Boiler system fault diagnosis method and system based on hierarchical directed graph

Abstract

The invention relates to the technical field of power generation boiler fault diagnosis, in particular to a boiler system fault diagnosis method and system based on a hierarchical directed graph, and the method comprises the steps: abstracting parameters in a boiler system process flow diagram into nodes, abstracting a parameter influence relation into directed edges, and constructing a symbol directed graph; layering the signed directed graph to obtain a layered directed graph; in the operation process of the boiler system, performing state monitoring on all nodes of each layer of the layered directed graph to obtain a real-time node state; determining a fault propagation path according to the real-time node state; determining a fault source node in the fault propagation path according to the membership degree; and verifying the correctness of the fault propagation path based on a reverse reasoning mechanism. When fault diagnosis is carried out by adopting the layered directed graph, nodes below the number of layers where fault nodes are located do not need to be diagnosed, so that the diagnosis efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power generation boiler fault diagnosis, and in particular to a boiler system fault diagnosis method and system based on a hierarchical directed graph. Background Art

[0002] With the continuous advancement of industrial technology, large-scale machinery is proliferating, and the complexity of these systems is increasing, leading to a higher probability of failure. When a complex system fails, simply replacing the damaged component will not resolve the underlying issue. Identifying the root cause requires significant manpower and time. Consequently, the resulting loss of production efficiency often results in losses for manufacturers. For example, a minor failure in a boiler system can lead to reduced production efficiency, increased costs, and, in severe cases, even complete plant shutdowns. Therefore, when designing large-scale systems, it is crucial not only to consider the system architecture but also to implement appropriate preventative measures against system failures. When a system failure occurs, quickly locating the source of the fault and promptly replacing or repairing the faulty component can improve production efficiency. Boiler systems have numerous parameters, and the relationships between these parameters are difficult to quantify using specific mathematical formulas. When a failure occurs, it is difficult to quickly locate the source of the fault using real-time observation data. Therefore, boiler fault diagnosis is a key factor in ensuring stable system production and a significant challenge that currently needs to be addressed.

[0003] Currently, commonly used fault diagnosis methods primarily include quantitative models, qualitative models, and empirical models. However, their application is often limited when dealing with large, complex systems. Quantitative models rely on precise mathematical relationships to represent the relationships between systems. However, in large, complex systems, due to the numerous and interdependent parameters, mathematical relationships between parameters are difficult to obtain. Qualitative models, on the other hand, do not rely on mathematical relationships between parameters, but their diagnostic accuracy is limited. Empirical models rely primarily on experience accumulated over long production processes for fault diagnosis, making them susceptible to subjective factors. Summary of the Invention

[0004] In order to address the deficiencies of the prior art, the present invention provides a boiler system fault diagnosis method and system based on a hierarchical directed graph; when using a hierarchical directed graph for fault diagnosis, it is not necessary to diagnose nodes below the layer where the fault node is located, thereby improving the diagnostic efficiency.

[0005] On the one hand, a fault diagnosis method for boiler system based on hierarchical directed graph is provided;

[0006] The boiler system fault diagnosis method based on hierarchical directed graph includes:

[0007] The parameters in the boiler system process flow diagram are abstracted as nodes, and the parameter influence relationships are abstracted as directed edges to construct a signed directed graph.

[0008] The symbol directed graph is layered to obtain a layered directed graph;

[0009] During the operation of the boiler system, the status of all nodes in each layer of the hierarchical directed graph is monitored to obtain the real-time node status; based on the real-time node status, the fault propagation path is determined;

[0010] According to the membership degree, the fault source node is determined in the fault propagation path;

[0011] Based on the reverse reasoning mechanism, the correctness of the fault propagation path is verified.

[0012] On the other hand, a boiler system fault diagnosis system based on hierarchical directed graph is provided;

[0013] The boiler system fault diagnosis system based on hierarchical directed graph includes:

[0014] A construction module is configured to: abstract parameters in the boiler system process flow diagram into nodes, abstract parameter influence relationships into directed edges, and construct a signed directed graph;

[0015] A processing module is configured to: perform hierarchical processing on the symbolic directed graph to obtain a hierarchical directed graph;

[0016] The propagation path determination module is configured to: monitor the status of all nodes in each layer of the hierarchical directed graph during the operation of the boiler system to obtain real-time node status; and determine the fault propagation path based on the real-time node status;

[0017] A fault source node determination module is configured to: determine the fault source node in the fault propagation path according to the membership degree;

[0018] The verification module is configured to verify the correctness of the fault propagation path based on a reverse reasoning mechanism.

[0019] In another aspect, an electronic device is provided, comprising:

[0020] a memory for non-transitory storage of computer-readable instructions; and

[0021] a processor for executing said computer-readable instructions,

[0022] When the computer-readable instructions are executed by the processor, the method described in the first aspect is executed.

[0023] On the other hand, a storage medium is provided, which non-temporarily stores computer-readable instructions, wherein when the non-temporary computer-readable instructions are executed by a computer, the method described in the first aspect is executed.

[0024] On the other hand, a computer program product is provided, comprising a computer program, wherein the computer program is configured to implement the method described in the first aspect when running on one or more processors.

[0025] The above technical solution has the following advantages or beneficial effects:

[0026] Hierarchical directed graphs inherit the excellent analytical completeness of symbolic directed graphs. This layered approach overcomes the difficulty of quickly identifying relationships between nodes in symbolic directed graphs when faced with numerous nodes. Because faults always propagate from higher layers to lower layers, fault diagnosis using hierarchical directed graphs eliminates the need to diagnose nodes below the faulty node's level, thereby improving diagnostic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0028] Figure 1 It is a flow chart of the present invention for executing boiler system fault diagnosis of hierarchical directed graph;

[0029] Figure 2 is a process flow chart of a boiler system established in an embodiment;

[0030] Figure 3 is a diagram of parameter values and parameter thresholds in a boiler system at a certain moment in one embodiment;

[0031] Figure 4 It is the boiler system symbolic directed graph established;

[0032] Figure 5 is Figure 4 After layered processing, a layered directed graph of the boiler system is obtained;

[0033] Figure 6 It is a diagram of the membership degree of a node. DETAILED DESCRIPTION

[0034] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0035] Example 1

[0036] This embodiment provides a boiler system fault diagnosis method based on a hierarchical directed graph;

[0037] like Figure 1As shown, the boiler system fault diagnosis method based on the hierarchical directed graph includes:

[0038] S101: abstracting the parameters in the boiler system process flow diagram into nodes, abstracting the parameter influence relationships into directed edges, and constructing a signed directed graph;

[0039] S102: performing layered processing on the symbolic directed graph to obtain a layered directed graph;

[0040] S103: During the operation of the boiler system, the status of all nodes in each layer of the layered directed graph is monitored to obtain real-time node status; and the fault propagation path is determined based on the real-time node status;

[0041] S104: Determine the fault source node in the fault propagation path based on the membership degree;

[0042] S105: Based on the reverse reasoning mechanism, verify the correctness of the fault propagation path.

[0043] It should be understood that a boiler system is comprised of a series of interconnected devices and components. Fuel combustion in the furnace releases heat energy, which heats water in the steam drum and converts it into steam. A ventilation system also facilitates air supply and exhaust gas discharge. Key components include the steam drum, water-cooled wall, superheater, economizer, preheater, burner, blower, and induced draft fan. The steam produced can be used for a variety of purposes, including power generation, heating, and industrial production, making it a crucial piece of infrastructure.

[0044] It should be understood that the boiler system process flow diagram includes the following: a blower first sends cold air from outside into the preheater, and the preheated hot air enters the furnace. Simultaneously, coal lumps are pulverized into pulverized coal, which is introduced into the furnace by a pulverizer. A water pump sends clean water to the economizer for heating before it is fed into the steam drum.

[0045] After preparations are complete, the induced draft fan is started to exhaust the remaining combustible gases in the furnace. The burner is then turned on to ignite the pulverized coal. The heat within the furnace heats the water in the water-cooled walls, forming a steam-water mixture that enters the steam drum. Within the drum, the water and steam separate. The water flows through the downcomer to the lower header, while the steam enters the steam pipe and reaches the superheater for secondary heating, becoming superheated steam. Finally, it flows to the steam turbine, driving its operation.

[0046] Furthermore, the S101: abstracting the parameters in the boiler system process flow diagram into nodes, abstracting the parameter influence relationships into directed edges, and constructing a signed directed graph;

[0047] Among them, the parameters include: fuel quantity, furnace temperature, furnace pressure, blower speed, drum water level, steam pressure, steam temperature, steam flow, economizer outlet water temperature, feedwater valve opening, exhaust temperature, and induced draft fan speed.

[0048] The parameter influence relationship includes: when there is an influence relationship between two nodes, a directed edge is used from the first node to the affected second node; there are two types of directed edges:

[0049] One is a directed edge with a positive influence relationship. When the parameter value of the first node increases, the parameter value of the second node on the other side of the directed edge also increases. The directed edge with a positive influence relationship is represented by a solid arrow.

[0050] The other is a negative impact directed edge. When the parameter value of the first node decreases, the parameter value of the second node on the other side of the directed edge increases. The directed edge of the negative impact relationship is represented by a dotted arrow.

[0051] When there is a positive influence relationship between node i and node j, use C ij =1 means that when there is a negative influence relationship between node i and node j, C ij = -1 means. When there is no influence relationship between node i and node j, that is, there is no arrow between the nodes in the directed graph, then the node influence relationship C ij =0.

[0052] The influencing relationships among boiler system parameters are obtained through expert systems and years of production and maintenance experience. Figure 2 This is a process flow chart of a boiler system established in an embodiment. Figure 3 is a diagram of parameter values and parameter thresholds in the boiler system at a certain moment in the embodiment. Figure 2 、 Figure 3 The boiler system directed graph is constructed as follows Figure 4 shown.

[0053] The nodes of the boiler system directed graph include: the boiler system fuel quantity is abstracted as node T01, the furnace temperature and pressure are abstracted as nodes T02 and T03, the blower speed is abstracted as node T04, the drum water level is abstracted as node T05, the steam pressure, temperature, and flow are abstracted as nodes T06, T07, and T08, the economizer outlet water temperature is abstracted as node T09, the feedwater valve opening is abstracted as node T10, the exhaust temperature is abstracted as node T11, and the induced draft fan speed is abstracted as node T12.

[0054] Furthermore, the step S102 of performing layered processing on the symbolic directed graph to obtain a layered directed graph includes:

[0055] (1-1): Construct an adjacency matrix based on the symbolic directed graph;

[0056] (1-2): Construct a reachability matrix based on the symbolic directed graph;

[0057] (1-3): According to the reachability matrix, the predecessor set and the reachable set are obtained; according to the intersection of the predecessor set and the reachable set, the number of logical layers of all nodes is determined, thus obtaining a hierarchical directed graph.

[0058] Furthermore, the (1-1): constructing an adjacency matrix according to the signed directed graph includes:

[0059] If there is a direct directed edge between node i and node j, then element a in the adjacency matrix ij is defined as 1; if there is no direct directed edge between node i and node j, then the element a in the adjacency matrix ij is defined as 0;

[0060] If the element a in the adjacency matrix ij If it is 0, it means that there is no direct directed edge between node i and node j, and the failure of node i does not directly affect node j; if the element a in the adjacency matrix is ij If it is 1, it means that there is a direct directed edge between node i and node j, and the failure of node i directly affects node j.

[0061] Furthermore, the step (1-2): constructing a reachability matrix based on the symbolic directed graph includes:

[0062] If there is at least one directed path from node i to node j, the element in the reachability matrix is defined as 1; if there is no directed path from node i to node j, the element in the reachability matrix is defined as 0;

[0063] If the element a in the reachable matrix ij If it is 0, it means that node i does not affect node j;

[0064] If the element a in the reachable matrix ij If it is 1, it means that node i affects node j.

[0065] If the system is complex and has many nodes, the Warshall algorithm can be used to obtain the reachability matrix. The input value of the Warshall algorithm is the adjacency matrix. After the Warshall algorithm is calculated, the reachability matrix is obtained. The output value of the Warshall algorithm is the reachability matrix.

[0066] Furthermore, the above (1-3): according to the reachable matrix, finding the predecessor set and the reachable set includes:

[0067] In the reachability matrix P ij In the reachable set R1, select all nodes corresponding to the first row element value 1. ij , select all nodes whose first column element value is 1 to form the predecessor set A1;

[0068] In the reachability matrix Pij In the reachable matrix P, select all nodes corresponding to the second row element value 1 to form the reachable set R2. ij , select all nodes whose second column element value is 1 to form the predecessor set A2;

[0069] In the reachability matrix P ij In the reachable matrix P, select all nodes corresponding to the third row element value 1 to form the reachable set R3. ij , select all nodes whose third column element value is 1 to form the predecessor set A3;

[0070] By analogy, find all predecessor sets and reachable sets in the symbolic directed graph.

[0071] Furthermore, (1-3): according to the intersection of the predecessor set and the reachable set, the number of logical layers of all nodes is determined, thereby obtaining a hierarchical directed graph, including:

[0072] Use the predecessor set Ai and the reachable set Ri,i∈{1,…,12} to perform stratification:

[0073] If L1={v i |Ri∩Ai=Ai}, then calculate the predecessor sets Ai and reachable sets Ri of other nodes in the symbolic directed graph in turn, and find the nodes belonging to the first layer of the hierarchical directed graph; where L1 represents the first layer, i represents the node number; L1={v i |Ri∩Ai=Ai} means that if the intersection of the reachable set of node i and the predecessor set of node i is the predecessor set of node i, node i is in the first layer of the hierarchical directed graph, that is, node i of the symbolic directed graph belongs to the first layer of the hierarchical directed graph;

[0074] After finding the nodes of the first layer of the layered directed graph, delete the rows and columns corresponding to the nodes of the first layer of the layered directed graph in the reachability matrix, and continue to find the nodes of the second layer of the layered directed graph;

[0075] If L2={v j |Rj∩Aj=Aj}, then iterate in sequence to find the logical layer number of each node in the symbolic directed graph; where L2 represents the second layer, represents the node name, L2={v j |Rj∩Aj=Aj} means that if the intersection of node j's reachable set and node j's predecessor set is node j's predecessor set, node j is in the second layer of the hierarchical directed graph, that is, node j belongs to the second layer of the hierarchical directed graph;

[0076] A layered directed graph is constructed according to the symbolic directed graph and the number of layers of the nodes in the symbolic directed graph, and there is at least one node in each layer.

[0077] According to this method, Figure 4After the symbolic directed graph is layered, a layered directed graph is constructed, such as Figure 5 As shown, in Figure 5 The hierarchical directed graph includes 7 logical layers.

[0078] Furthermore, S103: During the operation of the boiler system, the status of all nodes in each layer of the hierarchical directed graph is monitored to obtain real-time node status, wherein the real-time node status includes: fuel quantity, furnace temperature, furnace pressure, blower speed, drum water level, steam pressure, steam temperature, steam flow, economizer outlet water temperature, feedwater valve opening, exhaust temperature, and induced draft fan speed.

[0079] If the parameter state value of node i is higher than the set alarm threshold upper limit, the state of node i is ψi=1, where ψi represents the state of the node, which is higher than the alarm value upper limit, lower than the alarm value lower limit, or within the normal range;

[0080] For example, node i represents the furnace temperature. When the real-time monitored furnace temperature is higher than the upper alarm temperature limit, the state of node i is 1.

[0081] If the parameter state value of node i is within the normal operating range, the state of node i is ψi=0,

[0082] For example, node i represents the furnace temperature. When the real-time monitored furnace temperature is within the normal operating range, the state of node i is 0.

[0083] If the parameter state value of node i is lower than the set alarm threshold, the state of node i is obtained as ψi=-1.

[0084] Exemplarily, the node i represents the furnace temperature. When the real-time monitored furnace temperature is lower than the lower limit of the alarm temperature, the state of the node i is -1.

[0085] Furthermore, the step S103 of determining the fault propagation path according to the real-time node status includes:

[0086] If there is a positive influence relationship between node i and node j in the hierarchical directed graph, the directed edge between the nodes in the directed graph is a solid arrow. At this time, the node influence relationship C ij =1; wherein, a positive influence relationship means that when the value of one node increases, the value of another node also increases, or a positive influence relationship means that when the value of one node decreases, the value of another node also decreases;

[0087] If there is no influence relationship between node i and node j in the hierarchical directed graph, there is no arrow between the nodes in the directed graph. In this case, the node influence relationship C ij=0; where no influence relationship exists means that when the value of one node changes, the value of another node will not change, and node i cannot affect node j;

[0088] If there is a negative influence relationship between node i and node j in the hierarchical directed graph, the directed edge between the nodes in the directed graph is a dotted arrow. At this time, the node influence relationship C ij =-1; a negative influence relationship means that when the value of one node increases, the value of another node decreases, or a negative influence relationship means that when the value of one node decreases, the value of another node increases;

[0089] If the values of node i and node j satisfy ψ i C ij ψ j =1, then node i and node j are compatible nodes. A compatible node means that when the value of node i is abnormal, the value of node j will also change accordingly.

[0090] If the nodes meet ψ i C ij ψ j C jk ψ k C kl ψ l =1, then the path composed of node i, node j, node k, and node l becomes the fault propagation path, and the fault propagates in the fault propagation path. i C ij ψ j C jk ψ k C kl ψ l =1 means that the abnormal state of node i can affect node j, the abnormal state of node j can affect node k, and so on until the abnormal state of node k can affect node l.

[0091] Furthermore, the step S104 of determining the fault source node in the fault propagation path according to the membership degree includes:

[0092] The parameter state Ψi is assigned a membership degree u(v i ) calculation:

[0093]

[0094] v i is the state value of node i, The upper and lower limits of the node alarm thresholds are: are the upper limit and lower limit of normal operating conditions of node i.

[0095] In a hierarchical directed graph, if a child node l has two parent nodes i and j, the propagation source of node l is determined based on the membership degree of parent nodes i and j;

[0096] When the membership degree of node i is greater than that of node j, the fault source node of child node l is i;

[0097] When the membership of node j is greater than that of node i, the fault source node of child node l is j.

[0098] like Figure 6 As shown, the membership degree of the steam temperature represented by node T07 is 0.2.

[0099] Because faults always propagate from higher layers to lower layers, nodes at higher layers can influence nodes at lower layers. This influence path is called a compatible path. A compatible path is a path that can serve as a fault propagation path during fault diagnosis. When forward reasoning based on the compatible path principle finds multiple fault source nodes, the node with the highest layer is used as the fault source diagnosis result. If multiple fault sources are at the same layer, the node with the highest membership degree is used as the fault source diagnosis result.

[0100] Furthermore, the S105: verifying the correctness of the fault propagation path based on the reverse reasoning mechanism includes:

[0101] Forward reasoning starts from the node where the fault occurs and infers the fault source from the lower layer to the higher layer;

[0102] Reverse reasoning is to reason from the fault source to the fault node from the high level to the low level, which is used to verify correctness; reverse reasoning is to reason from the fault source to the fault node.

[0103] According to the principle of compatible paths and the membership function, after forward reasoning from the fault node to the fault source node, reverse reasoning is adopted to reason from the fault source node to the fault node to verify the correctness of the fault path.

[0104] For example, if the economizer outlet water temperature at node T09 falls below the lower temperature threshold, a low-temperature fault is generated. Forward reasoning, starting from the faulty node T09, reveals that nodes T02, T05, T07, and T09 form a compatible path. Ultimately, the fault source is found to be the low furnace temperature at node T02. Backward reasoning, starting from the fault source node T02, can be used to verify whether nodes T02, T05, T07, and T09 form a compatible path. This allows us to determine whether the status of node T02 can affect node T09, thereby verifying the correctness of the forward reasoning.

[0105] The present invention can be applied to boiler systems in the power generation industry. The method comprises: constructing a symbolic directed graph based on the boiler system parameters in the boiler system process flow chart, then extracting the predecessor set and the reachable set from the reachable matrix in the symbolic directed graph, and determining the number of logical layers of all nodes based on the intersection relationship between the two, thereby constructing a hierarchical directed graph. By real-time monitoring of the status of each operating parameter in the boiler system, the parameter operating status is mathematically defined, the compatible path principle and the membership function are used to locate the fault source node, and then reverse reasoning is used to verify the fault propagation path. The method establishes a reasonable fault logic reasoning method based on parameter status information and hierarchical directed graphs, prevents the combinatorial explosion caused by the large number of parameters in large and complex systems, and provides a real-time fault diagnosis method for the boiler system, which is convenient for technicians to take timely and effective measures to ensure the operation of the boiler system.

[0106] Example 2

[0107] This embodiment provides a boiler system fault diagnosis system based on a hierarchical directed graph;

[0108] The boiler system fault diagnosis system based on hierarchical directed graph includes:

[0109] A construction module is configured to: abstract parameters in the boiler system process flow diagram into nodes, abstract parameter influence relationships into directed edges, and construct a signed directed graph;

[0110] A processing module is configured to: perform hierarchical processing on the symbolic directed graph to obtain a hierarchical directed graph;

[0111] The propagation path determination module is configured to: monitor the status of all nodes in each layer of the hierarchical directed graph during the operation of the boiler system to obtain real-time node status; and determine the fault propagation path based on the real-time node status;

[0112] A fault source node determination module is configured to: determine the fault source node in the fault propagation path according to the membership degree;

[0113] The verification module is configured to verify the correctness of the fault propagation path based on a reverse reasoning mechanism.

[0114] It should be noted that the aforementioned construction module, processing module, propagation path determination module, fault source node determination module, and verification module correspond to steps S101 to S105 in Example 1. The examples and application scenarios implemented by these modules and the corresponding steps are the same, but are not limited to the contents disclosed in Example 1. It should be noted that the aforementioned modules, as part of a system, can be executed in a computer system, such as a set of computer-executable instructions.

[0115] The descriptions of the various embodiments in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0116] The proposed system can be implemented in other ways. For example, the system embodiment described above is merely illustrative. For example, the above module division is only a logical function division. In actual implementation, other division methods may be used. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not implemented.

[0117] Example 3

[0118] This embodiment also provides an electronic device, comprising: one or more processors, one or more memories, and one or more computer programs; wherein the processor is connected to the memory, and the one or more computer programs are stored in the memory. When the electronic device is running, the processor executes the one or more computer programs stored in the memory, so that the electronic device executes the method described in the above embodiment one.

[0119] It should be understood that in this embodiment, the processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0120] The memory may include a read-only memory and a random access memory, and provides instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0121] During implementation, each step of the above method may be completed by an integrated logic circuit of hardware in a processor or by instructions in the form of software.

[0122] The method in Example 1 can be directly implemented as being executed by a hardware processor, or by a combination of hardware and software modules within the processor. The software module can be located in a storage medium well-established in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not given here.

[0123] Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with this embodiment can be implemented using electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0124] Example 4

[0125] This embodiment further provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the method described in the first embodiment is performed.

[0126] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A boiler system fault diagnosis method based on a hierarchical directed graph is characterized by: include: The parameters in the boiler system process flow diagram are abstracted as nodes, and the parameter influence relationships are abstracted as directed edges to construct a signed directed graph. The symbol directed graph is layered to obtain a layered directed graph; During the operation of the boiler system, the status of all nodes in each layer of the layered directed graph is monitored to obtain the real-time node status; Determine the fault propagation path based on real-time node status; According to the membership degree, the fault source node is determined in the fault propagation path; Based on the reverse reasoning mechanism, the correctness of the fault propagation path is verified.

2. The boiler system fault diagnosis method based on hierarchical directed graph according to claim 1, characterized in that: The parameters in the boiler system process flow diagram are abstracted as nodes, and the parameter influence relationship is abstracted as directed edges to construct a symbolic directed graph. The parameter influence relationship includes: when there is an influence relationship between two nodes, a directed edge is used from the first node to the affected second node. There are two types of directed edges: One is a directed edge with a positive influence relationship. When the parameter value of the first node increases, the parameter value of the second node on the other side of the directed edge also increases. The directed edge of the positive influence relationship is represented by a solid arrow. The other is a negative impact directed edge. When the parameter value of the first node decreases, the parameter value of the second node on the other side of the directed edge increases. The directed edge of the negative impact relationship is represented by a dotted arrow. When there is a positive influence relationship between node i and node j, use C ij =1 means that when there is a negative influence relationship between node i and node j, C ij = -1 means; when there is no influence relationship between node i and node j, that is, there is no arrow between the nodes in the directed graph, then the node influence relationship C ij =0.

3. The boiler system fault diagnosis method based on hierarchical directed graph according to claim 1, characterized in that: The symbolic directed graph is layered to obtain a layered directed graph, including: According to the symbolic directed graph, construct the adjacency matrix; According to the symbolic directed graph, construct the reachability matrix; According to the reachability matrix, the predecessor set and the reachable set are obtained; according to the intersection of the predecessor set and the reachable set, the number of logical layers of all nodes is determined, thus obtaining a hierarchical directed graph; According to the symbolic directed graph, construct the adjacency matrix, including: If there is a direct directed edge between node i and node j, then element a in the adjacency matrix ij is defined as 1; if there is no direct directed edge between node i and node j, then the element a in the adjacency matrix ij is defined as 0; If the element a in the adjacency matrix ij If it is 0, it means that there is no direct directed edge between node i and node j, and the failure of node i does not directly affect node j; if the element a in the adjacency matrix is ij If it is 1, it means that there is a direct directed edge between node i and node j, and the failure of node i directly affects node j; According to the symbolic directed graph, a reachability matrix is constructed, including: If there is at least one directed path from node i to node j, the element in the reachability matrix is defined as 1; if there is no directed path from node i to node j, the element in the reachability matrix is defined as 0; If the element a in the reachable matrix ij If it is 0, it means that node i does not affect node j; If the element a in the reachable matrix ij If it is 1, it means that node i affects node j; According to the reachable matrix, find the predecessor set and reachable set, including: In the reachability matrix P ij In the reachable set R1, select all nodes corresponding to the first row element value 1, and in the reachable matrix P ij , select all nodes whose first column element value is 1 to form the predecessor set A1; In the reachability matrix P ij In the reachable matrix P, select all nodes corresponding to the second row element value 1 to form the reachable set R2. ij , select all nodes whose second column element value is 1 to form the predecessor set A2; In the reachability matrix P ij In the reachable matrix P, select all nodes corresponding to the third row element value 1 to form the reachable set R3. ij , select all nodes whose third column element value is 1 to form the predecessor set A3; By analogy, find all predecessor sets and reachable sets in the symbolic directed graph.

4. The boiler system fault diagnosis method based on hierarchical directed graph according to claim 3 is characterized in that: According to the intersection of the predecessor set and the reachable set, the logical number of layers of all nodes is determined, thus obtaining a hierarchical directed graph, including: Use the predecessor set Ai and the reachable set Ri,i∈{1,…,12} to perform stratification: If L1={v i |Ri∩Ai=Ai}, then calculate the predecessor sets Ai and reachable sets Ri of other nodes in the symbolic directed graph in turn, and find the nodes belonging to the first layer of the hierarchical directed graph; where L1 represents the first layer, i represents the node number; L1={v i |Ri∩Ai=Ai} means that if the intersection of the reachable set of node i and the predecessor set of node i is the predecessor set of node i, node i is in the first layer of the hierarchical directed graph, that is, node i of the symbolic directed graph belongs to the first layer of the hierarchical directed graph; After finding the nodes of the first layer of the layered directed graph, delete the rows and columns corresponding to the nodes of the first layer of the layered directed graph in the reachability matrix, and continue to find the nodes of the second layer of the layered directed graph; If L2={v j |Rj∩Aj=Aj}, then iterate in sequence to find the logical layer number of each node in the symbolic directed graph; where L2 represents the second layer, which represents the node name, L2={v j |Rj∩Aj=Aj} means that if the intersection of node j's reachable set and node j's predecessor set is node j's predecessor set, node j is in the second layer of the hierarchical directed graph, that is, node j belongs to the second layer of the hierarchical directed graph; A layered directed graph is constructed according to the symbolic directed graph and the number of layers of the nodes in the symbolic directed graph, and there is at least one node in each layer.

5. The boiler system fault diagnosis method based on hierarchical directed graph according to claim 1, characterized in that: During the operation of the boiler system, the status of all nodes in each layer of the hierarchical directed graph is monitored to obtain the real-time node status. If the parameter status value of node i is higher than the set alarm threshold upper limit, the state of node i is ψi=1, where ψi represents the state of the node, which is higher than the alarm value upper limit, lower than the alarm value lower limit, or within the normal range; if the parameter status value of node i is within the normal operating range, the state of node i is ψi=0, and if the parameter status value of node i is lower than the set alarm threshold lower limit, the state of node i is ψi=-1.

6. The boiler system fault diagnosis method based on hierarchical directed graph according to claim 1, characterized in that: Determine the fault propagation path based on real-time node status, including: If there is a positive influence relationship between node i and node j in the hierarchical directed graph, the directed edge between the nodes in the directed graph is a solid arrow. At this time, the node influence relationship C ij =1; wherein, a positive influence relationship means that when the value of one node increases, the value of another node also increases, or a positive influence relationship means that when the value of one node decreases, the value of another node also decreases; If there is no influence relationship between node i and node j in the hierarchical directed graph, there is no arrow between the nodes in the directed graph. In this case, the node influence relationship C ij =0; where no influence relationship exists means that when the value of one node changes, the value of another node will not change, and node i cannot affect node j; If there is a negative influence relationship between node i and node j in the hierarchical directed graph, the directed edge between the nodes in the directed graph is a dotted arrow. At this time, the node influence relationship C ij =-1; a negative influence relationship means that when the value of one node increases, the value of another node decreases, or a negative influence relationship means that when the value of one node decreases, the value of another node increases; If the values of node i and node j satisfy ψ i C ij ψ j =1, then nodes i and j are compatible nodes. A compatible node means that when the value of node i is abnormal, the value of node j will also change accordingly. If the nodes meet ψ i C ij ψ j C jk ψ k C kl ψ l =1, then the path composed of node i, node j, node k, and node l becomes the fault propagation path, and the fault propagates in the fault propagation path; i C ij ψ j C jk ψ k C kl ψ l =1 means that the abnormal state of node i can affect node j, the abnormal state of node j can affect node k, and so on until the abnormal state of node k can affect node l.

7. The boiler system fault diagnosis method based on hierarchical directed graph according to claim 1, characterized in that: According to the membership degree, the fault source node is determined in the fault propagation path, including: The parameter state Ψi is assigned a membership degree u(v i ) calculation: v i is the state value of node i, The upper and lower limits of the node alarm thresholds are: are the upper limit and lower limit of normal working condition for node i; In a hierarchical directed graph, if a child node l has two parent nodes, i and j, the propagation source of node l is determined based on the membership of parent nodes i and j. When the membership of node i is greater than that of node j, the fault source node of child node l is i. When the membership of node j is greater than that of node i, the fault source node of child node l is j.

8. A boiler system fault diagnosis system based on a hierarchical directed graph is characterized by: include: A construction module is configured to: abstract parameters in the boiler system process flow diagram into nodes, abstract parameter influence relationships into directed edges, and construct a signed directed graph; A processing module is configured to: perform hierarchical processing on the symbolic directed graph to obtain a hierarchical directed graph; The propagation path determination module is configured to: monitor the status of all nodes in each layer of the hierarchical directed graph during the operation of the boiler system to obtain real-time node status; Determine the fault propagation path based on real-time node status; A fault source node determination module is configured to: determine the fault source node in the fault propagation path according to the membership degree; The verification module is configured to verify the correctness of the fault propagation path based on a reverse reasoning mechanism.

9. An electronic device, comprising: a memory for non-transitory storage of computer-readable instructions; as well as a processor for executing said computer-readable instructions, When the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 7 is executed.

10. A storage medium, characterized in that: Non-transitory storage of computer-readable instructions, wherein when the non-transitory computer-readable instructions are executed by a computer, the method according to any one of claims 1 to 7 is performed.

Images