Pms-based device state monitoring and abnormality analysis strategy method and system

Abstract

The application discloses a PMS-based equipment state monitoring and abnormality analysis strategy method and system, belongs to the technical field of equipment management, and comprises the following steps: acquiring equipment basic information, real-time running state data, historical maintenance records and risk grade parameters in an equipment management system, generating a comprehensive equipment state data set, calculating a dynamic health attention index, generating a strategy chain instruction set, scheduling a sensor group to execute data collection operations, generating a target monitoring data set, inputting a multistage self-adaptive analysis tree in combination with an analysis strategy instruction, generating a structured analysis report, and matching a maintenance decision suggestion from a maintenance knowledge base and outputting when preset conditions of a decision trigger strategy instruction are met. The application adopts a dynamic health attention index to drive strategy chain instruction generation, combines a multistage self-adaptive analysis tree with a hierarchical diagnosis and closed-loop parameter optimization mechanism, and can realize accurate allocation of monitoring resources, accurate positioning of fault causes, economic optimization of maintenance decisions and self-adaptive continuous upgrading of the system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device management, in particular to a device state monitoring and abnormal analysis strategy method and system based on PMS. BACKGROUND

[0002] The device management system plays a core role in modern industrial operation and maintenance, and realizes state monitoring by integrating device basic information, real-time operation data and historical maintenance records. The existing technology generally relies on preset thresholds or fixed cycles for abnormal detection and maintenance decision, which belongs to the category of static management.

[0003] The current mainstream scheme adopts a unified monitoring strategy to collect data from all devices, analyzes real-time data through a single diagnostic model, and triggers maintenance recommendations according to fixed rules. Some systems introduce basic risk assessment, but the weight coefficients are mostly statically set, lacking a dynamic adjustment mechanism.

[0004] However, the existing scheme has rigid monitoring resource allocation, resulting in waste of resources for low-risk devices and insufficient monitoring of high-risk devices. The single diagnostic model is difficult to cope with complex fault scenarios, with high false positive and false negative rates. The maintenance decision lacks real-time operating condition adaptability, and is prone to over-maintenance or delayed processing. The system parameters are fixed and cannot be automatically optimized according to the operation effect, and the accuracy decreases over time. SUMMARY

[0005] To solve the above problems, the present application provides a device state monitoring and abnormal analysis strategy method and system based on PMS, which uses a dynamic health attention index to drive strategy chain instruction generation, combines a multi-level adaptive analysis tree hierarchical diagnosis and closed-loop parameter optimization mechanism, and can realize precise allocation of monitoring resources, accurate positioning of fault causes, economic optimization of maintenance decisions and self-adaptive continuous upgrading of the system.

[0006] The above object can be achieved by the following scheme:

[0007] The device state monitoring and abnormal analysis strategy method and system based on PMS comprises obtaining device basic information, real-time operation state data, historical maintenance records and preset risk level parameters in a device management system, and generating a comprehensive device state data set; calculating a dynamic health attention index according to the comprehensive device state data set; generating a strategy chain instruction set including monitoring strategy instructions, analysis strategy instructions and decision trigger strategy instructions based on the dynamic health attention index; scheduling a preset sensor group to perform a data collection operation according to the monitoring strategy instructions, and generating a target monitoring data set; inputting the target monitoring data set and the analysis strategy instructions into a preset multi-level adaptive analysis tree, performing a hierarchical diagnosis operation, and generating a structured analysis report; and when the structured analysis report meets the preset conditions of the decision trigger strategy instructions, matching a maintenance decision suggestion from a preset maintenance knowledge base and outputting.

[0008] Optionally, the calculating the dynamic health attention index according to the integrated equipment state data set comprises: extracting an abnormal feature parameter from the real-time operation state data, extracting a failure frequency parameter from the historical maintenance record, and extracting an equipment value parameter from the equipment basic information; generating a risk weight factor by using the abnormal feature parameter and the failure frequency parameter; generating a consequence severity coefficient by fusing the equipment value parameter and the risk level parameter; and performing weighted calculation on the risk weight factor and the consequence severity coefficient to generate the dynamic health attention index.

[0009] Optionally, the generating a strategy chain instruction set comprising a monitoring strategy instruction, an analysis strategy instruction, and a decision trigger strategy instruction based on the dynamic health attention index comprises: matching a preset monitoring intensity rule library according to the dynamic health attention index to generate a monitoring strategy instruction comprising a data acquisition frequency parameter, a sensor accuracy parameter, and a monitoring target parameter; activating a preset analysis depth rule library according to the dynamic health attention index to generate an analysis strategy instruction comprising a diagnosis model combination identifier and an analysis path identifier; calling a preset decision trigger rule library based on an equipment type identifier in the equipment basic information to generate a decision trigger strategy instruction comprising a confidence threshold and a failure mode matching condition; and integrating the monitoring strategy instruction, the analysis strategy instruction, and the decision trigger strategy instruction to obtain the strategy chain instruction set.

[0010] Optionally, the inputting the target monitoring data set and the analysis strategy instruction into a preset multi-level adaptive analysis tree to perform a hierarchical diagnosis operation to generate a structured analysis report comprises: selecting a primary analysis node or a high-level analysis node in the multi-level adaptive analysis tree according to an analysis path identifier in the analysis strategy instruction; performing a threshold check operation on the target monitoring data set through the primary analysis node to generate a primary diagnosis identifier; and when the primary diagnosis identifier satisfies a preset depth trigger condition, activating the high-level analysis node to perform multi-model fusion diagnosis to generate the structured analysis report.

[0011] Optionally, the activating the high-level analysis node to perform multi-model fusion diagnosis to generate the structured analysis report when the primary diagnosis identifier satisfies a preset depth trigger condition comprises: when the primary diagnosis identifier satisfies a preset depth trigger condition, activating the high-level analysis node and calling a preset machine learning model group according to the diagnosis model combination identifier; inputting the target monitoring data set into the machine learning model group to perform parallel analysis to generate a model output result set; and performing a confidence weighted fusion operation on the model output result set to generate a failure root cause parameter and an impact assessment parameter in the structured analysis report.

[0012] Optionally, the matching maintenance decision suggestion from the preset maintenance knowledge base comprises: extracting a fault mode code and a device type identifier from the structured analysis report; searching in the preset maintenance knowledge base according to the fault mode code to obtain a candidate maintenance scheme set; extracting an operating condition parameter from the device basic information, and performing cost optimization screening by using the operating condition parameter and the candidate maintenance scheme set to generate a maintenance decision suggestion.

[0013] Optionally, the method further comprises: collecting the policy chain instruction set, the target monitoring data set, the structured analysis report and a maintenance execution record to generate a policy execution chain data set; extracting a device state improvement parameter from the maintenance execution record; calculating a policy performance evaluation value according to the device state improvement parameter; and adjusting the weight coefficients of the risk weight factor and the consequence severity coefficient when the policy performance evaluation value is lower than a preset optimization trigger threshold.

[0014] Optionally, the adjusting the weight coefficients of the risk weight factor and the consequence severity coefficient comprises: extracting a false positive case and a false negative case from the policy execution chain data set to generate an optimization training sample set; correcting the weight coefficient of the risk weight factor according to the optimization training sample set; extracting an actual maintenance cost parameter from the maintenance execution record; and correcting the weight coefficient of the consequence severity coefficient by using the actual maintenance cost parameter.

[0015] Optionally, the method comprises: feeding back the policy performance evaluation value to the multi-level adaptive analysis tree; adjusting an activation priority of a machine learning model group in the high-level analysis node according to the policy performance evaluation value; and optimizing a confidence threshold in the decision trigger strategy instruction based on a fault verification result in the maintenance execution record.

[0016] Based on the same inventive concept, the application also provides a PMS-based equipment state monitoring and abnormality analysis system, which comprises: a state data acquisition module, configured to acquire equipment basic information, real-time running state data, historical maintenance records and preset risk level parameters in an equipment management system, and generate a comprehensive equipment state data set; a health index calculation module, configured to calculate a dynamic health attention index according to the comprehensive equipment state data set; a strategy instruction generation module, configured to generate a strategy chain instruction set comprising monitoring strategy instructions, analysis strategy instructions and decision trigger strategy instructions based on the dynamic health attention index; a monitoring data acquisition module, configured to execute data acquisition operations by preset sensors according to the monitoring strategy instructions, and generate a target monitoring data set; a report generation module, configured to input the target monitoring data set and the analysis strategy instructions into a preset multi-level adaptive analysis tree, execute hierarchical diagnosis operations, and generate a structured analysis report; and a decision output module, configured to match maintenance decision suggestions from a preset maintenance knowledge base and output the same when the structured analysis report meets preset conditions of the decision trigger strategy instructions.

[0017] Compared with the prior art, the application has the following advantages:

[0018] 1. The application quantifies the equipment risk level by the dynamic health attention index, automatically matches the differentiated monitoring strategy, significantly reduces the resource occupation of low-risk equipment, and at the same time guarantees the fine monitoring of high-risk equipment, effectively optimizing the utilization rate of sensor resources and computing resources;

[0019] 2. The application adopts the multi-level adaptive analysis tree to execute hierarchical diagnosis, the primary nodes rapidly screen the abnormality, the advanced nodes fuse multiple models to perform deep analysis, greatly reducing the real-time computing load and avoiding misjudgment and missed judgment, and ensuring the accurate positioning and root cause analysis of complex faults;

[0020] 3. The application combines the equipment real-time working condition and the maintenance knowledge base to perform cost optimization screening, generates the maintenance suggestions taking into account the reliability and economy, avoids excessive maintenance or insufficient maintenance, and at the same time dynamically controls the output time through the decision trigger strategy, guaranteeing the timely response of key faults;

[0021] 4. The application dynamically corrects the health index weight coefficient, model priority and confidence threshold based on the strategy performance evaluation value and the maintenance execution record, continuously improves the risk prediction accuracy and decision reliability, and forms a complete feedback closed loop from execution to optimization.

[0022] Other features and advantages of the application will be set forth in the specification, and in part will become apparent from the specification, or can be learned by practice of the application. The objectives and other advantages of the application can be realized and attained by the structure particularly pointed out in the specification, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings described below are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0024] Figure 1 is a flowchart of a PMS-based equipment state monitoring and abnormality analysis strategy method according to an embodiment of the present application.

[0025] Figure 2 is a structural diagram of a PMS-based equipment state monitoring and abnormality analysis system according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0027] With reference to Figure 1 , one embodiment of the present application proposes a PMS-based equipment state monitoring and abnormality analysis strategy method, adopts a dynamic health attention index driven strategy chain instruction generation, combines a multi-level adaptive analysis tree hierarchical diagnosis and closed-loop parameter optimization mechanism, and can realize accurate allocation of monitoring resources, accurate positioning of fault root causes, economic optimization of maintenance decisions and self-adaptive continuous upgrading of the system.

[0028] The method according to the embodiment specifically includes:

[0029] Obtaining equipment basic information, real-time running state data, historical maintenance records and preset risk level parameters in an equipment management system, and generating a comprehensive equipment state data set;

[0030] Specifically, the device model parameter, the installation date parameter, and the device value parameter are extracted from a device base information library of the device management system; the device running state time series data, including the vibration amplitude parameter, the temperature parameter, and the current parameter, are obtained from a real-time data interface; the historical failure frequency parameter, the maintenance time stamp sequence, and the maintenance measure description text are called from a maintenance record database; and the risk level parameter is read from a preset configuration file, which is predefined as a discrete numerical type identifier according to the importance of the device function. The four types of data are associated and mapped according to the unique identifier of the device, wherein the device base information and the risk level parameter are directly matched through the device code, the real-time running state data are aligned through the device physical location coordinates and the time stamp, and the historical maintenance record is associated through the device serial number and the maintenance event index. When performing the data fusion operation, the non-numerical type field is standardized and coded, and the numerical type field is dimensionally normalized, to generate a structured data table in the unit of the device, wherein each record in the table includes a device identifier field, a base information field set, a real-time state data matrix, a historical maintenance record array, and a risk level field. The operation ensures that the multi-source heterogeneous data is integrated under a unified time and space reference, to form a comprehensive device state data set that can be directly called by subsequent modules.

[0031] calculating a dynamic health attention index according to the comprehensive device state data set;

[0032] generating a strategy chain instruction set including a monitoring strategy instruction, an analysis strategy instruction, and a decision trigger strategy instruction based on the dynamic health attention index;

[0033] scheduling a preset sensor group to perform a data acquisition operation according to the monitoring strategy instruction, to generate a target monitoring data set;

[0034] Specifically, the analysis monitoring strategy instruction obtains a data acquisition frequency parameter, a sensor accuracy level parameter, and a monitoring target parameter list. The data acquisition frequency parameter is in units of hertz (Hz) and represents the number of samples per second, which is directly read from the monitoring strategy instruction. The sensor accuracy level parameter is a preset level identifier corresponding to different accuracy sensors, and the specific sensor model is determined by querying a preset sensor accuracy mapping table. The monitoring target parameter list includes physical quantity identifiers such as vibration and temperature that need to be monitored, and the identifiers are bound to device measurement point codes. According to the sensor accuracy level parameter, a sensor instance that meets the accuracy requirement is selected from a preset sensor group, and the sensor group is composed of sensor nodes deployed at different physical locations of the device, each node having a unique device code and accuracy attribute. According to the monitoring target parameter list, a sensor subset in the sensor group that matches the target physical quantity is screened. Control commands are sent to the screened sensor subset, and the commands include the data acquisition frequency parameter and a start timestamp. The sensor nodes acquire raw data at the frequency, perform timestamp alignment on the raw data, generate a time series dataset using a hardware clock synchronization protocol, perform normalization processing on the time series dataset according to the sensor raw readings and the upper limit value of the sensor range, and finally merge the normalized data of each sensor node according to the timestamp to generate a structured target monitoring dataset containing device codes, timestamp sequences, monitoring physical quantity identifiers, and corresponding normalized time series dataset values.

[0035] The target monitoring dataset and the analysis strategy instruction are input into a preset multi-level adaptive analysis tree to perform hierarchical diagnosis and generate a structured analysis report.

[0036] When the structured analysis report meets the preset conditions of the decision trigger strategy instruction, a maintenance decision suggestion is matched from a preset maintenance knowledge base and output.

[0037] Specifically, the monitoring strategy is adaptively adjusted through quantitative evaluation of the comprehensive device state, and precise positioning of abnormalities is achieved through a hierarchical diagnosis mechanism. Specifically, first, a dynamic health attention index is generated by fusing device static attributes, real-time running characteristics, historical maintenance records, and preset risk parameters, which quantitatively evaluates the device risk level in real time. Second, monitoring strategy instructions are dynamically generated according to the index value to accurately control the acquisition frequency, accuracy, and monitoring targets of the sensor group. Third, the collected data is input into a multi-level analysis tree for hierarchical diagnosis, with the primary nodes performing rapid threshold screening and the advanced nodes triggering multi-model fusion analysis. Finally, when the diagnosis result meets the preset trigger conditions, the most economical maintenance decision is matched from the knowledge base in combination with the device working condition, forming a closed-loop control from state perception to decision execution.

[0038] Optionally, the calculation of the dynamic health attention index according to the comprehensive device state dataset comprises:

[0039] extracting an abnormal feature parameter from the real-time running state data, extracting a failure frequency parameter from the historical maintenance record, and extracting a device value parameter from the device basic information;

[0040] generating a risk weight factor using the abnormal feature parameter and the failure frequency parameter;

[0041] generating a consequence severity coefficient by fusing the device value parameter and the risk level parameter;

[0042] generating a dynamic health concern index by performing weighted calculation on the risk weight factor and the consequence severity coefficient.

[0043] Specifically, first, specific data parameters need to be extracted from different sources. An abnormal feature parameter that can reflect whether the current running of the device is normal is extracted from the real-time running state data. For example, the abnormal increase of the vibration amplitude of the device or the degree of exceeding the normal range of the running temperature. At the same time, a failure frequency parameter is extracted from the historical maintenance record, which quantifies the frequency of failure of a specific device in the past period of time. Finally, a device value parameter is extracted from the device basic information, which represents the economic value of the device or its importance in the production process. Next is to generate a risk weight factor, which aims to quantify the possibility of device failure. Its calculation method is to combine the abnormal feature parameter monitored in real time with the historical failure frequency parameter. For the risk weight factor , there is:

[0044] ;

[0045] wherein, is the normalized abnormal feature parameter, reflecting the severity of the deviation of the current running state from the normal baseline, which is obtained from the real-time running state data; is the normalized failure frequency parameter, reflecting the frequency of historical failure of the device, which is obtained from the historical maintenance record. and are preset weights for adjusting the relative importance of real-time abnormalities and historical failures in evaluating risks. Then generate a consequence severity coefficient, which aims to quantify the loss or impact that may be caused by the failure of the device. Its calculation method is to fuse the value of the device with the preset risk level. The risk level parameter here is a preset value, which is usually defined by experts according to the criticality of the device in the entire system, such as general, important, and critical levels. For the consequence severity coefficient , there is:

[0046] ;

[0047] wherein, is the normalized equipment value parameter, representing the replacement cost or production loss caused by downtime, which is obtained from the equipment basic information; is the normalized preset risk level parameter, reflecting the functional importance of the equipment; and are preset weights, used to balance the economic value of the equipment itself and its criticality in function. Finally, the calculated risk weight factor and consequence severity coefficient are weighted to generate the final dynamic health attention index. This index combines the possibility of equipment failure and the severity of its consequences, serving as a comprehensive health status evaluation indicator. For the dynamic health attention index , there are:

[0048] ;

[0049] wherein, and are preset weight coefficients, used to balance the possibility of risk occurrence and the severity of consequences, and the sum of the two is usually one to ensure consistency of the results. Through this index, the system can dynamically evaluate and rank the priority of all equipment that needs attention. Instead of relying solely on a single real-time data threshold alarm, it constructs a more comprehensive and profound equipment health evaluation model by integrating multiple dimensions such as real-time state, historical performance, equipment value, and functional importance. This approach can accurately distinguish between equipment with different importance levels that have the same abnormal signals, for example, a slight vibration of an auxiliary cooling fan and a slight pressure fluctuation of a core reactor. The latter will have a much higher dynamic health attention index due to its high consequence severity coefficient. This enables maintenance resources to be accurately directed to the equipment with the highest current risk and the greatest potential loss, achieving a shift from passive response to proactive prevention, from extensive management to precise decision-making, and greatly improving the efficiency of equipment management and the overall reliability of the production system.

[0050] Optionally, generating a strategy chain instruction set containing monitoring strategy instructions, analysis strategy instructions, and decision trigger strategy instructions based on the dynamic health attention index includes:

[0051] matching a preset monitoring intensity rule library according to the dynamic health attention index to generate monitoring strategy instructions containing data acquisition frequency parameters, sensor accuracy parameters, and monitoring target parameters;

[0052] activating a preset analysis depth rule library according to the dynamic health attention index to generate analysis strategy instructions containing diagnosis model combination identifiers and analysis path identifiers;

[0053] Based on the device type identifier in the device basic information, a preset decision trigger rule library is called to generate a decision trigger strategy instruction including a confidence threshold and a fault mode matching condition;

[0054] The monitoring strategy instructions, the analysis strategy instructions and the decision triggering strategy instructions are integrated to obtain a strategy chain instruction set.

[0055] Specifically, the dynamic health concern index is first matched from a preset monitoring intensity rule library. This rule library is a two-dimensional mapping table that stores data acquisition frequency parameters, sensor accuracy parameters, and monitoring target parameters corresponding to different index intervals. The data acquisition frequency parameter is in Hz, the sensor accuracy parameter is a percentage value, and the monitoring target parameter is a string array that identifies key monitoring points. The matching process uses linear interpolation to determine the specific parameter values ​​to ensure the continuity of the index and parameter mapping. Next, the preset analysis depth rule library is activated. This rule library has a decision tree structure and stores diagnostic model combination identifiers and analysis path identifiers corresponding to different index thresholds. The diagnostic model combination identifier is a coded string pointing to a preset machine learning model group such as a random forest or neural network, and the analysis path identifier is a Boolean value indicating whether to select a primary or advanced node. During activation, the index is compared with the preset threshold set and the corresponding identifier is selected. Subsequently, the device type identifier (a classification code such as motor or transformer) is extracted from the basic device information. A preset decision trigger rule library, a hash table structure that stores confidence thresholds and fault mode matching conditions corresponding to different device types, is called. The confidence threshold is a floating-point number between 0 and 1, and the fault mode matching condition is a string pattern. The calling process directly retrieves the identifier to obtain parameters. Finally, the system integrates the generated monitoring strategy instructions, analysis strategy instructions, and decision trigger strategy instructions to form a policy chain instruction set. This integration method uses data structure serialization operations to encapsulate the instruction parameters as JSON-formatted objects. The health concern index dynamically quantifies device risks, adaptively adjusts monitoring and analysis resource allocation, optimizes device management efficiency, avoids resource waste on low-risk devices, and ensures timely diagnosis and accurate decision-making for high-risk devices, improving system response speed and maintenance accuracy.

[0056] Optionally, inputting the target monitoring data set and the analysis strategy instructions into a preset multi-level adaptive analysis tree, performing hierarchical diagnosis operations, and generating a structured analysis report includes:

[0057] selecting a primary analysis node or a high-level analysis node in the multi-level adaptive analysis tree according to the analysis path identifier in the analysis strategy instruction;

[0058] Performing a threshold check operation on the target monitoring data set by the primary analysis node to generate a primary diagnosis identifier;

[0059] When the preliminary diagnosis identifier meets a preset depth trigger condition, the senior analysis node is activated to perform multi-model fusion diagnosis to generate the structured analysis report.

[0060] Specifically, first, the analysis path identifier in the analysis strategy instruction is parsed. The identifier is a Boolean variable. When the value is true, the senior analysis node is selected. When the value is false, the preliminary analysis node is selected. In the data structure of the multi-level adaptive analysis tree, the preliminary analysis node stores a preset threshold rule library, and the senior analysis node stores a machine learning model group interface. When the preliminary analysis node is selected, the real-time monitoring parameters in the target monitoring data set are compared with the early warning threshold in the threshold rule library. The early warning threshold is determined by the device technical manual and historical operation data statistics. The comparison process uses a segmented function for processing. For temperature parameters When the preliminary diagnosis identifier meets a preset depth trigger condition, the senior analysis node is activated to perform multi-model fusion diagnosis to generate the structured analysis report. The overheating alarm identifier is for vibration parameters When the preliminary diagnosis identifier meets a preset depth trigger condition, the senior analysis node is activated to perform multi-model fusion diagnosis to generate the structured analysis report. The mechanical fault identifier, and are the upper threshold values of temperature and vibration, respectively. All alarm identifier combinations form a preliminary diagnosis identifier. When the severity level parameter of the preliminary diagnosis identifier exceeds a preset depth trigger threshold, the senior analysis node is activated. The severity level parameter is calculated according to the alarm number and type weighting. The depth trigger threshold is obtained from the decision trigger rule library according to the device type. Through layered diagnosis, resource dynamic allocation is realized, the preliminary node rapidly screens for abnormalities, the senior node accurately locates faults, the real-time calculation load is reduced, and the accuracy of deep analysis of key faults is ensured.

[0061] Optionally, when the preliminary diagnosis identifier meets a preset depth trigger condition, the senior analysis node is activated to perform multi-model fusion diagnosis to generate the structured analysis report, including:

[0062] When the preliminary diagnosis identifier meets a preset depth trigger condition, the senior analysis node is activated and a preset machine learning model group is called according to the diagnosis model combination identifier.

[0063] The target monitoring data set is input into the machine learning model group to perform parallel analysis, and a model output result set is generated.

[0064] A confidence weighted fusion operation is performed on the model output result set to generate fault root cause parameters and impact evaluation parameters in the structured analysis report.

[0065] Specifically, first verify whether the severity level parameter of the primary diagnosis identifier is greater than the deep trigger threshold, and if the condition is met, activate the advanced analysis node; according to the diagnosis model combination identifier in the analysis strategy instruction, call the corresponding model set from the preset machine learning model library, the identifier is a string array such as "CNN-SVM" pointing to the combination of convolutional neural network and support vector machine. The target monitoring data set is synchronously input into each model for parallel analysis, the convolutional neural network processes the vibration spectrum image data to output the fault probability , the support vector machine processes the temperature time series data to output the fault probability ; all model output results constitute a model output result set. Perform confidence weighted fusion on the result set, and each model weight is allocated according to the historical verification accuracy , and for the fault root cause probability

[0066] ;

[0067] wherein, is the weight of the i-th model, is the fault probability output by the i-th model. Set the fault root cause parameter as the fault type code corresponding to the maximum probability, and the influence evaluation parameter is the probability value itself. The structured analysis report encapsulates the above parameters in XML format, including the fault positioning field and the confidence field. Through multi-model complementary analysis, the diagnostic robustness is improved, the risk of single model misjudgment is reduced, and the recognition ability and result reliability of complex faults are enhanced.

[0068] Optionally, the matching of the maintenance decision suggestion from the preset maintenance knowledge base comprises:

[0069] extracting a fault mode code and a device type identifier from the structured analysis report;

[0070] searching in the preset maintenance knowledge base according to the fault mode code to obtain a candidate maintenance scheme set;

[0071] extracting an operating condition parameter from the device basic information, and performing cost optimization screening on the operating condition parameter and the candidate maintenance scheme set to generate a maintenance decision suggestion.

[0072] Specifically, first, the system parses the structured analysis report to extract the fault mode code and the equipment type identifier; the fault mode code is an alphanumeric combination in the standard fault classification system, such as BRG001 representing bearing wear, and the equipment type identifier is an equipment classification code, such as TM representing transformer. According to the fault mode code, a full-text search is performed in the maintenance knowledge base, which is stored in a graph database, with nodes being fault modes and edges connecting candidate maintenance solutions; the search returns a set of candidate maintenance solutions, including solution description, standard working hours, spare parts list, and historical effectiveness parameters. Subsequently, operating condition parameters are extracted from the equipment basic information, including quantitative indicators such as load rate, environmental temperature, and continuous operation time; cost optimization screening is performed on the set of candidate maintenance solutions, and the comprehensive cost of each solution is calculated :

[0073] ;

[0074] wherein is the spare part cost obtained in real time according to the spare parts list, is the labor cost calculated according to the standard working hours and the unit labor rate, is the downtime loss estimated according to the equipment capacity and the downtime duration; 、 、 are the operating condition adaptive weight coefficients, which are calculated by normalizing the load rate :

[0075] ;

[0076] The final maintenance decision suggestion is generated by selecting the solution with the lowest comprehensive cost. The economic evaluation of the fault characteristics and real-time operating conditions optimizes resource allocation while ensuring maintenance reliability, avoiding secondary losses caused by excessive maintenance or insufficient maintenance.

[0077] Optionally, the method further comprises:

[0078] Collecting the policy chain instruction set, the target monitoring data set, the structured analysis report, and the maintenance execution record to generate a policy execution chain data set;

[0079] Extracting the device state improvement parameters from the maintenance execution record;

[0080] Calculating a policy effectiveness evaluation value according to the device state improvement parameters;

[0081] When the policy effectiveness evaluation value is lower than a preset optimization trigger threshold, adjusting the weight coefficients of the risk weight factor and the consequence severity coefficient.

[0082] Optionally, the adjustment of the weight coefficients of the risk weight factor and the consequence severity coefficient comprises:

[0083] Extracting false positive cases and missed negative cases from the policy execution chain dataset to generate an optimized training sample set;

[0084] Modifying the weight coefficient of the risk weight factor according to the optimized training sample set;

[0085] extracting actual maintenance cost parameters from the maintenance execution record;

[0086] The actual maintenance cost parameter is used to modify the weight coefficient of the consequence severity coefficient.

[0087] Specifically, first collect the policy chain instruction set, target monitoring data set, structured analysis report and maintenance execution record, and serialize these data through the data integration module to generate the policy execution chain data set; the serialization process is encapsulated in JSON format to ensure that the data fields include instruction parameters, monitoring values, report content and maintenance logs. Then, extract the equipment status improvement parameters from the maintenance execution record, including the percentage of failure rate reduction and the average trouble-free operation time increment. The policy effectiveness evaluation value is obtained by calculating the difference between the equipment status improvement parameters before and after maintenance; the policy effectiveness evaluation value is defined as the normalized average value of multiple equipment status improvement parameters. For the policy effectiveness evaluation value ,have:

[0088] ;

[0089] in Indicates the status value after maintenance, Indicates the status value before maintenance. Improve the number of parameters for device status, and Extracted directly from the maintenance execution records, all values ​​are normalized to the range of [0,1]. When the strategy effectiveness evaluation value is lower than the preset optimization trigger threshold, false alarm cases and missed cases are extracted from the strategy execution chain data set. The optimization trigger threshold is a preset constant in the range of [0.6,0.8]. False alarm cases are defined as records that indicate faults in the structured analysis report but are actually fault-free, and missed cases are defined as records that are not reported but actually have faults. The extraction method is based on data field comparison to generate an optimized training sample set. The optimized training sample set contains case feature vectors and labels. The weight coefficient of the risk weight factor is corrected according to the optimized training sample set. The correction process includes calculating the false alarm rate FP and missed alarm rate FN of the optimized training sample set. The false alarm rate is the number of false alarm cases divided by the total number of cases, and the missed alarm rate is the number of missed alarm cases divided by the total number of cases. For the weight coefficient of the updated risk weight factor ,have:

[0090] ;

[0091] wherein, is a weight coefficient of a current risk weight factor, is a false alarm rate, is a missed alarm rate, is a preset learning rate range in [0.01, 0.1]. At the same time, an actual maintenance cost parameter is extracted from the maintenance execution record, which is a normalized comprehensive cost value including spare part cost, labor cost and downtime loss. For the actual maintenance cost parameter , there is:

[0092] ;

[0093] wherein, , , is extracted from the maintenance execution record field, is a preset maximum cost reference value. Finally, the weight coefficient of the consequence severity coefficient is corrected by using the actual maintenance cost parameter. For the weight coefficient of the corrected consequence severity coefficient , there is:

[0094] ;

[0095] wherein is a current weight coefficient, is a preset adjustment coefficient range in [0.05, 0.1], has been normalized. Through the closed-loop feedback mechanism, the weight coefficient is dynamically optimized to improve the accuracy of the dynamic health attention index; the adaptability of the system is enhanced, the false alarm and missed alarm rates are reduced, the maintenance resource allocation is optimized, the equipment failure prediction accuracy is improved, and the overall maintenance cost is reduced.

[0096] Optionally, the method comprises:

[0097] feedback the policy performance evaluation value to the multi-level adaptive analysis tree;

[0098] adjust the activation priority of the machine learning model group in the high-level analysis node according to the policy performance evaluation value;

[0099] optimize the confidence threshold in the decision trigger strategy instruction based on the failure verification result in the maintenance execution record.

[0100] Specifically, when feeding back the strategy effectiveness evaluation value to the multi-level adaptive analysis tree, it is necessary to open a shared memory area in the analysis tree data structure to store the strategy effectiveness evaluation value; this value is transmitted to the analysis tree control module through the data bus, and the control module parses the data according to the preset feedback interface protocol and updates the status register of the analysis tree. When adjusting the activation priority of the machine learning model group in the advanced analysis node according to the strategy effectiveness evaluation value, the following operations are performed: read the strategy effectiveness evaluation value from the status register of the analysis tree and compare it with the preset priority mapping table; the priority mapping table stores the correspondence between the strategy effectiveness evaluation value interval and the model group weight coefficient, and uses linear interpolation to calculate the new weight coefficient of each model. Updated model group weight coefficient ,have:

[0101] ;

[0102] in is the current weight coefficient of the i-th model (the initial value comes from the model historical accuracy database), is the strategy effectiveness evaluation value (calculated from the equipment status improvement parameters before and after maintenance), is the preset learning rate adjustment factor (range [0.01, 0.1]). The updated model group weight coefficients are used to reorder the model group calls in descending order, with high-weight models being prioritized for analysis. When optimizing the confidence threshold in the decision triggering strategy instruction based on the fault verification results in the maintenance execution record, the actual fault type and the diagnostic results of the structured analysis report are analyzed from the maintenance execution record, and the false alarm rate and missed alarm rate are calculated. For the optimized confidence threshold ,have:

[0103] ;

[0104] Where, is the current confidence threshold (the initial value comes from the decision trigger rule base), To optimize the step size coefficient (range [0.05, 0.2]), and By comparing actual maintenance records with diagnostic reports, ensure reasonable range.

[0105] Based on the same inventive concept, Figure 2 As shown, the present invention also provides a PMS-based equipment status monitoring and abnormality analysis system, the system comprising:

[0106] The status data acquisition module is used to obtain basic equipment information, real-time operating status data, historical maintenance records, and preset risk level parameters from the equipment management system to generate a comprehensive equipment status data set;

[0107] a health index calculation module configured to calculate a dynamic health attention index based on the comprehensive equipment state data set;

[0108] a policy instruction generation module configured to generate a policy chain instruction set including monitoring policy instructions, analysis policy instructions, and decision trigger policy instructions based on the dynamic health attention index;

[0109] a monitoring data collection module configured to perform data collection operations by scheduling a preset sensor group according to the monitoring policy instructions, and generate a target monitoring data set;

[0110] a report generation module configured to input the target monitoring data set and the analysis policy instructions into a preset multi-level adaptive analysis tree, perform hierarchical diagnosis operations, and generate a structured analysis report;

[0111] a decision output module configured to match a maintenance decision suggestion from a preset maintenance knowledge base and output the maintenance decision suggestion when the structured analysis report meets a preset condition of the decision trigger policy instructions.

[0112] In Example 1, in order to verify the effectiveness of the method of the present application, a regional natural gas distribution station is taken as an application scenario. The distribution station undertakes the key task of stably transporting natural gas to downstream city pipe networks and industrial users, and the safety and reliability of its operation are crucial. Any equipment failure, especially unplanned shutdown of core compressor units, can cause large-scale gas supply interruption and serious safety hazards. The traditional operation and maintenance mode of the station relies on fixed maintenance cycles and manual judgment based on SCADA system coarse-grained alarms, which is difficult to effectively prevent sudden events caused by early hidden failures. The present application aims to apply state monitoring and anomaly analysis to core equipment of the distribution station to realize predictive maintenance and ensure the safe operation of the pipe network.

[0113] In this embodiment, the system first acquires the basic information, historical maintenance records (there was an event of gas transmission efficiency reduction due to piston ring wear within the last 24 months), and preset risk level parameters of a reciprocating compressor unit (model: Ariel-JGK-4, asset value: 8 million RMB, key level: highest) from the station control SCADA system through the state data collection module. At the same time, by deploying sensors on the compressor body, inlet and outlet pipelines, real-time operation state data is collected, including body vibration, inlet and outlet pressure, outlet temperature, motor current, etc., forming a comprehensive equipment state data set.

[0114] The health index calculation module calculates the dynamic health attention index according to the data set. Taking a certain monitoring at 15:30 on September 20, 2024 as an example, the system extracts from the real-time running state data that the compressor outlet pressure appears abnormal micro fluctuation, and its pressure ratio with the inlet pressure deviates from the normal working curve by 3%, which is an abnormal characteristic parameter; it is extracted from the historical maintenance record that the fault frequency is 0.5 times / year as the fault frequency parameter; the equipment value parameter (normalized to 0.95) and the risk level parameter (normalized to 1.0) are extracted from the equipment basic information. Using the abnormal characteristic parameter (normalized to 0.5) and the fault frequency parameter (normalized to 0.6), the risk weight factor R = 0.5*0.5 + 0.5*0.6 = 0.55 is calculated. At the same time, the consequence severity coefficient C = 0.6*0.95 + 0.4*1.0 = 0.97 is generated by fusing the equipment value parameter and the risk level parameter. Finally, the dynamic health attention index H = 0.6*R + 0.4*C = 0.6*0.55 + 0.4*0.97 = 0.718 is calculated by weighting. The index is much higher than that of other devices in the station (a standby filter, whose dynamic health attention index is only 0.18), indicating that the compressor unit is the device that needs most attention at present.

[0115] Based on the health attention index as high as 0.718, the strategy instruction generation module generates a strategy chain instruction set. The monitoring strategy instruction: match the monitoring intensity rule base, increase the data acquisition frequency of the compressor outlet pressure and the machine body vibration sensor from the regular 1 time per minute to 1 time per second, and deploy the high-frequency acoustic sensor in the valve area for special monitoring. The analysis strategy instruction: activate the analysis depth rule base, select the advanced analysis path, and generate the diagnostic model combination identifier "Wavelet+GRU" for deep fault diagnosis. The decision trigger strategy instruction: call the decision trigger rule base of the compressor unit, set the confidence threshold of fault diagnosis to 0.90, and focus the fault mode matching conditions on "valve leakage, piston ring wear, bearing failure".

[0116] The monitoring data collection module schedules relevant sensors to perform a 15-minute high-frequency data collection according to the monitoring strategy instruction, generating a high-resolution target monitoring data set containing pressure, vibration, and acoustic signals. Subsequently, the report generation module inputs the data set and analysis strategy instruction into the multi-level adaptive analysis tree. Hierarchical diagnosis, as the analysis path is identified as high-level, the system directly activates the high-level analysis node. Multi-model fusion diagnosis, the system calls the preset "Wavelet+GRU" machine learning model group. Among them, the wavelet transform (Wavelet) is responsible for time-frequency analysis of high-frequency pressure and acoustic signals, extracting weak leakage characteristics; the gated recurrent unit (GRU) network models the time series of vibration and pressure ratio, analyzing its dynamic evolution trend. By confidence weighted fusion of model output results, the system generates a structured analysis report, with the conclusion: "Fault root cause: there is a slight internal leakage in the compressor secondary outlet valve plate; confidence: 0.92; impact assessment: currently causing a 4% decrease in gas transmission efficiency, if not handled, the leakage is expected to intensify in 15 days, causing efficiency to drop more than 10%, and there is a risk of valve plate rupture."

[0117] Since the confidence of the report is 0.92, which exceeds the threshold of 0.90 set by the decision trigger strategy instruction, the decision output module is activated. Retrieve the standard operating procedures related to "reciprocating compressor valve internal leakage" from the maintenance knowledge base, and combine the downstream gas load forecast provided by the SCADA system (flat peak period in the next week), and perform cost optimization screening, finally output the maintenance decision suggestion: "Suggest handling in the next planned maintenance window (expected in 7 days). Recommended maintenance plan: replace the secondary outlet valve assembly (spare parts number V-341-B), estimated downtime 2 hours. During this period, maintain high-frequency monitoring, if the efficiency drops more than 7%, immediately start the standby compressor and perform emergency maintenance."

[0118] In addition, the invention also exhibits closed-loop self-optimization capability. After the valve replacement maintenance is completed, the system collects maintenance execution records, confirms that the fault and diagnosis are consistent, and the device gas transmission efficiency returns to normal. The system calculates the strategy performance evaluation value of this strategy as 0.95, which is fed back to the system to improve the priority of the "Wavelet+GRU" model in diagnosing valve internal leakage. In another incident, the system issued an abnormal warning for a pressure sensor on a certain pipeline, but after on-site manual verification, it was confirmed as a false alarm caused by sensor zero drift. The system records this as a false alarm case, and automatically fine-tunes the risk weight factor calculation model according to the data, increasing the weight of multi-sensor data cross-validation, thereby effectively improving the robustness of the future health attention index.

[0119] From the data comparison, after the application of the present application, the safety management and operation and maintenance efficiency of the natural gas distribution station are significantly improved. The system can early warning the hidden failure of the compressor unit for 15 days, convert the potential unplanned shutdown risk into planned maintenance, and avoid major gas supply accidents. By dynamically adjusting the monitoring strategy, while ensuring the monitoring accuracy of the key equipment, the overall data transmission and calculation overhead is reduced by about 50%. Precise maintenance decisions not only reduce the cost of single repair, but also reduce the impact on downstream gas supply by optimizing the repair time.

[0120] Table 1 Dynamic health attention index of distribution station equipment:

[0121] ;

[0122] Table 2 Adaptive adjustment of strategy chain instructions:

[0123] ;

[0124] Table 3 Fault diagnosis and decision suggestions:

[0125] ;

[0126] As can be seen from the data of the above table 1 to table 3, the present application realizes differentiated intelligent monitoring of different equipment in the distribution station by dynamically quantifying risks. Table 1 clearly shows that the core compressor unit with early signs of failure has the highest attention. Table 2 shows how the system adaptively allocates monitoring and analysis resources accordingly. The results of table 3 prove the high precision of the present application in complex industrial equipment fault diagnosis and the foresight of the decision, which can provide specific maintenance solutions combined with working conditions and optimized, successfully upgrading the operation and maintenance mode from passive response to active prediction, providing a solid technical guarantee for the safe, stable and efficient operation of the natural gas pipeline network.

[0127] It should be noted that the above formulas can be translated into unitless standard values or parameters of the same dimension that can be superimposed by means of dimensional consistency principles and mathematical standardization methods (such as normalization processing, dimensionless parameter conversion or unit system unification), so as to eliminate the interference of different dimensions on the operation logic, make the formula retain the original data distribution characteristics, and have mathematical operation rationality and objective law adaptability. It is a conventional technical means, and will not be repeated here. The electrical connection between the above-mentioned units does not necessarily mean direct connection, and indirect connection mode can also be used as long as the purpose of the present application is achieved. The above-described is only an exemplary embodiment of the present application, and cannot limit the scope of the present application.

[0128] intended to encompass any and all embodiments of the application with equivalents as would be ascertained by those skilled in the art to which the application pertains. Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

Claims

1. A PMS-based device condition monitoring and anomaly analysis strategy method, characterized in that, The method comprises: obtaining equipment basic information, real-time running state data, historical maintenance records and preset risk level parameters in the equipment management system to generate comprehensive equipment state data set; calculating dynamic health attention index according to the comprehensive equipment state data set; generating strategy chain instruction set including monitoring strategy instruction, analysis strategy instruction and decision trigger strategy instruction based on the dynamic health attention index; wherein, generating monitoring strategy instruction including data acquisition frequency parameter, sensor precision parameter and monitoring target parameter according to matching preset monitoring intensity rule base with the dynamic health attention index; generating analysis strategy instruction including diagnosis model combination identifier and analysis path identifier according to activating preset analysis depth rule base with the dynamic health attention index; generating decision trigger strategy instruction including confidence threshold and fault mode matching condition based on calling preset decision trigger rule base with equipment type identifier in the equipment basic information; integrating the monitoring strategy instruction, the analysis strategy instruction and the decision trigger strategy instruction to obtain the strategy chain instruction set; scheduling preset sensor group to perform data acquisition operation according to the monitoring strategy instruction to generate target monitoring data set; inputting the target monitoring data set and the analysis strategy instruction into preset multi-level adaptive analysis tree to perform hierarchical diagnosis operation to generate structured analysis report; wherein, selecting primary analysis node or advanced analysis node in the multi-level adaptive analysis tree according to analysis path identifier in the analysis strategy instruction; performing threshold check operation on the target monitoring data set through the primary analysis node to generate primary diagnosis identifier; activating the advanced analysis node to perform multi-model fusion diagnosis to generate the structured analysis report when the primary diagnosis identifier meets preset depth trigger condition; matching maintenance decision suggestion from preset maintenance knowledge base and outputting when the structured analysis report meets preset condition of the decision trigger strategy instruction.

2. The PMS-based device condition monitoring and anomaly analysis strategy method according to claim 1, characterized in that, The calculation of dynamic health attention index according to the comprehensive equipment state data set comprises: extracting abnormal feature parameter from the real-time running state data, extracting fault frequency parameter from the historical maintenance records, and extracting equipment value parameter from the equipment basic information; generating risk weight factor using the abnormal feature parameter and the fault frequency parameter; generating consequence severity coefficient by fusing the equipment value parameter and the risk level parameter; generating dynamic health attention index by weighted calculation of the risk weight factor and the consequence severity coefficient.

3. The PMS-based device state monitoring and anomaly analysis strategy method according to claim 1, characterized in that, The activation of the advanced analysis node to perform multi-model fusion diagnosis to generate the structured analysis report when the primary diagnosis identifier meets preset depth trigger condition comprises: activating the advanced analysis node and calling preset machine learning model group according to the diagnosis model combination identifier when the primary diagnosis identifier meets preset depth trigger condition; inputting the target monitoring data set into the machine learning model group to perform parallel analysis to generate model output result set; Performing a confidence-weighted fusion operation on the model output result set to generate a fault root cause parameter and an impact assessment parameter in the structured analysis report.

4. The PMS-based device state monitoring and anomaly analysis strategy method according to claim 1, characterized in that, The matching of the maintenance decision suggestion from the preset maintenance knowledge base comprises: Extracting a fault mode code and a device type identifier from the structured analysis report; Retrieving in a preset maintenance knowledge base according to the fault mode code to obtain a candidate maintenance scheme set; Extracting an operating condition parameter from the device basic information, and performing cost optimization screening by using the operating condition parameter and the candidate maintenance scheme set to generate a maintenance decision suggestion.

5. The PMS-based device state monitoring and anomaly analysis strategy method according to claim 1, characterized in that, The method further comprises: Collecting the policy chain instruction set, the target monitoring data set, the structured analysis report, and a maintenance execution record to generate a policy execution chain data set; Extracting a device state improvement parameter from the maintenance execution record; Calculating a policy performance evaluation value according to the device state improvement parameter; When the policy performance evaluation value is lower than a preset optimization trigger threshold, adjusting the weight coefficients of the risk weight factor and the consequence severity coefficient.

6. The PMS-based device condition monitoring and anomaly analysis strategy method according to claim 5, characterized in that, The adjustment of the weight coefficients of the risk weight factor and the consequence severity coefficient comprises: Extracting a false positive case and a false negative case from the policy execution chain data set to generate an optimization training sample set; Correcting the weight coefficient of the risk weight factor according to the optimization training sample set; Extracting an actual maintenance cost parameter from the maintenance execution record; Correcting the weight coefficient of the consequence severity coefficient by using the actual maintenance cost parameter.

7. The PMS-based device state monitoring and anomaly analysis strategy method according to claim 5, characterized in that, The method comprises: Feeding back the policy performance evaluation value to the multi-level adaptive analysis tree; Adjusting the activation priority of the machine learning model group in the high-level analysis node according to the policy performance evaluation value; Optimizing the confidence threshold in the decision trigger strategy instruction based on the fault verification result in the maintenance execution record.

8. A PMS-based equipment condition monitoring and anomaly analysis system, applied to the PMS-based equipment condition monitoring and anomaly analysis strategy method according to any one of claims 1-7, characterized in that, The system comprises: A state data collection module configured to obtain device basic information, real-time operating state data, historical maintenance records, and a preset risk level parameter in a device management system, and generate a comprehensive device state data set; A health index calculation module configured to calculate a dynamic health attention index according to the comprehensive device state data set; A policy instruction generation module configured to generate a policy chain instruction set comprising monitoring strategy instructions, analysis strategy instructions, and decision trigger strategy instructions based on the dynamic health attention index; wherein the policy chain instruction set comprises: generating monitoring strategy instructions comprising data collection frequency parameters, sensor accuracy parameters, and monitoring target parameters according to the dynamic health attention index and by matching a preset monitoring intensity rule library; generating analysis strategy instructions comprising diagnosis model combination identifiers and analysis path identifiers according to the dynamic health attention index and by activating a preset analysis depth rule library; generating decision trigger strategy instructions comprising confidence thresholds and fault mode matching conditions based on a device type identifier in the device basic information and by calling a preset decision trigger rule library; and integrating the monitoring strategy instructions, the analysis strategy instructions, and the decision trigger strategy instructions to obtain the policy chain instruction set. The monitoring data collection module is configured to schedule a preset sensor group to perform a data collection operation according to the monitoring strategy instruction, and generate a target monitoring data set; The report generation module is configured to input the target monitoring data set and the analysis strategy instruction into a preset multi-level adaptive analysis tree, perform a hierarchical diagnosis operation, and generate a structured analysis report. The report generation module includes the following steps: according to an analysis path identifier in the analysis strategy instruction, selecting a primary analysis node or a high-level analysis node in the multi-level adaptive analysis tree; performing a threshold check operation on the target monitoring data set through the primary analysis node to generate a primary diagnosis identifier; when the primary diagnosis identifier satisfies a preset depth trigger condition, activating the high-level analysis node to perform a multi-model fusion diagnosis and generating the structured analysis report; The decision output module is configured to match a maintenance decision suggestion from a preset maintenance knowledge base and output the suggestion when the structured analysis report satisfies a preset condition of the decision trigger strategy instruction.

Images