mechanism diagnosis, early warning models for various types and systems were divided. The mechanism models were deployed to a big data platform and standardized and unified.

[0059] Figure 1 This is a flowchart of a method for testing an early warning model of a supercritical unit under mechanism diagnostic analysis, as described in an embodiment of this application.

[0060] like Figure 1 As shown, the early warning model test method for this supercritical unit under mechanism diagnostic analysis includes:

[0061] Step 101: Obtain the measurement data of each of the multiple mechanistic models.

[0062] In this embodiment, multiple mechanistic models are pre-built based on fault early warning data from multiple professional fields corresponding to supercritical units. Each mechanistic model consists of multiple interconnected operator blocks.

[0063] In this embodiment of the application, the measurement point data includes the measurement point name, measurement point identifier and monitoring value of each of the multiple measurement points.

[0064] In some embodiments of this application, the fault warning data from multiple professional fields includes any one or more of the following: fault warning information data from steam turbines, fault warning information data from boilers, fault warning information data from electrical systems, fault warning information data from thermal control systems, fault warning information data from chemical systems, fault warning information data from ash removal systems, fault warning information data from desulfurization systems, and fault warning information data from coal conveying systems.

[0065] For example, the fault early warning information database for steam turbines can include: early warning information for the main reheat steam and bypass system, high-pressure heater extraction steam and condensate system, low-pressure heater extraction steam and condensate system, small turbine steam system, auxiliary steam system, shaft sealing system, turbine feedwater system, condensate system, closed-loop water system, circulating water system, turbine vacuum system, turbine condensate system, turbine lubricating oil system, turbine lubricating oil purification system, turbine fire-resistant oil system, generator hydrogen system, sealing oil system, stator cooling water system, turbine safety monitoring system, steam-driven feedwater pump set oil system, electric feedwater pump body system, load center oil system, circulating water pump house system, heating network steam system, turbine-side important auxiliary machine temperature system, and generator body temperature system. The boundary conditions of its expert system may include: the valve position of the steam inlet valve, the speed of the feedwater pump, the differential pressure of the filter screen, the feedwater flow rate, the oil level of the lubricating oil tank, the oil temperature, the oil pressure, the condenser outlet pressure, the condenser water level, the condenser inlet filter differential pressure, the valve position of the recirculation valve, the water level of the deaerator, the valve position of the deaerator water supply regulating valve, the oil level of the fire-resistant oil tank, the oil temperature, the oil pressure, the high-pressure heater level, the normal drain valve position, the emergency drain valve position, the high-pressure heater extraction steam rate, the unit vacuum, the status of the circulating water inlet and outlet valves, the circulating water temperature, the main steam pipeline drain temperature and drain valve status, and the boundary conditions of the feedwater pump shaft seal pressure.

[0066] The boiler-specific fault early warning information database can include early warning information for: unit steam and water system, boiler start-up system, boiler condensate drainage system, boiler furnace body system, secondary air damper system, coal mill pulverizing system, furnace tube leakage status judgment system, combustion system, coal mill lubrication oil station system, coal mill hydraulic oil station system, plasma system, flue gas system, primary air and sealing air system, low temperature economizer system, forced draft fan body and oil station system, induced draft fan body and oil station system, primary air fan body system, air preheater body system, flame detection system, soot blowing system, denitrification soot blowing system, denitrification pyrolysis furnace system, MFT furnace purging system, furnace-side important auxiliary equipment temperature system, and boiler metal temperature system. Its expert system boundary conditions may include: valve opening degree, valve inlet and outlet pressure and temperature, primary air volume, total coal feed, coal mill inlet and outlet differential pressure, coal mill current, coal mill outlet pressure, furnace pressure, fan inlet and outlet differential pressure, fan inlet and outlet temperature, fan current, air preheater inlet and outlet temperature, air preheater inlet and outlet oxygen content, unit load, air volume, and boundary conditions of boiler flue gas duct friction pressure;

[0067] The electrical engineering fault early warning information database can include early warning information for: the main electrical wiring system, the auxiliary workshop 110V DC system, the 6KV boiler post-furnace centralized section system, the public UPS system, the public PC system, the unit 110V DC system, the 6KV 11-section plant power system, the unit UPS system, the boiler and turbine PC system, the coal conveying and desulfurization PC system, the 6KV 1-section plant power system, the unit 220KV DC system, the security PC section system, the water supply and plant front PC system, and the reclaimed water pumping station and railway PC system. Its expert system boundary conditions can include: boundary conditions for the 220kV main transformer high-voltage side circuit breaker, the generator demagnetizing switch, the plant fast-connector, the 6kV A and B section plant branch working power circuit breakers, and the 6kV A and B section standby branch power circuit breakers.

[0068] The fault early warning information database for thermal control can include early warning information from the primary frequency regulation system and the AGC (Automatic Generation Control, a paid ancillary service provided by grid-connected power plants, where generating units adjust their output in real time according to a certain rate of adjustment within a specified output adjustment range, following instructions issued by the power dispatching and trading agency, to meet the frequency and tie-line power control requirements of the power system) system. Its expert system boundary conditions include: primary frequency regulation amplitude, unit load, AGC instructions, main steam pressure, and load variation rate.

[0069] The fault early warning information database for the chemistry major can include early warning information for: ultrafiltration systems, secondary desalination systems, desalinated water tank systems, industrial wastewater dosing systems, oily wastewater systems, fine treatment auxiliary systems, pyrolysis furnace systems, steam and water sampling systems, integrated water pumping station systems, reclaimed water deep treatment, HVAC systems, primary desalination systems, reverse osmosis systems, water treatment dosing systems, water treatment regeneration systems, fine treatment systems, fine treatment regeneration systems, steam and water dosing systems, domestic sewage treatment systems, high-efficiency clarifier systems, refrigeration station systems, raw water heating and filtration, hydrogen supply station systems, urea solution preparation systems, oxygenation systems, circulating water treatment systems, reclaimed water dosing systems, and external greywater makeup water pumping station systems. Its expert system boundary conditions can include: operating time, water production flow rate, primary and secondary differential pressure, ultrafiltration system operating time, water production flow rate, ultrafiltration differential pressure, silicon value, sodium value, hydrogen conductivity, and mixed bed differential pressure boundary conditions.

[0070] The fault early warning information database of the ash removal system can include early warning information for the furnace ash conveying system, furnace slag removal system, dry ash sorting system, air compressor system, furnace ash hopper gasification system, and ash silo system.

[0071] The fault early warning information database for desulfurization systems can include early warning information for: zero wastewater discharge systems, absorption tower systems, gypsum dewatering systems, air oxidation systems, wastewater dosing systems, pulping systems, process water systems, and flue gas systems;

[0072] The fault early warning information database of the coal conveying system can include early warning information from the belt conveyor status system, coal crusher system, tipper PC section system, and coal-containing wastewater system.

[0073] Step 102: For each mechanism model, based on the measurement point name, measurement point identifier, and monitoring value of each measurement point of the mechanism model, the multiple measurement points are verified separately.

[0074] In some embodiments of this application, step 102 includes:

[0075] Step a1: Based on the measurement point names and measurement point identifiers of multiple measurement points in the mechanism model, search for the multiple measurement point identifiers of the mechanism model pre-stored in the database.

[0076] Optionally, the measuring point identifier can be the power plant identification system code of the measuring point.

[0077] Step a2: In response to the fact that the measurement point identifier of at least one of the multiple measurement points of the mechanism model does not correspond to the measurement point identifier of the mechanism model stored in the database, based on the measurement point name of at least one measurement point, determine whether the measurement point identifier of the related measurement point of at least one measurement point corresponds to the measurement point identifier of the mechanism model stored in the database.

[0078] As a possible implementation example, the historical data of the measurement points are verified within a big data platform using the power plant identification system code of the corresponding mechanism model measurement points to determine the correctness, accuracy, and authenticity of the measurement points. Based on the measurement point names and identifiers of multiple measurement points in the mechanism model, the system searches the database for multiple measurement point identifiers of the mechanism model stored in the database. If the measurement point identifier of at least one measurement point in the mechanism model does not correspond to the measurement point identifiers of the mechanism model stored in the database, it indicates that the measurement point does not meet the accuracy requirements. Based on the measurement point name of at least one measurement point, the system determines whether the measurement point identifiers of the related measurement points of at least one measurement point correspond to the measurement point identifiers of the mechanism model stored in the database.

[0079] Optionally, based on the measurement point name, keywords, and synonyms of the keywords can be determined. Related measurement points can then be found based on these keywords and synonyms, provided their detection value attributes match those of the aforementioned measurement points. It is understandable that data transmission may be incomplete or distorted due to network instability or other reasons. Therefore, it is necessary to verify the power plant identification system code of the measurement points. Multiple measurement points may monitor the same value, only with slight differences in their names; thus, the integrity of the data can be verified through related measurement points.

[0080] Step a3: In response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it is determined that at least one measurement point of the mechanism model has failed the verification. Parameter tuning is performed on at least one measurement point, and the steps of verifying multiple measurement points are repeated.

[0081] As an example of possible implementation, in response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it indicates that neither the measurement point nor its related measurement points meet the requirements. It is determined that at least one measurement point of the mechanism model has failed the verification. Parameter tuning is then performed on at least one measurement point, and the parameter-tuned measurement point is re-verified.

[0082] In some embodiments of this application, step 102 further includes:

[0083] Step b1: Based on the monitoring values ​​of multiple measurement points in the mechanistic model, determine the input attributes of each monitoring value.

[0084] Step b2: For each measuring point, obtain the actual attributes of the monitored value of the measuring point.

[0085] Step b3: Compare the input attributes with the actual attributes to obtain the second comparison result.

[0086] Step b4: In response to the second comparison result that the input attribute does not match the actual attribute, based on the measurement point name, determine the related measurement points of the measurement point, and determine whether the measurement point identifier of the related measurement points corresponds to the measurement point identifier of the mechanism model stored in the database.

[0087] Step b5: In response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it is determined that the measurement point has not passed the verification. The parameter adjustment process is performed on the measurement point, and the steps of verifying multiple measurement points are repeated.

[0088] For example, based on the monitored values ​​of the measurement points according to the mechanistic model, it is determined whether the input attribute of the detected value is an analog signal or a digital signal. If the actual attribute of the detected value is an analog signal and the input attribute is a digital signal, or vice versa, based on the measurement point name, related measurement points are identified, and it is determined whether the measurement point identifiers of the related measurement points correspond to the measurement point identifiers of the mechanistic model stored in the database. If the measurement point identifier of a related measurement point does not correspond to the measurement point identifiers of the mechanistic model stored in the database, it is determined that the measurement point has failed verification. Parameter tuning is then performed on the measurement point, and the steps of verifying multiple measurement points are repeated.

[0089] In some embodiments of this application, step 102 further includes:

[0090] Step c1: For each measuring point, compare the measuring point name with the actual measuring point name in the DCS (Distributed Control System).

[0091] Step c2: In response to the fact that the measurement point name does not match the actual measurement point name in the DCS, based on the measurement point name, determine the related measurement points of the measurement point, and determine whether the measurement point name of the related measurement points matches the actual measurement point name in the DCS.

[0092] Step c3: In response to the fact that the name of the relevant measuring point does not match the actual name of the measuring point in the DCS, it is determined that the measuring point has not passed the verification. The parameter adjustment process is performed on the measuring point, and the steps of verifying multiple measuring points are repeated.

[0093] As a possible implementation example, the actual required measurement point is obtained through relevant calculation formulas built within the DCS logic configuration. However, the measurement point found on the big data platform is not built within the DCS logic configuration. It needs to be built within the big data platform by combining various operator blocks to implement the mechanism calculation formulas. The final calculation result of the mechanism formula must be consistent with the measurement point on the actual DCS screen. Inconsistencies between the measurement point names on the big data platform and those on the platform will result in only similar measurement point names being searched on the platform. If the data of the found measurement point does not match the actual measurement point data, it is necessary to search for similar measurement point names again in the big data platform's point table and compare them until they match the actual situation.

[0094] As another possible example, when searching for measurement points through a big data platform, there may be instances where no historical data for the measurement points can be found. This may be due to current network issues or instability of the big data platform. This problem can be solved by refreshing the current webpage or reloading the page.

[0095] As another possible implementation example, if the required measurement points cannot be found through the above methods, the missing measurement points (i.e., measurement points that have not passed verification) used by each early warning model corresponding to each discipline (including: chemistry, boiler, steam turbine and thermal control) are counted, and relevant units are contacted to add the missing measurement points.

[0096] Step 103: In response to the verification of multiple measurement points, determine the values ​​of each of the multiple operator blocks in the mechanism model.

[0097] For example, the delay time can be set for the delay operator block: the set time is any number, and the corresponding delay time is set according to the logic requirements of the mechanism model diagram, thus realizing the delayed output of the switch signal; the pulse time can be set for the pulse operator block: the set time is any number, and the corresponding pulse time is set according to the logic requirements of the mechanism model diagram, thus realizing the pulse output of the switch signal. A comparison setpoint can be set within the greater than or less than operator block: the setpoint is any number, and the magnitude of the comparison setpoint is set according to the value of the input measurement point, thus realizing the comparison output of the switch signal; a comparison setpoint can be set within the large or small selection operator block: the setpoint is any number, and the magnitude of the comparison setpoint is set according to the value of the input measurement point, thus realizing the selection output of the analog signal. The calculation time for the rate operator block can be set: the set time is any number, and the rate value of the input measurement point during this time period is calculated based on the input value of the input measurement point, thus realizing the rate output of the analog signal. An arbitrary value can be set for the constant operator block: the corresponding value is set according to the logic requirements of the mechanism model diagram, thus realizing the output of the analog or switch signal. Set the corresponding calculation time for the accumulator block: The set time can be any number. Based on the logical requirements of the mechanism model diagram, set the corresponding total accumulation time or total cumulative time, and achieve analog signal output. Set the corresponding function relationship value for the function operator block: Based on the logical requirements of the mechanism model diagram, set the corresponding function relationship value, and achieve analog signal output. Set the set values ​​for input terminals 1 and 2 of the 2-to-1 multiplexer block: Based on the logical requirements of the mechanism model diagram, set the set values ​​for input terminals 1 and 2 of the 2-to-1 multiplexer block. When the enable terminal of the 2-to-1 multiplexer block is TRUE, use the set value set at input terminal 1; when the enable terminal of the 2-to-1 multiplexer block is FALSE, use the set value set at input terminal 2, and achieve different selection outputs of the analog signal. Set the fixed values ​​of input terminal 1 and input terminal 2 for the operation operator blocks (including addition, subtraction, multiplication, division, etc.): According to the logic requirements of the mechanism model diagram, set the fixed values ​​of input terminal 1 and input terminal 2 of the operation operator blocks, and realize the operation function of analog signals.

[0098] Optionally, when verifying the judgment logic one by one, since the input measurement point data is real-time measurement point data, it is usually difficult to trigger the early warning function of the mechanism model. Therefore, the setpoint can be manually changed to achieve the early warning effect. The function of directly triggering the early warning can be achieved by changing the judgment setpoint within the operator block. In this case, the time of changing the operator block setpoint needs to be recorded. If the judgment logic is triggered immediately after the setpoint is changed, the verification is successful. If the judgment logic is not triggered immediately, check if there is a delay operator block set in the mechanism model diagram. By checking the delay time, the early warning triggering time can be determined. If the judgment logic is triggered only after the set delay time, the verification is successful.

[0099] Step 104: Input the detection values ​​of multiple measuring points into the mechanism model to obtain the early warning results output by the mechanism model based on the fixed values ​​of multiple operator blocks and the detection values ​​of multiple measuring points.

[0100] In some embodiments of this application, after step 104, the method further includes:

[0101] Step d1: Obtain the trigger time of the mechanism model, and determine whether the mechanism model is triggered within the preset time based on the trigger time.

[0102] As a possible implementation example, the trigger time for the mechanistic model to issue an early warning based on the detection values ​​of multiple measurement points is obtained, and this trigger time is compared with a preset trigger duration to determine whether the mechanistic model has been triggered within the preset duration.

[0103] Step d2: In response to the mechanism model not being triggered within the preset time, determine whether there is a delay operator block in the mechanism model.

[0104] Step d3: In response to the existence of a delay operator block in the mechanism model, determine the delay time of the delay operator block.

[0105] It is understandable that if a delay operator block is set in the mechanism model, the mechanism model needs to be triggered according to the fixed value in the delay operator block.

[0106] Step d4: Based on the delay time and trigger time, determine whether the mechanism model should postpone triggering according to the delay time.

[0107] Step d5: In response to the mechanism model not delaying the trigger according to the delay time, parameter tuning is performed on the mechanism model.

[0108] Step 105: Compare the warning result with the actual result to obtain the first comparison result.

[0109] Step 106: In response to the fact that the first comparison results of each mechanism model meet the preset requirements, the multiple mechanism models are jointly tested based on the preset total model and the multiple measurement point data of each of the multiple mechanism models to obtain the test results.

[0110] In this embodiment, the overall model is a supercritical unit early warning model generated based on multiple mechanism models.

[0111] As an example of a possible implementation, in response to the fact that the first comparison results of each mechanism model meet the preset requirements, it indicates that all mechanism models have passed the test. Since there is a correlation between the output values ​​of multiple mechanism models, it is necessary to conduct further joint testing on multiple mechanism models using the overall model generated based on the above correlation.

[0112] In some embodiments of this application, step 106 includes:

[0113] Step e1 involves inputting the measurement data from multiple measurement points of each of the multiple mechanistic models into the overall model.

[0114] Step e2: Obtain the warning results of each of the multiple mechanistic models and the warning result of the overall model.

[0115] Step e3: In response to the fact that the warning results of each of the multiple mechanistic models and the warning result of the overall model are consistent with their respective actual results, the overall model is determined to have passed the test.

[0116] Step e4: In response to the fact that at least one of the warning results of the multiple mechanistic models and the warning result of the overall model is not consistent with their respective real results, it indicates that there is mutual interference among the warning input, judgment and output of the multiple mechanistic models, and the overall model is adjusted.

[0117] For example, by utilizing the expert boundary conditions of the fault early warning information database for a multi-disciplinary, multi-key system, a corresponding mechanism model is trained. The effectiveness of the input measurement point data for the mechanism model is verified on a big data platform, validating the functional integrity and authenticity of each operator block. Using the mechanism analysis technology of the fault early warning information database as guidance, a model diagram logic configuration is built using each operator. Mechanism logic analysis is completed by setting preset values, integrating multiple judgment conditions, and adding preset judgment conditions. The judgment logic is verified one by one and then merged for overall verification. Multiple types of mechanism models are combined for testing. Different types of mechanism models are selected for synchronous testing on the big data platform. If the early warning input, judgment, and output quantities do not interfere with each other, and each early warning can send results accurately and promptly, the test is considered successful and effective. Through the full-process verification of this mechanism model testing outline, accurate analysis and judgment of accident conditions are achieved, completing quantitative early warning for the system or equipment.

[0118] According to the method for testing early warning models of supercritical units under mechanism diagnostic analysis according to embodiments of this application, the following steps are taken: First, acquire measurement point data for multiple mechanism models. For each mechanism model, verify the multiple measurement points based on their names, identifiers, and monitoring values. If all measurement points pass verification, determine the setpoints of multiple operator blocks in the mechanism model. Input the detection values ​​of the multiple measurement points into the mechanism model to obtain an early warning result output by the mechanism model based on the setpoints of the multiple operator blocks and the detection values ​​of the multiple measurement points. Compare the early warning result with the actual result to obtain a first comparison result. If the first comparison result for each mechanism model meets preset requirements, perform joint testing on the multiple mechanism models based on a preset overall model and the measurement point data of each mechanism model to obtain test results. This verifies the accuracy, timeliness, and authenticity of the mechanism models and measurement point data. The problem of intelligent operation and smart safety system construction in thermal power plants was solved. A fault early warning information database for multiple key systems across all disciplines was created and formulated. Based on this mechanism diagnosis, early warning models for various types and systems were divided. The mechanism models were deployed to a big data platform and standardized and unified.

[0119] Figure 2 This is a structural block diagram of an early warning model testing device for a supercritical unit under mechanism diagnosis analysis, as described in an embodiment of this application.

[0120] like Figure 2 As shown, the early warning model test device for this supercritical unit under mechanism diagnostic analysis includes:

[0121] The acquisition module 201 is used to acquire the measurement point data of each of the multiple mechanism models; wherein, the multiple mechanism models are pre-built based on the fault early warning data of multiple professional fields corresponding to the supercritical unit; the measurement point data includes the measurement point name, measurement point identifier and monitoring value of each of the multiple measurement points;

[0122] The verification module 202 is used to verify multiple measurement points for each mechanism model based on their respective measurement point names, measurement point identifiers, and monitoring values.

[0123] Module 203 is used to determine the values ​​of each of the multiple operator blocks in the mechanism model in response to the fact that multiple measurement points have passed verification.

[0124] The input module 204 is used to input the detection values ​​of multiple measuring points into the mechanism model to obtain the early warning result output by the mechanism model based on the fixed values ​​of multiple operator blocks and the detection values ​​of multiple measuring points.

[0125] The comparison module 205 is used to compare the warning result with the actual result to obtain the first comparison result;

[0126] The test module 206 is used to jointly test multiple mechanism models based on the preset total model and multiple measurement point data of each mechanism model in response to the first comparison result of each mechanism model meeting the preset requirements, and obtain the test results; wherein, the total model is a supercritical unit early warning model generated based on multiple mechanism models.

[0127] According to the embodiments of this application, the early warning model testing device for supercritical units under mechanism diagnostic analysis acquires measurement point data for multiple mechanism models. For each mechanism model, based on the measurement point name, measurement point identifier, and monitoring value of each measurement point, multiple measurement points are verified. In response to multiple measurement points passing verification, the set values ​​of multiple operator blocks in the mechanism model are determined. The detection values ​​of each measurement point are input into the mechanism model to obtain the early warning result output by the mechanism model based on the set values ​​of each operator block and the detection values ​​of each measurement point. The early warning result is compared with the actual result to obtain a first comparison result. In response to the first comparison result of each mechanism model meeting preset requirements, multiple mechanism models are jointly tested based on a preset overall model and the multiple measurement point data of each mechanism model to obtain test results. This verifies the accuracy, timeliness, and authenticity of the mechanism models and measurement point data. The problem of intelligent operation and smart safety system construction in thermal power plants was solved. A fault early warning information database for multiple key systems across all disciplines was created and formulated. Based on this mechanism diagnosis, early warning models for various types and systems were divided. The mechanism models were deployed to a big data platform and standardized and unified.

[0128] Figure 3 This is a block diagram of an electronic device according to an embodiment of this application. For example... Figure 3 As shown, the electronic device may include: a transceiver 31, a processor 32, and a memory 33.

[0129] Processor 32 executes computer execution instructions stored in memory, causing processor 32 to perform the scheme in the above embodiments. Processor 32 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0130] The memory 33 is connected to the processor 32 via the system bus and completes communication between them. The memory 33 is used to store computer program instructions.

[0131] Transceiver 31 can be used to obtain the task to be run and its configuration information.

[0132] The system bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus. Transceivers are used to enable communication between database access devices and other computers (e.g., clients, read-write libraries, and read-only libraries). Memory may include random access memory (RAM) and may also include non-volatile memory.

[0133] The electronic device provided in this application embodiment can be the terminal device described in the above embodiments.

[0134] This application also provides a chip for executing instructions, which is used to execute the message processing method described in the above embodiments.

[0135] This application also provides a computer-readable storage medium storing computer instructions that, when executed on a computer, cause the computer to perform the message processing method described in the above embodiments.

[0136] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when the at least one processor executes the computer program, it can implement the technical solution of the message processing method in the above embodiments.

[0137] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0138] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for testing an early warning model of a supercritical unit under mechanistic diagnostic analysis, characterized in that, The method includes: The measurement point data of each of the multiple mechanistic models are obtained; wherein, the multiple mechanistic models are all pre-built based on fault early warning data of multiple professional fields corresponding to supercritical units; the measurement point data includes the measurement point name, measurement point identifier and monitoring value of each of the multiple measurement points; For each mechanism model, based on the measurement point name, measurement point identifier, and monitoring value of each of the multiple measurement points of the mechanism model, the multiple measurement points are respectively verified. The measurement point identifier is the power plant identification system code of the measurement point. In response to the fact that all the multiple measurement points have passed verification, the fixed values ​​of each of the multiple operator blocks in the mechanism model are determined; The monitoring values ​​of each of the multiple measuring points are input into the mechanism model to obtain the early warning result output by the mechanism model based on the fixed values ​​of each of the multiple operator blocks and the monitoring values ​​of each of the multiple measuring points; The warning result is compared with the actual result to obtain the first comparison result; In response to the fact that the first comparison results of each mechanism model meet the preset requirements, the multiple mechanism models are jointly tested based on the preset overall model and the multiple measurement point data of each of the multiple mechanism models to obtain the test results; wherein, the overall model is a supercritical unit early warning model generated based on the multiple mechanism models; The measurement point names, measurement point identifiers, and monitoring values ​​of the multiple measurement points based on the aforementioned mechanism model are used to verify the multiple measurement points, including: Based on the measurement point names and measurement point identifiers of multiple measurement points in the aforementioned mechanism model, search for the multiple measurement point identifiers of the aforementioned mechanism model that are pre-stored in the database. In response to the fact that the measurement point identifier of at least one of the multiple measurement points of the mechanism model does not correspond to the measurement point identifier of the mechanism model stored in the database, based on the measurement point name of the at least one measurement point, it is determined whether the measurement point identifier of the related measurement points of the at least one measurement point corresponds to the measurement point identifier of the mechanism model stored in the database, wherein the monitoring value attribute of the related measurement point is consistent with the monitoring value attribute of the at least one measurement point, and based on the measurement point name, the keywords, key words and synonyms of the keywords in the measurement point name are determined, and the related measurement points of the measurement point are searched based on the keywords, key words and synonyms of the keywords; In response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it is determined that at least one measurement point of the mechanism model has failed the verification. Then, parameter tuning is performed on the at least one measurement point, and the steps of verifying the multiple measurement points are re-executed.

2. The method according to claim 1, characterized in that, The method further includes verifying the measurement points based on the mechanistic model by specifying their names, identifiers, and monitored values, as well as verifying the measurement points themselves. Based on the monitoring values ​​of multiple measuring points in the aforementioned mechanism model, the input attributes of each monitoring value are determined. For each measuring point, obtain the actual attributes of the monitored value of that measuring point; The input attribute is compared with the actual attribute to obtain a second comparison result; In response to the second comparison result that the input attribute does not match the actual attribute, based on the measurement point name, the related measurement points of the measurement point are determined, and it is determined whether the measurement point identifier of the related measurement points corresponds to the measurement point identifier of the mechanism model stored in the database; In response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it is determined that the measurement point has failed the verification, and parameter tuning is performed on the measurement point. The steps of verifying the multiple measurement points are repeated.

3. The method according to claim 1, characterized in that, The method further includes verifying the measurement points based on the mechanistic model by specifying their names, identifiers, and monitored values, as well as verifying the measurement points themselves. For each measuring point, the measuring point name is compared with the actual measuring point name in the distributed control system (DCS). In response to the fact that the name of the measuring point does not match the actual name of the DCS measuring point, based on the name of the measuring point, the relevant measuring points of the measuring point are determined, and it is determined whether the name of the relevant measuring points matches the actual name of the DCS measuring point. If the name of the relevant measuring point does not match the actual measuring point name of the DCS, it is determined that the measuring point has failed the verification. The parameter adjustment process is performed on the measuring point, and the steps of verifying the multiple measuring points are re-executed.

4. The method according to claim 1, characterized in that, After inputting the monitoring values ​​of the plurality of measuring points into the mechanism model, the method further includes: The trigger time of the mechanism model is obtained, and based on the trigger time, it is determined whether the mechanism model is triggered within a preset time period; In response to the fact that the mechanism model is not triggered within a preset time period, it is determined whether there is a delay operator block in the mechanism model; In response to the presence of a delay operator block in the mechanism model, the delay time of the delay operator block is determined; Based on the delay time and the trigger time, determine whether the mechanism model should postpone triggering according to the delay time; In response to the mechanism model not delaying the trigger according to the delay time, the mechanism model is subjected to parameter tuning.

5. The method according to claim 1, characterized in that, The method involves jointly testing the multiple mechanistic models based on multiple measurement point data from a preset overall model and each of the multiple mechanistic models to obtain test results, including: The measurement data from multiple measurement points of each of the multiple mechanistic models are input into the overall model; The warning results of each of the multiple mechanistic models and the warning result of the overall model are obtained respectively; If the warning results of each of the multiple mechanistic models and the warning result of the overall model are consistent with their respective actual results, the overall model is determined to have passed the test. In response to at least one of the warning results of the multiple mechanistic models and the warning result of the overall model not being consistent with their respective actual results, the overall model is subjected to parameter tuning.

6. The method according to claim 1, characterized in that, The fault warning data in the multiple professional fields includes any one or more of the following: fault warning information data of steam turbine, fault warning information data of boiler, fault warning information data of electrical, fault warning information data of thermal control, fault warning information data of chemistry, fault warning information data of ash removal, fault warning information data of desulfurization, and fault warning information data of coal conveying system.

7. A test device for an early warning model of a supercritical unit under mechanism diagnostic analysis, characterized in that, The device includes: The acquisition module is used to acquire the measurement point data of each of the multiple mechanism models; wherein, the multiple mechanism models are all pre-built based on fault early warning data of multiple professional fields corresponding to supercritical units; the measurement point data includes the measurement point name, measurement point identifier and monitoring value of each of the multiple measurement points; The verification module is used to verify the multiple measuring points for each mechanism model based on their respective measuring point names, measuring point identifiers, and monitoring values. The measuring point identifier is the power plant identification system code of the measuring point. A determination module is used to determine the set value of each of the multiple operator blocks in the mechanism model in response to the fact that all the multiple measurement points have passed the verification. The input module is used to input the monitoring values ​​of the multiple measuring points into the mechanism model to obtain the early warning result output by the mechanism model based on the fixed values ​​of the multiple operator blocks and the monitoring values ​​of the multiple measuring points. The comparison module is used to compare the warning result with the actual result to obtain a first comparison result; The testing module is used to jointly test the multiple mechanistic models based on the preset overall model and the multiple measurement point data of each of the multiple mechanistic models in response to the fact that the first comparison result of each mechanistic model meets the preset requirements, and to obtain the test results; wherein, the overall model is a supercritical unit early warning model generated based on the multiple mechanistic models; The verification module is used to search for multiple measurement point identifiers of the mechanism model that are pre-stored in the database, based on the measurement point names and measurement point identifiers of the multiple measurement points of the mechanism model. In response to the fact that the measurement point identifier of at least one of the multiple measurement points of the mechanism model does not correspond to the measurement point identifier of the mechanism model stored in the database, based on the measurement point name of the at least one measurement point, it is determined whether the measurement point identifier of the related measurement points of the at least one measurement point corresponds to the measurement point identifier of the mechanism model stored in the database, wherein the monitoring value attribute of the related measurement point is consistent with the monitoring value attribute of the at least one measurement point, and based on the measurement point name, the keywords, key words and synonyms of the keywords in the measurement point name are determined, and the related measurement points of the measurement point are searched based on the keywords, key words and synonyms of the keywords; In response to the fact that the measurement point identifier of the relevant measurement point does not correspond to the measurement point identifier of the mechanism model stored in the database, it is determined that at least one measurement point of the mechanism model has failed the verification. Then, parameter tuning is performed on the at least one measurement point, and the steps of verifying the multiple measurement points are re-executed.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

Images