CN114881068A - Novel leakage-proof signal acquisition method for rotary machine - Google Patents

Abstract

The invention relates to a novel leak-proof sampling signal acquisition method for rotating machinery, which includes three kinds of leak-proof sampling logics, which are a fault acquisition mode, an abnormal working condition acquisition mode, a timing acquisition mode, and The signal sampling frequency is controllable and sampling is carried out in real time; the present invention establishes three acquisition modes by triggering acquisition of signals through various index parameters, namely, fault acquisition mode, abnormal working condition acquisition mode, and timing acquisition mode; The parameter builds an evaluation index library. The fault acquisition mode adopts 7 dimensional indicators and 7 non-dimensional indicators, and formulates a variety of algorithm judgment strategies according to the actual working conditions, and formulates the optimal acquisition strategy to achieve accurate acquisition of signals at the time of failure. Avoid the missed sampling of faulty signals to the greatest extent.

Description

A Novel Leak-proof Signal Acquisition Method for Rotating Machinery

technical field

The invention relates to the field of signal collection, in particular to a novel leak-proof signal collection method for rotating machinery.

Background technique

At present, the signal acquisition for mechanical structures usually adopts the strategy of continuous acquisition or fixed-point acquisition; continuous acquisition means continuous acquisition of vibration data, and fixed-point acquisition means acquisition at fixed time intervals; this type of acquisition strategy is relatively rigid, and the amount of data collected It is relatively large but contains a low amount of effective information, and it is easy to miss the data at the time of failure. It is not suitable for rotating machinery. In order to solve the above problems, practitioners in the industrial field have used vibration signal indicators as the standard to trigger signal acquisition, but the indicators used are generalized. The ability is poor, the number of indicators is small, and the trigger acquisition effect is poor, and it cannot be applied to multiple industrial scenarios.

SUMMARY OF THE INVENTION

The purpose of the present invention is to propose a novel leak-proof sampling signal acquisition method for rotating machinery to solve the above problems.

The present invention achieves the above objects through the following technical solutions: a novel leak-proof sampling signal acquisition method for rotating machinery, which is characterized in that it includes three kinds of leak-proof sampling logics, which are the fault sampling modes from high to low according to priority ordering , the abnormal working condition collection mode, the timing collection mode, and the signal sampling frequency is controllable, and the sampling is performed in real time. The specific collection method is as follows:

(1) Under the normal operation of the machine, according to the model and rated working condition of the motor, set the timing sampling time and collection time, and collect the signals regularly for storage and diagnosis;

(2) When the mechanical operating condition information exceeds 5% of the specified operating amount, it proves that the machinery is under abnormal operating conditions at this time, and the signal is continuously collected at this time;

(3) When the machine is running under the rated working condition, but the frequency domain index exceeds the standard during real-time operation, it proves that the machine may have a fault at this time, so the fault collection mode is activated to continuously collect signals.

As a preference of the present invention, the abnormal working condition collection mode mainly starts by comparing the working condition information and the rated standard value, and the specific logic is as follows:

(1) The working condition information is mainly based on speed, temperature, noise, and current, and its importance decreases in order;

(2) According to the motor factory manual and the experience of relevant experts, obtain the specific information of each operating condition of the motor;

(3) After setting the operating condition parameters, monitor the operating condition information of the motor in real time. As long as one of the operating conditions information exceeds 5% of the specified operating amount, it means that the machine is in an abnormal operating condition. The signal is continuously acquired.

As a preferred aspect of the present invention, the fault collection mode is implemented by an index feature method, and the fault collection mode adopts 7 dimensional indicators and 7 dimensionless indicators to form an evaluation index library.

As a preference of the present invention, the acquisition module of the fault acquisition mode collects in real time during operation, and the collected signal enters the algorithm module for evaluation, and its judgment logic is as follows:

(1) If the number of dimensional indicators and non-dimensional indicators exceeds the early warning threshold by less than 3, it proves that the mechanical structure is running well at this time, and signal acquisition is not required;

(2) If there are more than 3 dimensional indicators that exceed the warning threshold, but less than 3 non-dimensional indicators exceed the warning threshold, then the signal may be abnormal due to working conditions and other reasons, such as mechanical collision, There is no need to collect signals, but the current time stamp needs to be recorded;

(3) If there are no more than 3 dimensional indicators that exceed the warning threshold, but more than 3 non-dimensional indicators exceed the warning threshold, it means that the overall abnormality of the mechanical structure may occur at this time, and the vibration signal for 3 minutes is collected. Then stop the collection;

(4) If the number of dimensional indicators and non-dimensional indicators exceeding the warning threshold is greater than 3, it proves that failures may occur during the operation of the mechanical structure at this time, and continuous data collection is carried out.

Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention establishes three acquisition modes, namely, fault acquisition mode, abnormal working condition acquisition mode, and timing acquisition mode by triggering acquisition of signals through various index parameters; Different types of index parameters are used to build an evaluation index library. The fault acquisition mode adopts 7 dimensional indicators and 7 non-dimensional indicators, and formulates a variety of algorithm judgment strategies according to the actual working conditions, and formulates the optimal acquisition strategy to realize the fault time signal. Accurate collection of faulty signals to the greatest extent possible.

Description of drawings

Fig. 1 is the algorithm flow chart of the embodiment of the present invention;

Fig. 2 is the experimental motor rotational speed variation diagram of the embodiment of the present invention;

FIG. 3 is a signal acquisition vibration waveform diagram according to an embodiment of the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings:

A new leak-proof sampling signal acquisition method for rotating machinery, as shown in Figure 1-3, includes three leak-proof sampling logics. According to the priority order from high to low, the fault acquisition mode, the abnormal working condition acquisition mode, Timing acquisition mode, and the signal sampling frequency is controllable, real-time sampling, the specific acquisition method is as follows:

(1) Under the normal operation of the machine, according to the model and rated working condition of the motor, set the timing sampling time and collection time, and collect the signals regularly for storage and diagnosis;

(2) When the mechanical operating condition information exceeds 5% of the specified operating amount, it proves that the machinery is under abnormal operating conditions at this time, and the signal is continuously collected at this time;

(3) When the machine is running under the rated working condition, but the frequency domain index exceeds the standard during real-time operation, it proves that the machine may have a fault at this time, so the fault collection mode is activated to continuously collect signals.

Further, the abnormal working condition collection mode mainly starts by comparing the working condition information and the rated standard value, and the specific logic is as follows:

(1) The working condition information is mainly based on speed, temperature, noise, and current, and its importance decreases in order;

(2) According to the motor factory manual and the experience of relevant experts, obtain the specific information of each operating condition of the motor;

(3) After setting the operating condition parameters, monitor the operating condition information of the motor in real time. As long as one of the operating conditions information exceeds 5% of the specified operating amount, it means that the machine is in an abnormal operating condition. The signal is continuously acquired.

Further, the fault collection mode is implemented by an index feature method, and the fault collection mode adopts 7 kinds of dimensional indicators and 7 kinds of dimensionless indicators to form an evaluation index library.

Further, the collection module of the fault collection mode collects in real time during operation, and the collected signal enters the algorithm module for evaluation, and its judgment logic is as follows:

(1) If the number of dimensional indicators and non-dimensional indicators exceeds the early warning threshold by less than 3, it proves that the mechanical structure is running well at this time, and signal acquisition is not required;

(2) If there are more than 3 dimensional indicators that exceed the warning threshold, but less than 3 non-dimensional indicators exceed the warning threshold, then the signal may be abnormal due to working conditions and other reasons, such as mechanical collision, There is no need to collect signals, but the current time stamp needs to be recorded;

(3) If there are no more than 3 dimensional indicators that exceed the warning threshold, but more than 3 non-dimensional indicators exceed the warning threshold, it means that the overall abnormality of the mechanical structure may occur at this time, and the vibration signal for 3 minutes is collected. Then stop the collection;

(4) If the number of dimensional indicators and non-dimensional indicators exceeding the warning threshold is greater than 3, it proves that failures may occur during the operation of the mechanical structure at this time, and continuous data collection is carried out.

Embodiment 1.

Taking a company's full life cycle motor bearing as an example, the effectiveness of the patented method of the present invention is verified. The duration of this full life cycle experiment is one month. The motor bearing will be prefabricated with minor inner ring faults and run uninterruptedly on the test bench under actual working conditions.

The experimental group adopts the sampling method of the patented method, and the control group adopts the sampling interval of 0.5 hours (sampling time is 3 minutes) after expert decision, and the rest of the parts of the collection device are the same. The evaluation indicators are the number of samples collected and the number of valid fault information samples evaluated by expert experience.

In order to verify the correctness of the patented method, situations such as abnormal working conditions and excessive time-frequency domain indicators are simulated, as shown in Figure 2.

When the experiment runs to the 10th day, the rated speed should be 1980r/min, but the actual measured speed is 2300r/min.

It is judged that the motor is not under the rated working condition at this time, so the abnormal working condition acquisition mode is started, and the signal is continuously collected for 5 minutes. The signal vibration waveform is shown in Figure 3. However, the control group was not sampled during this period due to timing sampling. Therefore, the validity of the algorithm can be verified.

When the experiment ran to the 15th day, the mechanical real-time calculation time-frequency domain index values exceeded the standard, as shown in the following table.

It can be seen from the table that the number of dimensional indicators and non-dimensional indicators exceeding the warning threshold is greater than 3, which proves that failure may occur during the operation of the mechanical structure at this time, and continuous data collection is carried out; Timing sampling results in that only 3 minutes of data are collected during this period; the amount of information is small; therefore, the effectiveness of the algorithm can be verified.

The duration of this whole life cycle experiment is 1 month. All the data collected in the experimental group and the control group are verified and compared. The specific results are shown in the following table.

group Collect samples valid sample Efficient test group 950 827 87% control group 1440 560 39%

From the results in the table, the experimental group collects fewer samples than the control sample, but the number of effective samples far exceeds that of the control group, which proves that the method adopted in this patent can optimally collect data samples and reduce the pressure of sample storage. At the same time, the multi-strategy index-triggered collection method also avoids the leakage of effective information, and the experimental sample efficiency is as high as 87%, which proves the effectiveness and robustness of the patented method.

The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above-mentioned embodiments. The above-mentioned embodiments and descriptions only illustrate the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will also have Various changes and modifications fall within the scope of the claimed invention.

Claims (4)

1. a novel anti-leakage sampling signal acquisition method for rotating machinery, it is characterized in that, comprises three kinds of anti-leakage sampling logics, according to priority ordering from high to low successively is fault acquisition mode, abnormal working condition acquisition mode, timing acquisition mode, and the signal sampling frequency is controllable, and sampling is performed in real time. The specific acquisition method is as follows: (1) Under the normal operation of the machine, according to the model and rated working condition of the motor, set the timing sampling time and collection time, and collect the signals regularly for storage and diagnosis; (2) When the mechanical operating condition information exceeds 5% of the specified operating amount, it proves that the machinery is under abnormal operating conditions at this time, and the signal is continuously collected at this time; (3) When the mechanical operation is under the rated working condition, but the frequency domain index exceeds the standard during real-time operation, the fault collection mode is activated to continuously collect signals. 2. a kind of novel anti-leakage sampling signal collection method for rotating machinery according to claim 1, is characterized in that, described working condition abnormal collection mode mainly starts by comparing working condition information and rated standard value, and concrete logic is as follows: (1) The working condition information is mainly based on speed, temperature, noise, and current, and its importance decreases in order; (2) According to the motor factory manual and the experience of relevant experts, obtain the specific information of each operating condition of the motor; (3) After setting the operating condition parameters, monitor the operating condition information of the motor in real time. As long as one of the operating conditions information exceeds 5% of the specified operating amount, the signal will be continuously collected for 5 minutes. 3. a kind of novel anti-leakage signal acquisition method for rotating machinery according to claim 1, is characterized in that, described fault acquisition mode adopts index characteristic method to realize, and described fault acquisition mode adopts 7 kinds of dimension Indicators and 7 dimensionless indicators form an evaluation indicator library. 4. a kind of novel anti-leakage signal acquisition method for rotating machinery according to claim 3, is characterized in that, the acquisition module of described fault acquisition mode collects in real time during operation, and the collected signal enters the algorithm module to carry out The evaluation logic is as follows: (1) If there are no more than 3 dimensional indicators and non-dimensional indicators that exceed the warning threshold, no signal collection will be performed; (2) If there are more than 3 dimensional indicators that exceed the warning threshold, but the number of dimensionless indicators that exceed the warning threshold does not exceed 3, the signal will not be collected, and the current timestamp should be recorded; (3) If there are no more than 3 dimensional indicators that exceed the warning threshold, but the number of dimensionless indicators that exceed the warning threshold exceeds 3, the vibration signal will be collected for 3 minutes, and then the collection will be terminated; (4) If the number of dimensional indicators and non-dimensional indicators exceeds the warning threshold of more than 3, continuous data collection will be carried out.

Images