Fault diagnosis and positioning method and device based on intelligent irrigation gateway

Abstract

The invention provides a fault diagnosis and positioning method and device based on an intelligent irrigation gateway, and relates to the technical field of agricultural irrigation, and the method comprises the steps: obtaining multi-source sensor data collected from an irrigation system controlled by the irrigation gateway, and carrying out the condensation hierarchical clustering, and obtaining a hierarchical clustering tree; determining a cut-off point in the hierarchical clustering tree according to a preset distance threshold, and cutting the hierarchical clustering tree to obtain a plurality of clusters; abnormal data identification is carried out on the multi-source sensor data in the cluster, and an abnormal fault point is determined; and tracing the abnormal fault point in the tree topology structure network of the irrigation system, and carrying out path planning to obtain a shortest path for tracing the abnormal fault point so as to position the abnormal fault point in the irrigation system. Through the method and the device, the defects of low fault positioning precision and efficiency and influence on the maintenance efficiency of the irrigation system caused by possible misjudgment or missed judgment in fault detection of the irrigation system and difficulty in timely fault positioning are overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural irrigation, and in particular to a fault diagnosis and positioning method and device based on an intelligent irrigation gateway. Background Art

[0002] With the development of agricultural modernization, intelligent irrigation systems have become an important means to improve water resource utilization efficiency and crop yields. However, the equipment in the irrigation system may have various faults during long-term operation, such as pipe leakage, valve blockage or pump failure. If these faults cannot be detected and handled in time, they will seriously affect the irrigation effect and resource utilization. Therefore, it is very important to develop an efficient method for locating faults in irrigation systems.

[0003] The system data collected by the fault detection method of the irrigation system in the prior art is too single, and the judgment is made only based on pressure data or the status of equipment operation. In the judgment process, the fault is judged only by the preset threshold, which cannot fully detect the fault of the irrigation system and may also lead to misjudgment or missed judgment. In addition, after the fault is judged, it is difficult to locate the fault in time, and the fault location accuracy and efficiency are low, which affects the maintenance efficiency of the irrigation system. Summary of the invention

[0004] The present invention provides a fault diagnosis and positioning method and device based on an intelligent irrigation gateway, which is used to overcome the defects in the prior art that fault detection of the irrigation system may result in misjudgment or omission, and it is difficult to locate the fault in time, resulting in low fault positioning accuracy and efficiency, and affecting the maintenance efficiency of the irrigation system.

[0005] The present invention provides a fault diagnosis and positioning method based on an intelligent irrigation gateway, the method comprising the following steps: Acquire multi-source sensor data collected from an irrigation system controlled by an irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data, and device status data; Performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; According to a preset distance threshold, a cutoff point in the hierarchical clustering tree is determined, and the hierarchical clustering tree is cut by the cutoff point to obtain a plurality of clusters of the multi-source sensor data; Performing abnormal data identification on the multi-source sensor data in each of the clusters to determine abnormal fault points in the irrigation system; Tracing back the abnormal fault point in the tree topology network corresponding to the irrigation system, and performing path planning for the abnormal fault point to obtain the shortest path to trace back the abnormal fault point; The abnormal fault point is located in the irrigation system based on the shortest path.

[0006] In some embodiments, performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree includes: Each multi-source sensor data is taken as an initial data point cluster, and agglomerative hierarchical clustering is performed on the point cluster set formed by the initial data point cluster; The initial data point clusters to be aggregated in the agglomerative hierarchical clustering process and the merged data point clusters obtained after aggregation are used as nodes in the hierarchical clustering tree; A hierarchical clustering tree is constructed based on the nodes.

[0007] In some embodiments, performing agglomerative hierarchical clustering on the point cluster set formed by the initial data point clusters includes: In the point cluster set formed by the initial data point clusters, determining a distance matrix between any two initial data point clusters; Aggregate the two initial data point clusters with the smallest distance matrix to obtain a merged data point cluster; In each iteration process, two target data point clusters with the smallest distance matrix are selected for aggregation, wherein the target data point clusters include the initial data point clusters that have not been aggregated and the merged data point clusters obtained by aggregation; When the point cluster set formed by the initial data point cluster has been aggregated into a merged data point cluster, the iterative process ends.

[0008] In some embodiments, determining the cutoff point in the hierarchical clustering tree according to a preset distance threshold comprises: For two target nodes to be aggregated in the hierarchical clustering tree, determining a distance matrix between the two target nodes; When the distance matrix is ​​greater than a preset distance threshold, it is determined that there is a truncation point between two target nodes.

[0009] In some embodiments, the process of constructing a tree topology network corresponding to the irrigation system includes: Divide the irrigation system into a root node layer, a first child node layer, a second child node layer, and a leaf node layer in a tree shape; The irrigation gateway is used as the root node of the tree topology network and is deployed at the root node layer; The plurality of control area nodes managed by the irrigation gateway are deployed as first child nodes of the root node at the first child node layer; The pump node and the valve node controlled by the control area node are deployed as sub-nodes of the first sub-node in the second sub-node layer; Deploy the sensor nodes connected to the pump node and the valve node as leaf nodes of the second sub-node layer, wherein the sensors arranged in the sensor nodes are used to collect multi-source sensor data; The pipes connected between the irrigation gateway, the control area node, the pump node, the valve node and the sensor node are used as edges; A tree topology network is constructed based on the nodes in the root node layer, the first sub-node layer, the second sub-node layer, the leaf node layer, and the edges.

[0010] In some embodiments, tracing back to the abnormal fault point in the tree topology network corresponding to the irrigation system includes: The root node of the tree topology network is used as the starting node, and starting from the starting node, a depth-first search algorithm is called to search each node in the tree topology network, and during the search process, multi-source sensor data corresponding to each node is collected; For any current node searched, determine the difference value of the multi-source sensor data between the current node and the adjacent node. When the difference value is greater than a preset difference value threshold, determine the adjacent node as an abnormal fault point in the tree topology network, wherein the adjacent node is at least one node adjacent to the current node.

[0011] In some embodiments, the edges in the tree topology network are provided with edge weights, and the path planning for the abnormal fault point to obtain the shortest path to trace the abnormal fault point includes: Determine the root node in the tree topology network as the initial node, and determine the abnormal fault point as the target node; Invoking a dynamic programming algorithm to plan a path between the initial node and the target node, and determining a plurality of predecessor nodes between the initial node and the target node during the planning process; The multiple predecessor nodes are reversely traced back, and in the reverse traceback process, the shortest path for tracing back the abnormal fault point is constructed according to the edge weights between the predecessor nodes.

[0012] The present invention also provides a fault diagnosis and positioning device based on an intelligent irrigation gateway, which specifically includes the following modules: An acquisition module, used to acquire multi-source sensor data collected from an irrigation system controlled by an irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data, and equipment status data of the irrigation system; A clustering module, used for performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; A determination module, configured to determine a cutoff point in the hierarchical clustering tree according to a preset distance threshold, and cut the hierarchical clustering tree through the cutoff point to obtain a plurality of clusters of the multi-source sensor data; An identification module, used for identifying abnormal data of the multi-source sensor data in each of the clusters, and determining abnormal fault points existing in the irrigation system; A planning module, used to trace back the abnormal fault point in the tree topology network corresponding to the irrigation system, and perform path planning for the abnormal fault point to obtain the shortest path to trace back the abnormal fault point; A positioning module is used to locate the abnormal fault point in the irrigation system based on the shortest path.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a fault diagnosis and positioning method for the intelligent irrigation gateway as described above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the fault diagnosis and positioning method of the intelligent irrigation gateway as described in any one of the above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the fault diagnosis and positioning method of the intelligent irrigation gateway as described in any one of the above is implemented.

[0016] The fault diagnosis and positioning method and device based on the intelligent irrigation gateway provided by the present invention collects multi-source sensor data such as pressure data, flow data, environmental data and equipment status data for agglomerative hierarchical clustering, and divides the hierarchical clustering tree obtained by agglomerative hierarchical clustering by the distance threshold between clusters to obtain cluster clusters, and then identifies abnormal data one by one, thereby achieving a more comprehensive and faster diagnosis of the fault of the irrigation system, and avoiding the occurrence of misjudgment or missed judgment. After diagnosing the abnormal fault point, the shortest path planning is performed in the tree topology network corresponding to the irrigation system, so as to achieve real-time and rapid positioning of the abnormal fault point, thereby not only improving the fault positioning accuracy and efficiency, but also improving the maintenance efficiency of the irrigation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced one by one below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a flow chart of the fault diagnosis and positioning method based on the intelligent irrigation gateway provided by the present invention.

[0019] Figure 2 It is a schematic diagram of a tree topology network of the irrigation system provided by the present invention.

[0020] Figure 3 It is a structural schematic diagram of the fault diagnosis and positioning device based on the intelligent irrigation gateway provided by the present invention.

[0021] Figure 4 It is a schematic diagram of the physical structure of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Combine the following Figure 1-Figure 3 The present invention describes a fault diagnosis and positioning method and device based on an intelligent irrigation gateway. Figure 1 is a flow chart of the fault diagnosis and positioning method based on the intelligent irrigation gateway provided by the present invention, such as Figure 1 As shown, the method includes the following steps 101 to 106.

[0024] Step 101: Acquire multi-source sensor data collected from an irrigation system controlled by an irrigation gateway.

[0025] In an irrigation system, in order to diagnose abnormal fault points, it is necessary to obtain multi-source sensor data collected from the irrigation system as a basis for fault diagnosis. In the irrigation system of an embodiment of the present invention, it is a system controlled by an irrigation gateway. The irrigation gateway is a control module, such as a programmable logic controller (PLC) system, which is specifically responsible for controlling the irrigation work of each control area in the irrigation system. Each control area is provided with corresponding valves and water pumps to complete the irrigation work in the field. In addition, corresponding sensors are provided in the valves and pumps, such as pressure sensors, flow sensors (flow meters), environmental sensors, and equipment sensors. Environmental sensors (such as temperature sensors, humidity sensors, etc.) are used to be deployed in the field to measure indicators such as soil moisture, temperature, rainfall, etc. The equipment sensors include voltmeters, ammeters, and vibration meters, which are used to measure indicators such as current, voltage, and vibration data of valves and water pumps.

[0026] By deploying multi-source sensors in the irrigation system, it is possible to collect the pressure data, flow data, environmental data and equipment status data of the irrigation system in real time. As multi-source sensor data, the specific data overview is shown in Table 1 below: Table 1

[0027] Step 102: Perform agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree.

[0028] The embodiment of the present invention takes into account that there are many data types of the multi-source sensors shown in Table 1, and the differences between the sensor data of each data type are large, and the data formats are also different. Therefore, a hierarchical clustering method is used here to perform agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree. In this way, cluster analysis can be used to group the multi-source sensor data of the irrigation system, and similar data can be classified into one category, thereby identifying those abnormal data that significantly deviate from the normal data range, which is described in detail below.

[0029] First, the multi-source sensor data are combined into a comprehensive data point set. Each data point in the data point set, denoted as DP, represents a multi-source sensor data, which contains the data features of all sensors and is expressed as the following formula (1):

[0030] (1)

[0031] Next, before agglomerative hierarchical clustering, each multi-source sensor data (that is, each data point DP) is taken as an initial data point cluster, and then agglomerative hierarchical clustering is performed on the point cluster set formed by the initial data point cluster. Specifically, agglomerative hierarchical clustering includes the following two key steps, step 1 and step 2, which are explained one by one below.

[0032] Step 1: Determine the distance matrix between any two initial data point clusters in the point cluster set formed by the initial data point clusters. Here, in the point cluster set, calculate the distance matrix between each initial data point cluster, recorded as distance. The distance matrix is ​​used to measure the similarity between the initial data point clusters. The smaller the distance matrix, the higher the similarity, and vice versa.

[0033] The calculation formula of the distance matrix distance is expressed as the following formula (2): (2)

[0034] In the above formula (2), n represents the total number of initial data point clusters in the point cluster set. and They represent any two initial data point clusters in the point cluster set, and i represents the i-th initial data point cluster.

[0035] Step 2: Aggregate the two initial data point clusters with the smallest distance matrix to obtain a merged data point cluster. After calculating the distance matrix, the initial data point clusters can be aggregated according to the distance matrix. When performing aggregation, the two initial data point clusters with the smallest distance matrix are aggregated to obtain a merged data point cluster.

[0036] However, the process of agglomerative hierarchical clustering is a step-by-step aggregation process. Each aggregation process is an iterative process. In each iterative process, it is necessary to calculate the distance matrix between any two target data point clusters, and select the two target data point clusters with the smallest distance matrix for aggregation (that is, execute steps 1 and 2 above).

[0037] The target data point clusters here include the initial data point clusters that have not been aggregated and the merged data point clusters obtained by aggregation. Because after each aggregation, a new merged data point cluster will be obtained, and then combined with the remaining initial data point clusters that have not been aggregated, a new point cluster set will be formed, and then the above steps 1 and 2 will be repeated.

[0038] Through each iteration of aggregation, that is, continuously repeating the above steps 1 and 2, the point clusters formed by the initial data point clusters are finally aggregated into one cluster, and aggregation cannot be performed further. Therefore, when the point clusters formed by the initial data point clusters have been aggregated into a merged data point cluster, the iteration process ends, indicating that the agglomerative hierarchical clustering has been completed.

[0039] After the agglomerative hierarchical clustering is completed, the initial data point clusters that need to be aggregated in the agglomerative hierarchical clustering process and the merged data point clusters obtained after aggregation are used as nodes in the hierarchical clustering tree. Because each initial data point cluster needs to participate in aggregation, a new merged data point cluster will be generated in each iterative aggregation process. Therefore, when constructing the hierarchical clustering tree, each initial data point cluster and the merged data point clusters obtained by aggregation are used as nodes of the hierarchical clustering tree. Finally, a hierarchical clustering tree is constructed based on these nodes. Here, the aggregation relationship between the point clusters is used as the edge between the nodes. The hierarchical clustering tree can be constructed based on the nodes and edges. The hierarchical structure of the hierarchical clustering tree is used to show the aggregation process of the data point clusters.

[0040] For example, two initial data point clusters are aggregated as two child nodes on the hierarchical clustering tree.

[0041] Then the node of the merged data point cluster obtained by aggregation is the parent node corresponding to the two child nodes on the hierarchical clustering tree. Then the two child nodes are connected to the parent node with edges. And so on, finally a merged data point cluster is obtained, and the node corresponding to this merged data point cluster is the root node at the top of the hierarchical clustering tree.

[0042] Step 103: According to a preset distance threshold, a cutoff point in the hierarchical clustering tree is determined, and the hierarchical clustering tree is cut by the cutoff point to obtain a plurality of clusters of the multi-source sensor data.

[0043] Because in the process of aggregation, each iteration will calculate the distance matrix between each initial data point cluster in the current point cluster set (i.e., execute the above step 1). The smaller the distance matrix, the higher the similarity, indicating that the change difference between the multi-source sensor data is small and the possibility of abnormal failure is small. The larger the distance matrix, the lower the similarity, indicating that the change difference between the multi-source sensor data is large and the possibility of abnormal failure is high.

[0044] Therefore, the embodiment of the present invention presets a distance threshold here, and determines the truncation point in the hierarchical clustering tree according to the preset distance threshold. The determination process is specifically as follows: for the two target nodes to be aggregated in the hierarchical clustering tree, determine the distance matrix between the two target nodes. When the distance matrix is ​​greater than the preset distance threshold, it indicates that the multi-source sensor data corresponding to the two target nodes involved in the aggregation have large differences in variation and cannot be classified into the same group, that is, cannot be classified into one category, and need to be truncated. In this way, it can be determined that there is a truncation point between the two target nodes.

[0045] After traversing the entire hierarchical clustering tree, all existing truncation points can be determined, and then the hierarchical clustering tree can be cut through the truncation points to obtain multiple clusters of multi-source sensor data. Here, by cutting the hierarchical clustering tree, the merged clusters finally obtained by the agglomerative hierarchical clustering can be divided according to the basis of cutting, thereby obtaining multiple clusters of multi-source sensor data. Because the data change differences between these clusters are obvious, it is easy to identify anomalies. Therefore, for a large number of multi-source sensor data, agglomerative hierarchical clustering can be used to more comprehensively explore the data differences between multi-source sensor data, improve the accuracy of abnormal fault diagnosis, achieve faster diagnosis, and avoid misjudgment or missed judgment.

[0046] Step 104: perform abnormal data identification on the multi-source sensor data in each cluster to determine abnormal fault points in the irrigation system.

[0047] After dividing the multi-source sensor data into a plurality of clusters in step 103, abnormal data recognition is performed on the multi-source sensor data in each cluster to determine the abnormal fault points in the irrigation system. Here, the data of these clusters can be analyzed by a data analysis tool to represent the characteristics of the data, and the clusters are divided into normal clusters and abnormal clusters according to the data characteristics.

[0048] In a normal cluster, that is, a cluster without abnormal fault points, each multi-source sensor data in the cluster represents similar pressure, flow, environmental conditions and equipment status.

[0049] For example, in the normal cluster, the pressure data of each multi-source sensor data (i.e., pipeline pressure value, pump pressure value, valve pressure value) is displayed within the normal pressure range (e.g., 50-60 PSI), the water flow of the irrigation pipeline is also stable within the normal flow range (e.g., 100-120 L / min), and various environmental data and various equipment status data are all within the normal data range.

[0050] In the abnormal cluster, that is, the cluster with abnormal fault points, each multi-source sensor data in the cluster shows different pressure, flow, environmental conditions and equipment status.

[0051] For example, in a normal cluster, some pressure data of multi-source sensor data (i.e., pipeline pressure value, pump pressure value, valve pressure value) are all within the abnormal pressure range (for example, the pressure data suddenly drops to 30PSI), the water flow of the irrigation pipeline is also within the abnormal flow range (for example, the water flow suddenly drops to 50L / min), and other various environmental data and various equipment status data are also within the abnormal data range. This means that in the control area where these sensors are located, there may be pipeline leakage, pump failure or valve failure.

[0052] In the process of abnormal data identification, the difference value between these data features of the multi-source sensor data in the normal cluster and the abnormal cluster is determined, and then a difference threshold is preset. If the difference value is greater than the difference threshold, it means that the pump, valve, or pipeline corresponding to the multi-source sensor data has an abnormal fault. For example, the pressure data of a multi-source sensor data in the abnormal cluster is 30PSI, and the pressure data of the multi-source sensor data in the abnormal cluster are all 50PSI to 60PSI. The difference value between the data features is 20PSI (ie, 50PSI-30PSI), which is greater than the preset difference threshold of 10PSI, indicating that the pump, valve, or pipeline corresponding to the multi-source sensor data in the abnormal cluster has an abnormal fault.

[0053] By identifying the differences between the features of multi-source sensor data, all existing abnormal fault points can be quickly diagnosed from multiple clusters. This achieves a comprehensive and rapid diagnosis of abnormal faults in the irrigation system, improving the efficiency and accuracy of abnormal fault diagnosis.

[0054] Step 105: trace back to the abnormal fault point in the tree topology network corresponding to the irrigation system, and perform path planning for the abnormal fault point to obtain the shortest path to trace back to the abnormal fault point.

[0055] After all abnormal fault points in the irrigation system are diagnosed through step 104, it is necessary to locate the abnormal fault points in real time so as to quickly perform emergency maintenance on the abnormal fault points. Since the pipeline network of the irrigation system is complicated, the management control area is huge, and there may be more than one abnormal fault point, it is necessary to first trace the abnormal fault point from the irrigation system to search for the pipeline path direction of the abnormal fault point. Then, the shortest path planning is performed for each abnormal fault point to achieve the purpose of quickly locating the abnormal fault.

[0056] Therefore, the embodiment of the present invention constructs a tree topology network of the irrigation system based on the distribution characteristics of the irrigation system pipeline network structure, and then traces back to each abnormal fault point one by one from the tree topology network corresponding to the irrigation system. The construction process of the tree topology network is described below.

[0057] See also Figure 2 , Figure 2 Schematic diagram of the tree topology network of the irrigation system provided by the present invention. Figure 2 As shown, the embodiment of the present invention divides the irrigation system into a root node layer, a first sub-node layer, a second sub-node layer, and a leaf node layer in a tree shape. First, the irrigation gateway is used as the root node of the tree topology network and deployed in the root node layer. Because the irrigation gateway is the control center of the entire irrigation system, locating the abnormal fault point must start from here, because if the abnormal fault point appears in the irrigation gateway, the entire irrigation system will collapse, and there is no need to perform agglomerative hierarchical clustering.

[0058] Next, according to the tree-like hierarchical structure of the irrigation system, multiple control area nodes managed by the irrigation gateway are deployed as the first child nodes of the root node at the first child node layer, that is, Figure 2 The control area nodes A, B, and C are shown. Of course, Figure 2 This is just an example. In an actual irrigation system, the number of control areas managed by the irrigation gateway is greater than 3, for example, it may be 10 or 100. In each control area, there are corresponding valves and water pumps. Therefore, the pump node and valve node controlled by the control area node are used as sub-nodes of the first sub-node, also called second sub-nodes, and deployed in the second sub-node layer. Of course, Figure 2 This is also exemplary. In an actual irrigation system, the number of pumps and valves controlled by a control area is more than one, and may be 10 or 100.

[0059] In each pump or valve, there is a sensor connected, which is the sensor used to collect multi-source sensor data in step 101, such as a pressure sensor, a flow sensor (flow meter), an environmental sensor, and an equipment sensor. Figure 2 As shown, when constructing a tree topology network, the sensor nodes connected to the pump nodes and the valve nodes are deployed in the leaf node layer as leaf nodes of the second sub-node layer. Here, each leaf node is also a sub-node corresponding to each second sub-node in the second sub-node layer. According to the above, the sensors set in these sensor nodes are used to collect multi-source sensor data.

[0060] In the irrigation system, the irrigation gateway, control area, pump, valve and sensor are all connected by pipelines. Figure 2As shown in the figure, the pipes connecting the irrigation gateway, control area nodes, pump nodes, valve nodes and sensor nodes are taken as edges, and then a tree topology network is constructed based on the nodes and edges in the root node layer, the first child node layer, the second child node layer and the leaf node layer. Here, the irrigation gateway in the root node layer, the control area nodes in the first child node layer, the pump nodes in the second child node layer, the valve nodes and the sensor nodes in the leaf node layer are connected through edges (i.e., the pipes of the pipeline), so a tree topology network is constructed.

[0061] After constructing the tree topology network of the irrigation system, the abnormal fault point can be traced back in the tree topology network. Specifically, the root node of the tree topology network is used as the starting node. Here, the root node corresponding to the irrigation gateway is used as the starting point of the path search. In order to achieve a comprehensive positioning of the abnormal fault point without omission. The embodiment of the present invention starts from the starting node and calls the depth-first search (DFS) algorithm to search each node in the tree topology network. Because the abnormal fault point is determined by the change difference of multi-source sensors, in order to determine whether the abnormal fault point is found, it is also necessary to collect the multi-source sensor data corresponding to each node in real time during the search process.

[0062] In the process of tracing back by the depth-first search (DFS) algorithm, for any current node searched, the difference value of the multi-source sensor data between the current node and the adjacent node is determined to determine whether an abnormal fault point is found. Of course, the adjacent node is at least one node adjacent to the current node, that is, there are multiple adjacent nodes. If there are multiple adjacent nodes, the difference value can be calculated one by one, and the calculation order is not limited. When the difference value is greater than the preset difference value threshold, for example, the pressure data difference value in the multi-source sensor data of the current node and the adjacent node is 20PSI (that is, the pressure data corresponding to the current node 50PSI minus the pressure data corresponding to the adjacent node 30PSI), which is greater than the preset difference threshold 10PSI, it means that the pressure data has changed since this adjacent node, and the change difference is very large, then the corresponding adjacent node is determined as an abnormal fault point in the tree topology network, that is, it is determined that the pipeline corresponding to this adjacent node is leaking or the pump or valve is faulty.

[0063] In this way, the path between nodes is verified starting from the starting node, and all adjacent nodes are searched recursively until all abnormal fault points are searched out or there are no more adjacent nodes to continue exploring, then the exploration process ends.

[0064] By calling the depth-first algorithm to search for abnormal fault nodes, it is possible to ensure that the abnormal fault points existing in the tree topology network are fully located without omissions, thereby improving the efficiency of locating abnormal faults.

[0065] After tracing back to each abnormal fault point, the direction of the pipeline path from the irrigation gateway to the abnormal fault point is determined, so the path planning can be performed for the abnormal fault point to obtain the shortest path to trace the abnormal fault point for quick positioning. During the abnormal fault point search process, the path and path weight in the search process are also recorded. Because the edges in the tree topology network are set with edge weights, the path is also the edge in the tree topology network, and the path weight is also the edge weight, which is used to characterize the length of the connecting pipeline, for example Figure 2 For the tree topology network shown in the figure, the edge weight of each node can be set as shown in Table 2: Table 2

[0066] However, the paths obtained by searching for abnormal fault points are complex and diverse. There may be multiple paths from the irrigation gateway to the abnormal fault point, and the edge weights of each path are different. In order to improve positioning and fault maintenance efficiency, path planning is required for each abnormal fault point.

[0067] During the path planning process, the root node in the tree topology network is still determined as the initial node, and then the abnormal fault point is determined as the target node, that is, the pipeline path from the initial node to the target node is planned. The embodiment of the present invention here calls a dynamic programming algorithm to plan the path between the initial node and the target node. During the planning process, for each current node i, the shortest path from all possible predecessor nodes j to the current node i is calculated, and dp data is used to represent it, as shown in the following formula (3): (3)

[0068] In the above formula (3), represents the shortest path from the initial node to the current node i, represents the path weights from all possible predecessor nodes j to the current node i, and min represents the minimum value function.

[0069] At the beginning of planning, the dp data is initialized and defined. The path weight of the starting node is initialized to 0, that is, dp[initial node] = 0, and the other nodes are initialized to positive infinity. Then, during the planning process, the dp array is updated through state transfer to calculate the shortest path weight from the predecessor node j to the current node i. During the planning process, multiple predecessor nodes between the initial node and the target node are determined until the shortest path from the initial node to the target node is planned.

[0070] The predecessor nodes determined in the planning process are the nodes in the shortest path. Then, multiple predecessor nodes are traced back in reverse. During the reverse traceback process, the shortest path to trace the abnormal fault point is constructed based on the edge weights between the predecessor nodes.

[0071] The embodiment of the present invention, based on the comprehensive positioning of abnormal fault points existing in the tree topology network, constructs the shortest path from the starting node to the abnormal fault point through a dynamic programming algorithm, thereby improving the positioning reliability and efficiency of the abnormal fault point in a complex irrigation system.

[0072] Step 106: locate the abnormal fault point in the irrigation system based on the shortest path.

[0073] After determining the shortest path to the abnormal fault point, the abnormal fault point can be located in the tree topology network based on the shortest path. If the irrigation gateway has a visualization module, this shortest path can be highlighted. Accordingly, in the irrigation system mapped to the real scene, the corresponding abnormal fault location can be determined in the irrigation system based on the shortest path, which makes it convenient for maintenance personnel to go to the abnormal fault location as quickly as possible according to the shortest path to perform maintenance work on the irrigation system.

[0074] The fault diagnosis and positioning method based on the intelligent irrigation gateway provided by the present invention realizes an irrigation system fault detection and positioning mechanism of "cluster analysis-topology tracking-dynamic programming". By collecting multi-source sensor data such as pressure data, flow data, environmental data and equipment status data, agglomerative hierarchical clustering is performed, and the hierarchical clustering tree obtained by agglomerative hierarchical clustering is divided by the distance threshold between clusters to obtain cluster clusters, and then abnormal data is identified one by one, so as to realize a more comprehensive and faster diagnosis of the fault of the irrigation system, and also avoid the occurrence of misjudgment or missed judgment. After diagnosing the abnormal fault point, the shortest path planning is performed in the tree topology network corresponding to the irrigation system, so as to realize the real-time and rapid positioning of the abnormal fault point, thereby not only improving the fault positioning accuracy and efficiency, but also improving the maintenance efficiency of the irrigation system.

[0075] The fault diagnosis and positioning device based on the intelligent irrigation gateway provided by the present invention is described below. The fault diagnosis and positioning device based on the intelligent irrigation gateway described below and the fault diagnosis and positioning method based on the intelligent irrigation gateway described above can refer to each other.

[0076] See also Figure 3 , Figure 3 Schematic diagram of the structure of the fault diagnosis and positioning device based on the intelligent irrigation gateway provided by the present invention. Figure 3As shown, the fault diagnosis and positioning device based on the intelligent irrigation gateway provided by the present invention includes an acquisition module 301, a clustering module 302, a determination module 303, an identification module 304, a planning module 305, and a positioning module 306. Among them, the acquisition module 301 is used to acquire multi-source sensor data collected from the irrigation system controlled by the irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data and equipment status data of the irrigation system; the clustering module 302 is used to perform agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; the determination module 303 is used to determine the truncation point in the hierarchical clustering tree according to a preset distance threshold, and cut the hierarchical clustering tree through the truncation point to obtain multiple clusters of the multi-source sensor data; the identification module 304 is used to identify abnormal data of the multi-source sensor data in each of the clusters to determine the abnormal fault point in the irrigation system; the planning module 305 is used to trace back to the abnormal fault point in the tree topology network corresponding to the irrigation system, and perform path planning on the abnormal fault point to obtain the shortest path to trace back the abnormal fault point; the positioning module 306 is used to locate the abnormal fault point in the irrigation system based on the shortest path.

[0077] It should be noted that the beneficial effects of the fault diagnosis and positioning device based on the intelligent irrigation gateway here and the fault diagnosis and positioning method based on the intelligent irrigation gateway mentioned above can correspond to each other, so the beneficial effects of the fault diagnosis and positioning device based on the intelligent irrigation gateway will not be repeated here.

[0078] Figure 4 is a schematic diagram of the physical structure of an electronic device provided by the present invention, such as Figure 4As shown, the electronic device may include: a processor (processor) 410 , a communication interface (Communications Interface) 420 , a memory (memory) 430 and a communication bus 440 , wherein the processor 410 , the communication interface 420 , and the memory 430 communicate with each other through the communication bus 440 . The processor 410 can call the logic instructions in the memory 430 to execute the fault diagnosis and positioning method based on the intelligent irrigation gateway, which includes: obtaining multi-source sensor data collected from the irrigation system controlled by the irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data and equipment status data; performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; determining a truncation point in the hierarchical clustering tree according to a preset distance threshold, and cutting the hierarchical clustering tree through the truncation point to obtain multiple clusters of the multi-source sensor data; performing abnormal data identification on the multi-source sensor data in each of the clusters to determine the abnormal fault point in the irrigation system; tracing back to the abnormal fault point in the tree topology network corresponding to the irrigation system, and performing path planning on the abnormal fault point to obtain the shortest path to trace back to the abnormal fault point; locating the abnormal fault point in the irrigation system based on the shortest path.

[0079] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0080] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the fault diagnosis and positioning method based on the intelligent irrigation gateway provided by the above methods, the method comprising: obtaining multi-source sensor data collected from an irrigation system controlled by an irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data and equipment status data; performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; determining a truncation point in the hierarchical clustering tree according to a preset distance threshold, and cutting the hierarchical clustering tree through the truncation point to obtain multiple clusters of the multi-source sensor data; performing abnormal data identification on the multi-source sensor data in each of the clusters to determine the abnormal fault point in the irrigation system; tracing back to the abnormal fault point in the tree topology network corresponding to the irrigation system, and performing path planning on the abnormal fault point to obtain the shortest path to trace back to the abnormal fault point; locating the abnormal fault point in the irrigation system based on the shortest path.

[0081] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented by a processor to execute the fault diagnosis and positioning method based on the intelligent irrigation gateway provided by the above methods when the computer program is executed, the method comprising: acquiring multi-source sensor data collected from an irrigation system controlled by an irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data and equipment status data; performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; determining a truncation point in the hierarchical clustering tree according to a preset distance threshold, and cutting the hierarchical clustering tree through the truncation point to obtain a plurality of clustering clusters of the multi-source sensor data; performing abnormal data identification on the multi-source sensor data in each of the clustering clusters to determine the abnormal fault point in the irrigation system; tracing back to the abnormal fault point in the tree topology network corresponding to the irrigation system, and performing path planning on the abnormal fault point to obtain the shortest path to trace back to the abnormal fault point; locating the abnormal fault point in the irrigation system based on the shortest path.

[0082] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0083] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0084] Finally, it should be noted that 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fault diagnosis and positioning method based on an intelligent irrigation gateway, characterized in that: The method comprises: Acquire multi-source sensor data collected from an irrigation system controlled by an irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data, and device status data; Performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; According to a preset distance threshold, a cutoff point in the hierarchical clustering tree is determined, and the hierarchical clustering tree is cut by the cutoff point to obtain a plurality of clusters of the multi-source sensor data; Performing abnormal data identification on the multi-source sensor data in each of the clusters to determine abnormal fault points in the irrigation system; Tracing back the abnormal fault point in the tree topology network corresponding to the irrigation system, and performing path planning for the abnormal fault point to obtain the shortest path to trace back the abnormal fault point; The abnormal fault point is located in the irrigation system based on the shortest path.

2. The fault diagnosis and positioning method based on the intelligent irrigation gateway according to claim 1 is characterized in that: The performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree includes: Each multi-source sensor data is taken as an initial data point cluster, and agglomerative hierarchical clustering is performed on the point cluster set formed by the initial data point cluster; The initial data point clusters to be aggregated in the agglomerative hierarchical clustering process and the merged data point clusters obtained after aggregation are used as nodes in the hierarchical clustering tree; A hierarchical clustering tree is constructed based on the nodes.

3. The fault diagnosis and positioning method based on the intelligent irrigation gateway according to claim 2 is characterized in that: The step of performing agglomerative hierarchical clustering on the point cluster set formed by the initial data point cluster comprises: In the point cluster set formed by the initial data point clusters, determining a distance matrix between any two initial data point clusters; Aggregate the two initial data point clusters with the smallest distance matrix to obtain a merged data point cluster; In each iteration process, two target data point clusters with the smallest distance matrix are selected for aggregation, wherein the target data point clusters include the initial data point clusters that have not been aggregated and the merged data point clusters obtained by aggregation; When the point cluster set formed by the initial data point cluster has been aggregated into a merged data point cluster, the iterative process ends.

4. The fault diagnosis and positioning method based on the intelligent irrigation gateway according to claim 1 is characterized in that: Determining the cutoff point in the hierarchical clustering tree according to a preset distance threshold comprises: For two target nodes to be aggregated in the hierarchical clustering tree, determining a distance matrix between the two target nodes; When the distance matrix is ​​greater than a preset distance threshold, it is determined that there is a truncation point between two target nodes.

5. The fault diagnosis and positioning method based on the intelligent irrigation gateway according to claim 1 is characterized in that: The construction process of the tree topology network corresponding to the irrigation system includes: Divide the irrigation system into a root node layer, a first child node layer, a second child node layer, and a leaf node layer in a tree shape; The irrigation gateway is used as the root node of the tree topology network and is deployed at the root node layer; The plurality of control area nodes managed by the irrigation gateway are deployed as first child nodes of the root node at the first child node layer; The pump node and the valve node controlled by the control area node are deployed as subnodes of the first subnode in the second subnode layer; Deploy the sensor nodes connected to the pump node and the valve node as leaf nodes of the second sub-node layer, wherein the sensors arranged in the sensor nodes are used to collect multi-source sensor data; The pipes connected between the irrigation gateway, the control area node, the pump node, the valve node and the sensor node are used as edges; A tree topology network is constructed based on the nodes in the root node layer, the first sub-node layer, the second sub-node layer, the leaf node layer, and the edges.

6. The fault diagnosis and positioning method based on the intelligent irrigation gateway according to claim 1 is characterized in that: The tracing back to the abnormal fault point in the tree topology network corresponding to the irrigation system includes: The root node of the tree topology network is used as the starting node, and starting from the starting node, a depth-first search algorithm is called to search each node in the tree topology network, and during the search process, multi-source sensor data corresponding to each node is collected; For any current node searched, determine the difference value of the multi-source sensor data between the current node and the adjacent node. When the difference value is greater than a preset difference value threshold, determine the adjacent node as an abnormal fault point in the tree topology network, wherein the adjacent node is at least one node adjacent to the current node.

7. The fault diagnosis and positioning method based on the intelligent irrigation gateway according to claim 1 is characterized in that: The edges in the tree topology network are provided with edge weights, and the path planning for the abnormal fault point is performed to obtain the shortest path to trace back the abnormal fault point, including: Determine the root node in the tree topology network as the initial node, and determine the abnormal fault point as the target node; Invoking a dynamic programming algorithm to plan a path between the initial node and the target node, and determining a plurality of predecessor nodes between the initial node and the target node during the planning process; The multiple predecessor nodes are reversely traced back, and in the reverse traceback process, the shortest path for tracing back the abnormal fault point is constructed according to the edge weights between the predecessor nodes.

8. A fault diagnosis and positioning device based on an intelligent irrigation gateway, characterized in that: The device comprises: An acquisition module, used to acquire multi-source sensor data collected from an irrigation system controlled by an irrigation gateway, wherein the multi-source sensor data includes pressure data, flow data, environmental data, and equipment status data of the irrigation system; A clustering module, used for performing agglomerative hierarchical clustering on the multi-source sensor data to obtain a hierarchical clustering tree; A determination module, configured to determine a cutoff point in the hierarchical clustering tree according to a preset distance threshold, and cut the hierarchical clustering tree through the cutoff point to obtain a plurality of clusters of the multi-source sensor data; An identification module, used for identifying abnormal data of the multi-source sensor data in each of the clusters, and determining abnormal fault points existing in the irrigation system; A planning module, used to trace back the abnormal fault point in the tree topology network corresponding to the irrigation system, and perform path planning for the abnormal fault point to obtain the shortest path to trace back the abnormal fault point; A positioning module is used to locate the abnormal fault point in the irrigation system based on the shortest path.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the fault diagnosis and positioning method based on the intelligent irrigation gateway as described in any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the fault diagnosis and positioning method based on the intelligent irrigation gateway as described in any one of claims 1 to 7 is implemented.

Images