Monitoring management method for nuclear power equipment

Abstract

The invention discloses a monitoring management method for nuclear power equipment, which relates to the technical field of nuclear power equipment monitoring management, and is characterized in that a cloud monitoring management platform is arranged for accessing a plurality of pieces of nuclear power equipment and configuring a link environment of each piece of nuclear power equipment so as to obtain equipment live operation data of each piece of nuclear power equipment; the method comprises the following steps: acquiring equipment historical operation data of each piece of nuclear power equipment and performing tagging processing to generate an equipment operation tagging data set corresponding to each piece of nuclear power equipment, and performing tree mapping association between the equipment live operation data of each piece of nuclear power equipment and the corresponding equipment operation tagging data set; the method comprises the steps of generating a to-be-monitored data item corresponding to each piece of nuclear power equipment, establishing an equipment monitoring analysis model corresponding to each piece of nuclear power equipment according to the to-be-monitored data item corresponding to each piece of nuclear power equipment, performing equipment fault prediction of each piece of nuclear power equipment, and determining the fault type of each piece of nuclear power equipment according to different fault types of equipment fault prediction. And selecting to execute different equipment management business processes.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear power equipment monitoring and management, and specifically to a monitoring and management method for nuclear power equipment. Background Technique

[0002] Nuclear power equipment refers to the nuclear energy reactor system used for power generation, including nuclear reactors, nuclear fuel assemblies, cooling systems, control systems, etc. Nuclear power equipment uses the principle of nuclear energy to generate heat energy, which is then converted into electrical energy for supply to the power grid or specific users. Its core is the nuclear reactor, which heats the coolant by the energy released from nuclear fission or nuclear fusion reactions, generates steam to drive the turbine generator to rotate, and finally realizes power generation.

[0003] However, for the current monitoring and management of nuclear power equipment, there are problems that the relevant data collected from the operation monitoring of nuclear power equipment are not accurate enough, and the security and reliability of the data are difficult to guarantee. Accurate data is an important basis for subsequent monitoring and management. The accuracy of predicting the faults existing in nuclear power equipment is also too low, resulting in the inability to detect the corresponding faults of nuclear power equipment in time and solve the faults in time. These are all problems existing in the traditional monitoring and management methods of nuclear power equipment. Summary of the Invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a monitoring and management method for nuclear power equipment.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A monitoring and management method for nuclear power equipment includes the following steps: Step S1: Set up a cloud monitoring and management platform, connect several nuclear power equipment through the cloud monitoring and management platform, and configure the corresponding link environment for each nuclear power equipment, so as to obtain the actual operation data of each nuclear power equipment. Step S2: Obtain the historical operation data of each nuclear power equipment, perform tagging processing on the historical operation data of the equipment, and then generate an equipment operation annotation data set corresponding to each nuclear power equipment, and perform a tree-shaped mapping association between the actual operation data of each nuclear power equipment and the corresponding equipment operation annotation data set. Step S3: Generate a data item to be monitored corresponding to each nuclear power equipment through the tree-shaped mapping association, establish an equipment monitoring and analysis model corresponding to each nuclear power equipment according to the data item to be monitored corresponding to each nuclear power equipment, and perform equipment fault prediction for each nuclear power equipment through the equipment monitoring and analysis model. Step S4: Select and execute corresponding different equipment management business processes according to different fault types of equipment fault prediction corresponding to each nuclear power equipment.

[0006] Further, the process of setting up the cloud monitoring and management platform to connect several nuclear power equipment includes: Set up a cloud monitoring management platform and perform platform initialization operations on the cloud monitoring management platform. The platform initialization operations include server connection initialization and platform environment initialization; The server connection initialization is as follows: Set up a server. The server includes several gateway ports with different IP addresses. Each gateway port is set with different types of port statuses. The types of port statuses include port idle and port occupied. The cloud monitoring management platform is set with several platform ports. The several platform ports and several gateway ports are numbered respectively and are denoted as i and j. Then, i = 1, 2, 3,..., n, j = 1, 2, 3,..., m, where both n and m are natural numbers greater than 0; Starting from the platform port with the number i = 1, sequentially send respective port access requests to the server, and select any gateway port with the port status of port idle in the server. Perform port access according to the access information corresponding to the port access request. After the port access is completed, change the port status of the gateway port accessed by the port to port occupied, and generate a corresponding access sequence; The platform environment initialization is as follows: When the port access of each platform port is completed, obtain the port parameters of each platform port, and sequentially determine whether the platform parameters of each platform port are predefined standard parameters. If so, do not perform any operation. If not, associate the corresponding platform port with a preset port repair program. The port repair program obtains the port details information of the current platform port and performs port repair; When the server connection initialization and platform environment initialization of the cloud monitoring management platform are both completed, several platform ports corresponding to the cloud monitoring management platform are synchronously converted into security configuration ports. Number several nuclear power equipment and denote it as k, k = 1, 2, 3,..., p, where p is a natural number greater than 0. Access the corresponding nuclear power equipment through several security configuration ports. After the security configuration ports are accessed by the nuclear power equipment, generate corresponding access relation codes.

[0007] Furthermore, the process of configuring the connection environment corresponding to each nuclear power equipment and then obtaining the actual operation data of each nuclear power equipment includes: The cloud monitoring management platform is set with a control terminal and a terminal database. When several nuclear power equipment are all connected to the corresponding platform ports of the cloud monitoring management platform, send the access relation code corresponding to each nuclear power equipment to the control terminal. The control terminal sets the corresponding control priority levels according to the order of the access relation codes sent by the nuclear power equipment, and configures the connection environment of the corresponding nuclear power equipment in the order from the largest to the smallest of the control priority levels; The configuration of the link environment is as follows: A transmission link channel is established for each nuclear power equipment connected to the port through the platform port of the cloud monitoring and management platform, and the environmental information of the link environment corresponding to the transmission link channel is obtained. The environmental information includes the transmission packet loss rate, the transmission delay rate, and the transmission rate, and their respective values are denoted as D1, D2, and D3 respectively. The expected threshold values of the transmission packet loss rate, the transmission delay rate, and the transmission rate are set, and are denoted as Q1, Q2, and Q3 respectively; When D1 ≤ Q1 and D2 ≤ Q2 and D3 ≥ Q3 hold, the link environment of the transmission link channel corresponding to the current nuclear power equipment does not need to be configured; When any one of the conditions D1 ≤ Q1 or D2 ≤ Q2 or D3 ≥ Q3 does not hold, the channel bandwidth corresponding to the current transmission link channel is configured, or the channel signal strength corresponding to the transmission link channel is configured. Then, after D1 ≤ Q1 and D2 ≤ Q2 and D3 ≥ Q3 are satisfied, the configuration of the channel bandwidth and the channel signal strength corresponding to the transmission link channel is stopped. The device actual operation data corresponding to each nuclear power equipment is obtained through the transmission link channel corresponding to each nuclear power equipment. The device actual operation data corresponding to the nuclear power equipment numbered k is denoted as C[k]. The control terminal sends the device actual operation data corresponding to several nuclear power equipment to the terminal database for storage.

[0008] Furthermore, the process of obtaining the device historical operation data corresponding to each nuclear power equipment and performing tagging processing to generate the device operation annotation dataset corresponding to each nuclear power equipment includes: Obtain the device historical operation data corresponding to each nuclear power equipment. The device historical operation data corresponding to the nuclear power equipment numbered k is denoted as H[k]. The device historical operation data of each nuclear power equipment consists of the node operation data corresponding to several historical time nodes during the operation of the nuclear power equipment. Set several classification tags for tagging the device historical operation data; Each classification tag corresponds to a classification keyword set. The classification keyword set includes several classification keywords. The node operation data corresponding to several historical time nodes is segmented into several node vocabulary sets. The node vocabulary set includes several node keywords. The set similarity between the node vocabulary set and several classification keyword sets in each node operation data is statistically calculated. The classification tag of the classification keyword set with the highest set similarity value corresponding to each node operation data is used as the classification tag of the current node operation data. The node operation data in the same classification tag is classified into a group of device annotation data, and then several groups of device annotation data are classified. The several groups of device annotation data corresponding to each nuclear power equipment are summarized, and the device operation annotation dataset corresponding to each nuclear power equipment is generated.

[0009] Furthermore, the process of performing a tree - shaped mapping association between the actual operation data of each nuclear power equipment and the corresponding equipment operation annotation data set, and then generating the data items to be monitored for each nuclear power equipment includes: Obtain the actual operation data of each nuclear power equipment and the corresponding equipment operation annotation data set. Mark the actual operation data of the equipment as the tree source node, and mark the corresponding several groups of equipment annotation data in the equipment operation annotation data set as several leaf nodes respectively. Starting from the tree source node as the starting path point, and using the several leaf nodes as the ending path points, map several line segments from the starting path point to the several ending path points, and then construct the corresponding several tree - branch paths, and label each tree - branch path with the corresponding path weight. The path weight is the proportion of the number of groups of equipment annotation data corresponding to each tree - branch path in the total number of groups of equipment annotation data in the equipment operation annotation data set; When the tree - shaped mapping association between the actual operation data of each nuclear power equipment and the corresponding equipment operation annotation data set is completed, generate the data items to be monitored for each nuclear power equipment. The data items to be monitored include the actual operation data corresponding to the tree source node, and the equipment annotation data at the leaf nodes of several tree - branch paths with different path weights. Use the actual operation data of the equipment as the first modeling data, and use the equipment annotation data at the leaf nodes of several tree - branch paths as the second modeling data.

[0010] Furthermore, the process of establishing the equipment monitoring and analysis model for each nuclear power equipment includes: Establish an initial monitoring model through a convolutional neural network, and set the input elements and the corresponding input nodes of the input elements, and set the output elements and the corresponding output nodes of the output elements. Denote the number of input nodes and output nodes as N1 and N2 respectively. Input several groups of equipment annotation data corresponding to the current nuclear power equipment to the monitoring model for the N1 input nodes, and then generate N2 kinds of prediction result data. When several groups of equipment annotation data corresponding to each nuclear power equipment are all input to the monitoring model and all generate their corresponding N2 kinds of prediction result data, then convert the initial monitoring model corresponding to each nuclear power equipment into the equipment monitoring and analysis model corresponding to each nuclear power equipment.

[0011] Furthermore, the process of predicting equipment failures for each nuclear power equipment by the equipment monitoring and analysis model includes: Set the fault prediction time period, denoted as T 预测 , where T 预测 =[t1, t2]. At the start time point corresponding to t1, obtain the real - time equipment operation data of each nuclear power equipment, and use the equipment monitoring and analysis model corresponding to each nuclear power equipment to locate the fault type of the current nuclear power equipment according to the real - time equipment operation data corresponding to the current nuclear power equipment. The fault types include existing faults and unknown faults; Set up a fault storage pool, associate each nuclear power equipment with its corresponding equipment monitoring and analysis model. At the end time point corresponding to t2, each nuclear power equipment analyzes through its own equipment monitoring and analysis model to obtain its own fault type and the corresponding fault detail data of each nuclear power equipment. And allocate a corresponding storage partition for each nuclear power equipment in the fault storage pool, store the fault detail data corresponding to each nuclear power equipment through the storage partition, and use the fault type corresponding to each nuclear power equipment as the partition identifier of the current corresponding storage partition; When the fault type is an existing fault, the corresponding partition identifier is Str1. When the fault type is an unknown fault, the corresponding partition identifier is Str2. Combine all the storage partitions in the fault storage pool with the partition identifier Str1 to generate a first fault combined area, and combine all the storage partitions in the fault storage pool with the partition identifier Str2 to generate a second fault combined area.

[0012] Furthermore, the process of selecting and executing different equipment management business processes according to the different fault types predicted by each nuclear power equipment's equipment fault includes: Select and execute the corresponding different equipment management business processes according to different fault types. The equipment management business processes include a first business process and a second business process. The first fault combined area corresponds to storing the fault detail data of all nuclear power equipment with the fault type of existing faults, and the second fault combined area corresponds to storing the fault detail data of all nuclear power equipment with the fault type of unknown faults. Select and execute the first business process for all nuclear power equipment corresponding to the first fault combined area, and select and execute the second business process for all nuclear power equipment corresponding to the second fault combined area.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: By setting up a cloud monitoring and management platform to access a number of nuclear power equipment and configuring the corresponding link environment for each nuclear power equipment, the actual operation data of each nuclear power equipment can be obtained. To a certain extent, the configuration of the link environment ensures the accuracy and security of obtaining the actual operation data of the corresponding nuclear power equipment; By obtaining the historical operation data of each nuclear power equipment, performing tagging processing, and then generating the corresponding equipment operation annotation dataset, and establishing a tree-like mapping association between the actual operation data of each nuclear power equipment and the equipment operation annotation dataset, the data items to be monitored for each nuclear power equipment are generated. According to the data items to be monitored, an equipment monitoring and analysis model for each nuclear power equipment is established, and equipment failure prediction for each nuclear power equipment is carried out through the equipment monitoring and analysis model. According to different failure types predicted by the equipment failure prediction, corresponding different equipment management business processes are selected and executed. Among them, the establishment of the equipment monitoring and analysis model for each nuclear power equipment improves the accuracy of failure prediction. Executing different equipment management business processes according to different failure types enables timely discovery and solution of equipment failures, realizing the efficient and stable operation of nuclear power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is a flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] As Figure 1 shown, a monitoring and management method for nuclear power equipment includes the following steps: Step S1: Set up a cloud monitoring and management platform, and let the cloud monitoring and management platform access a number of nuclear power equipment and configure the corresponding link environment for each nuclear power equipment, so as to obtain the actual operation data of each nuclear power equipment; Step S2: Obtain the historical operation data of each nuclear power equipment, perform tagging processing on the historical operation data of the equipment, and then generate the equipment operation annotation dataset corresponding to each nuclear power equipment, and establish a tree-like mapping association between the actual operation data of each nuclear power equipment and the corresponding equipment operation annotation dataset; Step S3: Generate the data items to be monitored for each nuclear power equipment through the tree-like mapping association, establish an equipment monitoring and analysis model for each nuclear power equipment according to the data items to be monitored for each nuclear power equipment, and perform equipment failure prediction for each nuclear power equipment through the equipment monitoring and analysis model; Step S4: According to different failure types predicted by the equipment failure prediction of each nuclear power equipment, select and execute corresponding different equipment management business processes.

[0016] It should be further noted that in the specific implementation manner, the process of setting up a cloud monitoring and management platform and letting the cloud monitoring and management platform access a number of nuclear power equipment includes: Set up a cloud monitoring management platform and perform platform initialization operations on the cloud monitoring management platform. The platform initialization operations include server connection initialization and platform environment initialization, and the server connection initialization and platform environment initialization are carried out synchronously. The content of the server connection initialization is: Set up a server. The server includes several gateway ports with different IP addresses, and each gateway port is set with different types of port statuses. The types of the port statuses include port idle and port occupied. The cloud monitoring management platform is set with several platform ports. The several platform ports and several gateway ports are numbered respectively and are denoted as i and j. Then there are i = 1, 2, 3,..., n, j = 1, 2, 3,..., m, where both n and m are natural numbers greater than 0. The IP address of the gateway port and the corresponding number j of the gateway port are combined as the unique identity authentication and differentiation identifier for each gateway port. Starting from the platform port with the number i = 1, sequentially send the corresponding port access requests to the server, and select any gateway port with the port status being port idle in the server. Perform port access according to the access information corresponding to the port access request, and after the port access is completed, change the port status of the gateway port accessed by the port to port occupied. For example: The platform port with the number i performs port access and is port accessed to the gateway port with the number j, generating a corresponding access sequence, denoted as R, and R = <i, j>. The content of the platform environment initialization is: When the port access of each platform port is completed, it means that the current cloud monitoring management platform has been successfully connected to the server. Obtain the port parameters of each platform port, and sequentially determine whether the platform parameters of each platform port are predefined standard parameters. If so, it means it is in a safe environment and no operation is performed. If not, it means it is in a dangerous environment, and associate the corresponding platform port with a port repair program. The port repair program obtains the port details information of the current platform port, and matches the corresponding port correction information according to the port detailed information, and then repairs the corresponding platform port through the port correction information. When both the server connection initialization and the platform environment initialization of the cloud monitoring management platform are completed, several platform ports corresponding to the cloud monitoring management platform are synchronously converted into security configuration ports. The several nuclear power equipment are numbered and denoted as k, and there are k = 1, 2, 3,..., p, where p is a natural number greater than 0. And access several nuclear power equipment through several security configuration ports. After the security configuration ports are accessed by the nuclear power equipment, corresponding access relation codes are generated, denoted as Sign, and Sign = <i, k>. The meaning is: The security configuration port with the number i is accessed by the nuclear power equipment with the number k.

[0017] It should be further noted that in the specific implementation, the process of configuring the link environment corresponding to each nuclear power equipment and then obtaining the actual operation data of each nuclear power equipment includes: The cloud monitoring and management platform is provided with a control terminal and a terminal database. When several nuclear power equipment are all connected to the corresponding platform ports of the cloud monitoring and management platform, that is, the nuclear power equipment numbered k is connected to the platform port numbered i. It should be noted that at this time, all the platform ports are security configuration ports; Send the access relationship code corresponding to each nuclear power equipment to the control terminal. The control terminal sets the corresponding control priority level according to the order of the access relationship codes sent by the nuclear power equipment, and configures the link environment of the corresponding nuclear power equipment in descending order of the control priority level; The configuration content of the link environment for each nuclear power equipment is as follows: establish a transmission link channel for each nuclear power equipment connected to the corresponding port through the platform port of the cloud monitoring and management platform, obtain the environmental information of the link environment corresponding to the transmission link channel. The environmental information includes transmission packet loss rate, transmission delay rate, and transmission rate, and record the values of the transmission packet loss rate, transmission delay rate, and transmission rate as D1, D2, and D3 respectively. Set the expected threshold values of the transmission packet loss rate, transmission delay rate, and transmission rate, and record them as Q1, Q2, and Q3 respectively; When all of D1≤Q1, D2≤Q2, and D3≥Q3 are established, it means that the link environment of the transmission link channel corresponding to the current nuclear power equipment does not need to be configured; When any one of the conditions D1≤Q1, D2≤Q2, or D3≥Q3 does not hold, then configure the channel bandwidth corresponding to the current transmission link channel, or configure the channel signal strength corresponding to the transmission link channel, so as to satisfy the following conditions: after D1≤Q1, D2≤Q2, and D3≥Q3, stop configuring the channel bandwidth and channel signal strength corresponding to the transmission link channel, and obtain the actual operation data of each nuclear power equipment through the transmission link channel corresponding to each nuclear power equipment; Record the actual operation data of the nuclear power equipment numbered k as C[k]. The control terminal sends the actual operation data of several nuclear power equipment to the terminal database for storage.

[0018] It should be further noted that in the specific implementation, the process of obtaining the historical operation data of each nuclear power equipment and performing tagging processing on the historical operation data of the equipment, and then generating the equipment operation annotation data set corresponding to each nuclear power equipment includes: Obtain the device historical operation data corresponding to each nuclear power device. Denote the device historical operation data corresponding to the nuclear power device numbered k as H[k]. The device historical operation data of each nuclear power device consists of the node operation data corresponding to several historical time nodes during the operation of the nuclear power device. Set several classification labels for the labeling process of the device historical operation data; The content of the labeling process is as follows: Each classification label corresponds to a set of classification keywords. The set of classification keywords includes several classification keywords. Segment the node operation data corresponding to several historical time nodes into several sets of node vocabulary through semantic understanding technology. The set of node vocabulary includes several node keywords. Calculate the set similarity between the set of node vocabulary in each node operation data and several sets of classification keywords, denoted as G; Use the classification label of the set of classification keywords with the highest set similarity G value corresponding to each node operation data as the classification label of the current node operation data. Classify the node operation data under the same classification label into a group of device annotation data, and then classify several groups of device annotation data corresponding to several classification labels. Aggregate several groups of device annotation data corresponding to each nuclear power device, and then generate the device operation annotation data set corresponding to each nuclear power device.

[0019] It should be further noted that in a specific implementation manner, the process of generating the data items to be monitored corresponding to each nuclear power device by performing a tree - shaped mapping association between the device actual operation data of each nuclear power device and the corresponding device operation annotation data set includes: Obtain the device actual operation data and the device operation annotation data set corresponding to the nuclear power device numbered k. Mark the device actual operation data as the tree source node, and mark several groups of device annotation data corresponding to the device operation annotation data set as several leaf nodes respectively. Construct the corresponding tree - shaped mapping association through a tree source node and several leaf nodes; The content of the tree - shaped mapping association is: Use the tree source node as the starting path point, and use several leaf nodes as the ending path points. Map several line segments from the starting path point to several ending path points, and then construct several corresponding tree branch paths. Mark each tree branch path with a corresponding path weight. The path weight is the proportion of the number of groups of device annotation data corresponding to each tree branch path in the total number of groups of device annotation data in the device operation annotation data set. Among them, the higher the proportion, the greater the path weight; When the tree - shaped mapping association between the device actual operation data of each nuclear power device and the corresponding device operation annotation data set is completed, the data items to be monitored corresponding to each nuclear power device are generated. The data items to be monitored include the device actual operation data corresponding to the tree source node, and the device annotation data at the leaf nodes of several tree branch paths corresponding to different path weights; Among them, the actual operation data of the equipment is used as the first modeling data, and the equipment annotation data at the leaf nodes of the tree branch paths corresponding to several different path weights are used as the second modeling data.

[0020] It should be further noted that in a specific implementation manner, the process of establishing an equipment monitoring and analysis model for each nuclear power equipment according to the data items to be monitored corresponding to each nuclear power equipment includes: An initial monitoring model is established through a convolutional neural network. The input elements and output elements of the monitoring model are set, and the input nodes corresponding to the input elements and the output nodes corresponding to the output elements are set. The numbers of the input nodes and the output nodes are respectively denoted as N1 and N2. Among them, N1 is the number of groups of equipment annotation data in the data items to be monitored corresponding to each nuclear power equipment, and N2 = 3; By inputting several groups of equipment annotation data corresponding to the current nuclear power equipment to the monitoring model through N1 input nodes, N2 types of prediction result data are generated. When several groups of equipment annotation data corresponding to each nuclear power equipment are all input to the monitoring model and N2 types of prediction result data corresponding to each are generated, the initial monitoring model corresponding to each nuclear power equipment is converted into an equipment monitoring and analysis model corresponding to each nuclear power equipment.

[0021] It should be further noted that in a specific implementation manner, the process of predicting equipment failures of each nuclear power equipment through the equipment monitoring and analysis model includes: Set the fault prediction period corresponding to the equipment monitoring and analysis model, and denote the fault prediction period as T 预测 , there is T 预测 = [t1, t2], where t1 is the start time point of the fault prediction period and t2 is the end time point of the fault prediction period; During the fault prediction period, equipment failure prediction of each nuclear power equipment is carried out. At the start time point corresponding to t1, the real-time equipment operation data of each nuclear power equipment is obtained, and the equipment monitoring and analysis model corresponding to each nuclear power equipment locates the fault type of the current nuclear power equipment according to the real-time equipment operation data corresponding to the current nuclear power equipment; The fault types of the nuclear power equipment include existing faults and unknown faults; Set a fault storage pool, associate the equipment monitoring and analysis model corresponding to each nuclear power equipment with the fault storage pool. At the end time point corresponding to t2, each nuclear power equipment analyzes through its own equipment monitoring and analysis model to obtain its own fault type and the corresponding fault detail data of each nuclear power equipment, and a corresponding storage partition is allocated for each nuclear power equipment in the fault storage pool. The fault detail data corresponding to each nuclear power equipment is stored through the storage partition, and the fault type corresponding to each nuclear power equipment is used as the partition identifier of the current corresponding storage partition; When the fault type is an existing fault, the corresponding partition identifier is Str1; When the fault type is an unknown fault, the corresponding partition identifier is Str2; Combine the storage partitions in the fault storage pool with the partition identifier Str1 to generate a first fault combination area, and combine the storage partitions in the fault storage pool with the partition identifier Str2 to generate a second fault combination area.

[0022] It should be further noted that in the specific implementation, the process of selecting and executing different equipment management business processes according to different fault types corresponding to equipment fault prediction of each nuclear power equipment includes: Select and execute corresponding different equipment management business processes according to different fault types. The equipment management business processes include a first business process and a second business process. The first fault combination area corresponds to storing the fault detail data of all nuclear power equipment with the fault type of existing faults, and the second fault combination area corresponds to storing the fault detail data of all nuclear power equipment with the fault type of unknown faults; Select and execute the first business process for all nuclear power equipment corresponding to the first fault combination area. The content of the first business process is as follows: Obtain the fault detail data corresponding to all existing faults, and then formulate corresponding fault handling plans. The administrator hands over the fault handling plans to the corresponding fault maintenance personnel for processing, and records the operation records of the processing to generate corresponding processing reports; Select and execute the second business process for all nuclear power equipment corresponding to the second fault combination area. The content of the second business process is as follows: Obtain the fault detail data corresponding to all unknown faults and input them into the equipment monitoring and analysis model for model iteration. After each model iteration, judge the recognition rate of the unknown faults, denoted as λ[e], where e is the number of model iterations. The initial value of e is 0, and the value of e is incremented by one after each model iteration. e is an integer. Set a standard threshold for the recognition rate, denoted as μ; If λ[e]≥μ, then use the equipment monitoring and analysis model after this model iteration as the final equipment monitoring and analysis model. Locate the fault detail content corresponding to the unknown fault through the final equipment monitoring and analysis model, and establish an expert group to formulate corresponding troubleshooting plans according to the fault detail content, and hand over the troubleshooting plans to the corresponding fault maintenance personnel for processing; If λ[e]<μ, then continue the next model iteration until λ[e]≥μ is satisfied, and stop the model iteration of the current equipment monitoring and analysis model.

[0023] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A monitoring and management method for nuclear power equipment, characterized in that Including the following steps: Step S1: Set up a cloud monitoring and management platform. The cloud monitoring and management platform accesses a number of nuclear power equipment and configures the corresponding link environment for each nuclear power equipment, and then obtains the actual operation data of each nuclear power equipment; Step S2: Obtain the historical operation data of each nuclear power equipment, perform tagging processing on the historical operation data of the equipment, and then generate an equipment operation annotation data set corresponding to each nuclear power equipment, and perform a tree-like mapping association between the actual operation data of each nuclear power equipment and the corresponding equipment operation annotation data set; Step S3: Generate a data item to be monitored corresponding to each nuclear power equipment through the tree-like mapping association, establish an equipment monitoring and analysis model corresponding to each nuclear power equipment according to the data item to be monitored corresponding to each nuclear power equipment, and perform equipment fault prediction for each nuclear power equipment through the equipment monitoring and analysis model; Step S4: According to different fault types of equipment fault prediction corresponding to each nuclear power equipment, select and execute corresponding different equipment management business processes.

2. The monitoring and management method for nuclear power equipment according to claim 1, wherein The process of setting up a cloud monitoring and management platform to access a number of nuclear power equipment includes: Set up a cloud monitoring and management platform, and perform platform initialization operations on the cloud monitoring and management platform. The platform initialization operations include server link initialization and platform environment initialization; The server link initialization is as follows: Set up a server. The server includes a number of gateway ports with different IP addresses. Each gateway port is set with different types of port statuses. The types of port statuses include port idle and port occupied. The cloud monitoring and management platform is set with a number of platform ports. The number of platform ports and the number of gateway ports are numbered respectively, and are denoted as i and j respectively. Then there are i = 1, 2, 3,..., n, j = 1, 2, 3,..., m, where both n and m are natural numbers greater than 0; Starting from the platform port with the number i = 1, sequentially send the corresponding port access requests to the server, and select any gateway port with the port status of port idle in the server, and perform port access according to the access information corresponding to the port access request. After the port access is completed, change the port status of the gateway port accessed by the port to port occupied, and generate a corresponding access sequence; The platform environment initialization is as follows: When the port access of each platform port is completed, obtain the port parameters of each platform port, and sequentially determine whether the platform parameters of each platform port are predefined standard parameters. If so, do not perform any operations. If not, associate the corresponding platform port with a preset port repair program. The port repair program obtains the port details information of the current platform port and performs port repair; When both the server link initialization and the platform environment initialization of the cloud monitoring and management platform are completed, a number of platform ports corresponding to the cloud monitoring and management platform are synchronously converted into security configuration ports. A number of nuclear power equipment are numbered and denoted as k, k = 1, 2, 3,..., p, where p is a natural number greater than 0. Access the corresponding nuclear power equipment through a number of security configuration ports. After the security configuration ports are accessed by the nuclear power equipment, corresponding access relation codes are generated.

3. A monitoring and management method for nuclear power equipment according to claim 2, characterized in that, The process of configuring the link environment corresponding to each nuclear power equipment and then obtaining the actual operation data of each nuclear power equipment includes: The cloud monitoring and management platform is provided with a control terminal and a terminal database. After several nuclear power equipment are all connected to the corresponding platform ports of the cloud monitoring and management platform, the access relationship codes corresponding to each nuclear power equipment are sent to the control terminal. The control terminal sets the control priorities of corresponding levels according to the sequence of the access relationship codes sent by the nuclear power equipment, and configures the link environment of the corresponding nuclear power equipment in the order from large to small of the control priority levels; The configuration of the link environment is as follows: A transmission link channel is established for each nuclear power equipment accessed through the platform port of the cloud monitoring and management platform, and the environmental information of the link environment corresponding to the transmission link channel is obtained. The environmental information includes the transmission packet loss rate, the transmission delay rate, and the transmission rate, and their respective values are denoted as D1, D2, and D3 respectively. The expected threshold values of the transmission packet loss rate, the transmission delay rate, and the transmission rate are set, and are denoted as Q1, Q2, and Q3 respectively; When D1 ≤ Q1 and D2 ≤ Q2 and D3 ≥ Q3 are established, the link environment of the transmission link channel corresponding to the current nuclear power equipment does not need to be configured; When any one of the conditions D1 ≤ Q1 or D2 ≤ Q2 or D3 ≥ Q3 does not hold, the channel bandwidth corresponding to the current transmission link channel is configured, or the channel signal strength corresponding to the transmission link channel is configured. Then, after D1 ≤ Q1 and D2 ≤ Q2 and D3 ≥ Q3 are satisfied, the configuration of the channel bandwidth and the channel signal strength corresponding to the transmission link channel is stopped. The actual operation data of each nuclear power equipment is obtained through the transmission link channel corresponding to each nuclear power equipment. The actual operation data corresponding to the nuclear power equipment numbered k is denoted as C[k]. The control terminal sends the actual operation data of several nuclear power equipment to the terminal database for storage.

4. A monitoring and management method for nuclear power equipment according to claim 3, characterized in that, The process of obtaining the historical operation data of each nuclear power equipment and performing tagging processing to generate the equipment operation annotation dataset corresponding to each nuclear power equipment includes: Obtain the historical operation data of each nuclear power equipment. The historical operation data corresponding to the nuclear power equipment numbered k is denoted as H[k]. The historical operation data of each nuclear power equipment consists of the node operation data corresponding to several historical time nodes during the operation of the nuclear power equipment. Several classification tags are set for tagging the historical operation data of the equipment; Each classification label corresponds to a set of classification keywords. The set of classification keywords includes several classification keywords. The node operation data corresponding to several historical time nodes is segmented into several sets of node vocabulary. The set of node vocabulary includes several node keywords. The set similarity between the set of node vocabulary in each node operation data and several sets of classification keywords is calculated. The classification label of the set of classification keywords with the highest set similarity value corresponding to each node operation data is used as the classification label of the current node operation data. The node operation data under the same classification label is classified into a group of device annotation data, and then several groups of device annotation data are classified. The several groups of device annotation data corresponding to each nuclear power device are summarized, and a device operation annotation data set corresponding to each nuclear power device is generated.

5. A monitoring and management method for nuclear power equipment according to claim 4, characterized in that, The process of performing a tree mapping association between the device actual operation data of each nuclear power device and the corresponding device operation annotation data set, and then generating the data items to be monitored for each nuclear power device includes: Obtain the device actual operation data and the device operation annotation data set corresponding to each nuclear power device. Mark the device actual operation data as the tree source node, and mark the several groups of device annotation data corresponding to the device operation annotation data set as several leaf nodes. Starting from the tree source node as the starting path point and using the several leaf nodes as the ending path points, map several line segments from the starting path point to the several ending path points, and then construct several corresponding tree branch paths, and label each tree branch path with a corresponding path weight. The path weight is the proportion of the number of groups of device annotation data corresponding to each tree branch path in the total number of groups of device annotation data in the device operation annotation data set; When the tree mapping association between the device actual operation data of each nuclear power device and the corresponding device operation annotation data set is completed, generate the data items to be monitored for each nuclear power device. The data items to be monitored include the device actual operation data corresponding to the tree source node and the device annotation data at the leaf nodes of several tree branch paths with different path weights. Use the device actual operation data as the first modeling data and the device annotation data at the leaf nodes of several tree branch paths as the second modeling data.

6. The monitoring and management method for nuclear power equipment according to claim 5, characterized in that The process of establishing a device monitoring analysis model for each nuclear power device includes: Establish an initial monitoring model through a convolutional neural network, set the input elements and the input nodes corresponding to the input elements, set the output elements and the output nodes corresponding to the output elements, and record the number of input nodes and output nodes as N1 and N2 respectively. Input several groups of device annotation data corresponding to the current nuclear power device into the monitoring model for the N1 input nodes, and then generate N2 types of prediction result data. When several groups of device annotation data corresponding to each nuclear power device are all input into the monitoring model and each generates its corresponding N2 types of prediction result data, the initial monitoring model corresponding to each nuclear power device is converted into a device monitoring analysis model corresponding to each nuclear power device.

7. A monitoring and management method for nuclear power equipment according to claim 6, characterized in that, The process of predicting device failures for each nuclear power device by the device monitoring analysis model includes: Set a fault prediction period, denoted as T 预测 , with T 预测 = [t1, t2]. At the start time point corresponding to t1, obtain the real-time device operation data of each nuclear power equipment, and use the device monitoring and analysis model corresponding to each nuclear power equipment to locate the fault type of the current nuclear power equipment based on the real-time device operation data corresponding to the current nuclear power equipment. The fault types include existing faults and unknown faults; Set up a fault storage pool, associate each nuclear power equipment with its corresponding equipment monitoring and analysis model. At the end time point corresponding to t2, each nuclear power equipment analyzes through its own equipment monitoring and analysis model to obtain its own fault type and the corresponding fault detail data of each nuclear power equipment, and allocate a corresponding storage partition for each nuclear power equipment in the fault storage pool. Store the fault detail data corresponding to each nuclear power equipment through the storage partition, and use the fault type corresponding to each nuclear power equipment as the partition identifier of the current corresponding storage partition; When the fault type is an existing fault, the corresponding partition identifier is Str1. When the fault type is an unknown fault, the corresponding partition identifier is Str2. Combine all the storage partitions in the fault storage pool with the partition identifier Str1 to generate a first fault combination area, and combine all the storage partitions in the fault storage pool with the partition identifier Str2 to generate a second fault combination area.

8. A monitoring and management method for nuclear power equipment according to claim 7, characterized in that, The process of selecting and executing different equipment management business processes according to different fault types predicted by each nuclear power equipment's corresponding equipment faults includes: Select and execute corresponding different equipment management business processes according to different fault types. The equipment management business processes include a first business process and a second business process. The first fault combination area corresponds to storing the fault detail data of all nuclear power equipment with the fault type of existing faults, and the second fault combination area corresponds to storing the fault detail data of all nuclear power equipment with the fault type of unknown faults. Select and execute the first business process for all nuclear power equipment corresponding to the first fault combination area, and select and execute the second business process for all nuclear power equipment corresponding to the second fault combination area.

Images