A power plant system evaluation method, system, device and medium

Abstract

The application provides a power station system evaluation method, system, device and medium, the method comprises the following steps: S1, acquiring power station system data information, generating a power station system fault experience knowledge base, the system fault experience knowledge base comprises a plurality of types of collected data and a risk evaluation index set, the risk evaluation index set is generated according to a plurality of types of collected data; S2, real-time collection of actual data of the power station under the running state, according to the pre-processing and analysis of the actual data, a weight vector set is generated, and a fuzzy relation matrix is generated according to the membership function; S3, according to the weight vector set and the fuzzy relation matrix, a comprehensive evaluation matrix is calculated and obtained, and according to the comprehensive evaluation matrix result, information corresponding to the evaluation result is obtained. The application can effectively improve the safe and stable operation level of the unit and the integrated intelligent level of operation and maintenance, give the risk prediction of system failure in advance, and avoid the occurrence of sudden failure.

Description

Technical Field

[0001] This invention belongs to the field of electrical automation technology, and in particular relates to a power plant system evaluation method, system, equipment and medium. Background Technology

[0002] With rapid economic and social development, electricity load is growing rapidly, the peak-to-valley difference is widening, and the requirements for power grid stability are becoming increasingly stringent. Insufficient peak-shaving capacity will become a prominent problem restricting the development of the power system. Conventional hydropower stations, especially pumped storage power stations, with their unique peak-shaving and valley-filling operating characteristics, play a vital role in regulating load, promoting energy conservation in the power system, and maintaining the safe and stable operation of the power grid, gradually becoming an effective and indispensable means of power system regulation. Against the backdrop of continuously increasing installed capacity, the structure of generating units is becoming increasingly complex, and operating conditions are becoming more severe, leading to faster deterioration and a higher probability of failure. This places higher demands on the management, maintenance, monitoring, and diagnosis of generating units.

[0003] Pumped storage units are complex nonlinear power systems, and the formation and development of faults during their operation are highly random. However, the traditional fault diagnosis modeling theories and methods in the existing technology have long relied on time-frequency and time-space transformation analysis methods, which are difficult to accurately describe a large number of uncertain factors mathematically. This makes it difficult to obtain accurate diagnostic conclusions for actual condition-based maintenance systems, which greatly restricts the application of fault diagnosis theories and methods in engineering practice. Summary of the Invention

[0004] The first objective of this invention is to provide a power plant system evaluation method to improve diagnostic capabilities, enhance operational stability, and increase the precision of fault data classification.

[0005] Therefore, the above-mentioned objective of the present invention is achieved through the following technical solution:

[0006] A power plant system evaluation method includes the following steps:

[0007] S1. Acquire power plant system data information and generate a power plant system fault experience knowledge base. The system fault experience knowledge base includes multiple types of collected data and a risk assessment indicator set, which is generated based on multiple types of collected data.

[0008] S2. Real-time acquisition of actual power plant data under operating conditions; generation of weight vector set based on preprocessing and analysis of actual data; and generation of fuzzy relation matrix based on membership function.

[0009] The membership function is a three-dimensional membership function constructed from the two-dimensional relationship between at least two types of collected data, which are the data corresponding to the base events that have a transitive relationship between the at least two types of collected data but do not belong to the same cut set;

[0010] S3. Based on the weight vector set and fuzzy relation matrix, calculate and obtain the comprehensive evaluation matrix, and based on the results of the comprehensive evaluation matrix, obtain the information corresponding to the evaluation results.

[0011] While adopting the above technical solutions, the present invention may also adopt or combine the following technical solutions:

[0012] As a preferred embodiment of the present invention: in step S2, the membership function is:

[0013]

[0014] Among them, y lm x is the absolute value of the collected values ​​of the first type of data; lm x is the absolute value of the collected values ​​of the second type of data; min <x1<x2<x max The nodes represent the threshold range intervals corresponding to the collected values ​​of the second type of data; y min <y1<y2<y max Let be the node corresponding to the threshold range of the collected values ​​of the first type of data; k1 and k2 are the constant values ​​of the correlation coefficient between the first type of data and the second type of data in different threshold ranges; l is the dimension corresponding to the first type of data; m is the dimension corresponding to the second type of data, where a is the value of the correlation coefficient between the first type of data and the second type of data in (x2, x...). max ),(y2,y max The average slope constant of the data fitting in the interval is )

[0015] As a preferred embodiment of the present invention, the absolute value of k2 is greater than the absolute value of k1.

[0016] As a preferred technical solution of the present invention: the risk assessment index set includes the risk assessment index set determined in the first fault tree classification model trained by the system fault experience knowledge base;

[0017] The first fault tree classification model performs the following training steps:

[0018] S11. Qualitative analysis: The minimum cut set is obtained by using the descending method. The minimum cut set is the combination of the minimum number of bottom events that cause the top event of the system to occur.

[0019] S12. Quantitative analysis to obtain failure probability and bottom event importance;

[0020] S13. Determining the probability of the bottom event, which involves determining the failure distribution, estimating the distribution parameters based on the distribution model, and determining the probability of occurrence based on the distribution model.

[0021] The top event is the least desirable failure state;

[0022] The intermediate events are sub-level failure factors that lead to the occurrence of the top event;

[0023] The base event refers to all the direct factors that cause the intermediate event to occur, and all the direct factors are those factors that do not need to be investigated further.

[0024] As a preferred technical solution of the present invention: in the quantitative analysis, the importance analysis of the bottom event includes the probability importance analysis of the bottom event, the structural importance analysis of the bottom event, and the critical importance analysis of the bottom event;

[0025] The formula for analyzing the probability importance of bottom events is:

[0026]

[0027] The formula for analyzing the importance of bottom-event structures is:

[0028]

[0029] The formula for analyzing the criticality of bottom events is:

[0030]

[0031] The probability of system failure when the system is in the critical state of bottom event i is called the probability importance.

[0032] Q i (t) represents the failure probability of the i-th bottom event at time t, g[Q(t)] represents the failure probability of the top event at time t, g[1 i [Q(t)] represents the failure probability of the top event at time t when the i-th bottom event fails, g[0 i [Q(t)] represents the probability of failure of the top event at time t when the i-th bottom event is normal;

[0033] When the bottom event i changes from state 0 to 1, the proportion of the critical state number of bottom event i in the total number of states represents the structural importance. The effective calculation is as follows: The failure probability Q of all bottom events k Set to 0.5;

[0034] The criticality of a bottom event is represented by the relative value of the rate of change of the system failure probability to the rate of change of the failure probability of bottom event i, which reflects the probability that bottom event i will trigger a system failure, where g(t) is the failure probability at time t.

[0035] As a preferred embodiment of the present invention: the failure probability of the bottom event is calculated from the bottom up as the probability of failure of each component of the system.

[0036]

[0037] Where Ei represents the events that occur for all base events belonging to the minimal cut set Kj, j is the event dimension of the minimal cut set, i is the event dimension of the base events, and k is the total number of events in the minimal cut set, where F j Let P be the probability expression of the union of the totals of k events. r {E r Let} represent the probability of each base event out of the total number of k events, and r represent the dimension used to calculate the probability of a single base event. i {E i ∩E j Let} represent the intersection probability of the two events. Let r be the probability of the intersection of all r events.

[0038] As a preferred technical solution of the present invention, the process of obtaining the minimum cut set is as follows:

[0039] Starting from the top event of the fault tree, from top to bottom, the events of the previous level are replaced with the events of the next level. When an AND gate is encountered, the input events are written out horizontally in parallel. When an OR gate is encountered, the input events are written out vertically in series. This continues until all logic gates are replaced with the bottom events. At this point, the last column represents all cut sets. The cut sets are then simplified to obtain all the minimum cut sets Kj.

[0040] The second objective of this invention is to provide a power plant evaluation system, comprising the following modules:

[0041] The system fault data information acquisition module is used to acquire power plant system fault data information and compile it into a fault experience knowledge base;

[0042] The system fault tree construction module is used to classify the system fault diagnosis knowledge base and define top events, intermediate events and bottom events to build a tree-structured fault tree.

[0043] The system evaluation module collects real-time data of the power plant under operation, generates a weight vector set based on the preprocessing and parsing of the actual data, and generates a fuzzy relation matrix based on the membership function; it calculates and obtains a comprehensive evaluation matrix based on the weight vector set and the fuzzy relation matrix, and obtains the information corresponding to the evaluation results based on the comprehensive evaluation matrix results;

[0044] The minimum cut set corresponding to the membership function is constructed based on the fault experience knowledge base. The membership function is a three-dimensional membership function constructed from the two-dimensional relationship between at least two types of collected data. The data corresponding to the base events that have a transitive correlation between at least two types of collected data but do not belong to the same cut set.

[0045] A third objective of this invention is to provide an electronic device comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus, characterized in that:

[0046] The memory is used to store computer programs;

[0047] A processor for executing a computer program stored in a memory to implement the steps of the power plant system evaluation method as described above.

[0048] Another objective of this invention is to provide a non-transitory readable storage medium, characterized in that: the non-transitory readable storage medium stores a program, which, when executed by a processor, implements the steps of the power plant system evaluation method as described above.

[0049] This invention provides a method, system, equipment, and medium for evaluating power plant systems, which has the following beneficial effects:

[0050] 1) Based on the power plant fault diagnosis method that considers the membership function of multi-dimensional data in this invention, we can understand the possible deterioration of the system, effectively improve the safe and stable operation level of the unit and the intelligent level of integrated operation and maintenance, and provide early risk prediction of system faults to avoid the occurrence of sudden faults.

[0051] 2) It is proposed to consider the correlation between two cut sets in the membership function, thereby forming a membership function set with a data space. This can more accurately reflect the correlation and transitivity of data between cut sets, forming a data correlation analysis between cut sets in the evaluation system, improving the evaluation accuracy of the system, and thus facilitating a more accurate evaluation of the power plant system and outputting a more reliable maintenance strategy.

[0052] 3) It can effectively improve the safe and stable operation level of the unit and the intelligent level of integrated operation and maintenance, and provide early warning of system failure risks to avoid the occurrence of sudden failures. Attached Figure Description

[0053] Figure 1 The flowchart is for the power plant system evaluation method provided by the present invention.

[0054] Figure 2This is a schematic diagram of the sub-process steps for the cut set of the fault tree and the method for obtaining fault logic.

[0055] Figure 3 This is a schematic diagram of the composition space of membership functions. Detailed Implementation

[0056] The present invention will be described in further detail with reference to the accompanying drawings and specific examples.

[0057] like Figure 1 As shown, a power plant evaluation method includes:

[0058] S1. Acquire power plant system data information and generate a system fault experience knowledge base. The system fault experience knowledge base includes multiple types of collected data and a risk assessment indicator set generated based on the multiple types of collected data.

[0059] S2. Real-time acquisition of actual power plant data under operating conditions; generation of weight vector set based on preprocessing and analysis of actual data; and generation of fuzzy relation matrix based on membership function.

[0060] S3. Based on the weight vector set and fuzzy relation matrix, calculate and obtain the comprehensive evaluation matrix, and based on the comprehensive evaluation matrix result, obtain the information corresponding to the evaluation result; the information corresponding to the evaluation result in this invention refers to the level classification information of fault information based on the evaluation matrix.

[0061] The membership function is a three-dimensional membership function constructed from the two-dimensional relationship between at least two types of collected data, which are the data corresponding to the base events that have a transitive relationship between at least two types of collected data but do not belong to the same cut set.

[0062] The membership function is:

[0063]

[0064] Among them, y lm x is the absolute value of the collected values ​​of the first type of data; lm The absolute values ​​of the collected values ​​for the second type of data are the two types of correlated data initially selected through deep learning; x min <x1<x2<x max The nodes represent the threshold range intervals corresponding to the collected values ​​of the second type of data; y min <y1<y2<y max For the nodes corresponding to the threshold range of the collected values ​​of the first type of data; where a is the node in the range of (x2, x... max ),(y2,y maxThe average slope constant for data fitting within the specified interval. k1 and k2 are the constant values ​​of the correlation coefficients between the first and second types of data in different threshold intervals. In a specific implementation, k1 and k2 can be assigned empirical values ​​or obtained through correlation fitting calculations during actual computation. l is the dimension corresponding to the first type of data; m is the dimension corresponding to the second type of data. The absolute value of k2 is greater than the absolute value of k1. Generally, the constant value of the correlation coefficient changes within the data classification interval, and the correlation increases. The risk assessment index set includes the risk assessment index set determined in the first fault tree classification model trained by the system fault experience knowledge base.

[0065] According to the principle of the membership function implemented in this invention, as shown in Table 1 below, each cut set of the fault tree has its own corresponding monitoring parameters. The membership matrix is ​​derived from the actual measured values ​​of the data or the normalized coefficients of the data. Each fault tree cut set corresponds to a specific evaluation level, such as four levels: A, B, C, and D, corresponding to different states such as severe, under maintenance, key monitoring, and healthy operation. The membership function of this invention considers the correlation between two parameters between different cut sets. Therefore, when constructing the parameter matrix between each fault cut set, instead of establishing membership relationships independently, it relies on the parameter relationship between the two cut sets to form a correction bias in the membership function. This allows the membership function to link the relationship between the two cut sets. The distribution of coefficients and the calculation of the function are adjusted based on the training relationship between the two datasets under different data states. Figure 3 As shown, the spatial membership function established in this way expands the granularity of data correlation in a three-dimensional space within a two-dimensional space. By using data correlation under a three-dimensional spatial structure, the accuracy of data acquisition can be improved. With the development of big data processing technology, the granularity of the collected data analysis can be improved, more accurate fuzzy classification results can be obtained, and higher accuracy evaluation results can be achieved.

[0066] Table 1

[0067]

[0068] Acquire power plant system fault data and generate a system fault knowledge base;

[0069] Based on the fault knowledge base, the system fault data information is classified and trained to generate the first fault tree classification model;

[0070] Obtain the first input dataset and input it into the first fault tree classification model to complete the system's fault early warning and intelligent diagnosis; the first input dataset includes system body data, unit operation data and hydraulic device data.

[0071] The first fault tree classification model performs the following steps:

[0072] S11. Qualitative analysis: The downlink method is used to obtain the minimum cut set. The minimum cut set is the combination of the minimum number of bottom events that lead to the occurrence of the top event of the system.

[0073] S12. Quantitative analysis to obtain failure probability and bottom event importance;

[0074] S13. Determining the probability of the bottom event: The determination of the probability of the bottom event consists of determining the failure distribution, estimating the distribution parameters based on the distribution model, and determining the probability of occurrence based on the distribution model.

[0075] The top event is the least desirable failure state.

[0076] Intermediate events are sub-level failure factors that lead to the occurrence of the top event;

[0077] The bottom event is the entirety of the direct factors that cause the intermediate event to occur; all direct factors are those factors that do not need to be investigated further.

[0078] The acquisition of the fault knowledge base according to the present invention includes, but is not limited to, collecting data based on the actual operating status of the system, or using a deep learning model to accumulate and process fault-related data to achieve early warning and diagnosis.

[0079] This includes acquiring power plant fault data and compiling a system fault knowledge base; classifying system faults based on the acquired system fault knowledge base and constructing a system fault tree; and using the acquired system fault tree, combined with system body data, unit operation data, and hydraulic device data, employing a system fault tree model to perform fault analysis on the system operating status, thereby completing fault early warning and intelligent diagnosis of the power plant system.

[0080] Specifically, the research on power plant system diagnostic models is based on existing power plant system fault data information. It compiles the fault types of each subsystem and subcomponent under different operating conditions into a system fault knowledge base by collecting existing fault data and information.

[0081] Based on the aforementioned system fault knowledge base, and considering the common fault types in power plant systems, the system operation fault categories are divided into two main categories: body faults and hydraulic system faults. Sub-events that may cause a certain type of fault are listed, thus forming the system fault tree.

[0082] Based on the established system fault tree, a system fault tree and system model are established for all possible fault scenarios in the system. Combining system data, unit operation data and hydraulic device data, the system fault tree model is used to analyze the system operation status and realize power plant system fault early warning and intelligent diagnosis.

[0083] The fault tree model establishes a tree structure based on power plant system faults. The least desirable fault state is designated as the top event. All fault factors directly causing this fault are identified and treated as intermediate events in the tree. The model then identifies all direct factors causing the next event, tracing back to the factors that no longer require investigation; these are the bottom events. Each node in the fault tree structure includes: node number; node name: fault type name; node type: AND, OR, NOT logic gates; probability of occurrence: the bottom event is a set value, and the probabilities of the top and intermediate events are determined by the probability of the bottom event; parent node: the node numbers directly affected by the current node; child nodes: the set of node numbers directly affected by the current node, in array form.

[0084] In the fault tree model described above, all faults or events are connected through set logic gates, thereby reflecting the logical structural relationship between specific events of a system or device and the fault events of its various subsystems or components.

[0085] like Figure 2 As shown, the fault tree analysis of the system fault tree model mainly includes the following three parts: qualitative analysis, quantitative analysis, and determination of the probability of the bottom event.

[0086] The qualitative analysis of fault trees implemented according to this invention mainly involves finding the minimum cut sets, which are the combinations of base events that lead to the occurrence of the top event in the fault tree. A minimum cut set is a combination of base events that results in a minimum number of base events leading to the top event. It represents a fault mode that causes the top event to occur. Any fault tree consists of a finite number of minimum cut sets, which are unique for a given top event. A minimum cut set consisting of a single event indicates that the top event occurs as soon as that event occurs. A minimum cut set consisting of two events indicates that the top event occurs only when both events occur together. For a minimum cut set consisting of N events, all N events must occur simultaneously for the top event to occur.

[0087] The minimum cut set is usually found by the downward method, which starts from the top event of the fault tree and proceeds from top to bottom, replacing the previous level event with the next level event in turn. When an AND gate is encountered, the input events are written out horizontally in parallel, and when an OR gate is encountered, the input events are written out vertically in series, until all logic gates are replaced with the bottom events. At this point, the last column represents all cut sets. Then the cut sets are simplified to obtain all the minimum cut sets.

[0088] The quantitative fault tree analysis implemented according to this invention mainly includes failure probability analysis and bottom-event importance analysis. Failure probability analysis calculates the probability of failure of each component of the system using a bottom-up approach.

[0089] Let the minimum cut set expression of the fault tree be K.j (X), then the minimal cut set structure function is:

[0090] In the above formula, k is the total number of minimum cut sets, K j The definition of (X) is

[0091] To find the probability of the top event occurring, which is the probability that θ(X) = 1, we only need to take the expected value of both sides of the above equation, and the left side is the probability of the top event occurring:

[0092] Let Ei be the event that occurs when all base events of the minimum cut set Kj occur. Then the event that occurs when the top event occurs is the event that occurs when at least one of the k Ei occurs. Therefore...

[0093] If we write the event and probability as Fj, then F j =Σ 1＜j1＜…＜k Pr{E i1 ∩E i2 ∩…E ij} is used to calculate the probability of the union of multiple events. First, the sum of the probabilities of all individual events is calculated, then the sum of the probabilities of the intersection of all two events is subtracted, then the sum of the probabilities of the intersection of all three events is added, and so on, until the probability of the intersection of all k events is added or subtracted.

[0094] Expanding the above equation, we get:

[0095]

[0096] Where Ei represents the events that occur for all base events belonging to the minimal cut set Kj, j is the event dimension of the minimal cut set, i is the event dimension of the base events, and k is the total number of events in the minimal cut set, where F j Let P be the probability expression of the union of the totals of k events. r {E r Let} represent the probability of each bottom event out of the total number of k events, and r be the dimension used to calculate the probability of a single bottom event.

[0097] P i {E i ∩E j Let} represent the intersection probability of the two events. Let r be the probability of the intersection of all r events.

[0098] The above method can be used to complete the top event failure probability analysis. Similarly, failure probability analysis can be performed on each stage of the failure count using a similar method; these methods will not be elaborated upon here.

[0099] The bottom event importance analysis in the fault tree model implemented according to the present invention includes the analysis of the probability importance, structural importance, and critical importance of the bottom event.

[0100] The probability importance is defined as follows: when the system is in the critical state of component i (the state that leads to system failure when event i fails is called the critical state; among the 2n-1 possible failures of event i, only those that lead to system failure are the critical states of component i), the probability of system failure is called the probability importance I. i Pr (t).

[0101] If Q i Let g(t) represent the failure probability of the i-th bottom event at time t, and g[Q(t)] represent the failure probability of the top event at time t. Let g[1 i [Q(t)] represents the failure probability of the top event at time t when the i-th bottom event fails, g[0 i Let Q(t) represent the probability of the top event failing at time t when the i-th bottom event is normal. Then, the probability importance... for:

[0102] Structural importance: When a bottom event i changes from state 0 to 1, the proportion of critical states of bottom event i to the total number of states. In one specific implementation of the actual calculation of structural importance, the failure probability of all bottom events is set to 0.5 in the probabilistic importance expression for event i. Therefore, structural importance... The effective calculation is as follows:

[0103] Criticality: The relative value of the rate of change of the system failure probability to the rate of change of the failure probability of the underlying event i reflects the probability that the underlying event i will trigger a system failure. Its definition is as follows: For system fault diagnosis and inspection, determining the criticality of the bottom event is of great guiding significance, because once a system failure occurs, maintenance personnel have reason to suspect that the bottom event with the highest criticality triggered the system failure.

[0104] Among them, the quantitative analysis based on fault trees is based on the probability of occurrence of the bottom event, so the determination of the probability of the bottom event is crucial to the entire analysis.

[0105] Determining the probability of occurrence of a basic event can be mainly divided into three steps: determining the failure distribution, estimating the distribution parameters based on the distribution model, and determining the probability of occurrence based on the distribution model. In practice, the failure distribution curves of each basic event can be obtained by analyzing the occurrence of various faults during a major overhaul period, thereby determining a standard probability distribution model that conforms to the failure probability distribution. Based on the obtained distribution model, the distribution curve of the probability of occurrence of basic events is obtained using parameter identification methods. Using this curve, combined with the time since the last major overhaul, the probability of occurrence of each event can be determined.

[0106] Using the above methods, an open and scalable fault diagnosis knowledge base framework is established, and the fault diagnosis knowledge base and fault sample standard library are accumulated and updated in real time. A fault tree diagnosis model based on fault reasoning is constructed, realizing early warning and intelligent diagnosis of potential faults in power plant units. This provides theoretical guidance and technical support for the formulation of system maintenance strategies, ensuring the safety and reliability of system operation.

[0107] This invention also provides a power plant system fault diagnosis system. This system for implementing power plant fault diagnosis includes a system fault data information acquisition module, a system fault tree construction module, and a system fault tree analysis module. These modules are connected in series. The system fault data information acquisition module acquires power plant system fault data information and compiles a system fault knowledge base. The system fault tree construction module categorizes the system fault diagnosis knowledge base and defines top events, intermediate events, and bottom events to construct a tree-like system fault tree. The system fault tree analysis module, combined with system body data, unit operation data, and hydraulic device data, performs qualitative analysis, quantitative analysis, and bottom event probability determination to determine system fault warnings and intelligent diagnosis.

[0108] The description in this specification is merely illustrative of the invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the content of this specification or exceed the scope defined in the claims, they should all fall within the protection scope of this invention.

Claims

1. A power plant system evaluation method, characterized in that: Includes the following steps: S1. Acquire power plant system data information and generate a power plant system fault experience knowledge base. The system fault experience knowledge base includes multiple types of collected data and a risk assessment indicator set, which is generated based on multiple types of collected data. S2. Real-time acquisition of actual power plant data under operating conditions; generation of weight vector set based on preprocessing and analysis of actual data; and generation of fuzzy relation matrix based on membership function. The membership function is a three-dimensional membership function constructed from the two-dimensional relationship between at least two types of collected data, which are the data corresponding to the base events that have a transitive relationship between the at least two types of collected data but do not belong to the same cut set; S3. Based on the weight vector set and fuzzy relation matrix, calculate and obtain the comprehensive evaluation matrix, and based on the results of the comprehensive evaluation matrix, obtain the information corresponding to the evaluation results; In step S2, the membership function is: Among them, y lm x is the absolute value of the collected values ​​of the first type of data; lm x is the absolute value of the collected values ​​of the second type of data; min <x1<x2<x max The nodes represent the threshold range intervals corresponding to the collected values ​​of the second type of data; y min <y1<y2<y max Let be the node corresponding to the threshold range of the collected values ​​of the first type of data; k1 and k2 are constant values ​​of the correlation coefficient between the first type of data and the second type of data in different threshold ranges; l is the dimension corresponding to the first type of data; m is the dimension corresponding to the second type of data, where a is the value of the correlation coefficient between the first type of data and the second type of data in (x2, x...). max ),(y2,y max The data in the interval are fitted with the average slope constant; The risk assessment index set includes the risk assessment index set determined in the first fault tree classification model trained by the system fault experience knowledge base; The first fault tree classification model performs the following training steps: S11. Qualitative analysis: The minimum cut set is obtained by using the descending method. The minimum cut set is the combination of the minimum number of bottom events that cause the top event of the system to occur. S12. Quantitative analysis to obtain failure probability and bottom event importance; S13. Determining the probability of the bottom event, which involves determining the failure distribution, estimating the distribution parameters based on the distribution model, and determining the probability of occurrence based on the distribution model. The base event refers to all the direct factors that cause the intermediate event to occur, and all the direct factors are those factors that do not need to be investigated further. The intermediate events are sub-level failure factors that lead to the occurrence of the top event; The top event is the least desirable failure state.

2. The method according to claim 1, characterized in that: The absolute value of k2 is greater than the absolute value of k1.

3. The method according to claim 1, characterized in that: In quantitative analysis, the importance analysis of bottom events includes probability importance analysis, structural importance analysis, and critical importance analysis of bottom events; The formula for analyzing the probability importance of bottom events is: The formula for analyzing the importance of bottom-event structures is: The formula for analyzing the criticality of bottom events is: The probability of system failure when the system is in the critical state of bottom event i is called the probability importance. Q i (t) represents the failure probability of the i-th bottom event at time t, g[Q(t)] represents the failure probability of the top event at time t, g[1 i [Q(t)] represents the failure probability of the top event at time t when the i-th bottom event fails, g[0 i [Q(t)] represents the probability of failure of the top event at time t when the i-th bottom event is normal; When the bottom event i changes from state 0 to 1, the proportion of the critical state number of bottom event i in the total number of states represents the structural importance. The effective calculation is as follows: The failure probability Q of all bottom events k Set to 0.5; The criticality of a bottom event is represented by the relative value of the rate of change of the system failure probability to the rate of change of the failure probability of bottom event i, which reflects the probability that bottom event i will trigger a system failure, where g(t) is the failure probability at time t.

4. The method according to claim 3, characterized in that: The failure probability of the bottom event is calculated from the bottom up as the probability of failure of each component of the system: Where Ei represents the events that occur for all base events belonging to the minimal cut set Kj, j is the event dimension of the minimal cut set, i is the event dimension of the base events, and k is the total number of events in the minimal cut set, where F j Let P be the probability expression of the union of the totals of k events. r {E r Let} represent the probability of each base event out of the total number of k events, and r represent the dimension used to calculate the probability of a single base event. i {E i ∩E j Let} represent the intersection probability of the two events. Let r be the probability of the intersection of all r events.

5. The method according to claim 4, characterized in that: The process of obtaining the minimum cut set is as follows: Starting from the top event of the fault tree, from top to bottom, the events of the previous level are replaced with the events of the next level. When an AND gate is encountered, the input events are written out horizontally in parallel. When an OR gate is encountered, the input events are written out vertically in series. This continues until all logic gates are replaced with the bottom events. At this point, the last column represents all cut sets. The cut sets are then simplified to obtain all the minimum cut sets Kj.

6. A power plant evaluation system, characterized in that: The power plant evaluation system is based on the power plant system evaluation method as described in claim 1, and includes the following modules: The system fault data information acquisition module is used to acquire power plant system fault data information and compile it into a fault experience knowledge base; The system fault tree construction module is used to classify the system fault diagnosis knowledge base and define top events, intermediate events and bottom events to build a tree-structured fault tree. The system evaluation module collects real-time data of the power plant under operation, generates a weight vector set based on the preprocessing and parsing of the real data, and generates a fuzzy relation matrix based on the membership function. Based on the weight vector set and the fuzzy relation matrix, the comprehensive evaluation matrix is ​​calculated and obtained, and based on the results of the comprehensive evaluation matrix, the information corresponding to the evaluation results is obtained; The minimum cut set corresponding to the membership function is constructed based on the fault experience knowledge base. The membership function is a three-dimensional membership function constructed from the two-dimensional relationship between at least two types of collected data. The data corresponding to the base events that have a transitive correlation between at least two types of collected data but do not belong to the same cut set.

7. An electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus, characterized in that: The memory is used to store computer programs; A processor for executing a computer program stored in a memory to implement the steps of the power plant system evaluation method as described in any one of claims 1-5.

8. A non-transitory readable storage medium, characterized in that: The non-transitory readable storage medium stores a program, which, when executed by a processor, implements the steps of the power plant system evaluation method as described in any one of claims 1-5.

Images