Pump station engineering fault diagnosis and prediction system and method based on fault tree

Abstract

The invention discloses a pump station engineering fault diagnosis and prediction system and method based on a fault tree, and relates to the field of intelligent water conservancy. The system comprises a fault tree database, a data acquisition module, a data analysis module, a fault diagnosis module and a fault prediction module. The data analysis module transmits the abnormal data according to the following preset rules: when the associated parameter data exceeds a top event preset standard, the abnormal data is transmitted to the fault diagnosis module; when the associated parameter data exceed a sub-event preset standard, abnormal data are transmitted to a fault prediction module; when the trend of the associated parameter data deteriorates but does not exceed any preset standard, the preset standard in the fault tree database is dynamically adjusted, and then processing is carried out according to a preset rule; the fault diagnosis module is used for finding out a fault reason corresponding to the top event; and the fault prediction module is used for predicting whether the top event can occur or not. The invention aims to solve the problems that the fault reason positioning efficiency is low, the misjudgment rate is high, and missed judgment is easy to occur when a static threshold value is used for early warning.

Description

Technical Field

[0001] This invention relates to the field of intelligent water conservancy, and in particular to a system and method for fault diagnosis and prediction of pumping station projects based on fault tree. Background Technology

[0002] "Unmanned operation" and "reduced staffing" have always been the goals of intelligent construction in pumping station engineering. However, due to limitations in existing technologies such as fault handling procedures and equipment safety early warning, the reliance on on-site management personnel remains high. The current fault handling procedure for pumping station engineering typically involves on-site management personnel recording the alarm triggered by a sensor and notifying unit maintenance personnel to intervene and conduct fault screening according to the maintenance manual. This method has at least the following drawbacks:

[0003] First, the existing pump station early warning system relies on sensor threshold alarms to trigger manual intervention, which requires at least several steps, including transmitting the sensor alarm signal to management personnel, manually locating the cause of the fault, and maintenance personnel arriving at the scene. The fault cause is slow to be located, and critical equipment may be severely damaged when the fault is dealt with, increasing maintenance costs.

[0004] Second, the fault handling process relies on the personal experience of on-site personnel and maintenance manuals, resulting in a high rate of misjudgment. When complex faults occur, decision-making efficiency is further reduced, making it impossible to effectively uncover the root cause of the problem.

[0005] Third, static threshold early warning systems suffer from a lag. Alarms are issued after the threshold is exceeded, but by the time on-site management personnel at the pumping station address the issue, irreversible damage has already occurred to the equipment. Furthermore, the lack of effective means to identify gradual trends in data leads to missed diagnoses of potential faults. Current improvements involve lowering the threshold standard, which can achieve early warning and reduce fault losses, but significantly increases the probability of false alarms, impacting the normal scheduling of pumping stations. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a system and method for fault diagnosis and prediction of pump station engineering based on fault tree, which addresses the above-mentioned defects of the prior art. The aim is to solve the problems of low efficiency in fault cause location, high misjudgment rate and easy omission when using static threshold warning in the prior art.

[0007] The technical solution adopted by this invention to solve the problem is as follows:

[0008] In a first aspect, embodiments of the present invention provide a fault tree-based pump station engineering fault diagnosis and prediction system, including a fault tree database, a data acquisition module, a data analysis module, a fault diagnosis module, and a fault prediction module;

[0009] The fault tree database consists of multiple fault trees. Each fault tree has a top event as the root node and connects sub-events at each level through logic gates to form branches. The end of each branch is the cause of the fault. The top event and sub-events respectively contain the event name, associated parameters and their preset standards.

[0010] The data acquisition module is used to collect the associated parameter data of each fault tree top event and sub-event;

[0011] The data analysis module, based on the fault tree database, transmits abnormal data to the fault diagnosis module or the fault prediction module according to the following preset rules: when the associated parameter data exceeds the preset standard of the associated parameter of the top event, the associated parameter data and the corresponding top event are transmitted as abnormal data to the fault diagnosis module; when the associated parameter data exceeds the preset standard of the associated parameter of the sub-event, the associated parameter data and the corresponding sub-event are transmitted as abnormal data to the fault prediction module.

[0012] When the trend of the associated parameter data deteriorates but does not exceed any preset standard, the preset standard of the associated parameter of the top event or sub-event in the fault tree database is dynamically adjusted, and then the situation is handled according to the preset rules.

[0013] The fault diagnosis module is used to locate the specific fault tree based on the fault tree database and abnormal data; based on the fault tree, it judges whether each level of sub-event is valid from top to bottom until the cause of the fault is found.

[0014] The fault prediction module is used to locate a specific fault tree based on the fault tree database and abnormal data; predict whether the top event will occur based on the logic gates of the fault tree; and when the top event is predicted to occur, find the fault cause corresponding to the abnormal data based on the fault tree.

[0015] In one implementation, in the data analysis module, a deteriorating trend in the associated parameter data refers to:

[0016] Current value

[0017] In the formula, The average value of the data within the sliding window; k: adjustment factor; σ: standard deviation of the data within the sliding window;

[0018] The

[0019] In the formula, n: the size of the sliding window; x i : The associated parameter data at the i-th time point;

[0020] The

[0021] In the formula, n: the size of the sliding window; x i : The associated parameter data at the i-th time point; The average value of the data within the sliding window.

[0022] In one implementation, the calculation formula for dynamically adjusting the preset standard of the association parameters of the top event or sub-event in the fault tree database in the data analysis module is as follows:

[0023]

[0024] In the formula, A: the slope of the current associated parameter data; A max : The maximum slope of historical correlation parameter data; x t : Related parameter data for the current time t; The average value of the data within the sliding window; σ: the standard deviation of the data within the sliding window; ΔA: the difference between the current slope and the slope at the previous time step; α, β, γ: weighting coefficients, 0 ≤ α, β, γ ≤ 1, and α + β + γ = 1.

[0025] In one implementation, in the data analysis module, when the trend of the associated parameter data deteriorates and disappears, and remains stable within a preset time, the new preset standard of the top event or sub-event associated parameter in the dynamically adjusted fault tree database is restored to the original preset standard.

[0026] The disappearance of the trend in the coupled parameter data refers to the current value. In the formula, The average value of the data within the sliding window; k: adjustment factor; σ: standard deviation of the data within the sliding window.

[0027] In one implementation, the system further includes an information feedback module; the information feedback module generates a fault diagnosis report or a fault prediction warning based on the information transmitted by the fault diagnosis module and the fault prediction module, and compiles the entire fault diagnosis and fault prediction process into a fault report.

[0028] In one implementation, the system further includes a fault tree update module, which modifies and supplements the fault tree database based on the fault report generated by the information feedback module and in conjunction with the maintenance manual.

[0029] In one implementation, the step of determining whether each level of sub-event is valid from top to bottom according to the fault tree in the fault diagnosis module includes: requesting the data analysis module to provide the correlation parameter data of each level of sub-event in the fault tree; determining whether each level of sub-event is valid from top to bottom according to the correlation parameter data; if a second-level sub-event is valid, determining its next-level sub-event one by one, and transmitting one or more fault causes to the information feedback module; if a second-level sub-event is invalid, skipping the sub-event and its branch sub-events; if all second-level sub-events are invalid, transmitting all fault causes in the fault tree in descending order of frequency of occurrence to the information feedback module.

[0030] In one implementation, in the fault prediction module, if the top event is predicted to occur continuously within a preset time period, the name of the top event and the cause of the fault are transmitted to the information feedback module; otherwise, it is a false alarm and no information is transmitted to the information feedback module.

[0031] Secondly, embodiments of the present invention provide a fault diagnosis and prediction method for pump station engineering based on fault tree, comprising the following steps:

[0032] Step 1: Establish a fault tree database. The fault tree database consists of multiple fault trees. Each fault tree has a top event as the root node and connects sub-events at each level through logic gates to form branches. The end of each branch is the fault cause. The top event and sub-events respectively contain event names, associated parameters and their preset standards.

[0033] Step 2: Collect the correlation parameter data of each fault tree top event and sub-event;

[0034] Step 3: Based on the fault tree database, filter out abnormal data and perform fault diagnosis or fault prediction according to the following preset rules:

[0035] When the associated parameter data exceeds the preset standard of the top event associated parameter, the associated parameter data and the corresponding top event are regarded as abnormal data and fault diagnosis is performed: based on the fault tree database and the abnormal data, the specific fault tree is located; based on the fault tree, the validity of each level of sub-event is judged from top to bottom until the cause of the fault is found.

[0036] When the associated parameter data exceeds the preset standard of the sub-event associated parameter, the associated parameter data and the corresponding sub-event are regarded as abnormal data, and fault prediction is performed: based on the fault tree database and the abnormal data, the specific fault tree is located; based on the logic gates of the fault tree, it is predicted whether the top event will occur; when the top event is predicted to occur, the fault cause corresponding to the abnormal data is found based on the fault tree.

[0037] Step 4: When the trend of the associated parameter data deteriorates but does not exceed any preset standard, dynamically adjust the preset standard of the associated parameter of the top event or sub-event in the fault tree database, and then handle it according to the preset rules.

[0038] In one implementation, specific fault handling measures are associated with the cause of the fault to form a fault diagnosis report or fault prediction and early warning information; the entire process of fault diagnosis and fault prediction is compiled into a fault report, and the fault tree database is modified and supplemented in conjunction with the maintenance manual.

[0039] The beneficial effects of this invention are:

[0040] 1. This invention achieves efficient and accurate fault location by using a fault diagnosis module. Compared with the traditional manual fault diagnosis method, the fault tree database can quickly filter and judge possible fault paths according to preset logical rules, reducing the reliance on manual labor. This allows even relatively inexperienced staff to accurately find the root cause of the fault with the assistance of the system, avoiding subjective errors and omissions that may occur in the manual judgment process, thereby effectively improving the accuracy of fault cause location.

[0041] 2. The fault prediction module of the present invention is based on the logical gate relationship of the fault tree and can reverse the deduction of the risk of the top event. Traditional early warning methods often issue an alarm only when the fault has already occurred or is about to occur. The prediction module of the present invention can discover potential fault risks in advance and locate the fault cause that leads to the top event. Based on these prediction results, maintenance personnel can formulate maintenance plans in advance and carry out preventive maintenance on relevant equipment to avoid the occurrence of faults.

[0042] 3. In actual operation, the operating status of the equipment may be affected by various factors, and the data may show a gradual deterioration trend, but may not exceed the traditional fixed threshold for a period of time. This invention can keenly capture this data deterioration trend and dynamically reduce the warning threshold according to a preset dynamic threshold adjustment formula, thereby achieving accurate warning. At the same time, in order to avoid misjudgment caused by short-term fluctuations, this invention also sets up a threshold recovery mechanism.

[0043] 4. The multi-dimensional anomaly detection mechanism constructed in this invention forms an inseparable technical solution through a triple triggering mechanism, comprehensively solving the problems of delayed response, missed detections, false alarms, and blind maintenance inherent in traditional threshold-based early warning systems: the data analysis module can keenly capture data deterioration trends and detect equipment anomalies in advance; the fault diagnosis module can shorten the time for locating the cause of the fault and improve the accuracy of fault diagnosis; and the fault prediction module can provide early warnings, locate the cause of the fault, and guide preventive maintenance. These three modules work together in a monitoring-analysis-response closed loop, covering real-time anomalies, medium-term risks, and long-term trends, achieving dynamic collaboration among multiple modules and improving the intelligence level of pump stations. Attached Figure Description

[0044] The invention will now be further described with reference to the accompanying drawings.

[0045] Figure 1 This is a flowchart of a fault tree-based pump station engineering fault diagnosis and prediction system according to an embodiment of the present invention.

[0046] Figure 2 This is a schematic diagram of the data acquisition module of the fault diagnosis and prediction system for pumping station engineering based on fault tree according to an embodiment of the present invention.

[0047] Figure 3This is a schematic diagram of the data analysis module of the fault tree-based pump station engineering fault diagnosis and prediction system according to an embodiment of the present invention.

[0048] Figure 4 This is a schematic diagram of the fault diagnosis module of the fault diagnosis and prediction system for pump station engineering based on fault tree according to an embodiment of the present invention.

[0049] Figure 5 This is a schematic diagram of the fault prediction module of the fault tree-based pump station engineering fault diagnosis and prediction system according to an embodiment of the present invention.

[0050] Figure 6 This is a schematic diagram of the information feedback module of the fault tree-based pump station engineering fault diagnosis and prediction system according to an embodiment of the present invention.

[0051] Figure 7 This is a schematic diagram of the fault tree updating module of the fault tree-based pump station engineering fault diagnosis and prediction system according to an embodiment of the present invention.

[0052] Figure 8 This is a schematic diagram of a fault tree-based pump station engineering fault diagnosis and prediction system according to an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] This invention provides a fault tree-based fault diagnosis and prediction system for pump station engineering, such as... Figure 1-8 As shown, it includes a fault tree database, a data acquisition module, a data analysis module, a fault diagnosis module, and a fault prediction module;

[0055] The fault tree database consists of multiple fault trees. Each fault tree has a top event as the root node and connects sub-events at each level through logic gates to form branches. The end of each branch is the cause of the fault. The top event and sub-events respectively contain the event name, associated parameters and their preset standards.

[0056] The data acquisition module is used to collect the associated parameter data of each fault tree top event and sub-event;

[0057] The data analysis module, based on the fault tree database, transmits abnormal data to the fault diagnosis module or the fault prediction module according to the following preset rules: when the associated parameter data exceeds the preset standard of the associated parameter of the top event, the associated parameter data and the corresponding top event are transmitted as abnormal data to the fault diagnosis module; when the associated parameter data exceeds the preset standard of the associated parameter of the sub-event, the associated parameter data and the corresponding sub-event are transmitted as abnormal data to the fault prediction module.

[0058] When the trend of the associated parameter data deteriorates but does not exceed any preset standard, the preset standard of the associated parameter of the top event or sub-event in the fault tree database is dynamically adjusted, and then the situation is handled according to the preset rules.

[0059] The fault diagnosis module is used to locate the specific fault tree based on the fault tree database and abnormal data; based on the fault tree, it judges whether each level of sub-event is valid from top to bottom until the cause of the fault is found.

[0060] The fault prediction module is used to locate a specific fault tree based on the fault tree database and abnormal data; predict whether the top event will occur based on the logic gates of the fault tree; and when the top event is predicted to occur, find the fault cause corresponding to the abnormal data based on the fault tree.

[0061] Specifically, the fault tree database is primarily built based on the maintenance manual and can be supplemented based on management experience. Each fault tree uses a fault event as the top event, the main direct factors leading to the top event as second-level sub-events, and the main direct factors leading to N-level sub-events as N+1-level sub-events. Each branch ends with the fault cause as its termination point. The top event, sub-events, and fault causes are connected by logic gates. The information in each fault tree includes: the top event name, the corresponding operating condition parameters or monitoring data information, the preset standard for the corresponding operating condition parameters or monitoring data, the names of each sub-event, the corresponding operating condition parameters or monitoring data information, the preset standard for the corresponding operating condition parameters or monitoring data, and the relationships between the events in the fault tree. The preset standard for each operating condition parameter or monitoring data is different.

[0062] Specifically, the data acquisition module has the ability to process various API interfaces and can convert all operating parameters and monitoring data into a unified editable format for the data analysis module to call. The types of data acquired include: various sensor data, monitoring system data, and management system data. Sensor data includes: temperature sensor data, vibration sensor data, sway sensor data, sound sensor data, odor sensor data, and other sensor data.

[0063] Specifically, when the data analysis module determines that the associated parameter data exceeds the preset standard of the top event's associated parameter, in addition to transmitting the associated parameter data and the corresponding top event as abnormal data to the fault diagnosis module, it also transmits the associated operating parameters and monitoring data of the remaining sub-events in the fault tree associated with the top event to the fault diagnosis module, facilitating further analysis by the fault diagnosis module. The data list, preset standards, and data-related fault tree event relationships processed by the data analysis module all originate from the fault tree.

[0064] In one implementation, within the data analysis module, a deteriorating trend in the associated parameter data refers to:

[0065] Current value

[0066] In the formula, The average value of the data within the sliding window; k: adjustment factor; σ: standard deviation of the data within the sliding window;

[0067] The

[0068] In the formula, n: the size of the sliding window; x i : The associated parameter data at the i-th time point;

[0069] The

[0070] In the formula, n: the size of the sliding window; x i : The associated parameter data at the i-th time point; The average value of the data within the sliding window.

[0071] In one implementation, the calculation formula for dynamically adjusting the preset standard of the association parameters of the top event or sub-events in the fault tree database in the data analysis module is as follows:

[0072]

[0073] In the formula, A: the slope of the current associated parameter data; A max : The maximum slope of historical correlation parameter data; x t : Related parameter data for the current time t; The average value of the data within the sliding window; σ: the standard deviation of the data within the sliding window; ΔA: the difference between the current slope and the slope at the previous time step; α, β, γ: weighting coefficients, 0 ≤ α, β, γ ≤ 1, and α + β + γ = 1.

[0074] Specifically, The slope of the current data is taken into account relative to the historical maximum slope. The larger the slope, the greater the adjustment of the threshold should be. The degree of deviation of the current data from the mean is taken into account. The greater the deviation, the higher the degree of data abnormality, and the greater the adjustment of the threshold. γ×ΔA: The trend of slope change is taken into account. If the slope is increasing at an accelerating rate, it means that the data is deteriorating at an accelerated rate, and the threshold needs to be further reduced.

[0075] In one implementation, in the data analysis module, when the trend of the associated parameter data deteriorates and disappears, and remains stable within a preset time, the new preset standard of the top event or sub-event associated parameter in the dynamically adjusted fault tree database is restored to the original preset standard.

[0076] The disappearance of the trend in the coupled parameter data refers to the current value. In the formula, The average value of the data within the sliding window; k: adjustment factor; σ: standard deviation of the data within the sliding window.

[0077] In one implementation, the system also includes an information feedback module; the information feedback module generates a fault diagnosis report or a fault prediction early warning information based on the information transmitted by the fault diagnosis module and the fault prediction module, and compiles the entire fault diagnosis and fault prediction process into a fault report.

[0078] Specifically, after receiving data from the fault diagnosis module, the information feedback module associates the fault cause with the corresponding handling measures, generates a fault diagnosis report, and outputs the results. After receiving data from the fault prediction module, the information feedback module processes the fault prediction results into early warning information, associates the fault cause with the corresponding handling measures, generates early warning measures, and outputs the results. After the results are output, the information feedback module records the entire fault diagnosis and fault prediction process, generating a fault report.

[0079] Specifically, fault handling measures are stored in the information feedback module in the form of a database, and each measures correspond one-to-one with the cause of the fault. The fault diagnosis report includes: the time of occurrence of abnormal monitoring data, the name of the abnormal measuring point, the cause of the fault, the fault tree details, and the fault handling measures. Fault prediction and early warning information is determined by the fault prediction results. If the fault prediction result indicates that a fault is likely to occur, fault prediction and early warning information is generated. The fault prediction and early warning information includes the name of the early warning event, the time of occurrence of abnormal monitoring data, the name of the abnormal measuring point, the data value of the abnormal measuring point, and the fault handling measures.

[0080] In one implementation, the system also includes a fault tree update module, which modifies and supplements the fault tree database based on the fault report generated by the information feedback module and the maintenance manual.

[0081] Specifically, the preset standards for operating parameters or monitoring data correspond one-to-one with the top event or sub-event, and will be updated along with the information of the top event and sub-events in the fault tree.

[0082] In one implementation, the step of determining whether each level of sub-event is valid from top to bottom according to the fault tree in the fault diagnosis module includes: requesting the data analysis module to provide the correlation parameter data of each level of sub-event in the fault tree; determining whether each level of sub-event is valid from top to bottom according to the correlation parameter data; if a second-level sub-event is valid, determining its next-level sub-event one by one, and transmitting one or more fault causes to the information feedback module; if a second-level sub-event is invalid, skipping the sub-event and its branch sub-events; if all second-level sub-events are invalid, transmitting all fault causes in the fault tree in descending order of frequency of occurrence to the information feedback module.

[0083] In one implementation, if the fault prediction module continuously predicts that a top event will occur within a preset time period, the name of the top event and the cause of the fault will be transmitted to the information feedback module; otherwise, it is considered a false alarm and no information will be transmitted to the information feedback module.

[0084] This invention also provides a fault tree-based method for fault diagnosis and prediction in pump station engineering, comprising the following steps:

[0085] Step 1: Establish a fault tree database. The fault tree database consists of multiple fault trees. Each fault tree has a top event as the root node and connects sub-events at each level through logic gates to form branches. The end of each branch is the fault cause. The top event and sub-events respectively contain event names, associated parameters and their preset standards.

[0086] Step 2: Collect the correlation parameter data of each fault tree top event and sub-event;

[0087] Step 3: Based on the fault tree database, filter out abnormal data and perform fault diagnosis or fault prediction according to the following preset rules:

[0088] When the associated parameter data exceeds the preset standard of the top event associated parameter, the associated parameter data and the corresponding top event are regarded as abnormal data and fault diagnosis is performed: based on the fault tree database and the abnormal data, the specific fault tree is located; based on the fault tree, the validity of each level of sub-event is judged from top to bottom until the cause of the fault is found.

[0089] When the associated parameter data exceeds the preset standard of the sub-event associated parameter, the associated parameter data and the corresponding sub-event are regarded as abnormal data, and fault prediction is performed: based on the fault tree database and the abnormal data, the specific fault tree is located; based on the logic gates of the fault tree, it is predicted whether the top event will occur; when the top event is predicted to occur, the fault cause corresponding to the abnormal data is found based on the fault tree.

[0090] Step 4: When the trend of the associated parameter data deteriorates but does not exceed any preset standard, dynamically adjust the preset standard of the associated parameter of the top event or sub-event in the fault tree database, and then handle it according to the preset rules.

[0091] In one implementation, specific fault handling measures are associated with the fault cause, and a fault diagnosis report or fault prediction and early warning information is generated; based on the fault diagnosis report, the fault tree database is modified and supplemented in conjunction with the maintenance manual and maintenance report.

[0092] The advantages of this invention are:

[0093] 1. This invention achieves efficient and accurate fault location by using a fault diagnosis module. Compared with the traditional manual fault diagnosis method, the fault tree database can quickly filter and judge possible fault paths according to preset logical rules, reducing the reliance on manual labor. This allows even relatively inexperienced staff to accurately find the root cause of the fault with the assistance of the system, avoiding subjective errors and omissions that may occur in the manual judgment process, thereby effectively improving the accuracy of fault cause location.

[0094] 2. The fault prediction module of the present invention is based on the logical gate relationship of the fault tree and can reverse the deduction of the risk of the top event. Traditional early warning methods often issue an alarm only when the fault has already occurred or is about to occur. The prediction module of the present invention can discover potential fault risks in advance and locate the fault cause that leads to the top event. Based on these prediction results, maintenance personnel can formulate maintenance plans in advance and carry out preventive maintenance on relevant equipment to avoid the occurrence of faults.

[0095] 3. In actual operation, the operating status of the equipment may be affected by various factors, and the data may show a gradual deterioration trend, but may not exceed the traditional fixed threshold for a period of time. This invention can keenly capture this data deterioration trend and dynamically reduce the warning threshold according to a preset dynamic threshold adjustment formula, thereby achieving accurate warning. At the same time, in order to avoid misjudgment caused by short-term fluctuations, this invention also sets up a threshold recovery mechanism.

[0096] 4. The multi-dimensional anomaly detection mechanism constructed in this invention forms an inseparable technical solution through a triple triggering mechanism, comprehensively solving the problems of delayed response, missed detections, false alarms, and blind maintenance inherent in traditional threshold-based early warning systems: the data analysis module can keenly capture data deterioration trends and detect equipment anomalies in advance; the fault diagnosis module can shorten the time for locating the cause of the fault and improve the accuracy of fault diagnosis; and the fault prediction module can provide early warnings, locate the cause of the fault, and guide preventive maintenance. These three modules work together in a monitoring-analysis-response closed loop, covering real-time anomalies, medium-term risks, and long-term trends, achieving dynamic collaboration among multiple modules and improving the intelligence level of pump stations.

[0097] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fault tree-based fault diagnosis and prediction system for pump station engineering, characterized in that, It includes a fault tree database, a data acquisition module, a data analysis module, a fault diagnosis module, and a fault prediction module; The fault tree database consists of multiple fault trees. Each fault tree is rooted at the top event and connected to sub-events at each level through logic gates to form branches. The end of each branch is the cause of the fault. Top events and sub-events each contain an event name, associated parameters, and their preset standards; The data acquisition module is used to collect the associated parameter data of each fault tree top event and sub-event; The data analysis module, based on the fault tree database, transmits abnormal data to the fault diagnosis module or the fault prediction module according to the following preset rules: when the associated parameter data exceeds the preset standard of the associated parameter of the top event, the associated parameter data and the corresponding top event are transmitted as abnormal data to the fault diagnosis module; when the associated parameter data exceeds the preset standard of the associated parameter of the sub-event, the associated parameter data and the corresponding sub-event are transmitted as abnormal data to the fault prediction module. When the trend of the associated parameter data deteriorates but does not exceed any preset standard, the preset standard of the associated parameter of the top event or sub-event in the fault tree database is dynamically adjusted, and then the situation is handled according to the preset rules. The fault diagnosis module is used to locate the specific fault tree based on the fault tree database and abnormal data; based on the fault tree, it judges whether each level of sub-event is valid from top to bottom until the cause of the fault is found. The fault prediction module is used to locate a specific fault tree based on the fault tree database and abnormal data; predict whether the top event will occur based on the logic gates of the fault tree; and when the top event is predicted to occur, find the fault cause corresponding to the abnormal data based on the fault tree.

2. The fault tree-based pump station engineering fault diagnosis and prediction system as described in claim 1, characterized in that: In the data analysis module, a deteriorating trend in the associated parameter data indicates: Current value In the formula, The average value of the data within the sliding window; k: adjustment factor; σ: Standard deviation of the data within the sliding window; The In the formula, n: the size of the sliding window; x i : The associated parameter data at the i-th time point; The In the formula, n: the size of the sliding window; x i : The associated parameter data at the i-th time point; The average value of the data within the sliding window.

3. The fault tree-based pump station engineering fault diagnosis and prediction system as described in claim 2, characterized in that: In the data analysis module, the calculation formula for dynamically adjusting the preset standard of the association parameters of the top event or sub-events in the fault tree database is as follows: In the formula, A: the slope of the current associated parameter data; A max : The maximum slope of historical correlation parameter data; x t : Related parameter data for the current time t; The average value of the data within the sliding window; σ: the standard deviation of the data within the sliding window; ΔA: the difference between the current slope and the slope at the previous time step; α, β, γ: weighting coefficients, 0 ≤ α, β, γ ≤ 1, and α + β + γ = 1.

4. The fault tree-based pump station engineering fault diagnosis and prediction system as described in claim 3, characterized in that: In the data analysis module, when the trend of the associated parameter data deteriorates and disappears, and remains stable within a preset time, the new preset standard of the top event or sub-event associated parameter in the dynamically adjusted fault tree database will be restored to the original preset standard. The disappearance of the trend in the coupled parameter data refers to the current value. In the formula, The average value of the data within the sliding window; k: adjustment factor; σ: Standard deviation of the data within the sliding window.

5. The fault tree-based pump station engineering fault diagnosis and prediction system as described in claim 1, characterized in that: The system also includes an information feedback module; the information feedback module generates a fault diagnosis report or a fault prediction early warning information based on the information transmitted by the fault diagnosis module and the fault prediction module, and compiles the entire fault diagnosis and fault prediction process into a fault report.

6. The fault tree-based pump station engineering fault diagnosis and prediction system as described in claim 5, characterized in that: The system also includes a fault tree update module, which modifies and supplements the fault tree database based on the fault reports generated by the information feedback module and the maintenance manual.

7. The fault tree-based pump station engineering fault diagnosis and prediction system as described in claim 5, characterized in that: In the fault diagnosis module, the step of determining whether each level of sub-event is valid from top to bottom according to the fault tree includes: requesting the data analysis module to provide the correlation parameter data of each level of sub-event in the fault tree; determining whether each level of sub-event is valid from top to bottom according to the correlation parameter data; if a second-level sub-event is valid, determining its next-level sub-event one by one, and transmitting one or more fault causes to the information feedback module; if a second-level sub-event is invalid, skipping the sub-event and its branch sub-events; if all second-level sub-events are invalid, transmitting all fault causes in the fault tree in descending order of frequency of occurrence to the information feedback module.

8. The fault tree-based pump station engineering fault diagnosis and prediction system as described in claim 5, characterized in that: In the fault prediction module, if the top event is predicted to occur continuously within a preset time period, the name of the top event and the cause of the fault are transmitted to the information feedback module; otherwise, it is a false alarm and no information is transmitted to the information feedback module.

9. A fault tree-based method for fault diagnosis and prediction in pump station engineering, characterized in that, Includes the following steps: Step 1: Establish a fault tree database. The fault tree database consists of multiple fault trees. Each fault tree has the top event as the root node and connects the sub-events at each level through logic gates to form branches. The end of each branch is the cause of the fault. Top events and sub-events each contain an event name, associated parameters, and their preset standards; Step 2: Collect the correlation parameter data of each fault tree top event and sub-event; Step 3: Based on the fault tree database, filter out abnormal data and perform fault diagnosis or fault prediction according to the following preset rules: When the associated parameter data exceeds the preset standard of the top event associated parameter, the associated parameter data and the corresponding top event are regarded as abnormal data and fault diagnosis is performed: based on the fault tree database and the abnormal data, the specific fault tree is located; based on the fault tree, the validity of each level of sub-event is judged from top to bottom until the cause of the fault is found. When the associated parameter data exceeds the preset standard of the sub-event associated parameter, the associated parameter data and the corresponding sub-event are regarded as abnormal data, and fault prediction is performed: based on the fault tree database and the abnormal data, the specific fault tree is located; based on the logic gates of the fault tree, it is predicted whether the top event will occur; when the top event is predicted to occur, the fault cause corresponding to the abnormal data is found based on the fault tree. Step 4: When the trend of the associated parameter data deteriorates but does not exceed any preset standard, dynamically adjust the preset standard of the associated parameter of the top event or sub-event in the fault tree database, and then handle it according to the preset rules.

10. The fault tree-based pump station engineering fault diagnosis and prediction method as described in claim 9, characterized in that: Based on the causes of the faults, specific fault handling measures are associated to generate a fault diagnosis report or fault prediction and early warning information; the entire process of fault diagnosis and fault prediction is compiled into a fault report, and the fault tree database is modified and supplemented in conjunction with the maintenance manual.

Images