Operation safety early warning method and system suitable for garbage power plant

Abstract

The invention discloses an operation safety early warning method and system suitable for a garbage power plant, and relates to the technical field of safety early warning. The method comprises the following steps: performing operation monitoring on a target garbage power plant to obtain P operation monitoring data sets; performing outlier data identification to obtain P outlier operation monitoring data sets, and performing anomaly authentication to obtain Q abnormal outlier operation monitoring data sets; running exception type prediction is carried out, and Q prediction identification node exception types are determined; retrieving in a pre-constructed node association network, and determining Q association node sets and Q association node exception type sets; and carrying out safety abnormity identification on the Q prediction identification nodes, and obtaining operation safety early warning information according to a safety abnormity result. The technical problem of insufficient accuracy and real-time performance of operation safety monitoring of the garbage power plant in the prior art is solved, and the technical effect of improving the accuracy and real-time performance of operation safety early warning of the garbage power plant is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of safety early warning, and in particular to an operation safety early warning method and system applicable to a garbage power plant. Background Art

[0002] With the widespread application of waste incineration power generation technology, waste power plants play an important role in the energy production process. However, the operating environment of waste power plants is complex and susceptible to multiple factors, such as equipment failure, operational errors or changes in the external environment. These factors may cause abnormal equipment operation and even cause safety accidents. Therefore, how to effectively monitor and predict the operating status of waste power plants and identify potential safety risks in a timely manner has become a technical problem that needs to be solved urgently in the industry. Existing safety monitoring systems for waste power plants usually rely on traditional monitoring methods, such as regular inspections and manual inspections, but these methods often cannot provide real-time and effective early warnings when faced with complex operating conditions and sudden safety incidents. Summary of the invention

[0003] The present application provides an operation safety early warning method and system applicable to a waste power plant, which solves the technical problem of insufficient accuracy and real-time performance of operation safety monitoring of a waste power plant in the prior art.

[0004] In view of the above problems, the present application provides an operation safety early warning method and system applicable to a waste power plant.

[0005] In a first aspect of the present application, there is provided an operation safety early warning method applicable to a waste power plant, the method comprising: In a preset monitoring window, the sensor topology network is used to monitor the operation of the target waste power plant to obtain P operation monitoring data sets, where P is an integer greater than or equal to 1; the P operation monitoring data sets are traversed to identify outlier data to obtain P outlier operation monitoring data sets, and abnormal authentication is performed to obtain Q abnormal outlier operation monitoring data sets, where Q is a positive integer less than or equal to P; based on the Q abnormal outlier operation monitoring data sets, operation abnormality types are predicted to determine the abnormal types of Q prediction and identification nodes; with the Q prediction and identification node abnormal types and the corresponding Q prediction and identification nodes as indexes, a search is performed in the pre-constructed node association network to determine the Q associated node sets and the Q associated node abnormal type sets; the Q associated node sets, the Q associated node abnormal type sets and the Q prediction and identification node abnormal types are combined to perform safety abnormality identification on the Q prediction and identification nodes, and operation safety warning information is obtained according to the safety abnormality results.

[0006] The second aspect of the present application provides an operation safety early warning system applicable to a waste power plant, the system comprising: The monitoring module is used to monitor the operation of the target waste power plant using the sensor topology network within the preset monitoring window, and obtain P operation monitoring data sets, where P is an integer greater than or equal to 1; the data identification module is used to traverse the P operation monitoring data sets to identify outliers, obtain P outlier operation monitoring data sets, perform abnormal authentication, and obtain Q abnormal outlier operation monitoring data sets, where Q is a positive integer less than or equal to P; the abnormal type prediction module is used to predict the operation abnormality type based on the Q abnormal outlier operation monitoring data sets, and determine the abnormal types of Q prediction and identification nodes; the retrieval module is used to use the Q prediction and identification node abnormal types and the corresponding Q prediction and identification nodes as indexes to search in the pre-constructed node association network, and determine the Q associated node sets and the Q associated node abnormal type sets; the abnormal identification module is used to perform safety abnormality identification on the Q prediction and identification nodes in combination with the Q associated node sets, the Q associated node abnormal type sets and the Q prediction and identification node abnormal types, and obtain operation safety warning information according to the safety abnormality results.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, within the preset monitoring window, the sensor topology network is used to monitor the operation of the target waste power plant, and P operation monitoring data sets are obtained, where P is an integer greater than or equal to 1. Then, the P operation monitoring data sets are traversed to identify outlier data, obtain P outlier operation monitoring data sets, perform abnormal authentication, and obtain Q abnormal outlier operation monitoring data sets, where Q is a positive integer less than or equal to P. Further, based on the Q abnormal outlier operation monitoring data sets, the operation abnormality type prediction is performed to determine the abnormal types of Q predicted identification nodes. Then, with the Q predicted identification node abnormal types and the corresponding Q predicted identification nodes as indexes, a search is performed in the pre-constructed node association network to determine the Q associated node sets and the Q associated node abnormal type sets. Finally, the Q associated node sets, the Q associated node abnormal type sets and the Q predicted identification node abnormal types are combined to identify the safety abnormalities of the Q predicted identification nodes, and the operation safety warning information is obtained according to the safety abnormality results. The technical problem of insufficient accuracy and real-time performance of the operation safety monitoring of waste power plants in the prior art is solved, and the technical effect of improving the accuracy and real-time performance of the operation safety warning of waste power plants is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic diagram of a process flow of an operation safety early warning method applicable to a waste power plant provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of an operation safety warning system suitable for a waste power plant provided in an embodiment of the present application.

[0010] Explanation of the accompanying drawings: monitoring module 11, data identification module 12, abnormality type prediction module 13, retrieval module 14, abnormality identification module 15. DETAILED DESCRIPTION

[0011] The present application solves the technical problem of insufficient accuracy and real-time performance of operation safety monitoring of waste power plants in the prior art by providing an operation safety early warning method and system suitable for waste power plants.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides an operation safety early warning method applicable to a waste power plant, wherein the method comprises: In a preset monitoring window, the sensor topology network is used to monitor the operation of the target waste power plant to obtain P operation monitoring data sets, where P is an integer greater than or equal to 1.

[0015] Within the preset monitoring window, the real-time operation monitoring of the waste power plant is carried out through the sensor topology network. The sensor topology network refers to a network composed of multiple sensor nodes. These sensor nodes are responsible for real-time monitoring of various operation data of the waste power plant; the data collected by these sensor nodes form P operation monitoring data sets, where P is the number of sensor nodes in the sensor topology network, that is, each sensor node will collect and transmit certain operation data, and the data sets of all sensor nodes constitute P data sets.

[0016] P operation monitoring data sets are traversed to identify outlier data, P outlier operation monitoring data sets are obtained, anomaly authentication is performed, and Q outlier operation monitoring data sets are obtained, where Q is a positive integer less than or equal to P.

[0017] During the monitoring process, the operating data of the waste power plant usually has a certain fluctuation range. This fluctuation is normal. For example, the sensor may produce different measurement values ​​according to different working conditions or environmental factors. Under normal circumstances, the data should fluctuate within a specific range. These fluctuations reflect the normal operation status of the equipment. The purpose of identifying outliers on P sets of operation monitoring data is to find data points that exceed the normal fluctuation range from these data, that is, to obtain P sets of outlier operation monitoring data. These out-of-range data points may represent potential anomalies, such as equipment failures, sensor problems, etc. Next, the P sets of outlier operation monitoring data are anomaly authenticated, that is, to confirm whether these abnormal data really represent a safety issue. Usually, the anomaly authentication of outlier data can be verified by setting a certain threshold to verify whether these data points are beyond the normal range. After anomaly authentication, Q sets of abnormal outlier operation monitoring data are obtained, representing the outlier data points that are judged to be abnormal after authentication.

[0018] Further, P operation monitoring data sets are traversed to identify outlier data, and P outlier operation monitoring data sets are obtained, including: The P operation monitoring data sets are traversed to perform mean calculation to obtain the means of the P operation monitoring data; the P means of the P operation monitoring data are used as P aggregation starting points, the P aggregation starting points are used as neighborhood centers, and the preset aggregation bandwidth is used as the radius, and P aggregation starting point neighborhoods are constructed in the P operation monitoring data sets; the P aggregation starting points are iterated in the P operation monitoring data sets in a first direction and a second direction according to the preset aggregation bandwidth, and P update iteration directions and P update iteration starting points are determined in combination with the P aggregation starting point neighborhoods; based on the P update iteration directions, the P aggregation starting points are iterated in the P operation monitoring data sets in a first direction and a second direction respectively. In the operation monitoring data set, the P update iteration starting points are iterated according to the preset aggregation bandwidth until the preset number of iterations is met to obtain P target aggregation points and P target aggregation point neighborhoods; the P target aggregation points are used as the neighborhood center and the preset aggregation bandwidth is used as the radius to construct the P target aggregation point neighborhoods; the P target aggregation point neighborhoods are diffused outward according to the preset aggregation bandwidth to obtain P target aggregation point diffusion stop neighborhoods; the data in the P operation monitoring data sets that are outside the P target aggregation point diffusion stop neighborhoods are added to the P outlier operation monitoring data sets respectively.

[0019] Specifically, traverse P operation monitoring data sets, and calculate the mean for each data set to obtain P operation monitoring data means, which represent the average operation status of the node within the preset monitoring window, as the aggregation starting point of each sensor node; take these aggregation starting points as the neighborhood center, set a preset aggregation bandwidth as the radius, and construct P aggregation starting point neighborhoods in the operation monitoring data set. At this time, each data point will be divided into the corresponding neighborhood according to its distance from the aggregation starting point; according to the preset aggregation bandwidth, iteratively update in the first direction and the second direction, where the first direction and the second direction represent two different directions on the left and right, and gradually optimize the definition of the neighborhood by adjusting the position and direction of the aggregation starting point to ensure that the relevant data can be more accurately aggregated. Each iteration will combine the data of the current aggregation starting point neighborhood, calculate and update the new aggregation direction and position, until the preset number of iterations is met, and obtain P target aggregation points and P target aggregation point neighborhoods. With the determination of the target aggregation point, then based on these aggregation points, the neighborhood of the target aggregation point continues to be expanded with the same preset aggregation bandwidth to form a diffusion stop neighborhood of the target aggregation point. Finally, all data points outside these diffusion stop neighborhoods, i.e., data points that deviate significantly from the normal operation mode, will be identified as outliers and added to the P outlier operation monitoring data sets respectively.

[0020] Further, the P aggregation starting points are iterated in the P operation monitoring data sets in the first direction and the second direction respectively according to the preset aggregation bandwidth, and P update iteration directions and P update iteration starting points are determined in combination with the P aggregation starting point neighborhoods, including: According to the preset aggregation bandwidth, the P aggregation starting points are iterated in the P operation monitoring data sets in the first direction and the second direction respectively to obtain P first direction aggregation iteration starting points and P second direction aggregation iteration starting points; P first direction aggregation iteration starting point neighborhoods and P second direction aggregation iteration starting point neighborhoods are constructed, and the neighborhood density of the P first direction aggregation iteration starting points and the neighborhood density of the P second direction aggregation iteration starting points are calculated; the neighborhood density of the P aggregation starting point neighborhoods is calculated to obtain the neighborhood density of the P aggregation starting points; it is determined whether the neighborhood density of the P first direction aggregation iteration starting points and the neighborhood density of the P second direction aggregation iteration starting points are both greater than the neighborhood density of the P aggregation starting points, if not, the P aggregation starting points are used as P update iteration starting points, and iteration in any direction is used as P update iteration directions.

[0021] Specifically, according to the preset aggregation bandwidth, iterative updates are performed in the first direction and the second direction respectively. In each iteration process, for each aggregation starting point in the P operation monitoring data sets, it moves along the two directions respectively, and calculates P first direction aggregation iteration starting points and P second direction aggregation iteration starting points. These new iteration starting points represent the new position of each aggregation starting point after an iterative update in the two directions. Then, based on the new aggregation iteration starting point, P first direction aggregation iteration starting point neighborhoods and P second direction aggregation iteration starting point neighborhoods are constructed respectively; through these neighborhoods, the density of each aggregation iteration starting point neighborhood can be further calculated, that is, the density of the neighborhood of the P first direction aggregation iteration starting point and the density of the neighborhood of the P second direction aggregation iteration starting point. These density values ​​reflect the concentration of data in each neighborhood. A neighborhood with a larger density means that the data points in the area are more concentrated, which may represent the normal mode of the data. At the same time, it is also necessary to calculate the neighborhood density of each aggregation starting point neighborhood, and obtain the density of the neighborhood of the P aggregation starting points. These density values ​​are used to measure the normal data distribution near the aggregation starting point, providing a basis for the subsequent outlier data judgment. Then, compare the neighborhood densities of the P first-direction aggregation iteration starting points and the neighborhood densities of the P second-direction aggregation iteration starting points, and determine whether they are all greater than the neighborhood densities of the P aggregation starting points; if the neighborhood density of a certain direction does not meet this condition, it means that the direction may not effectively aggregate to reasonable data points, then the aggregation starting point needs to remain in the original position, and further iterations are selected in other directions. Therefore, these original aggregation starting points will be used as P update iteration starting points, and iterations in any direction are selected as P update iteration directions, so as to continue to optimize the position of the aggregation starting point.

[0022] Further, respectively judging whether the P first-direction aggregation iteration starting point neighborhood densities and the P second-direction aggregation iteration starting point neighborhood densities are both greater than the P aggregation starting point neighborhood densities, further includes: If so, when the neighborhood density of the P first-direction aggregation iteration starting points is greater than the neighborhood density of the P second-direction aggregation iteration starting points, the P first-direction aggregation iteration starting points are used as P update iteration starting points, and the P first directions are used as P update iteration directions; when the neighborhood density of the P first-direction aggregation iteration starting points is less than the neighborhood density of the P second-direction aggregation iteration starting points, the P second-direction aggregation iteration starting points are used as P update iteration starting points, and the P second directions are used as P update iteration directions; when the neighborhood density of the P first-direction aggregation iteration starting points is equal to the neighborhood density of the P second-direction aggregation iteration starting points, the P aggregation starting points are used as P update iteration starting points, and iteration in any direction is used as P update iteration directions.

[0023] In addition to judging whether the neighborhood densities of the P first-direction aggregation iteration starting points and the neighborhood densities of the P second-direction aggregation iteration starting points are both greater than the neighborhood densities of the P aggregation starting points, it is also necessary to further compare the two density values ​​to determine the update direction and starting point of the iteration. Specifically, when the neighborhood densities of the P first-direction aggregation iteration starting points are greater than the neighborhood densities of the P second-direction aggregation iteration starting points, it indicates that the aggregation update along the first direction is more in line with the normal distribution of the data, so the P first-direction aggregation iteration starting points will be selected as the starting point of the update iteration, and the first direction will be used as the direction of iteration, and the subsequent iterations will continue along the first direction. On the contrary, when the neighborhood densities of the P first-direction aggregation iteration starting points are less than the neighborhood densities of the P second-direction aggregation iteration starting points, it means that the aggregation effect of the second direction is better, so the P second-direction aggregation iteration starting points will be selected as the starting point of the update iteration, and the second direction will be used as the direction of iteration, and the update will continue along the second direction. When the neighborhood densities of the two are equal, it means that the aggregation effects of the two directions are not much different, and at this time, it is no longer limited to a certain direction, but any direction can be flexibly selected for iterative update. Therefore, the P aggregation starting points will be used as the starting points for update iterations, and the iteration direction can be any direction.

[0024] Further, the P target aggregation point neighborhoods are diffused outward according to the preset aggregation bandwidth until a preset diffusion stop condition is met, and the P target aggregation point diffusion stop neighborhoods are obtained, including: The P target aggregation point neighborhoods are diffused outward according to the preset aggregation bandwidth to obtain the P target aggregation point diffuse neighborhoods; it is determined whether the density of the P target aggregation point diffuse neighborhoods of the P target aggregation point diffuse neighborhoods is greater than or equal to the density of the P target aggregation point neighborhoods of the P target aggregation point neighborhoods, and if so, the P target aggregation point diffuse neighborhoods are continued to be diffused outward according to the preset aggregation bandwidth until the preset diffusion stop condition is met, and the diffuse neighborhood obtained by the last diffusion is used as the P target aggregation point stop diffusion neighborhood, wherein the preset diffusion stop condition is that the difference in the density of the two target aggregation point diffuse neighborhoods obtained by two adjacent diffusions is less than or equal to the preset density difference threshold.

[0025] Specifically, according to the preset aggregation bandwidth, the neighborhood of the P target aggregation points is diffused so as to extend the neighborhood of the target aggregation points outward, and more operation data is included by increasing the range of the neighborhood, so as to accurately determine which data points are consistent with the normal operation mode and which may be abnormal; next, it is necessary to determine whether the density of the diffusion neighborhood of the P target aggregation points after diffusion is greater than or equal to the density of the neighborhood of the target aggregation points; if the neighborhood density after diffusion is large, it means that the expanded data area still maintains a high degree of aggregation, indicating that the diffusion is effective and the data aggregation pattern has not changed significantly. In this case, the diffusion neighborhood of the target aggregation points will continue to be expanded outward according to the preset aggregation bandwidth until the preset diffusion stop condition is met. The condition for diffusion stop is that when the density difference of the diffusion neighborhood of the target aggregation point between two adjacent diffusion results is less than or equal to the preset density difference threshold, the diffusion is stopped, that is, when the diffusion effect tends to be stable and the density difference of the neighborhood becomes small, it means that the neighborhood of the target aggregation point has been stable and no further diffusion is required. Finally, when the diffusion stop condition is met, the diffusion neighborhood obtained by the last diffusion is used as the stop diffusion neighborhood of the P target aggregation points. At this point, the neighborhood range of the target gathering point has been determined, and the points included can be regarded as normal operating data, while the data points that are not within the diffusion stop neighborhood range are identified as outlier data.

[0026] Based on Q abnormal outlier operation monitoring data sets, operation abnormality type prediction is performed to determine Q prediction identification node abnormality types.

[0027] Q abnormal outlier operation monitoring data sets are obtained through outlier data identification and anomaly authentication. These data points represent abnormal data that deviate from the normal operation mode. Using data analysis methods (such as machine learning, statistical analysis, pattern recognition, etc.), the Q abnormal outlier data sets are deeply analyzed. These data sets contain multiple possible abnormal modes, and each data point may be related to different types of abnormalities, such as equipment overload, temperature abnormality, pressure abnormality, vibration abnormality, etc. By analyzing the characteristics of these abnormal data, the abnormal type corresponding to each data point can be predicted. Specifically, relevant feature values ​​are extracted from the Q abnormal outlier data sets, such as vibration frequency, temperature fluctuation, pressure change, etc. These feature values ​​can reflect the operation status of the equipment; according to historical operation data and known abnormal types, the prediction model is trained. Common models include support vector machine (SVM), random forest, decision tree, etc.; the Q abnormal outlier data are predicted by the model to determine the abnormal type of each data point; for each abnormal data point, the model outputs its corresponding abnormal type, such as equipment failure, over-temperature alarm, low pressure, etc. Finally, based on these prediction results, the abnormal types of Q predicted identification nodes can be clarified, thus providing clear guidance for subsequent safety warnings, fault location and repair.

[0028] Using the Q prediction and identification node anomaly types and the corresponding Q prediction and identification nodes as indexes, a search is performed in the pre-constructed node association network to determine a set of Q associated nodes and a set of Q associated node anomaly types.

[0029] By indexing the anomaly types of Q prediction and identification nodes with the corresponding Q prediction and identification nodes, retrieval can be performed in the pre-built node association network to deeply explore the correlation and propagation path of abnormal data. The node association network is a structured graph model. The nodes in the network represent sensors, and the edges between nodes represent the functional dependencies between them. The retrieval results include Q association node sets and Q association node anomaly type sets. The association node set is a set of nodes that are directly or indirectly connected to the prediction and identification nodes obtained by querying the node association network; the association node anomaly type set is the corresponding anomaly type set of these association nodes, indicating whether they also have anomalies and the specific types of anomalies.

[0030] Further, using the Q prediction and identification node anomaly types and the corresponding Q prediction and identification nodes as indexes, a search is performed in the pre-built node association network to determine Q association node sets and Q association node anomaly type sets, including: Obtain M historical operation safety abnormality logs of the target waste power plant, where M is an integer greater than or equal to 1; traverse the M historical operation safety abnormality logs to extract abnormal nodes, and obtain K historical abnormal node sets and K historical abnormal node type sets, where K is an integer greater than or equal to 1; based on the K historical abnormal node sets and the K historical abnormal node type sets, perform abnormal association on P nodes in the sensor topology network to obtain the node association network.

[0031] M historical operation safety anomaly logs are obtained from the historical data of the target waste power plant, where M is an integer greater than or equal to 1, indicating the number of logs. These logs record past operation safety anomaly events, including key information such as the type of anomaly, abnormal node, and time of occurrence. Next, these M historical operation safety anomaly logs are traversed to extract the abnormal node information. By analyzing the abnormal conditions recorded in the logs, K historical abnormal node sets and K historical abnormal node type sets can be obtained, where K is an integer greater than or equal to 1, indicating the number of abnormal nodes extracted from the logs. These historical abnormal nodes represent the key locations where abnormalities occurred in the operation of the waste power plant in the past, and the corresponding abnormal node types describe the specific fault types that occurred in these nodes, such as sensor failure, equipment failure, etc. After having these historical abnormal data, based on the K historical abnormal node sets and the K historical abnormal node type sets, the P nodes in the sensor topology network are analyzed for abnormal association. The P nodes in the sensor topology network represent the various sensors in the current monitoring system, which have clear associations and functional dependencies in normal operation. Through the correlation analysis with historical abnormal nodes, the potential relationship between the current node and the historical abnormal nodes can be identified, thereby constructing a node association network, which reveals which nodes have potential abnormal propagation paths and correlations, helping the system to better understand the mutual influence between different nodes.

[0032] The Q associated node sets, the Q associated node anomaly type sets and the Q predicted identification node anomaly types are combined to perform security anomaly identification on the Q predicted identification nodes, and operational safety warning information is obtained based on the security anomaly results.

[0033] The Q associated node sets represent other nodes associated with the prediction and identification node. These nodes may be affected by the abnormality of the prediction and identification node during operation, or there is a certain functional dependency between them. The Q associated node abnormality type sets describe whether these associated nodes also have abnormalities and the specific types of their abnormalities. The Q prediction and identification node abnormality types describe the abnormal situation of each prediction and identification node. These abnormality types are combined with the abnormality types of the associated nodes to form a more comprehensive fault identification framework. Based on the Q associated node sets, Q associated node abnormality type sets and Q prediction and identification node abnormality types, security abnormality identification is performed to analyze whether there are potential security risks or fault chains. If the propagation of abnormal patterns or the accumulation of potential faults is found, corresponding operation safety warning information will be issued. These warning information usually indicates the node where the fault occurs, the fault type and the possible impact range, helping operation and maintenance personnel to take timely measures to prevent the fault from expanding or causing more serious safety problems.

[0034] Furthermore, the Q associated node sets, the Q associated node anomaly type sets and the Q predicted identification node anomaly types are combined to perform security anomaly identification on the Q predicted identification nodes, and operation safety warning information is obtained according to the security anomaly results, including: Extracting a first associated node set, a first associated node anomaly type set, and a first predicted identification node from the Q associated node sets, the Q associated node anomaly type sets, and the Q predicted identification nodes; Determine whether the Q predicted identification nodes exist in the first associated node set, and if so, when the number of abnormal types of the Q predicted identification nodes that are the same as the corresponding first associated node abnormal type set exceeds a preset number threshold, take the first predicted identification node as the first abnormal node; By analogy, the Q prediction and identification nodes are traversed to perform security anomaly identification, an abnormal node set is obtained, the abnormal node set is used as a security anomaly result, and operation safety warning information is obtained according to the security anomaly result.

[0035] Specifically, a first associated node set, a first associated node abnormal type set and a first predicted identification node are extracted from Q associated node sets, Q associated node abnormal type sets and Q predicted identification nodes; then, it is determined whether there are Q predicted identification nodes in the first associated node set; if so, the matching of the Q predicted identification node abnormal types with the first associated node abnormal type set is further analyzed; if the number of abnormal types of the predicted identification nodes that are the same as the abnormal types of the first associated nodes exceeds a preset number threshold, it means that there is a strong abnormal correlation between these predicted identification nodes and the associated nodes, and they may be part of the same fault chain, so the first predicted identification node is marked as the first abnormal node, indicating that the The node is one of the key points where the anomaly occurs; then, all Q prediction and identification nodes are traversed in turn to perform the same safety anomaly identification process. If the anomaly type match between a certain prediction and identification node and its associated node exceeds the threshold, the node will be marked as an abnormal node, and other nodes will continue to be identified until all prediction and identification nodes have made corresponding matching judgments; finally, through this traversal and one-by-one judgment method, a set of abnormal nodes is obtained, and these nodes are considered to be key nodes in the entire device that may have safety hazards; based on these identified abnormal nodes, safety anomaly results are generated, and based on this safety anomaly result, operation safety warning information is issued in a timely manner, prompting operation and maintenance personnel to focus on monitoring these nodes or conduct timely troubleshooting.

[0036] In summary, the embodiments of the present application have at least the following technical effects: First, within the preset monitoring window, the sensor topology network is used to monitor the operation of the target waste power plant, and P operation monitoring data sets are obtained, where P is an integer greater than or equal to 1. Then, the P operation monitoring data sets are traversed to identify outlier data, obtain P outlier operation monitoring data sets, perform abnormal authentication, and obtain Q abnormal outlier operation monitoring data sets, where Q is a positive integer less than or equal to P. Further, based on the Q abnormal outlier operation monitoring data sets, the operation abnormality type prediction is performed to determine the abnormal types of Q predicted identification nodes. Then, with the Q predicted identification node abnormal types and the corresponding Q predicted identification nodes as indexes, a search is performed in the pre-constructed node association network to determine the Q associated node sets and the Q associated node abnormal type sets. Finally, the Q associated node sets, the Q associated node abnormal type sets and the Q predicted identification node abnormal types are combined to identify the safety abnormalities of the Q predicted identification nodes, and the operation safety warning information is obtained according to the safety abnormality results. The technical problem of insufficient accuracy and real-time performance of the operation safety monitoring of waste power plants in the prior art is solved, and the technical effect of improving the accuracy and real-time performance of the operation safety warning of waste power plants is achieved.

[0037] Embodiment 2 is based on the same inventive concept as the operation safety early warning method applicable to a waste power plant in the above embodiment. Figure 2 As shown, the present application provides an operation safety early warning system applicable to a waste power plant, wherein the system comprises: The monitoring module 11 is used to monitor the operation of the target waste-to-energy plant using the sensor topology network within a preset monitoring window to obtain P operation monitoring data sets, where P is an integer greater than or equal to 1; A data identification module 12 is used to traverse P operation monitoring data sets to identify outlier data, obtain P outlier operation monitoring data sets, perform anomaly authentication, and obtain Q outlier operation monitoring data sets, where Q is a positive integer less than or equal to P; An abnormality type prediction module 13 is used to predict the operation abnormality type based on Q abnormal outlier operation monitoring data sets, and determine Q predicted identification node abnormality types; A retrieval module 14 is used to search the pre-built node association network using the Q predicted identification node anomaly types and the corresponding Q predicted identification nodes as indexes to determine the Q associated node sets and the Q associated node anomaly type sets; The anomaly identification module 15 is used to perform safety anomaly identification on the Q predicted identification nodes in combination with the Q associated node sets, the Q associated node anomaly type sets and the Q predicted identification node anomaly types, and obtain operation safety warning information according to the safety anomaly results.

[0038] Furthermore, the data identification module 12 is used to perform the following method: The P operation monitoring data sets are traversed to perform mean calculation to obtain the means of the P operation monitoring data; the P means of the P operation monitoring data are used as P aggregation starting points, the P aggregation starting points are used as neighborhood centers, and the preset aggregation bandwidth is used as the radius, and P aggregation starting point neighborhoods are constructed in the P operation monitoring data sets; the P aggregation starting points are iterated in the P operation monitoring data sets in a first direction and a second direction according to the preset aggregation bandwidth, and P update iteration directions and P update iteration starting points are determined in combination with the P aggregation starting point neighborhoods; based on the P update iteration directions, the P aggregation starting points are iterated in the P operation monitoring data sets in a first direction and a second direction respectively. In the operation monitoring data set, the P update iteration starting points are iterated according to the preset aggregation bandwidth until the preset number of iterations is met to obtain P target aggregation points and P target aggregation point neighborhoods; the P target aggregation points are used as the neighborhood center and the preset aggregation bandwidth is used as the radius to construct the P target aggregation point neighborhoods; the P target aggregation point neighborhoods are diffused outward according to the preset aggregation bandwidth to obtain P target aggregation point diffusion stop neighborhoods; the data in the P operation monitoring data sets that are outside the P target aggregation point diffusion stop neighborhoods are added to the P outlier operation monitoring data sets respectively.

[0039] Furthermore, the data identification module 12 is used to perform the following method: According to the preset aggregation bandwidth, the P aggregation starting points are iterated in the P operation monitoring data sets in the first direction and the second direction respectively to obtain P first direction aggregation iteration starting points and P second direction aggregation iteration starting points; P first direction aggregation iteration starting point neighborhoods and P second direction aggregation iteration starting point neighborhoods are constructed, and the neighborhood density of the P first direction aggregation iteration starting points and the neighborhood density of the P second direction aggregation iteration starting points are calculated; the neighborhood density of the P aggregation starting point neighborhoods is calculated to obtain the neighborhood density of the P aggregation starting points; it is determined whether the neighborhood density of the P first direction aggregation iteration starting points and the neighborhood density of the P second direction aggregation iteration starting points are both greater than the neighborhood density of the P aggregation starting points, if not, the P aggregation starting points are used as P update iteration starting points, and iteration in any direction is used as P update iteration directions.

[0040] Furthermore, the data identification module 12 is used to perform the following method: If so, when the neighborhood density of the P first-direction aggregation iteration starting points is greater than the neighborhood density of the P second-direction aggregation iteration starting points, the P first-direction aggregation iteration starting points are used as P update iteration starting points, and the P first directions are used as P update iteration directions; when the neighborhood density of the P first-direction aggregation iteration starting points is less than the neighborhood density of the P second-direction aggregation iteration starting points, the P second-direction aggregation iteration starting points are used as P update iteration starting points, and the P second directions are used as P update iteration directions; when the neighborhood density of the P first-direction aggregation iteration starting points is equal to the neighborhood density of the P second-direction aggregation iteration starting points, the P aggregation starting points are used as P update iteration starting points, and iteration in any direction is used as P update iteration directions.

[0041] Furthermore, the data identification module 12 is used to perform the following method: The P target aggregation point neighborhoods are diffused outward according to the preset aggregation bandwidth to obtain the P target aggregation point diffuse neighborhoods; it is determined whether the density of the P target aggregation point diffuse neighborhoods of the P target aggregation point diffuse neighborhoods is greater than or equal to the density of the P target aggregation point neighborhoods of the P target aggregation point neighborhoods, and if so, the P target aggregation point diffuse neighborhoods are continued to be diffused outward according to the preset aggregation bandwidth until the preset diffusion stop condition is met, and the diffuse neighborhood obtained by the last diffusion is used as the P target aggregation point stop diffusion neighborhood, wherein the preset diffusion stop condition is that the difference in the density of the two target aggregation point diffuse neighborhoods obtained by two adjacent diffusions is less than or equal to the preset density difference threshold.

[0042] Furthermore, the retrieval module 14 is used to execute the following method: Obtain M historical operation safety abnormality logs of the target waste power plant, where M is an integer greater than or equal to 1; traverse the M historical operation safety abnormality logs to extract abnormal nodes, and obtain K historical abnormal node sets and K historical abnormal node type sets, where K is an integer greater than or equal to 1; based on the K historical abnormal node sets and the K historical abnormal node type sets, perform abnormal association on P nodes in the sensor topology network to obtain the node association network.

[0043] Furthermore, the abnormality identification module 15 is used to perform the following method: Extract the first associated node set, the first associated node abnormal type set and the first predicted identification node from the Q associated node sets, the Q associated node abnormal type sets and the Q predicted identification nodes; determine whether the Q predicted identification nodes exist in the first associated node set, and if so, when the number of abnormal types of the Q predicted identification nodes that are the same as the corresponding first associated node abnormal type set exceeds a preset number threshold, take the first predicted identification node as the first abnormal node; and so on, traverse the Q predicted identification nodes to perform security abnormality identification, obtain an abnormal node set, take the abnormal node set as a security abnormality result, and obtain operation safety warning information based on the security abnormality result.

[0044] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0045] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0046] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An operation safety early warning method applicable to a waste power plant, characterized in that: The method comprises: In a preset monitoring window, the sensor topology network is used to monitor the operation of the target waste power plant, and P operation monitoring data sets are obtained, where P is an integer greater than or equal to 1; Traverse P operation monitoring data sets to identify outlier data, obtain P outlier operation monitoring data sets, perform anomaly authentication, and obtain Q outlier operation monitoring data sets, where Q is a positive integer less than or equal to P; Based on Q abnormal outlier operation monitoring data sets, the operation abnormality type is predicted, and Q prediction identification node abnormality types are determined; Using the Q prediction and identification node anomaly types and the corresponding Q prediction and identification nodes as indexes, search in the pre-built node association network to determine the Q association node sets and the Q association node anomaly type sets; The Q associated node sets, the Q associated node anomaly type sets and the Q predicted identification node anomaly types are combined to perform security anomaly identification on the Q predicted identification nodes, and operational safety warning information is obtained based on the security anomaly results.

2. The operation safety early warning method applicable to a waste power plant according to claim 1, characterized in that: Traverse P operation monitoring data sets to identify outlier data, and obtain P outlier operation monitoring data sets, including: Traversing the P operation monitoring data sets to perform mean calculation to obtain the mean of the P operation monitoring data; Taking the mean of the P operation monitoring data as P aggregation starting points, taking the P aggregation starting points as neighborhood centers, taking the preset aggregation bandwidth as the radius, and constructing P aggregation starting point neighborhoods in the P operation monitoring data sets; Iterate the P aggregation starting points in the P operation monitoring data sets in a first direction and a second direction according to a preset aggregation bandwidth, and determine P update iteration directions and P update iteration starting points in combination with the neighborhoods of the P aggregation starting points; Based on the P update iteration directions, in the P operation monitoring data sets, the P update iteration starting points are iterated according to a preset aggregation bandwidth until a preset number of iterations is met, so as to obtain P target aggregation points and P target aggregation point neighborhoods; Taking P target aggregation points as the neighborhood center and the preset aggregation bandwidth as the radius, constructing the neighborhood of the P target aggregation points; Diffusion is performed on the P target aggregation point neighborhoods outward according to a preset aggregation bandwidth to obtain the P target aggregation point diffusion stop neighborhoods; The data in the P operation monitoring data sets that are outside the diffusion stop neighborhoods of the P target aggregation points are added into the P outlier operation monitoring data sets respectively.

3. The operation safety early warning method applicable to a waste power plant as claimed in claim 2, characterized in that: Iterating the P aggregation starting points in the P operation monitoring data sets in a first direction and a second direction according to a preset aggregation bandwidth, and determining P update iteration directions and P update iteration starting points in combination with the P aggregation starting point neighborhoods, including: Iterate the P aggregation starting points in the P operation monitoring data sets in a first direction and a second direction according to a preset aggregation bandwidth, respectively, to obtain P first direction aggregation iteration starting points and P second direction aggregation iteration starting points; Constructing P first-direction aggregation iteration starting point neighborhoods and P second-direction aggregation iteration starting point neighborhoods, and calculating the P first-direction aggregation iteration starting point neighborhood densities and the P second-direction aggregation iteration starting point neighborhood densities; Calculating the neighborhood density of the P aggregation starting point neighborhoods to obtain the P aggregation starting point neighborhood densities; Determine whether the neighborhood densities of the P first-direction aggregation iteration starting points and the neighborhood densities of the P second-direction aggregation iteration starting points are both greater than the neighborhood densities of the P aggregation starting points. If not, take the P aggregation starting points as P update iteration starting points, and take iteration in any direction as P update iteration directions.

4. The operation safety early warning method applicable to a waste power plant as claimed in claim 3, characterized in that: The method further comprises: determining whether the P first-direction aggregation iteration starting point neighborhood densities and the P second-direction aggregation iteration starting point neighborhood densities are both greater than the P aggregation starting point neighborhood densities. If so, when the neighborhood density of the P first-direction aggregation iteration starting points is greater than the neighborhood density of the P second-direction aggregation iteration starting points, the P first-direction aggregation iteration starting points are used as the P update iteration starting points, and the P first directions are used as the P update iteration directions; When the neighborhood density of the P first-direction aggregation iteration starting points is less than the neighborhood density of the P second-direction aggregation iteration starting points, the P second-direction aggregation iteration starting points are used as the P update iteration starting points, and the P second directions are used as the P update iteration directions; When the neighborhood density of P first-direction aggregation iteration starting points is equal to the neighborhood density of P second-direction aggregation iteration starting points, the P aggregation starting points are used as P update iteration starting points, and iteration in any direction is used as P update iteration directions.

5. The operation safety early warning method applicable to a waste power plant as claimed in claim 4, characterized in that: The P target aggregation point neighborhoods are diffused outward according to the preset aggregation bandwidth until a preset diffusion stop condition is met, and the P target aggregation point diffusion stop neighborhoods are obtained, including: Diffusion is performed on the P target aggregation point neighborhoods outward according to a preset aggregation bandwidth to obtain P target aggregation point diffusion neighborhoods; Determine whether the density of the P target cluster point diffusion neighborhoods of the P target cluster point diffusion neighborhoods is greater than or equal to the density of the P target cluster point neighborhoods of the P target cluster point neighborhoods. If so, continue to diffuse the P target cluster point diffusion neighborhoods outward according to the preset aggregation bandwidth until the preset diffusion stop condition is met, and use the diffusion neighborhood obtained by the last diffusion as the P target cluster point stop diffusion neighborhood, wherein the preset diffusion stop condition is that the difference in the density of the two target cluster point diffusion neighborhoods obtained by two adjacent diffusions is less than or equal to the preset density difference threshold.

6. The operation safety early warning method applicable to a waste power plant as claimed in claim 1, characterized in that: Using the Q prediction and identification node anomaly types and the corresponding Q prediction and identification nodes as indexes, a search is performed in the pre-built node association network to determine the Q association node sets and the Q association node anomaly type sets, including: Obtain M historical operation safety abnormality logs of the target waste-to-energy plant, where M is an integer greater than or equal to 1; Traversing the M historical operation safety abnormality logs to extract abnormal nodes, obtaining K historical abnormal node sets and K historical abnormal node type sets, where K is an integer greater than or equal to 1; Based on the K historical abnormal node sets and the K historical abnormal node type sets, abnormal association is performed on P nodes in the sensor topology network to obtain the node association network.

7. The operation safety early warning method applicable to a waste power plant as claimed in claim 1, characterized in that: Combine the Q associated node sets, the Q associated node anomaly type sets and the Q predicted identification node anomaly types to perform security anomaly identification on the Q predicted identification nodes, and obtain operation safety warning information according to the security anomaly results, including: Extracting a first associated node set, a first associated node anomaly type set, and a first predicted identification node from the Q associated node sets, the Q associated node anomaly type sets, and the Q predicted identification nodes; Determine whether the Q predicted identification nodes exist in the first associated node set, and if so, when the number of abnormal types of the Q predicted identification nodes that are the same as the corresponding first associated node abnormal type set exceeds a preset number threshold, take the first predicted identification node as the first abnormal node; By analogy, the Q prediction and identification nodes are traversed to perform security anomaly identification, an abnormal node set is obtained, the abnormal node set is used as a security anomaly result, and operation safety warning information is obtained according to the security anomaly result.

8. An operation safety early warning system suitable for a waste power plant, characterized in that: A system for implementing an operation safety early warning method applicable to a waste power plant according to any one of claims 1 to 7, the system comprising: A monitoring module, used to monitor the operation of the target waste power plant using a sensor topology network within a preset monitoring window, and obtain P operation monitoring data sets, where P is an integer greater than or equal to 1; A data identification module is used to traverse P operation monitoring data sets to identify outlier data, obtain P outlier operation monitoring data sets, perform anomaly authentication, and obtain Q outlier operation monitoring data sets, where Q is a positive integer less than or equal to P; An abnormality type prediction module is used to predict the operation abnormality type based on Q abnormal outlier operation monitoring data sets, and determine the abnormality types of Q predicted identification nodes; A retrieval module is used to retrieve the Q predicted and identified node anomaly types and the corresponding Q predicted and identified nodes in a pre-built node association network to determine the Q associated node sets and the Q associated node anomaly type sets; The anomaly identification module is used to identify safety anomalies of Q predicted identification nodes in combination with Q associated node sets, Q associated node anomaly type sets and Q predicted identification node anomaly types, and obtain operational safety warning information based on the safety anomaly results.

Images