SEA DC servo motor driver fault detection method

Abstract

The invention relates to the technical field of driver detection, and discloses an SEA direct current servo motor driver fault detection method, which comprises the steps of data acquisition, feature extraction, data analysis and fault early warning, and is characterized in that multiple signals such as input voltage, output current, motor rotating speed and driver internal temperature are acquired, and the operation state is monitored in an omnibearing manner; features such as mean value, variance and kurtosis are obtained through feature extraction, faults are judged based on Euclidean distance, the accuracy is high, fault types such as voltage, current, rotating speed and temperature can be accurately judged by combining feature vector elements after fault alarm, and targeted maintenance is facilitated; the audible and visual alarm is combined with a fault alarm signal to give an alarm in time, so that a worker can quickly know the fault; the method can be conveniently implemented in a plurality of application scenes, effectively guarantees the normal operation of equipment, and remarkably improves the fault detection, diagnosis and early warning capabilities of the direct current servo motor driver on the whole.

Description

Technical Field

[0001] The present invention relates to the technical field of driver detection, and in particular to a fault detection method for a SEA DC servo motor driver. Background Art

[0002] In the field of industrial automation control, SEA DC servo motor drives play a crucial role. They precisely control the operation of DC servo motors and are widely used in numerous devices requiring extremely high precision and reliability, such as CNC machine tools, robots, and automated production lines. With the continuous advancement of industrial technology, production processes are placing increasingly stringent demands on the stability and continuity of equipment operation. A failure in a DC servo motor drive can cause the entire production system to shut down, resulting in significant economic losses, including direct losses from production interruption, equipment repair costs, and potential liquidated damages due to delayed delivery.

[0003] Traditional fault detection methods often have numerous limitations. For example, some simple threshold judgment methods rely solely on the threshold range of a single parameter to determine faults, failing to comprehensively consider the interrelationships between multiple parameters and the overall system operating status. This can easily lead to misjudgments or missed faults. Furthermore, in complex industrial environments, motor drives may be subject to interference from a variety of factors, such as power supply fluctuations, sudden load changes, and electromagnetic interference. These factors can cause subtle changes in the drive's operating parameters, making it difficult for traditional detection methods to accurately capture these changes and promptly and effectively determine whether a potential fault exists. Furthermore, when a fault occurs, traditional methods typically only provide relatively general information, making it difficult to precisely determine the fault type and specific cause. This greatly complicates troubleshooting and repair efforts for maintenance personnel, prolonging equipment downtime and further exacerbating economic losses.

[0004] Therefore, in order to overcome the shortcomings of traditional fault detection methods, the present invention provides an efficient and reliable DC servo motor driver fault detection solution. Summary of the Invention

[0005] The purpose of the present invention is to provide a SEA DC servo motor driver fault detection method, which solves the technical problems raised in the background technology.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A SEA DC servo motor driver fault detection method comprises the following steps:

[0008] Step 1: Data Collection

[0009] Collect the input voltage signal, output current signal, motor speed signal of the DC servo motor driver and the temperature signal inside the driver;

[0010] Step 2: Feature Extraction

[0011] Extract the data collected within a preset time length, perform feature extraction processing on it, and obtain the corresponding mean feature, variance feature and kurtosis feature;

[0012] Step 3: Data Analysis

[0013] The data of the DC servo motor driver under normal state and current time length are respectively acquired through the data acquisition step, and the data features of the DC servo motor driver under normal state and current time length are subsequently extracted through the feature extraction step, and then whether the DC servo motor driver has a fault is determined based on the relationship between the data features under normal state and current time length;

[0014] Step 4: Fault Warning

[0015] Sound and light alarm is performed by combining sound and light alarm with fault alarm signal.

[0016] As a further solution of the present invention: wherein the sampling frequency is set to f, the acquisition time length is T, and the number of data points collected is n, and n=f×T.

[0017] As a further solution of the present invention: the feature extraction processing method is as follows:

[0018] Step K1. Vector marking

[0019] The input voltage signal, output current signal, motor speed signal and temperature signal collected within the time length T are marked as U i 、L i 、Z i and W i , i=1, 2, ... n;

[0020] Then U i 、L i 、Z i and W i As feature vectors, and combined into a feature vector set X i =[U i ,L i ,Z i ,W i ];

[0021] StepK2, vector mean calculation

[0022] pass:

[0023]

[0024] Calculate the mean XP of the corresponding eigenvector in the eigenvector set, and XP = [UP, LP, ZP, WP];

[0025] Among them, UP is U i The corresponding input voltage mean, LP is L i The corresponding output current mean, ZP is Z i The corresponding motor speed average, WP is W i The corresponding temperature mean;

[0026] Step K3, vector variance calculation

[0027]

[0028] Calculate the variance XF of the corresponding eigenvector in the eigenvector set, and XF = [UF, LF, ZF, WF];

[0029] Among them, UF is U i The corresponding input voltage variance, LF is L i The corresponding output current variance, ZF is Z i The corresponding motor speed variance, WF is W i The corresponding temperature variance;

[0030] Step K4, vector kurtosis calculation

[0031] pass:

[0032]

[0033] Calculate the kurtosis XD of the corresponding eigenvector in the eigenvector set, and XD = [UD, LD, ZD, WD];

[0034] Among them, UD is U i The corresponding input voltage kurtosis, LD is the output current kurtosis corresponding to Li, and ZD is the Z i The corresponding motor speed kurtosis, WD is W i The corresponding temperature kurtosis.

[0035] As a further solution of the present invention: the data analysis method is as follows:

[0036] Step H1, according to the data collection and feature extraction steps, obtain the feature vector and its mean, variance and kurtosis of the DC servo motor driver in the normal state;

[0037] And combine them into the total set of normal state eigenvectors:

[0038] G0=[G0 j]=[UP0, UF0, UD0, LF0, LD0, LP0, ZD0, ZF0, ZP0, WD0,

[0039] WF0, WP0];

[0040] Where, j = 1, 2, ..., 12;

[0041] and obtaining the characteristic vector of the DC servo motor driver at the current time length and its mean, variance and kurtosis;

[0042] And combine them into the total set of current time length feature vectors:

[0043] G1=[G1 j ]=[UP1, UF1, UD1, LF1, LD1, LP1, ZD1, ZF1, ZP1, WD1,

[0044] WF1, WP1];

[0045] in:

[0046] UP0, UF0, and UD0 are the mean, variance, and kurtosis of the input voltage under normal conditions, respectively;

[0047] UP1, UF1, and UD1 are the mean, variance, and kurtosis of the input voltage at the current time length, respectively;

[0048] LF0, LD0, and LP0 are the mean, variance, and kurtosis of the output current under normal conditions, respectively;

[0049] LF1, LD1, and LP1 are the mean, variance, and kurtosis of the output current at the current time length, respectively;

[0050] ZD0, ZF0, and ZP0 are the mean, variance, and kurtosis of the motor speed under normal conditions, respectively;

[0051] ZD1, ZF1, and ZP1 are the mean, variance, and kurtosis of the motor speed at the current time length, respectively;

[0052] WD0, WF0, and WP0 are the mean, variance, and kurtosis of the temperature under normal conditions, respectively;

[0053] WD1, WF1, and WP1 are the mean, variance, and kurtosis of the temperature at the current time length, respectively;

[0054] StepH2, then pass:

[0055]

[0056] Calculate the Euclidean distance Q between the total set of normal state feature vectors and the total set of current time length feature vectorsd ;

[0057] StepH3, the Euclidean distance Q will be obtained d And the corresponding preset Euclidean distance threshold QY d For comparison:

[0058] If Q d >QY d , it is determined that the DC servo motor driver has a fault, and a fault alarm signal is generated;

[0059] If Q d ≤QY d , it is determined that the DC servo motor driver is in normal operation and no fault alarm signal is generated.

[0060] As a further solution of the present invention: when the fault alarm signal is generated, the fault type determination analysis is performed by combining the normal state feature vector total set and the current time length feature vector total set of each feature vector element;

[0061] As a further solution of the present invention: the fault type determination and analysis method is as follows:

[0062] Step Y1: Determine the voltage fault type

[0063] Extract the mean UP0 and variance UF0 of the input voltage under normal conditions, as well as the mean UP1 and variance UF1 of the input voltage under the current time length;

[0064] Then through:

[0065]

[0066] Calculate the deviation ratio EP of the input voltage mean respectively U The deviation ratio EF of the input voltage variance U ;

[0067] Then the deviation ratio EP of the input voltage mean is U The deviation ratio EF of the input voltage variance U The corresponding preset output current mean deviation threshold EPy U and output current variance deviation threshold EFy U For comparison:

[0068] When EP U >EPy U and EF U >EFy U When , it is determined that there is an input voltage fault in the DC servo motor;

[0069] When EP U >EPyU and EF U >EFy U If at least one comparison formula is not true, it is not determined that the DC servo motor has an input voltage fault;

[0070] Step Y2: Determine the current fault type

[0071] Extract the mean LP0 and variance LF0 of the output current under normal conditions, as well as the mean LP1 and variance LF1 of the output current under the current time length;

[0072] Then through:

[0073]

[0074] Calculate the deviation ratio EP of the output current mean respectively L The deviation ratio of the output current variance EF L ;

[0075] Then the deviation ratio EP of the output current mean is L The deviation ratio of the output current variance EF L The corresponding preset output current mean deviation threshold EPy L and output current variance deviation threshold EFy L For comparison:

[0076] When EP L >EPy L and EF L >EFy L When , it is determined that the DC servo motor has an output current fault;

[0077] When EP L >EPy L and EF L >EFy L If at least one comparison formula in the formula is not true, it is not determined that the DC servo motor has an output current fault;

[0078] Step Y3: Determine the speed fault type

[0079] Extract the mean ZP0 and variance ZF0 of the motor speed under normal conditions, as well as the mean ZP1 and variance ZF1 of the motor speed under the current time length;

[0080] Then through:

[0081]

[0082] Calculate the deviation ratio EP of the mean motor speed respectively Z The deviation ratio EF of the motor speed varianceZ ;

[0083] Then the deviation ratio EP of the motor speed mean is Z The deviation ratio EF of the motor speed variance Z The corresponding preset output current mean deviation threshold EPy Z and output current variance deviation threshold EFy Z For comparison:

[0084] When EP Z >EPy Z and EF Z >EFy Z When , it is determined that the DC servo motor has a motor speed fault;

[0085] When EP Z >EPy Z and EF Z >EFy Z If at least one comparison formula in is not true, it is not determined that the DC servo motor has a motor speed fault;

[0086] Step Y4: Determine the temperature fault type

[0087] Extract the mean WP0 and variance WF0 of the temperature under normal conditions, as well as the mean WP1 and variance WF1 of the temperature under the current time length;

[0088] Then through:

[0089]

[0090] Calculate the deviation ratio EP of the temperature mean respectively W The deviation ratio EF from the temperature variance W ;

[0091] Then the deviation ratio EP of the temperature mean is W The deviation ratio EF from the temperature variance W The corresponding preset output current mean deviation threshold EPy W and output current variance deviation threshold EFy W For comparison:

[0092] When EP W >EPy W and EF W >EFy W When , it is determined that the DC servo motor has a temperature fault;

[0093] When EP W >EPy W and EF W >EFyW When at least one comparison formula is not true, it is not determined that the DC servo motor has a temperature fault.

[0094] Beneficial effects of the present invention:

[0095] Multi-signal acquisition: By collecting the input voltage signal, output current signal, motor speed signal and internal temperature signal of the DC servo motor driver, it is possible to fully obtain key information during the driver's operation and monitor the driver's operating status from multiple angles, avoiding missing potential faults due to the limitations of single signal monitoring.

[0096] Rich feature extraction: Feature extraction is performed on the collected data to obtain mean, variance, and kurtosis features. These features can more carefully depict the distribution and changes of the data, providing a more comprehensive and in-depth basis for accurately judging the status of the drive and helping to discover some relatively hidden fault signs.

[0097] Fault detection based on Euclidean distance: By calculating the Euclidean distance between the total set of normal state feature vectors and the total set of current time length feature vectors and comparing it with a preset threshold, the system effectively determines whether the drive is faulty. This method, based on data feature relationships, comprehensively considers the changes in multiple features, improving the accuracy and reliability of fault detection and reducing the possibility of misjudgments and missed detections.

[0098] Detailed Fault Analysis: Once a fault alarm signal is generated, the fault type is further analyzed by combining the normal state and the characteristic vector elements of the current time duration. Calculations and comparisons are performed on various aspects, such as input voltage, output current, motor speed, and temperature, to accurately determine the specific fault type, such as input voltage fault, output current fault, motor speed fault, or temperature fault. This helps maintenance personnel quickly locate the problem and implement targeted repair measures, improving maintenance efficiency and reducing maintenance costs and downtime.

[0099] Sound and light alarms: Sound and light alarms are combined with fault alarm signals to provide immediate and intuitive warnings to relevant personnel once a fault is detected. This intuitive early warning method ensures that personnel are immediately notified of drive faults, allowing them to respond quickly and take appropriate measures to address the fault, preventing further deterioration and the resulting greater impact on equipment and production.

[0100] A clear technical solution: This invention details a complete technical solution from data acquisition, feature extraction, data analysis, to fault warning. Each step is clearly defined and highly operational. Whether in industrial production or other scenarios involving DC servo motor drives, this fault detection method can be easily implemented to effectively ensure the normal operation of equipment.

[0101] In summary, the present invention provides a comprehensive, accurate and practical DC servo motor driver fault detection method, which has many beneficial effects and can significantly improve the ability to detect, diagnose and warn of DC servo motor driver faults, thereby ensuring the stable operation of related equipment and production processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0102] The present invention will be further described below with reference to the accompanying drawings.

[0103] Figure 1 The present invention is a system block diagram of a SEA DC servo motor driver fault detection method.

[0104] Figure 2 It is a flow chart of feature extraction processing in a SEA DC servo motor driver fault detection method of the present invention.

[0105] Figure 3 The present invention is a schematic flow chart of a fault type determination method for a SEA DC servo motor driver. DETAILED DESCRIPTION

[0106] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0107] Example 1

[0108] See also Figure 1 、 Figure 2 and Figure 3 As shown, the present invention is a SEA DC servo motor driver fault detection method, comprising the following steps:

[0109] Step 1: Data Collection

[0110] Collect the input voltage signal, output current signal, motor speed signal of the DC servo motor driver and the temperature signal inside the driver;

[0111] The sampling frequency is set to f, the acquisition time is T, and the number of data points collected is n, and n = f × T;

[0112] Step 2: Feature Extraction

[0113] Step K1: Mark the input voltage signal, output current signal, motor speed signal and temperature signal collected within the time length T as U i , L i , Z i and W i , i=1, 2, ... n;

[0114] Then U i , L i , Z i and W i As feature vectors, and combined into a feature vector set X i =[U i ,L i ,Z i ,W i ];

[0115] Step K2, by:

[0116]

[0117] Calculate the mean XP of the corresponding eigenvector in the eigenvector set, and XP = [UP, LP, ZP, WP];

[0118] Among them, UP is U i The corresponding input voltage mean, LP is L i The corresponding output current mean, ZP is Z i The corresponding motor speed average, WP is W i The corresponding temperature mean;

[0119] Step K3, by:

[0120]

[0121] Calculate the variance XF of the corresponding eigenvector in the eigenvector set, and XF = [UF, LF, ZF, WF];

[0122] Among them, UF is U i The corresponding input voltage variance, LF is L i The corresponding output current variance, ZF is Z i The corresponding motor speed variance, WF is W i The corresponding temperature variance;

[0123] Step K4, through:

[0124]

[0125] Calculate the kurtosis XD of the corresponding eigenvector in the eigenvector set, and XD = [UD, LD, ZD, WD];

[0126] Among them, UD is U i The corresponding input voltage kurtosis, LD is the output current kurtosis corresponding to Li, and ZD is the Z i The corresponding motor speed kurtosis, WD is W i The corresponding temperature kurtosis;

[0127] Step 3: Data Analysis

[0128] Step H1, according to the data collection and feature extraction steps, obtain the feature vector and its mean, variance and kurtosis of the DC servo motor driver in the normal state;

[0129] And combine them into the total set of normal state eigenvectors:

[0130] G0=[G0 j ]=[UP0, UF0, UD0, LF0, LD0, LP0, ZD0, ZF0, ZP0, WD0,

[0131] WF0, WP0];

[0132] Where, j = 1, 2, ..., 12;

[0133] and obtaining the characteristic vector of the DC servo motor driver at the current time length and its mean, variance and kurtosis;

[0134] And combine them into the total set of current time length feature vectors:

[0135] G1=[G1 j ]=[UP1, UF1, UD1, LF1, LD1, LP1, ZD1, ZF1, ZP1, WD1,

[0136] WF1, WP1];

[0137] in:

[0138] UP0, UF0, and UD0 are the mean, variance, and kurtosis of the input voltage under normal conditions, respectively;

[0139] UP1, UF1, and UD1 are the mean, variance, and kurtosis of the input voltage at the current time length, respectively;

[0140] LF0, LD0, and LP0 are the mean, variance, and kurtosis of the output current under normal conditions, respectively;

[0141] LF1, LD1, and LP1 are the mean, variance, and kurtosis of the output current at the current time length, respectively;

[0142] ZD0, ZF0, and ZP0 are the mean, variance, and kurtosis of the motor speed under normal conditions, respectively;

[0143] ZD1, ZF1, and ZP1 are the mean, variance, and kurtosis of the motor speed at the current time length, respectively;

[0144] WD0, WF0, and WP0 are the mean, variance, and kurtosis of the temperature under normal conditions, respectively;

[0145] WD1, WF1, and WP1 are the mean, variance, and kurtosis of the temperature at the current time length, respectively;

[0146] StepH2, then pass:

[0147]

[0148] Calculate the Euclidean distance Q between the total set of normal state feature vectors and the total set of current time length feature vectors d ;

[0149] StepH3, the Euclidean distance Q will be obtained d And the corresponding preset Euclidean distance threshold QY d For comparison:

[0150] If Q d >QY d , it is determined that the DC servo motor driver has a fault, and a fault alarm signal is generated;

[0151] If Q d ≤QY d , it is determined that the DC servo motor driver is in normal operation and no fault alarm signal is generated;

[0152] Step 4: Fault Warning

[0153] Sound and light alarm is performed by combining sound and light alarm with fault alarm signal.

[0154] This embodiment collects the input voltage signal, output current signal, motor speed signal, and internal temperature signal of the DC servo motor driver and performs feature extraction to obtain various feature information of the driver under different operating states, including mean, variance, and kurtosis, providing a rich data foundation for subsequent accurate fault detection. By calculating the Euclidean distance between the total set of normal state feature vectors and the total set of current time length feature vectors and comparing it with a preset Euclidean distance threshold, it is possible to effectively determine whether the DC servo motor driver has a fault. This method is relatively simple and intuitive, and can promptly detect whether the driver is in an abnormal operating state. When a fault occurs, a fault alarm signal can be quickly generated so that timely measures can be taken to reduce the losses that may result from the failure to detect the fault in time. After the fault is determined, an audible and visual alarm is issued in combination with the fault alarm signal, which can intuitively remind relevant personnel that the driver has a fault, allowing staff to know and deal with the problem immediately, thereby improving the timeliness of fault handling.

[0155] Example 2

[0156] As the second embodiment of the present invention, when this application is specifically implemented, compared with the first embodiment, the technical solution of this embodiment differs from that of the first embodiment only in that:

[0157] When a fault alarm signal is generated, the fault type determination analysis is also performed by combining the normal state feature vector total set with each feature vector element in the current time length feature vector total set;

[0158] The specific method is as follows:

[0159] Step Y1, extract the mean UP0 and variance UF0 of the input voltage under normal conditions, as well as the mean UP1 and variance UF1 of the input voltage under the current time length;

[0160] Then through:

[0161]

[0162] Calculate the deviation ratio EP of the input voltage mean respectively U The deviation ratio EF of the input voltage variance U ;

[0163] Then the deviation ratio EP of the input voltage mean is U The deviation ratio EF of the input voltage variance U The corresponding preset input voltage mean deviation threshold EPy U and input voltage variance deviation threshold EFy U For comparison:

[0164] When EPU >EPy U and EF U >EFy U When , it is determined that there is an input voltage fault in the DC servo motor;

[0165] When EP U >EPy U and EF U >EFy U If at least one comparison formula is not true, it is not determined that the DC servo motor has an input voltage fault;

[0166] In this example:

[0167] If UP1 is significantly higher than the normal average UP0 and UF1 is large, the input voltage may be too high and fluctuate greatly. Check whether the power supply is overvoltage or the voltage regulator is faulty.

[0168] If UP1 is significantly lower than the normal average UP0 and UF1 is large, the input power line may have poor contact, resulting in unstable voltage or too low input voltage. Check the line connection and the output voltage of the power supply equipment.

[0169] Step Y2: Extract the mean LP0 and variance LF0 of the output current under normal conditions, as well as the mean LP1 and variance LF1 of the output current under the current time length;

[0170] Then through:

[0171]

[0172] Calculate the deviation ratio EP of the output current mean respectively L The deviation ratio EF of the output current variance L ;

[0173] Then the deviation ratio EP of the output current mean is L The deviation ratio EF of the output current variance L The corresponding preset output current mean deviation threshold EPy L and output current variance deviation threshold EFy L For comparison:

[0174] When EP L >EPy L and EF L >EFy L When , it is determined that the DC servo motor has an output current fault;

[0175] When EP L >EPy L and EF L >EFyL If at least one comparison formula in the formula is not true, it is not determined that the DC servo motor has an output current fault;

[0176] In this example:

[0177] If LP1 is much larger than the normal average value LP0 and LF1 is large, it may be that the motor load has suddenly increased or the motor has stalled. Check whether the mechanical load part of the motor is stuck or overloaded.

[0178] If EF L If the value of LP1 is large and deviates from the normal average value of LP0, the driver's power transistor module may be partially damaged or its performance has degraded, resulting in unstable output current. The power transistor needs to be inspected and replaced.

[0179] Step Y3, extract the mean ZP0 and variance ZF0 of the motor speed under normal conditions, as well as the mean ZP1 and variance ZF1 of the motor speed under the current time length;

[0180] Then through:

[0181]

[0182] Calculate the deviation ratio EP of the mean motor speed respectively Z The deviation ratio EF of the motor speed variance Z ;

[0183] Then the deviation ratio EP of the motor speed mean is Z The deviation ratio EF of the motor speed variance Z Respectively correspond to the preset motor speed mean deviation threshold EPy Z and motor speed variance deviation threshold EFy Z For comparison:

[0184] When EP Z >EPy Z and EF Z >EFy Z When , it is determined that the DC servo motor has a motor speed fault;

[0185] When EP Z >EPy Z and EF Z >EFy Z If at least one comparison formula in is not true, it is not determined that the DC servo motor has a motor speed fault;

[0186] In this example:

[0187] If ZP1 is significantly lower than the normal average ZP0 and ZF1 is large, the motor's transmission components, such as loose belts or worn gears, may be causing the speed to become unstable and drop. Check the transmission connection between the motor and the load.

[0188] If EF Z If the value ZP1 is large and deviates greatly from the normal mean value ZP0, the motor's feedback device (such as an encoder) may be faulty, resulting in inaccurate speed measurement and abnormal control. Calibrate or replace the encoder.

[0189] Step Y4, extract the mean WP0 and variance WF0 of the temperature under normal conditions, as well as the mean WP1 and variance WF1 of the temperature under the current time length;

[0190] Then through:

[0191]

[0192] Calculate the deviation ratio EP of the temperature mean respectively W The deviation ratio EF from the temperature variance W ;

[0193] Then the deviation ratio EP of the temperature mean is W The deviation ratio EF from the temperature variance W The corresponding preset temperature mean deviation threshold EPy W and temperature variance deviation threshold EFy W For comparison:

[0194] When EP W >EPy W and EF W >EFy W When , it is determined that the DC servo motor has a temperature fault;

[0195] When EP W >EPy W and EF W >EFy W If at least one comparison formula is not true, it is not determined that the DC servo motor has a temperature fault;

[0196] In this example:

[0197] If WP1 is too high and WF1 is large, it may be that the internal heat dissipation of the drive is poor. Check whether the cooling fan is running normally and whether the heat sink is dusty.

[0198] If EF W If the value of WP1 is large and the deviation between WP1 and the normal average value WP0 is large, there may be a local short circuit or overcurrent inside the driver, which causes the temperature to rise unstable. The driver circuit needs to be tested in detail to find the potential electrical fault point.

[0199] Based on the first embodiment, when the fault alarm signal is generated, this embodiment further combines the normal state and the current time length of each characteristic vector element to perform fault type determination analysis. By calculating the deviation ratio of the mean and variance of different signals respectively and comparing them with the corresponding preset thresholds, it is possible to more accurately determine the specific type of fault, such as input voltage fault, output current fault, motor speed fault, temperature fault, etc., which helps to take targeted maintenance measures and improve maintenance efficiency. The possible causes of different types of faults are analyzed and prompted. For example, for input voltage faults, based on the input voltage mean and variance, it can be inferred that the possible causes are excessively high and fluctuating input voltage, poor contact of the input power line, etc., and the corresponding inspection direction is given; similar analysis and guidance are also provided for other types of faults, which enables maintenance personnel to carry out troubleshooting more targetedly when facing faults, reducing the time and workload of troubleshooting.

[0200] Example 3

[0201] As the third embodiment of the present invention, when this application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments.

[0202] This embodiment combines the solutions of Example 1 and Example 2. This method not only uses the method of Example 1 to quickly and accurately detect whether a DC servo motor driver has a fault and promptly issues an alarm, but also uses the method of Example 2 to further accurately determine the fault type and obtain preliminary troubleshooting guidance for the fault cause after the alarm is issued. By combining the advantages of both embodiments, the entire fault detection and diagnosis system is more complete and efficient, better ensuring the normal operation of the DC servo motor driver and reducing the various adverse effects caused by faults.

[0203] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0204] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A SEA DC servo motor driver fault detection method, characterized in that: The following steps are involved: Step 1: Data Collection Collect the input voltage signal, output current signal, motor speed signal of the DC servo motor driver and the temperature signal inside the driver; Step 2: Feature Extraction Extract the data collected within a preset time length, perform feature extraction processing on it, and obtain the corresponding mean feature, variance feature and kurtosis feature; Step 3: Data Analysis The data of the DC servo motor driver under normal conditions and the current time length are respectively acquired through the data acquisition step, and the data features of the DC servo motor driver under normal conditions and the current time length are subsequently extracted through the feature extraction step. Then, it is determined whether the DC servo motor driver has a fault based on the relationship between the data features under normal conditions and the current time length, and if the DC servo motor driver has a fault, a fault alarm signal is generated; Step 4: Fault Warning Sound and light alarm is performed by combining sound and light alarm with fault alarm signal.

2. A SEA DC servo motor driver fault detection method according to claim 1, characterized in that: The feature extraction process is as follows: Step K1. Vector marking The input voltage signal, output current signal, motor speed signal and temperature signal collected within the time length T are marked as U i 、L i 、Z i and W i , i=1, 2, ... n; Then U i 、L i 、Z i and W i As feature vectors, and combined into a feature vector set X i =[U i ,L i ,Z i ,W i ]; StepK2, vector mean calculation Calculate the mean of the corresponding eigenvectors in the eigenvector set and record it as XP; Step K3, vector variance calculation Calculate the variance of the corresponding eigenvector in the eigenvector set and record it as XF; Step K4, vector kurtosis calculation pass: Calculate the kurtosis XD of the corresponding eigenvector in the eigenvector set.

3. A SEA DC servo motor driver fault detection method according to claim 2, characterized in that: in, XF=[UF,LF,ZF,WF], XP=[UP,LP,ZP,WP], XD=[UD,LD,ZD,WD]; Among them, UF is U i The corresponding input voltage variance, LF is L i The corresponding output current variance, ZF is Z i The corresponding motor speed variance and WF are W i Corresponding temperature variance; UP is U i The corresponding input voltage mean, LP is L i The corresponding output current mean, ZP is Z i The corresponding motor speed mean and WP are W i The corresponding temperature mean; UD is U i The corresponding input voltage kurtosis, LD is L i The corresponding output current kurtosis, ZD is Z i The corresponding motor speed kurtosis and WD are W i The corresponding temperature kurtosis.

4. A SEA DC servo motor driver fault detection method according to claim 3, characterized in that: The data analysis method is as follows: Step H1, according to the data collection and feature extraction steps, obtain the feature vector and its mean, variance and kurtosis of the DC servo motor driver in the normal state; And combine them into the total set of normal state feature vectors: G0=[G0 j ]; Where, j = 1, 2, ..., 12; and obtaining the characteristic vector of the DC servo motor driver at the current time length and its mean, variance and kurtosis; And combine them into the total set of current time length feature vectors: G1=[G1 j ]; StepH2, then pass: Calculate the Euclidean distance Q between the total set of normal state feature vectors and the total set of current time length feature vectors d ; StepH3, the Euclidean distance Q will be obtained d And the corresponding preset Euclidean distance threshold QY d Compare and determine whether to generate a fault alarm signal based on the comparison result.

5. A SEA DC servo motor driver fault detection method according to claim 4, characterized in that: In StepH1: [G0 j ]=[UP0,UF0,UD0,LF0,LD0,LP0,ZD0,ZF0,ZP0,WD0,WF0,WP0]; [G1 j ]=[UP1,UF1,UD1,LF1,LD1,LP1,ZD1,ZF1,ZP1,WD1,WF1,WP1]; UP0, UF0, and UD0 are the mean, variance, and kurtosis of the input voltage under normal conditions, respectively; UP1, UF1, and UD1 are the mean, variance, and kurtosis of the input voltage at the current time length, respectively; LF0, LD0, and LP0 are the mean, variance, and kurtosis of the output current under normal conditions, respectively; LF1, LD1, and LP1 are the mean, variance, and kurtosis of the output current at the current time length, respectively; ZD0, ZF0, and ZP0 are the mean, variance, and kurtosis of the motor speed under normal conditions, respectively; ZD1, ZF1, and ZP1 are the mean, variance, and kurtosis of the motor speed at the current time length, respectively; WD0, WF0, and WP0 are the mean, variance, and kurtosis of the temperature under normal conditions, respectively; WD1, WF1, and WP1 are the mean, variance, and kurtosis of the temperature at the current time length, respectively.

6. A SEA DC servo motor driver fault detection method according to claim 4, characterized in that: In StepH3: If Q d >QY d , it is determined that the DC servo motor driver has a fault, and a fault alarm signal is generated; If Q d ≤QY d , it is determined that the DC servo motor driver is in normal operation and no fault alarm signal is generated.

7. A SEA DC servo motor driver fault detection method according to claim 6, characterized in that: When a fault alarm signal is generated, the fault type determination analysis is performed by combining the normal state feature vector total set with each feature vector element in the current time length feature vector total set.

8. A SEA DC servo motor driver fault detection method according to claim 7, characterized in that: The fault type determination and analysis method is as follows: Step Y1: Determine the voltage fault type Extract the mean UP0 and variance UF0 of the input voltage under normal conditions, as well as the mean UP1 and variance UF1 of the input voltage under the current time length; Then through: Calculate the deviation ratio EP of the input voltage mean respectively U The deviation ratio EF of the input voltage variance U ; Then the deviation ratio EP of the input voltage mean is U The deviation ratio EF of the input voltage variance U The corresponding preset output current mean deviation threshold EPy U and output current variance deviation threshold EFy U Perform comparison and determine whether the DC servo motor has an input voltage fault type based on the comparison result; Step Y2: Determine the current fault type Extract the mean LP0 and variance LF0 of the output current under normal conditions, as well as the mean LP1 and variance LF1 of the output current under the current time length; Then through: Calculate the deviation ratio EP of the output current mean respectively L The deviation ratio EF of the output current variance L ; Then the deviation ratio EP of the output current mean is L The deviation ratio EF of the output current variance L The corresponding preset output current mean deviation threshold EPy L and output current variance deviation threshold EFy L Perform comparison and determine whether the DC servo motor has an output current fault based on the comparison result; Step Y3: Determine the speed fault type Extract the mean ZP0 and variance ZF0 of the motor speed under normal conditions, as well as the mean ZP1 and variance ZF1 of the motor speed under the current time length; Then through: Calculate the deviation ratio EP of the motor speed mean value respectively Z The deviation ratio EF of the motor speed variance Z ; Then the deviation ratio EP of the motor speed mean is Z The deviation ratio EF of the motor speed variance Z The corresponding preset output current mean deviation threshold EPy Z and output current variance deviation threshold EFy Z Perform comparison and determine whether the DC servo motor has a motor speed fault type based on the comparison result; Step Y4: Determine the temperature fault type Extract the mean WP0 and variance WF0 of the temperature under normal conditions, as well as the mean WP1 and variance WF1 of the temperature under the current time length; Then through: Calculate the deviation ratio EP of the temperature mean respectively W The deviation ratio EF from the temperature variance W ; Then the deviation ratio EP of the temperature mean is W The deviation ratio EF from the temperature variance W The corresponding preset output current mean deviation threshold EPy W and output current variance deviation threshold EFy W Perform a comparison and determine whether the DC servo motor has a temperature fault based on the comparison result.

9. A SEA DC servo motor driver fault detection method according to claim 8, characterized in that: In StepY1: When EP U >EPy U and EF U >EFy U When , it is determined that there is an input voltage fault in the DC servo motor; When EP U >EPy U and EF U >EFy U If at least one comparison formula is not true, it is not determined that the DC servo motor has an input voltage fault; In StepY2: When EP L >EPy L and EF L >EFy L When , it is determined that the DC servo motor has an output current fault; When EP L >EPy L and EF L >EFy L If at least one comparison formula in the formula is not true, it is not determined that the DC servo motor has an output current fault; In Step Y3: When EP Z >EPy Z and EF Z >EFy Z When , it is determined that the DC servo motor has a motor speed fault; When EP Z >EPy Z and EF Z >EFy Z If at least one comparison formula in is not true, it is not determined that the DC servo motor has a motor speed fault; In Step Y4: When EP W >EPy W and EF W >EFy W When , it is determined that the DC servo motor has a temperature fault; When EP W >EPy W and EF W >EFy W When at least one comparison formula is not true, it is not determined that the DC servo motor has a temperature fault.

Images