Expert system for diagnosing faults of textile slurry production equipment

Abstract

The invention discloses an expert system for diagnosing faults of textile slurry production equipment, and the system comprises a data collection module which collects the operation data of the textile slurry production equipment through a sensor network; the signal input end of the data transmission module is connected with the signal output end of the data acquisition module, and the data transmission module can convert the electric signals acquired by the data acquisition module into digital signals which can be processed by a computer; the signal input end of the expert system module is connected with the signal output end of the data transmission module, and the expert system module can compare the digital signals transmitted by the data transmission module to complete fault diagnosis; the signal input end of the user interface module is connected with the signal output end of the expert system module; the whole system can be directly designed for industry application requirements at the same time, and can be widely applied to fault diagnosis of textile slurry production equipment.

Description

Technical Field

[0001] The invention relates to the technical field of fault diagnosis systems, and in particular to an expert system for diagnosing faults of textile pulp production equipment. Background Art

[0002] During the operation of textile pulp production equipment, various faults and abnormal conditions may occur. Traditional maintenance methods mainly rely on the experience and judgment of maintenance personnel. This method has the limitation of relying on personal experience and the problem of low accuracy of diagnostic results. In view of this, developing an expert system that can automatically diagnose equipment faults, using existing expert knowledge to establish a rule base, matching the detected abnormal conditions with the knowledge in the rule base, inferring the cause of the fault, and guiding subsequent maintenance is an important means to improve the quality and efficiency of equipment maintenance.

[0003] At present, the development of Internet technology has provided a technical foundation for building a network-based expert system. Various sensors are used to monitor the equipment operation status data in real time, and the data is transmitted to the back-end server through the network. An expert system is built on the server side, and the equipment data is analyzed using the expert knowledge base to achieve remote online intelligent equipment status monitoring and fault diagnosis. This intelligent equipment fault diagnosis solution based on the combination of network and expert system is gaining more and more applications.

[0004] However, for industries such as textile pulp production equipment, due to its technical threshold and industry specificity, there is a lack of relevant intelligent equipment diagnostic solutions and commercial systems. Therefore, the development of an expert system that can realize status monitoring, anomaly detection and intelligent fault reasoning diagnosis of the entire textile pulp production process is of great significance for improving the reliability and service life of equipment in this industry and reducing maintenance costs. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide an expert system for diagnosing faults in textile pulp production equipment in order to better and effectively solve the above problems.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: an expert system for diagnosing faults of textile pulp production equipment, comprising:

[0007] Data acquisition module, which collects the operation data of textile pulp production equipment through a sensor network;

[0008] The data transmission module has a signal input end connected to a signal output end of the data acquisition module and can convert the electrical signal collected by the data acquisition module into a digital signal that can be processed by a computer;

[0009] The expert system module, whose signal input terminal is connected to the signal output terminal of the data transmission module, can compare the digital signal transmitted by the data transmission module to complete fault diagnosis;

[0010] The signal input terminal of the user interface module is connected with the signal output terminal of the expert system module.

[0011] A method for diagnosing a fault of a textile pulp production equipment, the method is implemented by using the expert system for diagnosing a fault of a textile pulp production equipment, and the specific steps of the method include:

[0012] Step A, knowledge base construction: relevant books on textile pulp production equipment, knowledge of maintenance experts in the field of textile pulp production equipment and results obtained from analyzing previous maintenance cases are summarized to generate a knowledge base;

[0013] Step B, establishment of fault tree: determine which event combinations will cause faults based on the knowledge base, determine the minimum cut set, calculate the probability of each top event, analyze and obtain the prior probability, posterior probability and probability importance information, and establish a fault tree based on the prior probability, posterior probability and probability importance information;

[0014] Step C, fault judgment: Take the digital signal of the data transmission module as the characteristic information, match the information in the knowledge base, locate the fault tree branches and leaf nodes corresponding to the characteristic information, and deduce the system fault type that may be caused by the characteristic information based on the logical relationship reflected by the fault tree.

[0015] Preferably, the knowledge base in step A adopts forward chain reasoning, starting from the data stored in the knowledge base, and gradually inferring conclusions according to the rules. The conclusions obtained will transmit the diagnostic results to the user interface module via the interpreter to display the final results to the user.

[0016] Preferably, the data transmission module converts the electrical signal into a digital signal that can be processed by a computer through an analog-to-digital converter, and after the conversion into the digital signal, the digital signal needs to be preprocessed.

[0017] Preferably, if the fault judgment in step C fails to deduce the system fault type that may be caused by the characteristic information, the fault judgment includes the following steps:

[0018] Step C1, prompt: If the fault judgment fails to deduce the type of system fault that may be caused by the characteristic information, it will be regarded as an unknown fault, and the user interface module will display an unknown fault to prompt relevant staff;

[0019] Step C2, record and update: unknown faults will be recorded in the log, and the log data will be analyzed to obtain new fault modes and symptoms. According to the new fault modes and symptoms, the corresponding branch points will be found in the fault tree, and the new fault results will be added as new leaf nodes of the fault tree.

[0020] Preferably, the digital signal preprocessing specifically includes denoising, filtering, signal calibration and feature extraction.

[0021] Preferably, the preprocessed digital signal uses a LoRa gateway to transmit data to the expert system module.

[0022] The beneficial effects of the present invention are:

[0023] 1. It improves the operational reliability of textile pulp production equipment, reduces the occurrence of unexpected failures, and effectively prevents economic losses and social impacts caused by sudden equipment shutdown.

[0024] 2. It realizes intelligent monitoring of equipment operation status and fault prediction, greatly reducing the dependence on professional maintenance personnel, making equipment maintenance simple and efficient.

[0025] 3. Through continuous learning of unknown faults and optimization and updating of the knowledge base, the accuracy of diagnosis and maintenance is continuously improved, making the system adapt to the continuous development of technology and equipment in this field.

[0026] 4. The system integrates a variety of intelligent, networked and modular technologies, has good versatility, and can be extended to condition monitoring and fault diagnosis of other industrial process equipment.

[0027] In summary, the entire system can be designed directly for industry application needs at the same time, has good economic value and application prospects, and can be widely used in the diagnosis of textile pulp production equipment faults. DETAILED DESCRIPTION

[0028] The present invention will be further described below in conjunction with specific embodiments.

[0029] The expert system for diagnosing textile pulp production equipment faults of the present invention comprises:

[0030] Data acquisition module, which collects the operation data of textile pulp production equipment through a sensor network;

[0031] Among them, in the sensor network, a variety of sensors are used, such as temperature sensors, flow sensors, vibration sensors, current sensors, etc., which are installed in appropriate places. For example, the current sensor should be installed on the power supply line of the motor to detect the current consumption of the motor; the vibration sensor is installed near the bearing seat and motor of the pump to detect the vibration during operation and identify faults such as bearing wear, imbalance or mechanical looseness; the pressure sensor is installed at the inlet and outlet pipes of the slurry pump to monitor the slurry pressure changes to determine whether the pump is operating normally, etc.

[0032] The data transmission module has a signal input end connected to a signal output end of the data acquisition module and can convert the electrical signal collected by the data acquisition module into a digital signal that can be processed by a computer;

[0033] The expert system module, whose signal input terminal is connected to the signal output terminal of the data transmission module, can compare the digital signal transmitted by the data transmission module to complete fault diagnosis;

[0034] A user interface module, a signal input terminal of which is connected to a signal output terminal of the expert system module;

[0035] Among them, the user interface module is set up to facilitate operators to interact with the system, input equipment data, and view diagnostic results.

[0036] A method for diagnosing a fault of a textile pulp production equipment, the method is implemented by using the expert system for diagnosing a fault of a textile pulp production equipment, and the specific steps of the method include: step A, knowledge base construction: summarizing the contents of books on textile pulp production equipment, the knowledge of maintenance experts in the field of textile pulp production equipment, and the results obtained by analyzing previous maintenance cases to generate a knowledge base;

[0037] Among them, common failure modes and their corresponding symptoms related to textile pulp production equipment are collected from these knowledge sources, and the manifestations of different failure modes and how to perform successful repairs are understood, such as pump failure (leakage, increased noise, insufficient pressure), heating equipment failure (unstable temperature, slow heating speed), mixing equipment failure (uneven mixing, unstable operation), etc., thereby forming an expert system rule base for diagnosing such equipment failures; then, knowledge engineers encode this knowledge into the expert system and convert the knowledge into a form that can be understood by the computer (the system can process).

[0038] Step B, establishment of fault tree: determine which event combinations will cause faults based on the knowledge base, determine the minimum cut set, calculate the probability of each top event, analyze and obtain the prior probability, posterior probability and probability importance information, and establish a fault tree based on the prior probability, posterior probability and probability importance information;

[0039] Among them, the minimum cut set is the minimum combination that leads to the occurrence of the top event; the prior probability is the preliminary estimate of the probability of an event based on historical data or experience in the absence of new information; the posterior probability is the re-estimation of the probability of an event after the new information is obtained and the probability importance, that is, the contribution of each bottom event to the occurrence of the top event.

[0040] Step C, fault judgment: using the digital signal of the data transmission module as feature information, matching the information in the knowledge base, locating the fault tree branches and leaf nodes corresponding to the feature information, and deriving the system fault type that may be caused by the feature information according to the logical relationship reflected by the fault tree;

[0041] Among them, a range is set for the data collected by each sensor. If the sensor value is within its corresponding range, it means that the production equipment is in normal operation. Otherwise, it means that a fault has occurred and further diagnosis is required.

[0042] Specifically, the knowledge base in step A adopts forward chain reasoning, starting from the data stored in the knowledge base, and gradually inferring the conclusion according to the rules. The conclusion obtained will transmit the diagnosis result to the user interface module through the interpreter to display the final result to the user;

[0043] The final result displayed in the user module includes the fault name, the location of the abnormal sensor and its reading, the solution and the generation time of the result.

[0044] Specifically, the data transmission module converts the electrical signal into a digital signal that can be processed by a computer through an analog-to-digital converter, and after the digital signal is converted, the digital signal needs to be preprocessed.

[0045] Specifically, if the fault judgment in step C fails to deduce the system fault type that may be caused by the characteristic information, the following steps are included:

[0046] Step C1, prompt: If the fault judgment fails to deduce the type of system fault that may be caused by the characteristic information, it will be regarded as an unknown fault, and the user interface module will display an unknown fault to prompt relevant staff;

[0047] Step C2, record and update: unknown faults will be recorded in the log, and the log data will be analyzed to obtain new fault modes and symptoms. According to the new fault modes and symptoms, the corresponding branch points will be found in the fault tree, and the new fault results will be added as new leaf nodes of the fault tree;

[0048] By continuously adding new fault knowledge, the quality of the expert system and the accuracy of diagnosis will gradually improve.

[0049] Specifically, the digital signal preprocessing includes denoising, filtering, signal calibration and feature extraction to ensure the accuracy and clarity of the information.

[0050] Specifically, the preprocessed digital signal uses a LoRa gateway to transmit data to the expert system module.

[0051] In order to better understand the solution, three specific embodiments are now proposed in combination with actual practice.

[0052] Embodiment 1:

[0053] At the inlet of the main delivery pump pipeline of a slurry production line, an electromagnetic flowmeter with a diameter of 1 inch is installed, with a measuring range of 0.1-2 cubic meters / hour and an accuracy of plus or minus 0.5%; a pressure transmitter with a pressure measuring range of 0-2.5 MPa and an accuracy level of 0.5 is installed at the same pipeline position; a switch-type current transformer is installed on the three-phase power supply cable of the main motor to test the motor phase current.

[0054] The analog output signals of the three sensors are collected by the analog-to-digital conversion module with a conversion accuracy of 16 bits. The collected digital signals are input into the industrial host for filtering and smoothing, and then sent to the monitoring host 300 meters away from the site through the LoRa wireless transmission module with an RF power of 27dBm. The host is loaded with a knowledge base and fault decision tree logic built based on years of production data training. When the flow meter and pressure gauge data move down abnormally at the same time, exceeding the threshold corresponding to the fault library, and the motor current value is normal, the expert system determines that it is a pipeline blockage fault and recommends checking the filter or pipe bend.

[0055] Embodiment 2:

[0056] A harmonic vibration sensor with a measurement range of 100 g-force is installed on the side bearing of a certain type of slurry mixer, and the sampling frequency is set to 2KHz to collect vibration signals when the box and rotor are moving; a photoelectric encoder is connected to the main motor shaft end through a coupling to measure the main shaft speed with an accuracy level of 0.05. After the two signals are collected and processed by the data acquisition card, they are sent to the monitoring terminal via WiFi.

[0057] When the vibration signal amplitude exceeds the standard for a long time and the speed is stable within the standard operating range, the expert system matches the "normal speed-excessive vibration" rule and infers that it is a mechanical jam, prompting you to check whether the mixing blades are blocked or the main bearing is damaged.

[0058] Embodiment 3:

[0059] An ultrasonic level transmitter is installed at three height positions, namely the middle, top and bottom, of the side wall of a certain type of 500 cubic meter textile pulp storage box. The liquid level change range is 0-15 meters with an accuracy of ±0.2%. A thermocouple temperature measuring element is installed at the center of the box top to measure the pulp temperature range of 0-100℃ with an accuracy of ±0.5℃.

[0060] The analog output signals of the two sensors are filtered and amplified by the signal conditioning module. The processed two signals are respectively input into the analog-to-digital conversion module with a sampling accuracy of 24 bits for digital conversion. The converted digital signals are input into the industrial field bus processor to collect liquid level data and temperature data at a frequency of 1Hz.

[0061] When the collected data of the liquid level sensor shows that the slurry level in the storage box continues to drop rapidly, for example, it drops by more than 2 meters within 5 minutes; and the temperature data is stably maintained at the standard operating temperature of 65°C, the expert system knowledge base input by the industrial field bus processor will match the "rapid drop in liquid level - normal temperature" rule, and judge it as a pipeline or storage box leakage fault, and send a maintenance prompt or alarm to the maintenance personnel through the human-machine interface software.

[0062] Example 4

[0063] In the mixing device system of a textile pulp production line, various sensors such as temperature, vibration, and noise are installed in key parts such as the main motor, gear box, and stirring shaft to monitor the operating parameters in real time. The collected data is input into the expert system module for comparison and analysis.

[0064] During the operation of a certain device, the monitoring data showed abnormal fluctuations in the current and power of the main motor, and the vibration amplitude was significantly higher than the standard working condition. This combined abnormal mode was not in the knowledge base of the expert system and did not match the existing fault mode. At this time, the system determined it to be an "unknown fault" and popped up a prompt box through the human-machine interface software, instructing the maintenance personnel to record abnormal data and analyze parameters.

[0065] According to the prompts output by the system, the on-site staff collected and extracted the monitoring data and parameters of each sensor when the unknown fault occurred, and stored them in the system database. In the subsequent system optimization and knowledge update process, the expert system developers retrieved the data samples corresponding to the unknown fault, analyzed and identified them, and summarized a new fault mode characterized by motor fluctuation-increased vibration, namely, motor track fault. The new fault was added as an independent knowledge point in the fault tree model to improve the system knowledge base.

[0066] In summary, the expert system for diagnosing textile pulp production equipment faults of the present invention has the following effects:

[0067] 1. It improves the operational reliability of textile pulp production equipment, reduces the occurrence of unexpected failures, and effectively prevents economic losses and social impacts caused by sudden equipment shutdown.

[0068] 2. It realizes intelligent monitoring of equipment operation status and fault prediction, greatly reducing the dependence on professional maintenance personnel, making equipment maintenance simple and efficient.

[0069] 3. Through continuous learning of unknown faults and optimization and updating of the knowledge base, the accuracy of diagnosis and maintenance is continuously improved, making the system adapt to the continuous development of technology and equipment in this field.

[0070] 4. The system integrates a variety of intelligent, networked and modular technologies, has good versatility, and can be extended to condition monitoring and fault diagnosis of other industrial process equipment.

[0071] In summary, the entire system can be designed directly for industry application needs at the same time, has good economic value and application prospects, and can be widely used in the diagnosis of textile pulp production equipment faults.

[0072] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An expert system for diagnosing faults in textile pulp production equipment, characterized in that: include, Data acquisition module, which collects the operation data of textile pulp production equipment through a sensor network; The data transmission module has a signal input end connected to a signal output end of the data acquisition module and can convert the electrical signal collected by the data acquisition module into a digital signal that can be processed by a computer; The expert system module, whose signal input terminal is connected to the signal output terminal of the data transmission module, can compare the digital signal transmitted by the data transmission module to complete fault diagnosis; The signal input terminal of the user interface module is connected with the signal output terminal of the expert system module.

2. The expert system for diagnosing textile pulp production equipment faults according to claim 1, characterized in that: The data transmission module converts the electrical signal into a digital signal that can be processed by a computer through an analog-to-digital converter, and after the digital signal is converted, the digital signal needs to be preprocessed.

3. The expert system for diagnosing textile pulp production equipment faults according to claim 2, characterized in that: The digital signal preprocessing specifically includes denoising, filtering, signal calibration and feature extraction.

4. The expert system for diagnosing textile pulp production equipment faults according to claim 2, characterized in that: The pre-processed digital signal uses the LoRa gateway to transmit data to the expert system module.

5. A method for diagnosing a fault in a textile pulp production equipment, the method being implemented by using the expert system for diagnosing a fault in a textile pulp production equipment according to claim 1, characterized in that: The specific steps of the method include: Step A, knowledge base construction: relevant books on textile pulp production equipment, knowledge of maintenance experts in the field of textile pulp production equipment and results obtained from analyzing previous maintenance cases are summarized to generate a knowledge base; Step B, establishment of fault tree: determine which event combinations will cause faults based on the knowledge base, determine the minimum cut set, calculate the probability of each top event, analyze and obtain the prior probability, posterior probability and probability importance information, and establish a fault tree based on the prior probability, posterior probability and probability importance information; Step C, fault judgment: Take the digital signal of the data transmission module as the characteristic information, match the information in the knowledge base, locate the fault tree branches and leaf nodes corresponding to the characteristic information, and deduce the system fault type that may be caused by the characteristic information based on the logical relationship reflected by the fault tree.

6. The method for diagnosing a fault in a textile pulp production device according to claim 2, characterized in that: The knowledge base in step A adopts forward chain reasoning, starting from the data stored in the knowledge base, and gradually inferring conclusions according to the rules. The conclusions obtained will be transmitted to the user interface module via the interpreter to display the final results to the user.

7. The method for diagnosing a fault in a textile pulp production device according to claim 2, characterized in that: If the fault judgment in step C fails to deduce the system fault type that may be caused by the characteristic information, the following steps are included: Step C1, prompt: If the fault judgment fails to deduce the type of system fault that may be caused by the characteristic information, it will be regarded as an unknown fault, and the user interface module will display an unknown fault to prompt relevant staff; Step C2, record and update: unknown faults will be recorded in the log, and the log data will be analyzed to obtain new fault modes and symptoms. According to the new fault modes and symptoms, the corresponding branch points will be found in the fault tree, and the new fault results will be added as new leaf nodes of the fault tree.

8. The method for diagnosing a fault in a textile pulp production device according to claim 4, characterized in that: The digital signal preprocessing specifically includes denoising, filtering, signal calibration and feature extraction.

9. The method for diagnosing a fault in a textile pulp production device according to claim 4, characterized in that: The pre-processed digital signal uses the LoRa gateway to transmit data to the expert system module.