CN-121658980-B - Abnormal detection system for building drainage pipe network

CN121658980BCN 121658980 BCN121658980 BCN 121658980BCN-121658980-B

Abstract

The invention relates to the technical field of fault detection, in particular to a system for detecting abnormality of a building drainage pipe network, which comprises a hydraulic pressure chain construction module, a sudden drop identification module, a flow resistance classification module, a structure mapping module and a attribution verification module. According to the invention, a continuous change chain is constructed among water pressure monitoring points of a drainage pipe network, space correlation analysis is realized according to the water pressure direction and amplitude change of adjacent nodes, an abnormal region can be rapidly identified by taking chain type water pressure gradient fracture as a clue, in abnormal sudden drop identification, the timing sequence matching of upstream rebound and downstream delay is used as a verification condition, so that abnormal judgment has a dynamic response basis, flow resistance type division is formed by water pressure curve trend segmentation matching, the hierarchy of an abnormal position and pipe section attribution positioning are realized, a closed loop detection mechanism from signal chain change identification to space path confirmation is integrally realized, and the abnormal diagnosis of the drainage system has continuity, traceability and positioning precision improvement.

Inventors

YUWEN WEI
FANG YUAN
YU JIAN
GENG XIN
Shen Qingzong
WU DONG
ZHAO PEIXIANG

Assignees

山东中水管道工程有限公司

Dates

Publication Date: 20260505
Application Date: 20260205

Claims (8)

1. A system for detecting anomalies in a building drainage network, the system comprising: the hydraulic pressure chain construction module acquires real-time hydraulic pressure sequence data of hydraulic pressure monitoring points distributed in a vertical pipe, a transverse branch pipe and an inspection well section in a building drainage pipe network, invokes the hydraulic pressure change direction between the adjacent three monitoring points and forms a continuous chain type relationship according to the pipe section space arrangement sequence to generate a broken chain segment initialization identification list; The hydraulic chain construction module includes: The hydraulic sequence extraction submodule acquires real-time hydraulic sequence data of hydraulic monitoring points distributed in a vertical pipe, a transverse branch pipe and an inspection well section in a building drainage pipe network, and extracts the hydraulic change direction between three adjacent monitoring points according to the arrangement sequence of the monitoring points in space to obtain a hydraulic direction sequence chain group; The segment continuity judging submodule calls the water pressure direction sequence chain group, takes continuous three segments of water pressure monitoring points as segment units, analyzes the consistency of the water pressure change direction, judges whether the direction inside the segment has a reverse state, and marks the segment as a direction reverse segment if the direction reverse state exists, so as to obtain a direction reverse segment list; The gradient fracture identification submodule extracts the difference of the water pressure change amplitude values of nodes in the chain segments according to the direction reversal chain segment list, carries out synchronous deviation judgment according to the space synchronism between the direction reversal position and the water pressure amplitude abrupt change position, calculates a water pressure deviation synchronism index value, judges the chain segments to be in a water pressure gradient fracture state when the water pressure deviation synchronism index value is larger than a set water pressure deviation synchronism threshold value, and acquires a fracture chain segment initialization identification list; the sudden drop identification module extracts segment nodes marked as broken according to the broken segment initialization mark list, acquires a time sequence water pressure fluctuation path of adjacent upstream and downstream pipe segments, and generates a space sudden drop verification mark set by judging whether water pressure rebound occurs in a time window upstream of a sudden drop point or not; the flow resistance classification module extracts a continuous water pressure change trend curve according to the space sudden drop verification mark set, and divides the time span and the change slope of the ascending and descending paragraphs to obtain a flow resistance structure type identification table; And the structure mapping module is used for calling node coordinate information in the building structure plane and elevation layout model according to the flow resistance structure type identification table, positioning the belonging building hierarchy structure relation, analyzing the position and connection sequence of the corresponding pipe section between nodes on the vertical path in the building model, and generating an abnormal node space attribution information list.
2. The system of claim 1, wherein the list of fracture segment initialization identifiers comprises a direction reversal location, a hydraulic gradient fracture node, a hydraulic offset synchronization point, and a hydraulic segment number, the set of spatial dip verification markers comprises a dip point node location, an upstream and downstream response characteristic, a time sequence verification identifier, and a segment correspondence index, the list of flow resistance structure type identifiers comprises a transient release type number, a stagnation accumulation type number, a change slope value, and a time span label, and the list of abnormal node spatial attribution information comprises floor location information, a trunk branch type, a node spatial coordinate, and a connection sequence path.
3. The system for detecting anomalies in a building drainage network of claim 1, wherein the dip identification module comprises: The broken node extraction submodule extracts chain segment nodes marked as broken states according to the broken chain segment initialization mark list, adjusts the numbers of upstream and downstream adjacent pipe segments to the chain segment nodes, acquires time sequence water pressure fluctuation paths of the corresponding nodes, analyzes the continuous fluctuation amplitude and direction change trend in the water pressure data, and acquires a broken node index structure; The hydraulic response analysis submodule selects a fracture node to correspond to an upstream adjacent pipe section and a downstream adjacent pipe section based on the fracture node index structure, respectively collects hydraulic data sequences in a fixed time window before and after a sudden drop point, calls pressure amplitude and rebound rate of the node in the time window and time delay interval values, calculates a spatial response joint characteristic value, marks node hydraulic response behaviors and generates a hydraulic linkage marking vector; And the space verification sub-module screens the broken nodes with upstream pressure reverse rebound and downstream response lag states according to the water pressure linkage mark vector, counts the joint behavior sections of the nodes meeting the conditions, identifies the response relation between the broken nodes and the adjacent upstream and downstream, integrates the node sudden drop type, the water pressure response direction and the time lag state, and acquires a space sudden drop verification mark set.
4. A building drainage pipe network anomaly detection system according to claim 3 wherein the spatial response joint characteristic value employs the formula: ; Wherein, the Representing the spatial response joint characteristic values, Represent the first The variation of the water pressure at the upstream of each node, Represent the first The amount of water pressure fluctuation downstream of the individual nodes, Represent the first The amount of reverse amplification of water pressure caused by the response delay of each node, Represent the first The water pressure fluctuation amplitude of each node sudden drop point, Representing the total number of nodes, The broken nodes are indexed.
5. A building drainage network anomaly detection system according to claim 3 wherein the flow resistance classification module comprises: The trend curve extraction submodule extracts corresponding continuous water pressure time sequence data based on the space sudden drop verification mark set, identifies the water pressure breakpoint position in the time sequence, limits the data segments before and after marking, extracts the continuous water pressure sequence of the adjacent measuring point positions, carries out time sequence reconstruction through time indexes, and generates a water pressure change trend data sequence; The path characteristic calculation submodule invokes the water pressure change trend data sequence, searches continuous data sections before and after the sudden drop occurrence point, delimits a time index interval of a water pressure rising section and a water pressure falling section, calculates the duration of a corresponding section according to an index difference value, calculates the difference value according to the water pressure value in the sections, and generates a water pressure path change characteristic parameter set; And the type label classifying sub-module selects the duration time of the ascending segment and the change slope of the descending segment as classification indexes according to the characteristic parameter set of the water pressure path change, sets a time span threshold value and a slope judgment reference value for classifying the type of the sudden drop chain segment, compares the relation between the classification indexes and the corresponding judgment reference values, judges the type of the sudden drop chain segment and gives a corresponding type label, and a flow resistance structure type identification table is generated.
6. The system for anomaly detection of a building drainage network of claim 5, wherein the structural mapping module comprises: The node coordinate calling sub-module is used for matching and extracting corresponding node coordinate numbers in the building structure plane and facade layout model according to the sudden drop chain segment indexes recorded in the flow resistance structure type identification table, calling three-dimensional coordinate data associated with the node numbers, integrating the spatial positions and the associated numbers of components to which the nodes belong, and generating an abnormal node coordinate information set; The path structure positioning sub-module calls the abnormal node coordinate information set, analyzes the communication relation between the floor index number of the node in the building structure model and the upper and lower nodes, judges whether adjacent nodes are positioned in the same drainage partition or a cross-layer channel according to node grouping information set in the building hierarchical structure, and identifies the connection sequence on the vertical path according to the sequence relation of the upstream and downstream nodes to acquire vertical pipe section connection sequence data; and the space attribution marking sub-module reads the pipe section identification corresponding to the node and the floor number to which the node belongs according to the vertical pipe section connection sequence data, calls a drainage direction field recorded in the flow direction information, compares the drainage direction field with the upstream and downstream positions of the node, judges the floor position of the abnormal node in a drainage path, matches the structural classification fields of a main pipe and a branch pipe in building drainage, marks the corresponding pipe section type code, and generates an abnormal node space attribution information list.
7. The system for anomaly detection of a building drainage network of claim 1, further comprising a home verification module: The attribution verification module extracts response time records of the node and the upstream and downstream liquid level monitoring points according to the abnormal node spatial attribution information list, judges whether the liquid level change of the node accords with the response sequence from upstream to downstream when an abnormal event occurs, verifies that attribution positions are reasonable if downstream response is lagged, and re-extracts adjacent gradient segments to carry out reverse tracing judgment if the response time is advanced, and outputs a pipe network abnormal verification node set; the pipe network abnormal verification node set comprises a home verification success node, a home verification failure node, a reverse tracing result node and a response sequence consistency identifier.
8. The system for anomaly detection of a building drainage network of claim 7, wherein the home verification module comprises: The response sequence judging submodule invokes the abnormal node space attribution information list, extracts the abnormal node and the related upstream and downstream liquid level monitoring point numbers, retrieves the liquid level monitoring data record of the corresponding numbers, determines the liquid level change starting time of three monitoring points in the time period corresponding to the abnormal event, compares the response sequence of the upstream and downstream monitoring points through the timestamp field, screens the node group with the downstream response time positioned behind the upstream, and generates a liquid level response sequence verification result; And the tracing judging and updating sub-module identifies the number of a downstream node with response time earlier than that of an upstream monitoring point according to the response time distribution in the liquid level response sequence verification result, retrieves adjacent gradient segment information corresponding to the node, extracts continuous connection nodes of the upstream pipe section through reverse sequence retrieval, judges whether a liquid level fluctuation record consistent with an abnormal event period exists in the real-time upstream node, screens the number of the node meeting the condition, and generates a pipe network abnormal verification node set.

Description

Abnormal detection system for building drainage pipe network Technical Field The invention relates to the technical field of fault detection, in particular to a system for detecting abnormality of a building drainage pipe network. Background The technical field of fault detection relates to the identification, analysis and positioning of abnormal states of various systems, devices or components in the operation process, and core matters comprise the monitoring of operation parameters, data acquisition and comparison analysis, abnormal mode identification, fault source tracing, state evaluation and the like. The system comprises a device and a method for detecting problems of blockage, leakage, water volume abnormality and the like of a drainage system in a building in the operation process, and aims at technical matters of structural or operational abnormality identification of the drainage system of the building, real-time monitoring of the drainage state is completed by arranging a liquid level sensor, a flow sensor, a water pressure sensor and the like to collect internal operation parameters of the drainage system, abnormal conditions existing in the drainage system are identified by means of threshold comparison, trend analysis and the like, and the traditional method adopts subclasses based on sensor signal collection and simple threshold judgment to carry out abnormality detection work. The existing building drainage pipe network anomaly detection relies on a single-point sensor to collect signals and perform threshold comparison, a time sequence and space association is lacking among different monitoring points in a complex pipe network structure, so that anomaly judgment only reflects a local state, continuous feature recognition is difficult to form at the early stage of pipe blockage or leakage, a fixed threshold is easy to be subjected to misjudgment or missed detection caused by environmental fluctuation interference, when synchronous fluctuation occurs in different pipe sections, a flow direction relationship cannot be distinguished, abnormal positioning is fuzzy, fault judgment is delayed, an investigation period is prolonged, maintenance cost is increased, the integral operation risk of the pipe network is increased, and the requirements of fine monitoring and real-time diagnosis are difficult to meet. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides a system for detecting abnormality of a building drainage pipe network. In order to achieve the above object, the present invention adopts the following technical scheme that a system for detecting abnormality of a drainage pipe network of a building comprises: the hydraulic pressure chain construction module acquires real-time hydraulic pressure sequence data of hydraulic pressure monitoring points distributed in a vertical pipe, a transverse branch pipe and an inspection well section in a building drainage pipe network, invokes the hydraulic pressure change direction between the adjacent three monitoring points and forms a continuous chain type relationship according to the pipe section space arrangement sequence to generate a broken chain segment initialization identification list; the sudden drop identification module extracts segment nodes marked as broken according to the broken segment initialization mark list, acquires a time sequence water pressure fluctuation path of adjacent upstream and downstream pipe segments, and generates a space sudden drop verification mark set by judging whether water pressure rebound occurs in a time window upstream of a sudden drop point or not; the flow resistance classification module extracts a continuous water pressure change trend curve according to the space sudden drop verification mark set, and divides the time span and the change slope of the ascending and descending paragraphs to obtain a flow resistance structure type identification table; And the structure mapping module is used for calling node coordinate information in the building structure plane and elevation layout model according to the flow resistance structure type identification table, positioning the belonging building hierarchy structure relation, analyzing the position and connection sequence of the corresponding pipe section between nodes on the vertical path in the building model, and generating an abnormal node space attribution information list. As a further scheme of the invention, the broken chain segment initialization identification list comprises a direction reversal position, a water pressure gradient broken node, a water pressure deviation synchronization point and a water pressure chain segment number, the space sudden drop verification mark set comprises a sudden drop point node position, an upstream and downstream response characteristic, a time sequence verification mark and a chain segment corresponding index, the flow resistance structure type identification list comprises