A hydraulic fault diagnosis expert system with priority

Abstract

The application belongs to the technical field of hydraulic fault detection, and specifically relates to a hydraulic fault diagnosis expert system with priority, which comprises a man-machine interactive interface module, a component failure rate database module, a dynamic fault tree algorithm module, a fault probability sorting module and a fault maintenance strategy module. The man-machine interactive interface module comprises an editing module and a fault diagnosis module. The editing module edits the component failure rate database module, and the fault diagnosis module selects a top event required for diagnosis to obtain a corresponding maintenance strategy of the fault maintenance strategy module. The component failure rate database module provides bottom event failure probability data for the dynamic fault tree algorithm module. The dynamic fault tree algorithm module calculates and processes the bottom event failure probability data to obtain a fault top event probability required by the fault probability sorting module. The fault probability sorting module sorts the fault sub-event probability in size. The fault maintenance strategy module proposes a maintenance strategy with priority.

Description

Technical Field

[0001] This invention belongs to the field of hydraulic fault detection technology, specifically a priority-based hydraulic fault diagnosis expert system. Background Technology

[0002] Hydraulic systems, as the power source in fields such as construction machinery, are the core of the entire equipment. A major malfunction in the hydraulic system will lead to equipment downtime, disrupting production activities and potentially causing significant economic losses. Due to the complexity of hydraulic systems and the limited professional skills of on-site personnel, many hydraulic system malfunctions cannot be resolved promptly, posing challenges to maintenance and necessitating on-site service from design personnel.

[0003] Current fault diagnosis expert systems mostly focus on areas such as engines, transformers, and diesel engines, with relatively little research on hydraulic systems. Furthermore, the analysis of potential faults after fault diagnosis lacks prioritization, preventing staff from promptly investigating high-probability faults and impacting maintenance efficiency.

[0004] Hydraulic fault diagnosis expert systems can use databases and algorithms to calculate complex fault diagnosis problems, use expert thinking to find the cause and source of the fault, and propose maintenance strategies. This is of great significance in ensuring the stable operation of equipment hydraulic systems, timely maintenance, and safe production. Summary of the Invention

[0005] In order to solve the complex problem of fault diagnosis, this invention provides a priority hydraulic fault diagnosis expert system by using expert thinking to find the cause and source of the fault and propose maintenance strategies.

[0006] This invention adopts the following technical solution: a priority-based hydraulic fault diagnosis expert system, comprising a human-machine interface module, a component failure rate database module, a dynamic fault tree algorithm module, a fault probability ranking module, and a fault repair strategy module. The human-machine interface module includes an editing module and a fault diagnosis module. The editing module edits the component failure rate database module, and the fault diagnosis module selects the top event to be diagnosed, obtaining a priority-based repair strategy corresponding to the fault repair strategy module. The component failure rate database module includes statistically analyzed failure probabilities of various bottom events, providing bottom event failure probability data for the dynamic fault tree algorithm module. The dynamic fault tree algorithm module calculates and processes the bottom event failure probability data to deduce the probability of the top fault event required by the fault probability ranking module. The fault probability ranking module sorts the probabilities of the fault sub-events by size. The fault repair strategy module proposes priority-based repair strategies based on the fault sub-event probabilities, which are displayed through the human-machine interface module.

[0007] The failure probability of the basic event at the beginning of the component failure rate database module comes from query data. As the system running time increases, the failure probability of the basic event under actual working conditions is obtained by statistically analyzing the replacement rate of each component. Then, the database is modified through the editing module.

[0008] The dynamic fault tree algorithm module constructs a fault tree that includes the probability of a top fault event, i.e., the probability of a certain fault phenomenon in the hydraulic system; the probability of a sub-fault event, i.e., the probability of a possible fault subsystem corresponding to the fault phenomenon; and the failure rate of a possible faulty component corresponding to the fault subsystem.

[0009] The calculation process of the dynamic fault tree algorithm module is as follows:

[0010] A fault top event has multiple fault sub-events F;

[0011] The sub-events of fault sub-event F are: ,

[0012] The severity of the fault is described as follows:

[0013] , , , ..., ;

[0014] The fault severity of fault sub-event F is described as follows: The severity of the fault is between 0 and 1. , m This represents the total number of all rules.

[0015] When the fault level of x1 is The fault level of x2 is , ..., The degree of failure is Then the fault severity of fault sub-event F is: The failure probability is The fault severity of fault sub-event F is: The failure probability is The fault severity of fault sub-event F is... Failure probability ;in, , , ;

[0016] If the fuzzy probabilities of various faults in fault sub-event F are The probability of the rule being executed is: ;

[0017] The probabilities of different fault degrees in the top fault event are:

[0018]

[0019] This module uses the probability of a complete failure of the top event as the basis for calculation comparison, that is, to obtain the failure probability of the fault sub-event F when F=1.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1. A priority-based hydraulic fault diagnosis expert system that can guide on-site personnel to troubleshoot and repair faults. The repair strategies are prioritized, with higher priority strategies having a higher probability of failure and lower priority strategies having a lower probability of failure, thus saving repair time and improving repair efficiency.

[0022] 2. By combining the hydraulic fault diagnosis expert system with the TS dynamic fault tree algorithm, the algorithm's basic event fault probabilities are modified by comparing actual fault conditions, thereby improving the probability of subsequent fault diagnosis. This enables fault diagnosis to adapt to actual working conditions, and the longer the maintenance and usage time, the better the fault diagnosis effect. Attached Figure Description

[0023] Figure 1 This is a flowchart of the hydraulic fault diagnosis expert system with priority according to the present invention;

[0024] Figure 2 A schematic diagram of a human-computer interaction interface module;

[0025] Figure 3 This is a schematic diagram of the component failure rate database module;

[0026] Figure 4 A flowchart showing the maintenance strategy sequence for a sliding cylinder malfunction displayed on a human-machine interface;

[0027] Figure 5 A flowchart outlining the maintenance strategy sequence for a sliding cylinder that is not operating. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are some embodiments of the present invention, but 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.

[0029] like Figure 1As shown, a priority-based hydraulic fault diagnosis expert system includes a human-machine interface module, a component failure rate database module, a dynamic fault tree algorithm module, a fault probability ranking module, and a fault repair strategy module.

[0030] The human-computer interaction interface module includes an editing module and a fault diagnosis module. The editing module edits the component failure rate database module, and the fault diagnosis module selects the top event to be diagnosed to obtain the priority maintenance strategy corresponding to the fault maintenance strategy module.

[0031] like Figure 3 As shown, the component failure rate database module includes the statistical failure probability of each basic event, providing basic event failure probability data for the dynamic fault tree algorithm module.

[0032] For example, the top event of a hydraulic cylinder not moving can be categorized into several fault sub-events: multi-way valve failure, hydraulic cylinder failure, hydraulic pump failure, actuator failure, and pipeline failure. Among these, the bottom events corresponding to the multi-way valve failure sub-event include secondary pressure regulating valve failure, shuttle valve failure, and valve core failure. The component failure rate database module statistically analyzes the failure probability of each bottom event, which is used as input for algorithm calculations. Initially, the bottom event failure rates in the component failure rate database module are derived from queried data, and these rates may vary depending on the operating conditions. As the system's operating time increases, the actual operating condition bottom event failure rates are obtained by statistically analyzing the replacement rates of each component. The database is then modified through the editing module to increase the accuracy of fault diagnosis calculations.

[0033] The dynamic fault tree algorithm module calculates and processes the failure probability data of the bottom events to deduce the failure probability of the top event required by the fault probability ranking module.

[0034] The dynamic fault tree algorithm module applies fault tree theory to calculate and process component failure rate data. Dynamic fault trees utilize fuzzy mathematics theory, describing the failure state of components through fuzzy fault numbers: for example, 1 represents complete failure of a hydraulic component, 0.5 represents partial failure, and 0 represents no failure. The membership function for the fuzzy numbers is a trapezoidal function.

[0035]

[0036] In the formula, μ represents the membership degree; F represents the fuzzy number; m represents the center of the support set; a represents the support radius; and b represents the fuzzy region.

[0037] like Figure 4 As shown, construct the fault tree:

[0038] Where F represents the probability of the top fault event, that is, the probability of a certain fault phenomenon in the hydraulic system. F1, F2, F3, etc. represent the probability of the sub-fault events, that is, the probability of the possible fault subsystems corresponding to the fault phenomena. X1~X10, etc. represent the failure rates of the bottom fault events, that is, the failure rates of the possible faulty components corresponding to the faulty subsystems.

[0039] By using the dynamic fault tree algorithm and the membership function of fuzzy numbers, the failure probability of the top fault event is calculated from the failure probability of the bottom fault event.

[0040] The calculation process of the dynamic fault tree algorithm module is as follows:

[0041] A fault top event has multiple fault sub-events F;

[0042] The sub-events of fault sub-event F are: ,

[0043] The severity of the fault is described as follows:

[0044] ,

[0045] , ..., ;

[0046] The fault severity of fault sub-event F is described as follows: The severity of the fault is between 0 and 1. , m This represents the total number of all rules.

[0047] When the fault level of x1 is The fault level of x2 is , ..., The degree of failure is Then the fault severity of fault sub-event F is: The failure probability is The fault severity of fault sub-event F is: The failure probability is The fault severity of fault sub-event F is... Failure probability ;in, , , ;

[0048] If the fuzzy probabilities of various faults in fault sub-event F are The probability of the rule being executed is: .

[0049] The probabilities of different fault degrees in the top fault event are:

[0050]

[0051] This module uses the probability of a complete failure of the top event as the basis for calculation and comparison, that is, to obtain the failure probability of the failure sub-event F when F=1. By comparison, the probability of each failure possibility can be compared.

[0052] The fault probability sorting module sorts the fault sub-event probabilities by size.

[0053] The dynamic fault tree algorithm calculates the probability of a complete subsystem failure, a partial failure, and a zero failure. The fault probability sorting module sorts the complete failure probability calculation results from the dynamic fault tree algorithm module, obtaining the probability of a complete subsystem failure from highest to lowest.

[0054] The fault repair strategy module proposes priority repair strategies based on the probability of fault events, which are displayed through the human-machine interface module to guide on-site personnel in troubleshooting and repair.

[0055] The fault repair strategy module contains all the troubleshooting solutions for all common faults in the hydraulic system, and can be modified according to the actual situation of the entire hydraulic system.

[0056] This section uses a hydraulic system as an example for illustration. If a hydraulic system is found to be malfunctioning during operation, the sliding cylinder can be diagnosed using the "Priority-Based Hydraulic Fault Diagnosis Expert System" of this invention.

[0057] Select the top fault event through the human-machine interface module. The human-machine interface module includes fault types related to this hydraulic system, such as the sliding cylinder not moving, the travel motor not moving, the leveling cylinder not moving, etc. Here, select the fault type of the sliding cylinder not moving.

[0058] The failure of the sliding cylinder to move can be categorized into several fault events: multi-way valve failure, cylinder failure, pump failure, actuator failure, and pipeline failure. The basic events for multi-way valve failure include secondary pressure regulating valve failure, shuttle valve failure, and valve core failure; the basic events for cylinder failure include: sealing failure and gas presence in the cylinder; the basic events for pump failure include: pump reversal, gas presence in the pump, and pump wear; the basic event for actuator failure is: actuator jamming; and the basic event for pipeline failure is: gas presence in the pipeline. The initial probability data for each of these basic events has been established in a database during the design of the fault diagnosis expert system.

[0059] Based on the internal dynamic fault tree algorithm, the probability of actuator failure is calculated as follows: > pipeline failure > multi-way valve failure > hydraulic cylinder failure > hydraulic pump failure. At this point, the human-machine interface displays a sequence diagram of maintenance strategies corresponding to the sliding hydraulic cylinder not moving, as shown in Figure 5.

[0060] On-site staff followed the troubleshooting strategy sequence diagram displayed on the human-computer interaction interface and proceeded to troubleshoot in the order of maintenance strategy 1, maintenance strategy 2, maintenance strategy 3, maintenance strategy 4, and maintenance strategy 5.

[0061] As equipment maintenance time increases, the probability of fault events corresponding to the sliding cylinder not moving in actual conditions can be statistically obtained. If the probability of multi-way valve failure is > pipeline failure > actuator failure > cylinder failure > pump failure, the probability of the bottom event failure can be modified again through editing module 1.1 to adapt to the fault diagnosis calculation under actual working conditions and improve the accuracy of subsequent fault diagnosis.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A priority-based hydraulic fault diagnosis expert system, characterized in that: It includes a human-computer interaction interface module, a component failure rate database module, a dynamic fault tree algorithm module, a fault probability ranking module, and a fault repair strategy module. The human-computer interaction interface module includes an editing module and a fault diagnosis module. The editing module edits the component failure rate database module, and the fault diagnosis module selects the top event to be diagnosed to obtain the priority maintenance strategy corresponding to the fault maintenance strategy module. The component failure rate database module includes the statistical failure probability of each basic event, providing basic event failure probability data for the dynamic fault tree algorithm module; The dynamic fault tree algorithm module calculates and processes the bottom event failure probability data to deduce the top event failure probability required by the fault probability ranking module. The fault probability sorting module sorts the fault sub-event probabilities by size. The fault repair strategy module proposes priority repair strategies based on the probability of fault events, which are displayed through the human-machine interface module. In the dynamic fault tree algorithm module, the constructed fault tree includes the probability of a top fault event, which is the probability of a certain fault phenomenon in the hydraulic system; a probability of a top fault event corresponds to multiple probabilities of sub-fault events, which are the probabilities of possible faulty subsystems corresponding to the fault phenomenon; a probability of a sub-fault event corresponds to multiple bottom fault events, which are the failure rates of possible faulty components corresponding to the faulty subsystem. The calculation process of the dynamic fault tree algorithm module is as follows: A fault top event has multiple fault sub-events F; The sub-events of fault sub-event F are: , The severity of the fault is described as follows: ， ， ，…, ； The fault severity of fault sub-event F is described as follows: The severity of the fault is between 0 and 1. , m This represents the total number of all rules. When the fault level of x1 is The fault level of x2 is , ..., The degree of failure is Then the fault severity of fault sub-event F is: The failure probability is The fault severity of fault sub-event F is: The probability of failure is The fault severity of fault sub-event F is... Failure probability ;in, , , ; If the fuzzy probabilities of various faults in fault sub-event F are The probability of the rule being executed is: ; The probabilities of different fault degrees in the top fault event are: This module uses the probability of a complete failure of the top event as the basis for calculation comparison, that is, to obtain the failure probability of the fault sub-event F when F=1.

2. The priority-based hydraulic fault diagnosis expert system according to claim 1, characterized in that: The failure probability of the basic event at the beginning of the component failure rate database module comes from query data. As the system running time increases, the failure probability of the basic event under actual working conditions is obtained by statistically analyzing the replacement rate of each component. Then, the database is modified through the editing module.

Images