The method of operating status parameter collection and fault diagnosis of three-phase horizontal screw centrifuge
By collecting and analyzing the bearing vibration, temperature and noise data of three-phase scatter centrifuge, generating abnormal coefficients and building a fault model, it solves the problem that the type and degree of fault cannot be accurately diagnosed in the existing technology, and achieves more efficient fault detection and maintenance.
Patent Information
- Application Number
- CN202410922912.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-10
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-07-10
AI Technical Summary
The existing three-phase horizontal screw centrifuge fault diagnosis methods cannot accurately diagnose the type and degree of fault, and cannot fully capture the complex dynamic changes when the fault occurs, resulting in unfavorable maintenance and repair work.
By collecting time domain characteristic data of bearing vibration, temperature and noise, vibration abnormality coefficients, temperature abnormality coefficients and noise abnormality coefficients are generated, and imbalance, lubrication and wear fault models are constructed to diagnose fault types and degrees using these coefficients.
It improves the sensitivity and accuracy of fault detection, can diagnose fault types in detail, provide intuitive information on fault severity, and help formulate corresponding maintenance strategies.
Smart Images

Figure CN118503882B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of centrifuge diagnosis, and in particular to a method for collecting operating state parameters and diagnosing faults of a three-phase horizontal screw centrifuge. Background Art
[0002] The three-phase decanter centrifuge is a separation equipment widely used in chemical, pharmaceutical, food processing and other industries. Its main function is to separate different components in the mixture through the centrifugal force generated by high-speed rotation. Due to its complex working environment, various faults may occur during operation, such as imbalance, poor lubrication, wear, etc. These faults not only affect the normal operation of the equipment, but may also cause serious safety problems. Therefore, it is particularly important to monitor the operating status of the three-phase decanter centrifuge in real time and diagnose faults.
[0003] Traditional fault diagnosis methods usually rely on manual inspections and regular maintenance. This method is not only inefficient, but also fails to monitor the operating status of the equipment in real time, and is prone to miss early signs of faults. With the development of sensor technology and data processing technology, fault diagnosis methods based on real-time data collection and analysis have gradually become mainstream. This method uses sensors installed on the equipment to collect key parameters such as vibration, temperature, and noise in real time, and then uses data analysis technology to extract features and compare them with preset thresholds to determine whether the equipment is faulty.
[0004] In the prior art, the publication number CN114594721A discloses a three-legged centrifuge self-learning fault monitoring and alarm system, wherein an acceleration sensor is installed on the centrifuge motor or a base near the motor, a temperature sensor is installed on the side of the centrifuge body, and a data acquisition card collects data from the acceleration sensor and the temperature sensor, and transmits the data to the industrial computer, which runs the fault monitoring and alarm system to monitor the centrifuge in real time. The method is to automatically set the vibration and temperature judgment thresholds by collecting vibration and temperature signals when the centrifuge is just put into use; collect vibration data after the centrifuge has worked for a period of time, and predict the vibration threshold of the centrifuge in the next time period by self-learning, so as to avoid misjudgment caused by too harsh initial threshold setting. However, the prior art also has some shortcomings. First, the prior art system only diagnoses faults through simple threshold comparison, without considering the complex relationship between vibration and temperature data. In practical applications, the occurrence of faults is often accompanied by changes in multiple parameters, and there may be mutual influence and dependence between these changes. Since the complex relationship between the various parameters when the fault occurs is not taken into account, although the system can detect the occurrence of the fault, it cannot accurately diagnose the specific type of the fault, and it cannot quantify the degree of the fault. This is disadvantageous for subsequent maintenance and repair work, because different types of failures may require different treatment methods and parts replacement.
[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention
[0006] The object of the present invention is to provide a method for collecting operating status parameters and diagnosing faults of a three-phase horizontal screw centrifuge, so as to solve the problems raised in the above-mentioned background technology.
[0007] To achieve the above object, the present invention provides the following technical solutions:
[0008] A method for collecting operating status parameters and diagnosing faults of a three-phase horizontal screw centrifuge, the specific steps comprising:
[0009] Step 1: Collecting the working state time series data of the three-phase horizontal screw centrifuge during operation, wherein the working state time series data includes the amplitude time series data of the bearing vibration, the bearing temperature time series data and the noise time series data of the three-phase horizontal screw centrifuge;
[0010] Step 2: Divide the bearing vibration amplitude time series data into The time series data is converted from time domain to frequency domain using Fourier transform, and analyzed and processed to obtain the power spectrum density corresponding to each time series data segment. The power spectrum density corresponding to each amplitude time series number segment is comprehensively analyzed to obtain the maximum power spectrum density and the average power spectrum density.
[0011] Step 3: Analyze and process the amplitude time series data to obtain amplitude time domain feature data of the amplitude time series data, wherein the amplitude time domain feature data includes the vibration amplitude average value and the vibration amplitude root mean square value, and construct a mathematical analysis model based on the vibration amplitude average value and the generated vibration amplitude root mean square value to generate a vibration abnormality coefficient; then construct an unbalanced fault model through the vibration abnormality coefficient and the vibration frequency domain feature data to generate an unbalanced fault coefficient;
[0012] Step 4: Analyze and process the temperature time series data to obtain temperature time domain characteristic data of the temperature time series data, wherein the temperature time domain characteristic data includes the maximum temperature change rate and the temperature fluctuation rate, and perform mathematical analysis on the maximum temperature change rate and the temperature fluctuation rate to generate a temperature anomaly coefficient, and then construct a lubrication fault model through the temperature anomaly coefficient and the vibration anomaly coefficient to generate a lubrication fault coefficient;
[0013] Step 5: Analyze and process the noise time series data to obtain noise time domain feature data of the noise time series data, wherein the noise time series feature data includes a maximum noise level and an equivalent continuous noise level; and construct a noise anomaly model according to the maximum noise level and the equivalent continuous noise level to generate a noise anomaly coefficient, and then construct a wear fault model through the temperature anomaly coefficient and the noise anomaly coefficient to generate a wear fault coefficient;
[0014] Step 6: Compare the imbalance fault coefficient, lubrication fault coefficient and wear fault coefficient with the preset threshold value, and diagnose the fault type according to the comparison result.
[0015] Furthermore, the specific logic for obtaining the maximum power spectral density and the average power spectral density is as follows: the amplitude time series data is divided into Each segment of amplitude time series data is transformed into the form of the sum of a series of complex numbers by Fourier transform to obtain the spectrum. , the power spectrum density of each time series data segment is calculated through the spectrum information. The specific formula for calculating the power spectrum density is:
[0016] ;
[0017] in, For the The power spectrum density of the vibration signal is is the number of frequencies, For the The length of the vibration signal. is the vibration spectrum;
[0018] Then, the maximum power spectrum density and the average power spectrum density are obtained by calculating the power spectrum density. The specific formula is as follows:
[0019] ;
[0020] in, is the maximum power spectral density, is the average power spectral density, is the number of signal segments.
[0021] Furthermore, the specific logic for generating the unbalanced fault coefficient is as follows: the vibration abnormality coefficient is generated by multiplying the vibration amplitude average value and the generated vibration amplitude root mean square, and then the unbalanced fault coefficient is generated by the vibration abnormality coefficient and the vibration frequency domain characteristic data. The specific formula for generating the dynamic abnormality coefficient is as follows:
[0022] ;
[0023] in, is the vibration anomaly coefficient, is the average value of the vibration, is the root mean square of vibration;
[0024] The specific formula based on which the unbalanced fault coefficient is generated is:
[0025] ;
[0026] is the unbalanced fault coefficient, is the maximum power spectral density, is the average power spectral density.
[0027] Furthermore, the specific logic for generating the temperature fluctuation rate is: calculating the average value of the temperature time series data, integrating the square of the difference between the temperature value at each moment and the average value, and then dividing it by the duration of the time period to obtain the temperature fluctuation rate; the specific formula is:
[0028] ;
[0029] in, is the temperature fluctuation rate, is the total time of temperature time series data, It is the temperature time series data.
[0030] Furthermore, the specific logic for generating the lubrication fault coefficient is as follows: the temperature anomaly coefficient is generated by analyzing the maximum temperature change rate and the temperature fluctuation rate, and the lubrication fault coefficient is generated by analyzing the temperature anomaly coefficient and the vibration anomaly coefficient; the specific formula for generating the vibration anomaly coefficient is as follows:
[0031] ;
[0032] is the temperature anomaly coefficient, is the maximum temperature change rate, is the temperature fluctuation rate;
[0033] The specific formula upon which the lubrication failure factor is generated is:
[0034] ;
[0035] is the lubrication failure coefficient, is the vibration abnormality coefficient.
[0036] Furthermore, the specific logic for generating the wear fault coefficient is as follows: the noise anomaly coefficient is generated by the maximum noise level and the equivalent continuous noise level, and the wear fault coefficient is then generated by analyzing the temperature anomaly coefficient and the noise anomaly; the specific formula for generating the noise anomaly coefficient is as follows:
[0037] ;
[0038] is the noise anomaly coefficient, is the equivalent continuous noise level, is the maximum noise level;
[0039] The specific formula used to generate the wear failure coefficient is:
[0040] ;
[0041] is the wear failure coefficient, is the temperature anomaly coefficient. Further, the specific logic for the three-phase horizontal screw centrifuge is: , then the three-phase horizontal screw centrifuge has an unbalanced fault; if The three-phase horizontal screw centrifuge has a lubrication failure; if ; The three-phase horizontal screw centrifuge will have a wear failure;
[0042] in, is the unbalanced fault coefficient, is the unbalance fault threshold, is the lubrication failure coefficient, is the lubrication failure threshold, is the wear failure coefficient, is the wear failure threshold.
[0043] Compared with the prior art, the present invention has the following beneficial effects:
[0044] The present invention generates vibration anomaly coefficients, temperature anomaly coefficients and noise anomaly coefficients by collecting the time domain characteristics of vibration, temperature and noise. These coefficients can comprehensively reflect the abnormal state of the equipment during operation, rather than just the threshold comparison of a single parameter. This method can more comprehensively capture the complex dynamic changes when a fault occurs, and improve the sensitivity and accuracy of fault detection.
[0045] On this basis, the present invention further analyzes the vibration frequency domain characteristics, vibration abnormality coefficient, temperature abnormality coefficient and noise abnormality coefficient. Unbalance fault coefficient, lubrication fault coefficient and noise fault coefficient are generated. These fault coefficients can help diagnose specific fault types, such as equipment imbalance, poor lubrication or abnormal noise. This detailed fault classification is helpful for subsequent maintenance and repair work, because different types of faults may require different treatment methods and component replacement.
[0046] The present invention also quantifies the degree of abnormality and fault by generating various abnormality coefficients and fault coefficients, which provides intuitive information about the severity of the fault to maintenance personnel, so that they can formulate corresponding maintenance strategies according to the actual situation of the fault. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION
[0048] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.
[0049] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0050] Example:
[0051] See also Figure 1 , the present invention provides a technical solution:
[0052] A method for collecting operating status parameters and diagnosing faults of a three-phase horizontal screw centrifuge, the specific steps comprising:
[0053] Step 1: Collect the working state time series data of the three-phase horizontal screw centrifuge when it is working, wherein the working state time series data includes the amplitude time series data of the bearing vibration, the bearing temperature time series data and the noise time series data of the three-phase horizontal screw centrifuge.
[0054] The working status time series data includes bearing vibration amplitude time series data, bearing temperature time series data and three-phase horizontal screw centrifuge noise time series data collected by sensors. The bearing vibration amplitude time series data is collected by a vibration sensor installed on the bearing seat; the bearing temperature time series data is collected by a temperature sensor installed in the bearing raceway; the three-phase horizontal screw centrifuge noise time series data is collected by a noise sensor installed on the casing of the three-phase horizontal screw centrifuge.
[0055] Step 2: Divide the bearing vibration amplitude time series data into The time series data are converted from time domain to frequency domain using Fourier transform, and analyzed and processed to obtain the power spectrum density corresponding to each segment of time series data. The power spectrum density corresponding to each segment of amplitude time series data is comprehensively analyzed to obtain the maximum power spectrum density and the average power spectrum density.
[0056] The amplitude time series data of bearing vibration is divided into In order to ensure that the segmented data has sufficient resolution, the present invention chooses to position the length of each segment of the amplitude timing number to 1024 sampling points, and the segment number is obtained. The specific formula is:
[0057] ;
[0058] in, is the number of segments, is the signal length.
[0059] The specific logic for obtaining the maximum power spectral density and the average power spectral density is: divide the amplitude time series data into Each segment of amplitude time series data is transformed into the form of the sum of a series of complex numbers by Fourier transform to obtain the spectrum. , the power spectrum density of each time series data segment is calculated through the spectrum information. The specific formula for calculating the power spectrum density is:
[0060] ;
[0061] in, For the The power spectrum density of the vibration signal is is the number of frequencies, For the The length of the vibration signal. is the vibration spectrum;
[0062] Then, the maximum power spectrum density and the average power spectrum density are obtained by calculating the power spectrum density. The specific formula is as follows:
[0063] ;
[0064] in, is the maximum power spectral density, is the average power spectral density, is the number of signal segments. Maximum power spectral density Reflects the maximum power of vibration; the larger the value, the greater the power of vibration in a certain period of time, the vibration also reaches the maximum value in that period of time, and unbalanced faults are most likely to occur. The average power spectrum density Reflects the overall power of vibration. The larger the value, the greater the overall power of vibration, the more violent the vibration, and the more likely it is that an unbalanced fault will occur.
[0065] Step 3: Analyze and process the amplitude time series data to obtain the amplitude time domain feature data of the amplitude time series data, wherein the amplitude time domain feature data includes the vibration amplitude average value and the vibration amplitude root mean square value, and construct a mathematical analysis model based on the vibration amplitude average value and the generated vibration amplitude root mean square value to generate a vibration abnormality coefficient; then construct an unbalanced fault model through the vibration abnormality coefficient and the vibration frequency domain feature data to generate an unbalanced fault coefficient.
[0066] The specific formula for generating the average value and root mean square of vibration is:
[0067] ;
[0068] is the average value of the vibration, is the total time of the time series data, is the amplitude time series data, is the root mean square of the vibration. The average value of the vibration Can reflect the floating level of vibration, the root mean square of vibration Reflecting the average energy of the signal, both play a huge role in evaluating vibration abnormalities.
[0069] The specific logic for generating the unbalanced fault coefficient is as follows: multiply the average value of the vibration amplitude and the root mean square of the generated vibration amplitude to generate the vibration abnormality coefficient, and then generate the unbalanced fault coefficient through the vibration abnormality coefficient and the vibration frequency domain characteristic data. The specific formula for generating the dynamic abnormality coefficient is as follows:
[0070] ;
[0071] in, is the vibration anomaly coefficient, is the average value of the vibration, is the root mean square of vibration; vibration abnormality coefficient Reflects the degree of vibration abnormality. The larger the value, the higher the degree of vibration abnormality. The average value of vibration Reflects the overall floating level of vibration. The larger the value, the higher the overall floating level of vibration. Reflects the average energy of the signal. The larger the value, the higher the overall energy level of the vibration. This formula comprehensively analyzes the abnormality of the vibration and reflects the abnormal degree of the vibration as a whole by constructing the vibration abnormality coefficient.
[0072] The specific formula based on which the unbalanced fault coefficient is generated is:
[0073] ;
[0074] is the unbalanced fault coefficient, is the maximum power spectral density, is the average power spectrum density. Unbalanced fault coefficient Reflects the tendency of unbalanced failure of the three-phase decanter centrifuge. The larger the value, the greater the tendency of unbalanced failure of the three-phase decanter centrifuge. Abnormal vibration coefficient Reflects the degree of vibration abnormality. The larger the value, the higher the degree of vibration abnormality and the greater the probability of unbalanced fault. Maximum power spectrum density Reflects the maximum power of vibration; the larger the value, the greater the power of vibration in a certain period of time, the vibration also reaches the maximum value in that period of time, and unbalanced faults are most likely to occur. The average power spectrum density It reflects the overall power of vibration. The larger the value is, the greater the overall power of vibration is, the more violent the vibration is, and the more likely it is to have an unbalanced fault. This formula obtains the unbalanced fault coefficient by analyzing the vibration abnormality coefficient and the vibration frequency domain characteristics, providing a basis for judging unbalanced faults.
[0075] Step 4: Analyze and process the temperature time series data to obtain temperature time domain characteristic data of the temperature time series data, wherein the temperature time domain characteristic data includes the maximum temperature change rate and the temperature fluctuation rate, and perform mathematical analysis on the maximum temperature change rate and the temperature fluctuation rate to generate a temperature anomaly coefficient, and then construct a lubrication fault model through the temperature anomaly coefficient and the vibration anomaly coefficient to generate a lubrication fault coefficient;
[0076] The specific logic for extracting the maximum temperature change rate is: differentiate the temperature time series data, and its maximum differential value is the maximum temperature change rate. The specific formula is:
[0077] ;
[0078] in, is the maximum temperature change rate, is the time series data of temperature. The temperature change rate is an important basis for evaluating temperature anomalies. The greater the temperature change rate, the more serious the temperature anomaly. It can reflect the extreme changes in temperature and plays an important role in judging temperature anomalies.
[0079] The specific logic for generating the temperature fluctuation rate is: calculate the average value of the temperature time series data, integrate the square of the difference between the temperature value at each moment and the average value, and then divide it by the duration of the time period to obtain the temperature fluctuation rate; the specific formula is:
[0080] ;
[0081] in, is the temperature fluctuation rate, is the total time of temperature time series data, is the temperature time series data; temperature fluctuation rate Reflects the temperature fluctuation range. The larger the value, the greater the temperature fluctuation range and the greater the probability of temperature anomaly. Temperature fluctuation rate The generation of plays an important role in the assessment of temperature anomalies.
[0082] The specific logic for generating the lubrication failure coefficient is as follows: the temperature anomaly coefficient is generated by analyzing the maximum temperature change rate and the temperature fluctuation rate, and the lubrication failure coefficient is generated by analyzing the temperature anomaly coefficient and the vibration anomaly coefficient; the specific formula for generating the vibration anomaly coefficient is as follows:
[0083] ;
[0084] is the temperature anomaly coefficient, is the maximum temperature change rate, is the temperature fluctuation rate; temperature anomaly coefficient Reflects the degree of temperature abnormality. The larger the value, the higher the temperature abnormality. The maximum temperature change rate Reflects the maximum temperature change rate. The larger the value, the faster the temperature mutation rate and the greater the probability of temperature anomaly. Temperature fluctuation rate Reflects the temperature fluctuation range. The larger the value, the greater the temperature fluctuation range and the greater the probability of temperature abnormality. Lubrication failure is usually accompanied by temperature abnormality. By comprehensively analyzing the maximum temperature change rate and temperature fluctuation rate The generated temperature anomaly coefficient It plays an important role in diagnosing lubrication failures.
[0085] The specific formula upon which the lubrication failure factor is generated is:
[0086] ;
[0087] is the lubrication failure coefficient, is the vibration abnormality coefficient; lubrication failure coefficient Reflects the tendency of lubrication failure of the three-phase decanter centrifuge. The larger the value, the greater the tendency of lubrication failure of the three-phase decanter centrifuge. Temperature anomaly coefficient Reflects the abnormal degree of temperature. The larger the value, the higher the abnormal degree of temperature. Reflects the degree of vibration abnormality. The larger the value, the higher the degree of vibration abnormality and the greater the probability of lubrication failure. Lubrication failure is usually accompanied by temperature abnormality and vibration abnormality. This formula comprehensively analyzes the temperature abnormality coefficient. and vibration anomaly coefficient Generate a lubrication failure coefficient, which provides an important basis for the diagnosis of lubrication failures.
[0088] Step 5: Analyze and process the noise time series data to obtain noise time domain feature data of the noise time series data, wherein the noise time series feature data includes a maximum noise level and an equivalent continuous noise level; and construct a noise anomaly model according to the maximum noise level and the equivalent continuous noise level to generate a noise anomaly coefficient, and then construct a wear fault model through the temperature anomaly coefficient and the noise anomaly coefficient to generate a wear fault coefficient;
[0089] The maximum noise level is the maximum noise level collected by the noise sensor, which can reflect the sound level with the highest noise. Generally, the greater the maximum noise level, the higher the probability of noise abnormality.
[0090] The specific logic for generating the equivalent continuous noise level is: calculate the sound energy at each moment through the noise time series data, then calculate the average value of the sound energy, and finally convert the sound energy calculation into the equivalent sound level through logarithmic operation; the specific formula is:
[0091] ;
[0092] in, is the equivalent continuous noise level, is the total time of the noise time series data, is the noisy time series data, is the sound energy, equivalent continuous noise level Reflects the average level of noise. The larger the value, the higher the average level of noise, generating an equivalent continuous noise level It can provide important basis for the evaluation of noise anomalies.
[0093] The specific logic for generating the wear failure coefficient is as follows: the noise anomaly coefficient is generated by the maximum noise level and the equivalent continuous noise level, and then the wear failure coefficient is generated by analyzing the temperature anomaly coefficient and the noise anomaly; the specific formula for generating the noise anomaly coefficient is as follows:
[0094] ;
[0095] is the noise anomaly coefficient, is the equivalent continuous noise level, is the maximum noise level; noise anomaly coefficient Reflects the degree of noise anomaly. The larger the value, the higher the degree of noise anomaly. Reflects the average level of noise. The larger the value, the higher the average level of noise. Maximum noise level It can reflect the maximum noise level. Generally, the greater the maximum noise level, the higher the probability of noise abnormality.
[0096] The specific formula used to generate the wear failure coefficient is:
[0097] ;
[0098] is the wear failure coefficient, is the temperature anomaly coefficient; wear failure coefficient Reflects the tendency of wear failure of the three-phase horizontal screw centrifuge. The larger the value, the greater the tendency of wear failure of the three-phase horizontal screw centrifuge. Temperature anomaly coefficient Reflects the degree of temperature abnormality. The larger the value, the higher the temperature abnormality and the more likely it is to cause wear and tear failure; Noise Abnormality Coefficient Reflects the degree of noise abnormality. The larger the value, the higher the degree of noise abnormality; the more likely it is to cause wear failure. The present invention analyzes the temperature abnormality coefficient and noise anomaly coefficient Generate wear failure coefficient , providing an important basis for evaluating wear failures.
[0099] Step 6: Compare the imbalance fault coefficient, lubrication fault coefficient and wear fault coefficient with the preset threshold value, and diagnose the fault type according to the comparison result.
[0100] The specific logic for the three-phase decanter centrifuge is: , then the three-phase horizontal screw centrifuge has an unbalanced fault; if The three-phase horizontal screw centrifuge has a lubrication failure; if ; The three-phase horizontal screw centrifuge will have a wear failure;
[0101] in, is the unbalanced fault coefficient, is the unbalance fault threshold, is the lubrication failure coefficient, is the lubrication failure threshold, is the wear failure coefficient, is the wear failure threshold.
[0102] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0103] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.
[0104] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0105] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.
Claims
1. A method for collecting operating status parameters and diagnosing faults of a three-phase horizontal screw centrifuge, characterized in that: Specifically include: Step 1: collecting the working state time series data of the three-phase horizontal screw centrifuge when it is working, wherein the working state time series data includes the amplitude time series data of the bearing vibration, the temperature time series data of the bearing and the noise time series data of the three-phase horizontal screw centrifuge; Step 2: Divide the amplitude time series data of bearing vibration into N segments, and use Fourier transform to convert the time series data from time domain to frequency domain, analyze and process them, obtain the power spectrum density corresponding to each segment of time series data, and comprehensively analyze the power spectrum density corresponding to each segment of amplitude time series data to obtain the maximum power spectrum density and average power spectrum density; Step 3: Analyze and process the amplitude time series data to obtain amplitude time domain feature data of the amplitude time series data, wherein the amplitude time domain feature data includes the vibration amplitude average value and the vibration amplitude root mean square value, and construct a mathematical analysis model based on the vibration amplitude average value and the generated vibration amplitude root mean square value to generate a vibration abnormality coefficient; then construct an unbalanced fault model through the vibration abnormality coefficient and the vibration frequency domain feature data to generate an unbalanced fault coefficient; The specific logic for generating the unbalanced fault coefficient is as follows: multiply the average value of the vibration amplitude and the root mean square of the generated vibration amplitude to generate the vibration abnormality coefficient, and then generate the unbalanced fault coefficient through the vibration abnormality coefficient and the vibration frequency domain characteristic data. The specific formula for generating the dynamic abnormality coefficient is as follows: Among them, ZD is the vibration abnormality coefficient, is the average value of vibration, Z RMS is the root mean square of vibration; The specific formula based on which the unbalanced fault coefficient is generated is: Bl is the unbalanced fault coefficient, P max is the maximum power spectral density, is the average power spectral density; Step 4: Analyze and process the temperature time series data to obtain temperature time domain characteristic data of the temperature time series data, wherein the temperature time domain characteristic data includes the maximum temperature change rate and the temperature fluctuation rate, and perform mathematical analysis on the maximum temperature change rate and the temperature fluctuation rate to generate a temperature anomaly coefficient, and construct a lubrication fault model through the temperature anomaly coefficient and the vibration anomaly coefficient to generate a lubrication fault coefficient; The specific logic for generating the temperature fluctuation rate is: calculate the average value of the temperature time series data, integrate the square of the difference between the temperature value at each moment and the average value, and then divide it by the duration of the time period to obtain the temperature fluctuation rate; the specific formula is: Among them, T B is the temperature fluctuation rate, T is the total time of the temperature time series data, and T(t) is the temperature time series data; The specific logic for generating the lubrication failure coefficient is as follows: the temperature anomaly coefficient is generated by analyzing the maximum temperature change rate and the temperature fluctuation rate, and the lubrication failure coefficient is generated by analyzing the temperature anomaly coefficient and the vibration anomaly coefficient; the specific formula for generating the vibration anomaly coefficient is as follows: TD is the temperature anomaly coefficient, is the maximum temperature change rate, T B is the temperature fluctuation rate; The specific formula upon which the lubrication failure factor is generated is: Lu=ZD*TD Lu is the lubrication failure coefficient, ZD is the vibration abnormality coefficient; Step 5: Analyze and process the noise time series data to obtain noise time domain feature data of the noise time series data, wherein the noise time series feature data includes a maximum noise level and an equivalent continuous noise level; and construct a noise anomaly model according to the maximum noise level and the equivalent continuous noise level to generate a noise anomaly coefficient, and construct a wear fault model through the temperature anomaly coefficient and the noise anomaly coefficient to generate a wear fault coefficient; The specific logic for generating the wear failure coefficient is as follows: the noise anomaly coefficient is generated by the maximum noise level and the equivalent continuous noise level, and then the wear failure coefficient is generated by analyzing the temperature anomaly coefficient and the noise anomaly; the specific formula for generating the noise anomaly coefficient is as follows: LD=L eq 5 *L max LD is the noise anomaly coefficient, L eq is the equivalent continuous noise level, L max is the maximum noise level; The specific formula used to generate the wear failure coefficient is: Mo=TD+LD 2 Mo is the wear failure coefficient, TD is the temperature anomaly coefficient; Step 6: Compare the imbalance fault coefficient, lubrication fault coefficient and wear fault coefficient with preset thresholds respectively, and diagnose the fault type according to the comparison results; The specific logic for evaluating the three-phase decanter centrifuge is: if Bl>Bl0, the three-phase decanter centrifuge has an unbalance fault; if Lu>Lu0, the three-phase decanter centrifuge has a lubrication fault; if Mo>Mo0; the three-phase decanter centrifuge has a wear fault; Wherein, Bl is the imbalance fault coefficient, Bl0 is the imbalance fault threshold, Lu is the lubrication fault coefficient, Lu0 is the lubrication fault threshold, Mo is the wear fault coefficient, and Mo0 is the wear fault threshold.
2. The method for collecting operating status parameters and diagnosing faults of a three-phase horizontal screw centrifuge according to claim 1, characterized in that: The specific logic for obtaining the maximum power spectrum density and the average power spectrum density is as follows: the amplitude time series data is divided into N segments, each segment of the amplitude time series data is transformed into the form of the sum of a series of complex numbers by Fourier transform, and the spectrum X(k) is obtained. The power spectrum density of each segment of the time series data is calculated through the spectrum information. The specific formula for calculating the power spectrum density is: Among them, P(k) i is the power spectrum density of the vibration signal in the i-th segment, n is the number of frequencies, n i is the length of the i-th vibration signal, X(k) i is the vibration spectrum; Then, the maximum power spectrum density and the average power spectrum density are obtained by calculating the power spectrum density. The specific formula is as follows: Among them, P max is the maximum power spectral density, is the average power spectral density, and N is the number of signal segments.
