Motor long-term cold state management method and system based on AI intelligent control

Abstract

The invention discloses a motor long-term cold state management method and system based on AI intelligent control. Comparing the obtained current static data of the motor with the original static data, if a motor winding is normal, triggering insulation detection, and if the motor winding is abnormal, outputting an alarm command; if insulation detection is triggered, detected motor insulation data is compared with a preset motor insulation threshold value, and if an insulation detection result is that insulation is qualified, a pulse starting command is output; if the insulation detection result is that the insulation is unqualified, outputting an alarm command; if the pulsation starting command is received, the dynamic data of the motor during operation are compared with the original dynamic data, whether the motor and the load thereof are in a normal state or an abnormal state is judged, and a pulsation stopping command or an alarm command is correspondingly output; when a pulsation stop command is received, the action time for maintaining the motor next time is intelligently decided. According to the invention, faults of the motor can be checked, and maintenance of the motor can be completed efficiently.

Description

TECHNICAL FIELD

[0001] The application relates to the field of operation management of electric motors, in particular to an AI intelligent control-based management method and system for long-term cold state of electric motors. BACKGROUND

[0002] At present, in recent years, with the rapid development of China's manufacturing industry, industrial enterprises are booming, municipal projects are increasing, and people's livelihood projects are penetrating into all parts of the country. In the above places, the application of electric motors is also increasing, and various use environments and use conditions exist. Among the many use environments, there is a kind of electric motor that is special, long-term cold state, occasional or intermittent use. According to the specific process requirements or use conditions, this kind of electric motor is not continuously operated or long-term operated, and some are long-term in corrosive environment or various liquids. Due to the long-term cold state of the electric motor, there are problems such as insulation decline due to environmental humidity, lubricating grease solidification and insulation aging. At the same time, the electric motor bearing or the load such as water pump, fan and speed reducer receives corrosion or dampness and appears rust and jamming and other phenomena, so the cold state starting is easy to cause mechanical jamming, winding breakdown and other faults, so that normal production cannot be realized and even damage and maintenance occur. In addition, the traditional manual operation depends on experience judgment, and has the defects of low efficiency and high risk. Therefore, in view of the above problems, an electric motor management method is particularly suitable for long-term cold state electric motor.

[0003] The application No. CN201811266088.6 provides a generator health management method and equipment. The generator health management method is used for real-time monitoring of the winding temperature, bearing temperature, insulation resistance and output voltage of the generator, and performing health management on the generator according to the monitoring result. The generator health management method mainly includes the steps of parameter presetting, data acquisition, data comparison and management indication. The generator health management method can realize real-time monitoring of the performance of the generator, and give corresponding health management indication according to the monitoring result, but focuses on solving the problem after finding it, and cannot be applied from the scientificity and implementability of fault prediction, prevention and maintenance. SUMMARY

[0004] The purpose of the application is to provide an AI intelligent control-based management method and system for long-term cold state of electric motors, which can automatically investigate electric motor faults, realize necessary maintenance of electric motors while preventing the occurrence of faults, reduce labor cost and improve economic benefit.

[0005] The application adopts the following technical scheme:

[0006] An AI intelligent control-based management method for long-term cold state of electric motors comprises the following steps:

[0007] S1: constructing an original database of the target motor, the original database including original static data and original dynamic data;

[0008] S2: acquiring current static data of the motor and comparing the current static data with the original static data to determine whether the motor winding is normal or abnormal;

[0009] S3: if the motor winding is normal, triggering insulation detection, comparing the detected motor insulation data with a preset motor insulation threshold in the original database, determining whether to perform insulation promotion operation, if insulation promotion is needed, outputting the insulation detection result after insulation promotion, if insulation promotion is not needed, outputting the insulation detection result, and if the motor winding is abnormal, outputting an alarm command;

[0010] S4: if the insulation detection result is insulation qualified, outputting a pulsation start command and comparing dynamic data of the motor during operation with the original dynamic data to determine whether the motor and its load are in normal state or abnormal state, and if the insulation detection result is insulation unqualified, outputting an alarm command;

[0011] S5: if the motor and its load are in normal state, outputting a pulsation stop command and intelligently determining the next maintenance action time, and if the motor and its load are in abnormal state, outputting an alarm command.

[0012] Further, the original static data in step S1 includes calibrated motor phase winding resistance, winding inductance, magnetic flux and winding-to-ground capacity resistance.

[0013] Further, in step S2, the current static data of the motor is obtained by applying a 5-30V DC voltage between any two windings of the motor and between each winding and ground.

[0014] Further, the motor insulation data includes motor internal rotor insulation value, winding insulation value and turn-to-turn insulation value.

[0015] Further, in step S3, the motor insulation data of the motor is compared with the preset motor insulation threshold to perform insulation detection, determine and output the insulation detection result, and the insulation detection process is as follows:

[0016] Performing first insulation detection, if the motor insulation data is all greater than the preset motor insulation threshold, determining that the insulation detection result is insulation qualified;

[0017] If any one of the motor insulation data is less than or equal to the preset motor insulation threshold value, the next insulation detection time is obtained according to the insulation detection interval; in the insulation detection process, when the number of times that the insulation deterioration acceleration is greater than the insulation deterioration acceleration threshold value is greater than the insulation detection threshold value, an insulation improvement command is output to start the motor heating device, and the current heating number is marked as 0.

[0018] Further, the insulation improvement operation includes the following processes:

[0019] The insulation improvement number threshold value is set, that is, when the insulation improvement number is greater than the insulation improvement number threshold value, the heating cycle is exited, the motor heating device is heated according to the preset heating condition, and after the heating is stopped, the insulation detection is performed again,

[0020] If the current obtained motor insulation data is greater than the preset motor insulation threshold value, a stop heating command is output and the insulation detection result is determined to be insulation qualified;

[0021] If any one of the current obtained motor insulation data is less than or equal to the preset motor insulation threshold value, it is judged whether the current heating number is greater than the insulation improvement number threshold value,

[0022] If the current heating number is less than or equal to the insulation improvement number threshold value, the current heating number is marked as one, the motor heating device is started again, and the next detection heating cycle is entered;

[0023] If the current heating number is greater than the insulation improvement number threshold value, the heating cycle is exited and the insulation detection result is determined to be insulation unqualified.

[0024] Further, the preset heating condition is that the heating device is heated by a heating coil pre-installed in the motor, the heating time is less than or equal to 30 minutes each time, the heating temperature is less than or equal to 85 degrees, and the temperature sensor installed in the motor is used for real-time temperature measurement.

[0025] Further, the judgment logic for judging whether the motor and its load are in a normal state or an abnormal state in step S4 is as follows:

[0026] In each motor and load state judgment, a plurality of groups of real-time dynamic data of the motor are continuously collected to obtain a corresponding number of dynamic data detection samples; in the plurality of groups of dynamic data detection samples:

[0027] If the number of dynamic data detection samples in which the real-time dynamic data is completely consistent with the original dynamic data is greater than the set state judgment number threshold value, and the total number of times of motor and load state result judgment is less than the set retry detection threshold value, the state result is output as the motor and its load being in a normal state;

[0028] If the number of dynamic data detection samples that are completely consistent with the real-time dynamic data and the original dynamic data is less than or equal to the set state judgment number threshold, and the total number of motor and load state result judgments is less than the set retry detection threshold, the motor and load state result judgment is performed again.

[0029] If the number of dynamic data detection samples that are completely consistent with the real-time dynamic data and the original dynamic data is less than or equal to the set state judgment number threshold, and the total number of motor and load state result judgments is greater than or equal to the set retry detection threshold, an alarm command is output.

[0030] Further, in step S5, the next maintenance action time is intelligently decided according to the motor correction coefficient, the time average of historical maintenance, the comprehensive health degree variance at the sampling time, the optimal state reference vector, and the motor current state feature vector.

[0031] An AI intelligent control-based motor long-term cold state management system, comprising a data acquisition unit, a data storage unit, a data operation unit, a communication interface unit, and an execution unit; wherein,

[0032] The data acquisition unit is configured to acquire the current static data, the current dynamic data, and the insulation detection data of the motor.

[0033] The data storage unit is configured to store an original database of the target motor, the original database including original static data and original dynamic data, and to further store the acquired current static data, current dynamic data, and insulation detection data of the motor.

[0034] The data operation unit is configured to compare the acquired current static data of the motor with the original static data to determine whether the motor winding is normal or abnormal, to perform insulation detection determination on the acquired motor insulation data and a preset motor insulation threshold value and output an insulation detection result, and to correspondingly output an insulation improvement command, a pulsation start command, and an alarm command; the data operation unit is further configured to compare the received dynamic data of the motor during operation with the original dynamic data, and to determine whether the motor and the load are in a normal state or an abnormal state according to a load detection logic, and to correspondingly output a pulsation stop command or an alarm command.

[0035] The communication interface unit is configured to perform network communication and data processing externally, and is connected to the execution unit and the data operation unit.

[0036] The execution unit is configured to perform torque pulsation if the pulsation start command is received, to perform alarm processing or output a warning information if the alarm command is received, to start the motor heating device and enter an insulation detection heating cycle if the insulation improvement command is received, and to mark the current heating number as 0, and to control the pulsation process to end if the pulsation stop command is received.

[0037] The weak current generated by the low-frequency pulse in the application can generate a small amount of Joule heat in the winding, which can gradually dissipate the internal moisture, reduce the moisture in the motor, and indirectly improve the insulation performance. In addition, through slight friction, the lubricating grease in the motor is redistributed and activated, avoiding mechanical damage caused by jamming during starting. It can ensure that the motor system in long-term cold state is always in good state, ensure that the motor is always in a startable state, improve the stable operation rate of the equipment, prolong the maintenance and scrap cycle of the equipment, reduce the maintenance and depreciation cost, thereby reducing maintenance and maintenance, reducing maintenance cost, artificial maintenance cost and equipment replacement cost, and improving economic benefits. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings needed to be used in the embodiment or related art description. Obviously, the drawings in the following description are only part of the embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of the provided drawings.

[0039] Figure 1 The management method flow chart of the long-term cold state of the motor based on AI intelligent control provided by the present application is provided.

[0040] Figure 2 The management system schematic diagram of the long-term cold state of the motor based on AI intelligent control provided by the present application is provided. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical scheme and advantages of the embodiments herein clearer, the technical scheme in the embodiments herein will be described clearly and completely in the following with reference to the drawings in the embodiments herein. Obviously, the described embodiments are part of the embodiments herein, not all. Based on the embodiments herein, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection. It should be noted that the embodiments herein and the features in the embodiments can be combined with each other arbitrarily without conflict.

[0042] The application will be described in detail below with reference to the drawings and embodiments:

[0043] As shown in the drawings, Figure 1 A specific embodiment provided by the present application is a management method of long-term cold state of motor based on AI intelligent control, which includes the following steps:

[0044] S1: Constructing the original database of the target motor, wherein the original database includes original static data and original dynamic data.

[0045] The original database of the target motor includes original static data and original dynamic data; the original static data includes the resistance value, inductance, magnetic flux and ground capacitance of each phase winding of the motor at the time of manufacturing qualification detection; the original dynamic data includes the dynamically changed voltage, current, magnetic flux, torque, temperature rise and vibration data of the motor at the time of normal operation at a preset frequency.

[0046] The original dynamic data includes the voltage, current, magnetic flux, torque, temperature rise and vibration data of the motor at the time of normal operation at a preset frequency of 0HZ, 0.5HZ, 1.0HZ, 2HZ and 5HZ,

[0047] S2: Obtain the current static data of the motor, and compare the current static data with the original static data to determine whether the motor winding is normal.

[0048] The current static data of the motor is obtained by applying a direct current voltage of 5-30V between any two windings of the motor and between each winding and the ground wire, and the resistance value, inductance and magnetic flux of the winding are obtained according to Ohm's law I=U / R, the inductance voltage and the current change rate L=Ψ / I; wherein I is the current, U is the voltage, R is the resistance value; V represents the self-induced electromotive force, L is the self-induction coefficient, dI is the change amount of the current, dt is the time of the current change, and Ψ represents the magnetic flux.

[0049] In this embodiment, the time for applying voltage between any two windings of the motor and between each winding and the ground wire is 60 seconds; according to the manufacturing principle of the motor and the voltage and current characteristic curve, applying a voltage of 30V or less on the two windings for 60 seconds will not cause any damage to the motor, and can also reduce the humidity inside the motor and improve the insulation inside the motor.

[0050] If it is detected that the resistance value of a certain phase winding is significantly higher than the preset standard value, it may be that the winding is partially open-circuited, the cross-sectional area of the wire is reduced, indicating that the motor winding has physical damage.

[0051] S3: If the motor winding is normal, triggering insulation detection, comparing the detected motor insulation data with the preset motor insulation threshold in the original database, and simultaneously determining whether to perform insulation improvement, if insulation improvement is required, outputting the insulation detection result after insulation improvement, if insulation improvement is not required, outputting the insulation detection result; if the motor winding is abnormal, outputting an alarm command.

[0052] The data operation unit is used for performing insulation detection on the internal rotor, winding and turn-to-turn of the target motor, and the data acquisition unit is used for obtaining motor insulation data including the internal rotor insulation value, winding insulation value and turn-to-turn insulation value of the motor.

[0053] The maximum period of insulation detection is set, if the insulation detection is triggered, the detected motor insulation data is compared with the preset motor insulation threshold in the original database respectively, and the insulation detection result is output; the insulation detection process is as follows:

[0054] The first insulation detection is performed, if the motor insulation data is greater than the preset motor insulation threshold, the insulation detection result is determined to be insulation qualified;

[0055] If one or more insulation values in the motor insulation data is less than or equal to the preset motor insulation threshold, the next insulation detection time is obtained according to the insulation detection interval; if the number of times that the insulation deterioration acceleration is greater than the insulation deterioration acceleration threshold ε is greater than the insulation detection threshold during the insulation detection process, the data operation unit outputs the insulation promotion command to start the motor heating device, enters the insulation promotion operation, and marks the current heating number as 0.

[0056] Each group of indicators in the insulation detection data is assigned a risk weight for weighted integration, and the insulation detection interval is calculated by the integrated insulation detection value after integration. detect (n) with insulation deterioration acceleration decreases, and their relationship is expressed as follows:

[0057]

[0058] wherein, T detect (n) represents the insulation detection interval calculated for the nth time; T detect (n-1) represents the detection interval actually executed last time; represents the insulation deterioration acceleration, ΔR ins represents the change amount of insulation resistance, represents the derivative of ΔR ins with respect to time; ε represents a comprehensive normalization coefficient, which is an insulation deterioration acceleration threshold, when the acceleration reaches ε / d 2 , the detection interval is minimum; d represents the minimum value of the insulation detection interval.

[0059] In actual application scenarios, sudden changes in environmental humidity may cause temporary decrease in insulation resistance, such as sensor moisture in rainy days, but it will automatically recover after the sun comes out. If the heating and dehumidification is started as soon as the single standard is exceeded, the insulation may be damaged due to thermal stress, and the insulation promotion needs to inject direct current heating, frequent operation not only consumes energy but also accelerates material aging, therefore, the insulation detection threshold is set, when the number of times that the insulation deterioration acceleration is greater than the insulation deterioration acceleration threshold ε is greater than the insulation detection threshold, the motor heating device is started, and the insulation promotion operation is entered.

[0060] The insulation promotion times threshold is further provided in the application, and the heating cycle is exited when the insulation promotion times is greater than the insulation promotion times threshold, and the insulation promotion operation is performed as follows:

[0061] The motor heating device heats according to the preset heating condition, and the insulation detection is performed again after the heating is stopped:

[0062] If the current obtained motor insulation data is all greater than the preset motor insulation threshold, a stop heating command is outputted, and the insulation detection result is determined as insulation qualified;

[0063] If the current obtained motor insulation data is less than or equal to the preset motor insulation threshold, whether the current heating times is greater than the insulation promotion times threshold is judged;

[0064] If the current heating times is less than or equal to the insulation promotion times threshold, the current heating times is marked as one more, the motor heating device is started again, and the next detection heating cycle is entered;

[0065] If the current heating times is greater than the insulation promotion times threshold, the heating cycle is exited, and the insulation detection result is determined as insulation unqualified.

[0066] According to an example embodiment, the preset heating condition is that the heating device performs low-voltage and small-current heating through the heating coil pre-installed in the motor, each heating does not exceed 30 minutes, and the temperature does not exceed 85 degrees, and the temperature is measured and controlled through the PT100 or thermocouple pre-installed in the motor; and the preset motor insulation threshold is set as 2 mega ohms.

[0067] S4: If the insulation detection result is insulation qualified, a pulsation start command is outputted, and the dynamic data of the motor during the operation is compared with the original dynamic data to judge whether the motor and the load are in normal state or abnormal state; if the insulation detection result is insulation unqualified, an alarm command is outputted.

[0068] According to an example embodiment, if the insulation detection result is insulation qualified, a pulsation start command is outputted to the frequency converter of the motor, the frequency converter performs torque pulsation on the motor, and during the pulsation control process, 0-speed servo control and direct current braking control are performed in sequence, the output frequency of the frequency converter or servo driver is 0 Hz, that is, the motor theoretically does not rotate, but through closed-loop feedback control, the stationary state of the motor shaft is maintained, and the output torque is adjusted in real time to resist the control mode of external disturbance; the sending period of the pulsation signal is 0.5 seconds of sending and 1 second of stopping, 100 periods are continuously sent each time, and when the pulsation signal is sent, the torque boosting function of the frequency converter can be enabled, and the torque boosting amount is 10%-20%.

[0069] The real-time dynamic data includes voltage, current, magnetic flux, torque, temperature rise and vibration data of the motor when the motor is normally running at a preset frequency of 0 Hz, 0.5 Hz, 1.0 Hz, 2 Hz and 5 Hz. The above data is obtained by the frequency converter vector control and dynamic data link system at each pulse of the motor.

[0070] The torque calculation formula is T = CT * Φ * Ia, wherein CT is a torque constant, Φ is a main magnetic flux per pole, and Ia is an armature current.

[0071] In the motor and load state judgment, since the collection of a sample for state judgment has contingency, a plurality of groups of real-time dynamic data of the motor are continuously collected to obtain a corresponding number of dynamic data detection samples, and the motor and load state are determined by a plurality of groups of dynamic data detection samples. A state determination quantity threshold is set, the state determination quantity threshold represents the minimum value of the number of dynamic data detection samples for determining that the motor and load are in a normal state in a plurality of groups of detection samples for each state result determination, the number of motor and load state result determinations is 0, and the number of motor and load state result determinations is increased by one each time the state result determination is performed. In a plurality of groups of dynamic data detection samples:

[0072] If the number of dynamic data detection samples in which the real-time dynamic data is completely consistent with the original dynamic data is greater than the set state determination quantity threshold, and the total number of motor and load state result determinations is less than the set retry detection threshold, the state result is output as the motor and load being in a normal state;

[0073] If the number of dynamic data detection samples in which the real-time dynamic data is completely consistent with the original dynamic data is less than or equal to the set state determination quantity threshold, and the total number of motor and load state result determinations is less than the set retry detection threshold, the motor and load state result determination is performed again;

[0074] If the number of dynamic data detection samples in which the real-time dynamic data is completely consistent with the original dynamic data is less than or equal to the set state determination quantity threshold, and the total number of motor and load state result determinations is greater than or equal to the set retry detection threshold, an alarm command is output.

[0075] According to an exemplary embodiment, the number of times of continuously collecting real-time dynamic data of the motor is 100, the state determination quantity threshold is 90, and the retry detection threshold is 2.

[0076] A plurality of groups of real-time dynamic data of the motor are continuously collected to obtain a corresponding number of dynamic data detection samples; in the plurality of groups of dynamic data detection samples:

[0077] If the number of dynamic data detection samples that are completely consistent between the real-time dynamic data and the original dynamic data is greater than the set state judgment number threshold 90, and the total number of motor and load state result judgments is less than or equal to the set retry detection threshold 2, the motor and load are output as being in a normal state.

[0078] If the number of dynamic data detection samples that are completely consistent between the real-time dynamic data and the original dynamic data is less than or equal to the set state judgment number threshold 90, and the total number of motor and load state result judgments is less than the set retry detection threshold 2, the motor and load state result judgment is performed again.

[0079] If the number of dynamic data detection samples that are completely consistent between the real-time dynamic data and the original dynamic data is less than or equal to the set state judgment number threshold 90, and the total number of motor and load state result judgments is less than the set retry detection threshold 2, the motor and load state result judgment is performed again.

[0080] In the embodiment, the above operation allows the rotor to rotate slowly at a speed close to peristalsis, without causing obvious mechanical impact, and enables the lubricating grease to be redistributed and activated through slight friction, thereby avoiding rusting of the components or mechanical damage caused by jamming during startup. In addition, in a long-term cold state of the motor, water vapor is easily condensed in the motor due to air humidity, which causes the winding insulation resistance to decrease. The weak current generated by the pulsation signal produces a small amount of Joule heat (controllable at low frequency), and the 0.5 Hz and 1 Hz pulsation signals are each sent 100 times, with a "breathing" time of 0.5 seconds of operation and 1 second of stop. The internal moisture is gradually dispersed, and the insulation performance is indirectly improved. At the same time, the alternating magnetic field formed by the low-frequency current can reduce the local charge accumulation that may occur in the winding due to long-term static, stabilize the electrical parameters such as resistance and inductance of the winding, and avoid current surges caused by abnormal parameters during startup. The 10%-20% torque boosting function is enabled to ensure that the motor can overcome the static friction and reliably rotate at a low frequency and low speed, thereby avoiding "pulsation invalidity". The multiple actions of 100 cycles can ensure the stability of the motor's conversion capability from static to dynamic response through repeated verification, thereby reducing the risk of subsequent failures (such as startup jamming and overload tripping) during formal startup. The entire process is small in motion and low in energy consumption, but can effectively solve the problem of damaged long-term static motor.

[0081] In the embodiment, if the insulation detection result is unqualified, an alarm command is output, and the alarm or early warning process is performed through the setting of warning lights, built-in buzzers, and / or visual panels.

[0082] According to an exemplary embodiment, the frequency converter uploads the collected dynamic data of actual operation of the motor to the data operation unit through communication; the communication can be through serial communication such as RS485, Ethernet such as Modbus-TCP or based on CAN bus protocol.

[0083] S5: If the motor and its load are in a normal state, output a pulsation stop command and intelligently determine the next maintenance action time; if the motor and its load are in an abnormal state, output an alarm command.

[0084] If the motor and its load are in a normal state, output a pulsation stop command, the pulsation stop command controls the end of pulsation, and the system intelligently determines the next maintenance action time T next The calculation formula is as follows:

[0085]

[0086] Wherein, η represents the motor correction coefficient; T history represents the preset time average value of historical maintenance; is the variance of the comprehensive health degree at the current sampling time; X current represents the current state feature vector of the motor; X ref represents the optimal state reference vector; ||| 2 represents the square of the Euclidean norm.

[0087] In this embodiment, the output alarm command can be alarmed or warned through setting warning lights, built-in buzzers and / or visual panels.

[0088] As shown in the accompanying Figure 2 As another specific embodiment provided by the present application, the long-term cold-state management system of the motor based on AI intelligent control is applicable to the long-term cold-state management method of the motor based on AI intelligent control; comprising a data acquisition unit, a data storage unit, a data operation unit, a communication interface unit and an execution unit; wherein,

[0089] The data acquisition unit is used to collect the current static data, current dynamic data and insulation detection data of the motor;

[0090] The data storage unit is used to store the original database of the target motor, the original database including original static data and original dynamic data, and also store the collected current static data, current dynamic data and insulation detection data of the motor;

[0091] The data operation unit is used for comparing the current static data of the motor obtained with the original static data, judging whether the motor winding is normal or abnormal; is used for performing insulation detection judgment on the motor insulation data obtained and the preset motor insulation threshold value, and outputting the insulation detection result, corresponding to outputting an insulation improvement command, a pulse start command and an alarm command; is also used for comparing the dynamic data received when the motor is running with the original dynamic data, and judging whether the motor and its load are in a normal state or an abnormal state according to a load detection logic, corresponding to outputting a pulse stop command or an alarm command;

[0092] The communication interface unit is used for external network communication and data processing, connecting the execution unit and the data operation unit;

[0093] In the embodiment, the long-term cold-state management system of the motor based on AI intelligent control is connected with the production system of the whole workshop, the MES system of the whole factory and the production database through the communication interface unit, so that the whole production and equipment management are in an unmanned, digitized and intelligent state.

[0094] The execution unit performs torque pulsation if the pulse start command is received, performs alarm processing or outputs a warning information if the alarm command is received, starts the motor heating device if the insulation improvement command is received, enters the insulation improvement operation, and marks the current heating number as 0, and controls the end of the pulse process if the pulse stop command is received.

[0095] In the embodiment, the frequency converter in the motor receives the pulse start command to perform torque pulsation as the execution unit, and can perform alarm or warning processing through setting of an alarm lamp, a built-in buzzer and / or a visual panel.

[0096] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A method for managing a motor in a long-term cold state based on AI intelligent control, characterized in that: The method comprises the following steps: S1: constructing an original database of the target motor, the original database comprising original static data and original dynamic data; S2: acquiring current static data of the motor and comparing the current static data with the original static data to determine whether the motor winding is normal or abnormal; S3: if the motor winding is normal, triggering insulation detection, comparing the detected motor insulation data with a preset motor insulation threshold in the original database, and determining whether insulation improvement operation is needed, if insulation improvement is needed, outputting the insulation detection result after insulation improvement, if insulation improvement is not needed, outputting the insulation detection result, and if the motor winding is abnormal, outputting an alarm command; S4: if the insulation detection result is insulation qualified, outputting a pulsation start command and comparing dynamic data of the motor during operation with the original dynamic data to determine whether the motor and its load are in a normal state or an abnormal state, if the insulation detection result is insulation unqualified, outputting an alarm command; S5: if the motor and its load are in a normal state, outputting a pulsation stop command and intelligently determining the next maintenance action time, if the motor and its load are in an abnormal state, outputting an alarm command.

2. The method of claim 1, wherein the method is based on AI intelligent control of a long-term cold state of the motor. The original static data in step S1 comprises calibrated motor phase winding resistance, winding inductance, magnetic flux and winding-to-ground capacity resistance. 3.The method of claim 1, wherein the method further comprises: determining a temperature of the motor; and determining whether the temperature of the motor is less than a threshold temperature. In step S2, the current static data of the motor is obtained by applying a 5-30V DC voltage between any two windings of the motor and between each winding and the ground wire. 4.The method of claim 1, wherein the method further comprises: determining a temperature of the motor; and determining whether the temperature of the motor is less than a threshold temperature. The motor insulation data comprises motor internal rotor insulation value, winding insulation value and turn-to-turn insulation value.

5. The method of claim 4, wherein the method is based on AI intelligent control of a long-term cold state of the motor. In step S3, the motor insulation data of the motor is compared with the preset motor insulation threshold for insulation detection, and the insulation detection result is determined and output, and the insulation detection process is as follows: For the first insulation detection, if all the motor insulation data are greater than the preset motor insulation threshold, the insulation detection result is determined to be insulation qualified; If any one of the motor insulation data is less than or equal to the preset motor insulation threshold, the next insulation detection time is obtained according to the insulation detection interval, and when the number of times that the insulation deterioration acceleration is greater than the insulation deterioration acceleration threshold is greater than the insulation detection threshold during the insulation detection process, an insulation improvement command is output to start the motor heating device, the insulation improvement operation is entered, and the current heating number is marked as 0.

6. The method of claim 5, wherein the method is based on AI intelligent control of a long-term cold state of the motor. The insulation improvement operation comprises the following processes: setting an insulation improvement number threshold, i.e. when the insulation improvement number is greater than the insulation improvement number threshold, the heating cycle is exited, the motor heating device heats according to the preset heating condition, and after the heating is stopped, the insulation detection is performed again, if all the current obtained motor insulation data are greater than the preset motor insulation threshold, the heating is stopped and the insulation detection result is determined to be insulation qualified, if any one of the current obtained motor insulation data is less than or equal to the preset motor insulation threshold, it is determined whether the current heating number is greater than the insulation improvement number threshold, If the current heating number is less than or equal to the insulation promotion number threshold, the current heating number is marked as one plus, the motor heating device is started again, and the next detection heating cycle is entered; If the current heating number is greater than the insulation promotion number threshold, the heating cycle is exited, and the insulation detection result is determined as unqualified insulation.

7. The method of claim 6, wherein the method is based on AI intelligent control of a long-term cold state of the motor. The preset heating condition is that the heating device heats through a heating coil pre-installed in the motor, each heating time is less than or equal to 30 minutes, the heating temperature is less than or equal to 85 degrees, and real-time temperature measurement is performed through a temperature sensor installed in the motor.

8. The method of claim 1, wherein the method is based on AI intelligent control of a long-term cold state of the motor. The judgment logic for determining whether the motor and its load are in a normal state or an abnormal state in step S4 is as follows: In each motor and load state judgment, a plurality of groups of real-time dynamic data of the motor are continuously collected to obtain a corresponding number of dynamic data detection samples; in the plurality of groups of dynamic data detection samples: If the number of dynamic data detection samples in which the real-time dynamic data is completely consistent with the original dynamic data is greater than the set state judgment number threshold, and the total number of motor and load state result judgments is less than the set retry detection threshold, the state result is output as the motor and its load being in a normal state; If the number of dynamic data detection samples in which the real-time dynamic data is completely consistent with the original dynamic data is less than or equal to the set state judgment number threshold, and the total number of motor and load state result judgments is less than the set retry detection threshold, the motor and load state result judgment is performed again; If the number of dynamic data detection samples in which the real-time dynamic data is completely consistent with the original dynamic data is less than or equal to the set state judgment number threshold, and the total number of motor and load state result judgments is greater than or equal to the set retry detection threshold, an alarm command is output. 9.The method of claim 1, wherein the method further comprises: determining a temperature of the motor; and determining whether the temperature of the motor is less than a threshold temperature. In step S5, the next maintenance action time is intelligently determined according to the motor correction coefficient, the time average of historical maintenance, the comprehensive health degree variance of the sampling time, the optimal state reference vector, and the motor current state feature vector.

10. An AI intelligent control-based motor long-term cold-state management system suitable for the management method of any one of claims 1-9, characterized in that, It comprises: a data acquisition unit, a data storage unit, a data operation unit, a communication interface unit, and an execution unit; wherein the data acquisition unit is used to collect the current static data, the current dynamic data, and the insulation detection data of the motor; the data storage unit is used to store the original database of the target motor, the original database including the original static data and the original dynamic data, and also store the collected current static data, the current dynamic data, and the insulation detection data of the motor; the data operation unit is used to compare the current static data of the motor with the original static data to determine whether the motor winding is normal or abnormal; is used to perform insulation detection judgment on the collected motor insulation data and the preset motor insulation threshold and output the insulation detection result, corresponding to outputting the insulation promotion command, the pulsation start command, and the alarm command; and is also used to compare the received dynamic data of the motor during operation with the original dynamic data, and determine whether the motor and its load are in a normal state or an abnormal state according to the load detection logic, corresponding to outputting the pulsation stop command or the alarm command; Communication interface unit: for external network communication and data processing, connecting execution unit and data operation unit; Execution unit: if receiving pulsation start command, it carries out torque pulsation; if receiving alarm command, it carries out alarm processing or outputs early warning information; If receiving insulation promotion command, it starts motor heating device, enters insulation detection heating cycle, and marks the current heating number as 0; if receiving pulsation stop command, it controls the end of pulsation process.

Images