Intelligent prediction method and system for mine electromechanical system failure

Abstract

The application discloses a mine electromechanical system fault intelligent prediction method and system. The prediction method comprises the following steps: establishing a common fault database based on common fault influencing factors of mine electromechanical equipment, and generating a common fault data interval through the common fault database; establishing a corresponding normal data database based on the common fault database; collecting real-time data of the common fault influencing factors in real time through a computer, comparing the real-time data with corresponding equipment data, obtaining a data deviation value, and then calculating a data deviation rate of the data deviation value relative to the equipment data; outputting fault prediction data results based on the weight proportion of the common fault influencing factors and the data deviation rate, reminding when the fault prediction data results are greater than a first set threshold, and outputting real-time data with a data deviation rate greater than a second set threshold and data information of a corresponding common fault database, so that the possible faults of the mine electromechanical system can be predicted.

Description

Technical Field

[0001] This invention relates to the field of electromechanical system fault prediction technology, and in particular to intelligent fault prediction methods and systems for mining electromechanical systems. Background Technology

[0002] In the mining process, the normal operation of underground electromechanical systems plays a crucial role. With the rapid development of industrial systems, the complexity of industrial equipment operation scenarios has also increased significantly. However, traditional equipment management methods rely too heavily on manual labor, resulting in poor efficiency and accuracy.

[0003] In manual inspections, the detection and discovery of electromechanical system faults are often delayed. After a fault is detected or discovered, maintenance is often required, which affects production efficiency. In addition, there is also the phenomenon of fault data silos, where environmental parameter collection and equipment status monitoring are isolated from each other, making it difficult to quickly and accurately deduce the specific fault factors of the electromechanical system. Summary of the Invention

[0004] This application provides a method and system for intelligent prediction of faults in mine electromechanical systems. It can establish an intelligent fault prediction model based on environmental, electrical, and mechanical factors related to the electromechanical system through a computer, predict the faults that the underground electromechanical system may face, and quickly and accurately infer the specific fault situations that may occur in the electromechanical system. This can assist workers in quickly carrying out targeted repairs or maintenance.

[0005] The first aspect of this application provides a method for intelligent prediction of faults in a mining electromechanical system, comprising the following steps: S10, establish a common fault database based on the common fault influencing factors of mining electromechanical equipment. The common fault database includes one-to-one corresponding equipment name, fault factor, fault code, fault data and fault description information. Generate common fault data range through the common fault database. The common fault influencing factors include environmental factors, electrical factors and mechanical factors. S20, establish a corresponding normal data database based on the common fault database, the normal data database including the equipment name, the fault factor and the equipment data corresponding to the fault data; S30, the computer, in conjunction with sensors, collects real-time data of the common fault influencing factors, compares the real-time data with the corresponding equipment data, obtains the data deviation value, and then calculates the data deviation rate of the data deviation value relative to the equipment data; S40, based on the weight ratio of the common fault influencing factors and the data deviation rate, output the fault prediction data result. When the fault prediction data result is greater than a first set threshold, issue a reminder and output the real-time data where the data deviation rate is greater than a second set threshold, as well as the data information of the common fault database corresponding to the real-time data.

[0006] In one possible implementation, the environmental factors include humidity, dust concentration, and corrosive gases; the electrical factors include voltage and current; and the mechanical factors include vibration, noise, temperature, rotational speed, and torque.

[0007] In one possible implementation, among the common fault-influencing factors, the weight percentages of humidity, dust concentration, corrosive gas, voltage, current, vibration, noise, temperature, rotational speed, and torque are 5%, 20%, 10%, 9%, 9%, 14%, 6%, 10%, 8%, and 9%, respectively.

[0008] In one possible implementation, the first set threshold is 70%-80% of the lower limit of the common fault data range.

[0009] In one possible implementation, the second set threshold corresponding to each of the common fault influencing factors is different.

[0010] In one possible implementation, the normal data database further includes fault solution information corresponding to the fault description information. When the fault prediction data result is greater than a first set threshold, the following are output: the real-time data with a data deviation rate greater than a second set threshold, the data of the common fault database corresponding to the real-time data, and the fault solution information corresponding to the fault description information in the common fault database.

[0011] In one possible implementation, the weight percentages of the common fault data, the normal data database, and the common fault influencing factors are updated at predetermined intervals.

[0012] In one possible implementation, in step S40, when the fault prediction data result is greater than a first preset threshold, a reminder is given simultaneously through a display screen and a voice speaker, and the fault solution information corresponding to the fault description information in the common fault database is displayed on the display screen.

[0013] In one possible implementation, a linear model is established based on the data deviation rate of real-time data of the common fault influencing factors and the corresponding weight ratio of the common fault influencing factors. In the linear model, the features are multiplied by their weights, summed, and then input into an activation function to obtain the fault prediction probability. The calculation formula is as follows: Fault prediction probability P = σ ( w 1 * x 1+ w 2 * x 2+ ⋯+ w n * x n+ b ); in wi It is the weighting percentage. xi It is the data deviation rate. σ It is the sigmoid function. b It is a bias term.

[0014] A second aspect of this application provides an intelligent prediction system for mine electromechanical system faults, used to implement the intelligent prediction method for mine electromechanical system faults as described above, the intelligent prediction system comprising: The common fault database establishment unit is used to establish a common fault database based on the common fault influencing factors of mining electromechanical equipment. The common fault database includes a one-to-one corresponding equipment name, fault factor, fault code, fault data and fault description information. The common fault database generates a common fault data range. The common fault influencing factors include environmental factors, electrical factors and mechanical factors. A normal data database establishment unit is used to establish a corresponding normal data database based on the common fault database. The normal data database includes the device name, the fault factor, and device data corresponding to the fault data. The data deviation rate acquisition unit is used to collect real-time data of the common fault influencing factors in real time through a computer and sensors, compare the real-time data with the corresponding equipment data, obtain the data deviation value, and then calculate the data deviation rate of the data deviation value relative to the equipment data. The prediction information output unit is used to output fault prediction data results based on the weight ratio of the common fault influencing factors and the data deviation rate. When the fault prediction data result is greater than a first set threshold, an alert is issued, and the real-time data with the data deviation rate greater than a second set threshold, as well as the data information of the common fault database corresponding to the real-time data, are output.

[0015] Beneficial Effects: Compared with existing technologies, the intelligent fault prediction method and system for mining electromechanical systems provided in this application comprehensively judges the probability of future faults in mining electromechanical equipment by simultaneously evaluating common fault databases, normal data databases, data deviation rates calculated from real-time data collected by computers, and the weight ratio of common fault influencing factors. This forms an intelligent fault prediction model, which can quickly and accurately predict the specific fault situations that may occur in the electromechanical system. This assists staff in quickly carrying out targeted repairs or maintenance, thereby solving the problem of delayed fault detection to a certain extent. By predicting and preventing in advance, the frequency of downtime maintenance can be reduced, or maintenance can be carried out in advance during non-working hours, ensuring mine production efficiency and extending the service life of mining electromechanical equipment.

[0016] These and other objects, features and advantages of the present invention will become fully apparent from the following detailed description. Attached Figure Description

[0017] Figure 1 A flowchart illustrating the intelligent fault prediction method for mine electromechanical systems presented in this application is shown. Detailed Implementation

[0018] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0019] Those skilled in the art should understand that, in the disclosure of this specification, the terms "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the above terms should not be construed as limiting the present invention.

[0020] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0021] In the mining production process, the normal operation of underground mining electromechanical systems plays a crucial role. The operation of these systems involves multiple factors, including environmental, electrical, and mechanical factors. Manual inspections require comprehensive attention to all these factors, which is not only labor-intensive and demands a high level of experience from workers, but also suffers from a certain lag in fault detection. Downtime for maintenance can severely impact production efficiency. Furthermore, there is the problem of isolated fault data; environmental parameter collection and equipment status monitoring are not interconnected. Because underground inspections generally follow a certain time pattern, it is difficult to quickly and accurately deduce the specific faults in the electromechanical system when problems occur outside of inspection periods. This delay can easily lead to missing the optimal maintenance window, further deteriorating the equipment, increasing losses, and severely impacting its normal service life. With the development of computer equipment, intelligent management of mining electromechanical equipment has become possible. It can quickly and accurately predict potential future faults in mining electromechanical equipment, thereby assisting workers in the repair, maintenance, and prevention of equipment failures, and providing scientific management basis and guidance.

[0022] refer to Figure 1 The first aspect of this application provides a method for intelligent prediction of faults in a mining electromechanical system, comprising the following steps: S10. Establish a common fault database based on the common influencing factors of mining electromechanical equipment failures. This database includes a one-to-one correspondence of equipment name, fault factor, fault code, fault data, and fault description information. Since different equipment manufacturers establish their own equipment fault codes, even for the same equipment from different manufacturers, the fault codes may still be different. Therefore, the common fault database can establish new standards (referring to relevant national or industry standards). For example, equipment name: motor; fault factor: bearing end cover; fault code: ZCD032; fault data: 4.8 mm / s (referencing the ISO 10816-3 specification for "Class I" equipment (motors with power greater than 15kW and speed between 120-15000 rpm); fault description information: unacceptable, severe bearing vibration may cause equipment damage, immediate shutdown and inspection are required. Common fault data ranges are generated through the aforementioned common fault database. These ranges are primarily established based on fault data, with each fault factor corresponding to a specific range. For example, after statistically analyzing numerous fault data points related to bearing end caps, the established fault data range for motor bearing end caps is 4.6 mm / s - 5.1 mm / s. The common fault influencing factors include environmental, electrical, and mechanical factors. Environmental factors include humidity, dust concentration, and corrosive gases; electrical factors include voltage and current; and mechanical factors include vibration, noise, temperature, rotational speed, and torque. S20, a corresponding normal data database is established based on the common fault database. The normal data database includes the equipment name, the fault factor, and the equipment data corresponding to the fault data. Since the intelligent fault prediction method and system for mining electromechanical systems provided in this application are for fault prediction, and new faults lack relevant data, they cannot be predicted. Therefore, the equipment name and fault factor are the same and known in the common fault database and the normal data database, while the fault data and equipment data are different. The fault data deviates significantly from the equipment data, either higher or lower than the equipment data range value. The equipment data refers to the normal operation data of the fault factor. S30: A computer, in conjunction with sensors, collects real-time data on common fault influencing factors. Simultaneously, it compares the real-time data with the corresponding equipment data to determine the data deviation value, and then calculates the data deviation rate relative to the equipment data. When the real-time data deviates from the equipment data, it may deviate upwards (positive value) or downwards (negative value). Therefore, any deviation in real-time data from the equipment data can be counted as a positive deviation using absolute values. In the process of using the computer with sensors, conventional sensors can be selected, such as: temperature and humidity sensors to collect temperature and humidity data of the electromechanical system environment; laser dust sensors (PM2.5 / PM10) or photoelectric dust sensors to collect dust concentration data of the electromechanical system; electrochemical sensors to collect data on toxic and corrosive gases (such as H2S, SO2, NOx, Cl2); voltage and current sensors to collect voltage and current data respectively; vibration sensors to collect equipment vibration data; acoustic sensors to collect noise data of the electromechanical system; speed sensors to collect equipment rotation speed data; and reactive torque sensors or rotary torque sensors to collect equipment torque data, etc. Based on the aforementioned fault data of 4.8 mm / s, and referring to ISO 10816-3 for "Class 1" equipment (power greater than 15 kW, speed between 120-15000 rpm)... According to the specifications for motors with a rated speed of rpm, the normal vibration speed range of the motor bearing end cover is 0-2.8 mm / s. If the corresponding data collected in real time is 3.1 mm / s (less than 4.6 mm / s, not reaching the fault level), the data deviation value can be calculated as: 3.1-2.8=0.3, and the data deviation rate is: 0.3 / 2.8=10.71%. For example, if the rated speed of the motor rotor is 1480 RPM and the real-time speed is 1350 RPM, the data deviation value is: 1350-1480=-130. Taking the absolute value as 130, the data deviation rate is: 130 / 1480=8.78%. As can be seen from the above, the difference between the data deviation values ​​of different fault factors can be very large. If the data deviation value is directly used as the product base of the weight ratio, it will obviously distort the final result of fault prediction. Therefore, the data deviation value must be converted into a data deviation rate. S40, based on the weight proportions of the common fault influencing factors and the data deviation rate, output the fault prediction data result. The fault prediction data result reflects the fitted result of the weight proportions of all fault factors and the corresponding data deviation rates. Specifically, a linear model is established based on the data deviation rate of the real-time data of the common fault influencing factors and the corresponding weight proportions of the common fault influencing factors. In the linear model, the features are multiplied by their weights and summed, then input into an activation function to obtain the fault prediction probability. The calculation formula is as follows: Fault prediction probability P= σ ( w 1 * x 1+ w 2 * x 2+ ⋯+ w n * x n+ b ); in wi It is the weighting percentage. xi It is the data deviation rate. σ It is the sigmoid function. b This is a bias term; * indicates multiplication. Fault prediction probability. P This system generates fault predictions by capturing real-time data from all fault factors. The results are presented under the premise that no common faults have occurred for any of the equipment's fault factors, and are used for probability prediction before a fault occurs. Simultaneously, when the fault prediction data exceeds a first preset threshold, an alert is issued, and the real-time data showing a deviation rate exceeding a second preset threshold, along with corresponding data from the common fault database, is output. b This is the bias term, also known as the intercept term, in fault prediction scenarios. b The "basic failure tendency" representing the failure factor is used in actual prediction processes. b It is a constant value that can be set as needed. For example, when the device itself is relatively reliable and a strong abnormal signal is required to predict a fault, it can be set to... b When a negative value is assigned, and the equipment itself has a high failure tendency, there is still a certain risk of failure even if all factors appear normal, then... b The value is assigned as a positive value. b Weights of each influencing factor wi Together, they determine the final failure probability prediction result. Meanwhile, during model training, b It can also be related to weights wi They are learned and optimized together to achieve the best predictive performance.

[0023] The intelligent fault prediction method for mining electromechanical systems in this application is set as follows: when the data deviation rate of at least one fault factor is greater than a second preset threshold, the overall fault prediction probability will be increased. PIf the data deviation exceeds a first preset threshold, an alert is issued, and real-time data on fault factors with a deviation rate exceeding a second preset threshold, along with data from a common fault database, is output. This includes equipment name, fault factor, fault code, fault data, and fault description information. This allows staff to quickly locate and view relevant information for the fault factor, enabling repairs or maintenance to be performed before the fault factor malfunctions. This addresses the issue of delayed fault detection to some extent. Early prediction and prevention can reduce downtime for maintenance or allow for maintenance to be performed during off-peak hours, ensuring mine production efficiency and extending the lifespan of mine electromechanical equipment. The fault prediction probability is also included. P The following settings can be made: .

[0024] Based on the above settings, the first threshold can be 0.5% or 50%. Of course, this also applies to the fault prediction probability. P When the deviation rate exceeds the first set threshold, it's also possible that two fault factors simultaneously exceed their corresponding second set threshold. In this case, staff can handle each fault factor separately. However, if the fault prediction probability... P If the deviation rate is not greater than the first set threshold, it means that the data deviation rate of all fault factors is not greater than their corresponding second set threshold.

[0025] In practical applications, the actual conditions of each mine may differ. For example, some mines have high dust levels but relatively new equipment; in this case, the weight of environmental factors in the common fault influencing factors needs to be higher. Conversely, some mines have relatively good environments but older equipment; in this case, the weight of electrical and mechanical factors in the common fault influencing factors needs to be relatively higher. Therefore, in one embodiment, the weights of humidity, dust concentration, corrosive gases, voltage, current, vibration, noise, temperature, rotational speed, and torque in the common fault influencing factors are 5%, 20%, 10%, 9%, 9%, 14%, 6%, 10%, 8%, and 9%, respectively. This weighting is mainly for mines with high dust levels but moderate other influencing factors.

[0026] In one embodiment, the first set threshold is 70%-80% of the lower limit of the common fault data range. This is a proportional value used to assess the degree to which the fault factor is close to the actual fault. Based on the magnitude of this degree value, a certain amount of time can be reserved for the fault factor to ensure that staff can deal with it in a timely manner before the fault factor fails, such as through repair or maintenance, making fault prediction more practical.

[0027] Furthermore, the fault tolerance or fault tolerance of each fault factor in the electromechanical system may vary significantly. Fault tolerance refers to the ability of a fault factor to maintain normal operation even when a fault occurs. Therefore, the second set threshold corresponding to each of the common fault influencing factors is different. This allows for tailored settings based on the fault factor and actual working conditions. Thus, this differentiated second set threshold setting method is more in line with the intelligent management strategy of mines, improving the probability of fault prediction. P It has a higher accuracy rate and can make more accurate fault predictions for various devices or fault factors in electromechanical systems.

[0028] For example, if the second set threshold for the vibration velocity of the motor bearing end cover is 12%, and the data deviation rate for the vibration velocity of the motor bearing end cover is 0.3 / 2.8 = 10.71%, it falls below its corresponding second set threshold. Similarly, if the second set threshold for the motor rotor speed is 8%, and the data deviation rate for the motor rotor speed is 130 / 1480 = 8.78%, it falls above its corresponding second set threshold. In this case, the fault prediction probability... P A reminder will definitely be given.

[0029] In one embodiment, the normal data database further includes fault solution information corresponding to the fault description information. Simultaneously, when the fault prediction data result exceeds a first preset threshold, the system outputs: the real-time data where the data deviation rate exceeds a second preset threshold, the data from the common fault database corresponding to the real-time data, and the fault solution information corresponding to the fault description information in the common fault database. This allows staff to quickly obtain the corresponding fault solution information when a fault prediction alert or alarm occurs, facilitating targeted and rapid processing. This significantly reduces the experience requirements for staff and minimizes the risk of incorrect handling.

[0030] Over time, not only does the mine environment change, but the fault tolerance of electrical and mechanical equipment may also change with the use of these devices. Furthermore, with the continuous development of technology, the databases of common faults and normal data may also evolve. Therefore, in one embodiment, the weights of the common fault data, the normal data database, and the influencing factors of common faults are updated at predetermined intervals. This allows for dynamic adjustment of the weights, thereby ensuring the accuracy of fault prediction. P The accuracy rate is improved, enabling more precise fault prediction for various devices or fault factors in electromechanical systems.

[0031] In one embodiment, in step S40, when the fault prediction data result is greater than a first set threshold, a reminder is given simultaneously through a display screen and a voice speaker, and the fault solution information corresponding to the fault description information in the common fault database is displayed on the display screen, so that staff can clearly receive and view the fault solution information, and thus quickly repair or maintain the corresponding fault factor.

[0032] Based on a largely similar working principle, a second aspect of this application provides an intelligent prediction system for mine electromechanical system faults, used to implement the intelligent prediction method for mine electromechanical system faults as described above. The intelligent prediction system includes: The common fault database establishment unit is used to establish a common fault database based on the common fault influencing factors of mining electromechanical equipment. The common fault database includes a one-to-one corresponding equipment name, fault factor, fault code, fault data and fault description information. The common fault database generates a common fault data range. The common fault influencing factors include environmental factors, electrical factors and mechanical factors. A normal data database establishment unit is used to establish a corresponding normal data database based on the common fault database. The normal data database includes the device name, the fault factor, and device data corresponding to the fault data. The data deviation rate acquisition unit is used to collect real-time data of the common fault influencing factors in real time through a computer and sensors, compare the real-time data with the corresponding equipment data, obtain the data deviation value, and then calculate the data deviation rate of the data deviation value relative to the equipment data. The prediction information output unit is used to output fault prediction data results based on the weight ratio of the common fault influencing factors and the data deviation rate. When the fault prediction data result is greater than a first set threshold, an alert is issued, and the real-time data with the data deviation rate greater than a second set threshold, as well as the data information of the common fault database corresponding to the real-time data, are output.

[0033] It should be noted that the terms "first" and "second" used in this application are for descriptive purposes only and do not indicate any order. They should not be construed as indicating or implying relative importance, and can be interpreted as names.

[0034] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the invention. The advantages of the present invention have been fully and effectively realized. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments; any variations or modifications can be made to the implementation of the present invention without departing from these principles.

Claims

1. A method for intelligent prediction of faults in mine electromechanical systems, characterized in that, Includes the following steps: S10, establish a common fault database based on the common fault influencing factors of mining electromechanical equipment. The common fault database includes one-to-one corresponding equipment name, fault factor, fault code, fault data and fault description information. Generate common fault data range through the common fault database. The common fault influencing factors include environmental factors, electrical factors and mechanical factors. S20, establish a corresponding normal data database based on the common fault database. The normal data database includes the device name, the fault factor, and the device data corresponding to the fault data, wherein the device data refers to the normal operation data of the fault factor. S30, the computer, in conjunction with sensors, collects real-time data of the common fault influencing factors, compares the real-time data with the corresponding equipment data, obtains the data deviation value, and then calculates the data deviation rate of the data deviation value relative to the equipment data; S40, based on the weight ratio of the common fault influencing factors and the data deviation rate, output the fault prediction data result, and when the fault prediction data result is greater than the first set threshold, issue a reminder, and output the real-time data when the data deviation rate is greater than the second set threshold and the data information of the common fault database corresponding to the real-time data; A linear model is established based on the data deviation rate of real-time data of the common fault influencing factors and the corresponding weight ratio of the common fault influencing factors. In the linear model, the features are multiplied by their weights and summed, then input into an activation function to obtain the fault prediction probability, i.e., the fault prediction data result. The calculation formula is as follows: Fault prediction probability P = σ ( w 1 * x 1+ w 2 * x 2+ … + w n * x n+ b ); in wi It is the weighting percentage. xi It is the data deviation rate. σ It is the sigmoid function. b It is a bias term.

2. The intelligent fault prediction method for mine electromechanical systems as described in claim 1, characterized in that, The environmental factors include humidity, dust concentration, and corrosive gases; the electrical factors include voltage and current; and the mechanical factors include vibration, noise, temperature, rotational speed, and torque.

3. The intelligent fault prediction method for mine electromechanical systems as described in claim 2, characterized in that, Among the common factors affecting failure, the weight percentages of humidity, dust concentration, corrosive gas, voltage, current, vibration, noise, temperature, speed, and torque are 5%, 20%, 10%, 9%, 9%, 14%, 6%, 10%, 8%, and 9%, respectively.

4. The intelligent fault prediction method for mine electromechanical systems as described in claim 1, characterized in that, The first set threshold is 70%-80% of the lower limit of the common fault data range.

5. The intelligent fault prediction method for mine electromechanical systems as described in claim 4, characterized in that, The second set threshold corresponding to each of the common fault influencing factors is different.

6. The intelligent fault prediction method for mine electromechanical systems as described in claim 5, characterized in that, The normal data database also includes fault solution information corresponding to the fault description information. When the fault prediction data result is greater than a first set threshold, the following are output: the real-time data with a data deviation rate greater than a second set threshold, the data of the common fault database corresponding to the real-time data, and the fault solution information corresponding to the fault description information in the common fault database.

7. The intelligent fault prediction method for mine electromechanical systems as described in claim 6, characterized in that, The weight percentages of the common fault database, the normal data database, and the common fault influencing factors are updated at predetermined intervals.

8. The intelligent fault prediction method for mine electromechanical systems as described in claim 6, characterized in that, In step S40, when the fault prediction data result is greater than the first set threshold, a reminder is given simultaneously through the display screen and the voice speaker, and the fault solution information corresponding to the fault description information in the common fault database is displayed on the display screen.

9. A mine electromechanical system fault intelligent prediction system, used to implement the mine electromechanical system fault intelligent prediction method as described in any one of claims 1 to 8, characterized in that, The intelligent prediction system includes: The common fault database establishment unit is used to establish a common fault database based on the common fault influencing factors of mining electromechanical equipment. The common fault database includes a one-to-one corresponding equipment name, fault factor, fault code, fault data and fault description information. The common fault database generates a common fault data range. The common fault influencing factors include environmental factors, electrical factors and mechanical factors. A normal data database establishment unit is used to establish a corresponding normal data database based on the common fault database. The normal data database includes the device name, the fault factor, and device data corresponding to the fault data. The data deviation rate acquisition unit is used to collect real-time data of the common fault influencing factors in real time through a computer and sensors, compare the real-time data with the corresponding equipment data, obtain the data deviation value, and then calculate the data deviation rate of the data deviation value relative to the equipment data. The prediction information output unit is used to output fault prediction data results based on the weight ratio of the common fault influencing factors and the data deviation rate. When the fault prediction data result is greater than a first set threshold, an alert is issued, and the real-time data with the data deviation rate greater than a second set threshold, as well as the data information of the common fault database corresponding to the real-time data, are output.

Images