An online assessment method for the health status of mining belt conveyors
Through the multi-dimensional time series clustering algorithm TICC and real-time operation data, the shortcomings of relying on physical models in the health status evaluation of belt conveyors are solved, and accurate online evaluation of belt conveyor systems is realized, and false alarms and missed reports are reduced. Improved the practicality and accuracy of the evaluation.
Patent Information
- Application Number
- CN202111373681.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-19
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2041-11-19
AI Technical Summary
The existing belt conveyor fault diagnosis and health status assessment methods rely on physical models, making it difficult to effectively evaluate the health status of multi-subsystem coupled devices, and the monitoring system has problems such as false alarms, missed alarms and protective device startup lags.
Using multi-dimensional time series clustering algorithm TICC combined with real-time operation data, the operating conditions characteristic parameters of the belt conveyor are obtained, and the working conditions are identified through multi-dimensional time series clustering algorithm, the system efficiency is horizontally compared, and the health status changes are vertically analyzed, and the health status changes are established. Health status evaluation model to realize online evaluation of belt conveyor system.
No need to rely on accurate physical models, reduce false positives and missed reports, improve the accuracy and real-time evaluation, is easy to apply on field devices, and reduces the impact of protection device startup lag.
Smart Images

Figure CN114662834B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of intelligent detection and evaluation of belt conveyors, and in particular relates to an online evaluation technology for the health status of a belt conveyor. Background Art
[0002] Mining belt conveyors are key electrical equipment in the process of coal production and transportation. Their reliable and efficient operation has an important impact on the production safety and economic benefits of coal enterprises. In order to improve its operational reliability, it is necessary to establish indicators that effectively evaluate the health of the equipment, so that maintenance and care can be performed when the equipment is not faulty but is in a low health state, to ensure the reliable and efficient operation of the belt conveyor system.
[0003] There are still some problems in the existing research on fault diagnosis, fault prediction and health status assessment of mining equipment systems such as belt conveyors. The proposed methods mainly rely on physical models. Since mining equipment is often composed of complex coupling of multiple subsystems, the mechanism modeling of many equipment subsystems has not yet been completed. Verification is mostly based on the fault status data of a single subsystem or the standard data set measured on the test bench, and the theoretical effect is often not obtained when migrating to field equipment. The quality of the monitoring system depends on the setting of the threshold, and there are problems such as false alarms, missed alarms and delayed start-up of protection devices. Summary of the invention
[0004] In order to overcome the problems existing in the prior art, the present invention proposes an online assessment method for the health status of a mining belt conveyor, establishes a belt conveyor health status assessment model, and realizes online assessment of its health status. The technical solution of the present invention includes the following steps:
[0005] S1. Obtaining characteristic parameters of operating conditions, including: belt conveyor speed, instantaneous coal flow rate and driving motor current;
[0006] S2. Using the acquired operating condition characteristic parameters, a multidimensional time series clustering algorithm TICC (Toeplitz Inverse Covariance-based Clustering, TICC for short) is used to identify the operating condition of the belt conveyor system and divide it into n operating conditions;
[0007] S3. Compare the efficiency of belt conveyor systems under different working conditions horizontally, and fit the probability density of system operating efficiency corresponding to each working condition by normal distribution;
[0008] S4. Conduct longitudinal analysis on the sampling data under the same working conditions to obtain the trend of the belt conveyor's health status change, and divide the health status evaluation level based on the actual on-site experience;
[0009] S5. Establish a belt conveyor health status evaluation model and combine it with real-time operation data to evaluate the health status of the belt conveyor system.
[0010] Furthermore, in step S2, the specific process of the TICC algorithm used includes:
[0011] The historical operation data of the equipment with a time length of T is:
[0012] X=[x1,x2,x3,…x t ,…x T ]
[0013] In the formula, T is the number of samples in the time series dataset, x t The n-dimensional operating data vector measured by the n sensors installed on the equipment at time t:
[0014]
[0015] Taking each moment t as the benchmark, we extract a data subsequence X with a window length of k. t , recorded as:
[0016] X t =[x t ,x t+1 ,x t+2 ,…x t+k-1 ]
[0017] Assuming that the data set is divided into M categories, the set of p sensor signals contained in the mth cluster is recorded as:
[0018] W m =[W m1 ,W m2 ,…W mp ]
[0019] Assume that each cluster uses the inverse Gaussian covariance matrix defined as Θ m =R nk*nk , the TICC solution model is:
[0020]
[0021] In the formula, represents a set of symmetric Teoplitz matrices; represents the L1 norm penalty of the Hadamard product, using the incentive sparse quasi-covariance; α represents the temporal consistency coefficient; is used to check the last subsequence X t-1 The logarithmic likelihood of belonging to the mth category is expressed as:
[0022]
[0023] In the formula, β m represents the empirical mean of category m.
[0024] Furthermore, in step S3, the process of obtaining the efficiency of the belt conveyor system is as follows:
[0025] From the energy perspective, the efficiency η of the belt conveyor system is expressed as:
[0026]
[0027] In the formula, E out Refers to the energy output of the belt conveyor, that is, the work done during the operation of the equipment; E in Refers to the total energy input of the belt conveyor system, that is, the electrical energy consumed. This part is obtained from the voltage and current data of the belt conveyor system:
[0028]
[0029] In the formula, U refers to the line voltage of the belt conveyor system, I refers to the line current of the belt conveyor system, It refers to the phase angle between the line voltage and the line current, and t refers to the time range corresponding to the energy input to be calculated;
[0030] From the perspective of energy conversion, the operating power P provided by the main roller of the belt conveyor A It is mainly consumed by the following four parts:
[0031] ① No-load operation power consumption P nl (W):
[0032] P nl =gf(L h +L0)qv b
[0033] In the formula, g is the acceleration due to gravity, which is 9.8 m / s 2 , f is the simulated friction factor, L h is the horizontal length of the belt conveyor roller center distance (m), L0 is the length compensation coefficient (m), which is used to compensate for the friction force component that is not related to the belt length, v b is the conveyor belt speed (m / s), q is the unit mass of the moving part of the equipment (kg / m), and the calculation formula is as follows:
[0034]
[0035] In the formula, is the unit mass of the rotating part of the supporting roller (kg / m), is the unit mass of the rotating part of the return roller (kg / m), qbelt is the mass of the conveyor belt per unit length (kg / m);
[0036] ②Power consumed in horizontal material transport P l (W):
[0037] P l =gf(L h +L0)q m v b
[0038] Where H is the height of the material being lifted or below (m), q m is the mass of material per unit length on the conveyor belt (kg / m);
[0039] ③Power consumed in vertical material transport P h (W):
[0040] P h =gHq m v b
[0041] Where H is the lifting or lowering height of the material (m);
[0042] ④ The power consumed by the friction between the accessories and the conveyor belt when the belt conveyor is running P ACS (W):
[0043] P ACS =0.2gd 2 L h γv b
[0044] Where d is the material cross-sectional depth (m), γ is the material bulk density (kg / m 3 ).
[0045] In summary, based on the perspective of energy conversion, the total effective energy E consumed by the belt conveyor for continuous material transportation for t seconds is out (J) is:
[0046]
[0047] Further simplified to:
[0048]
[0049] At the same time, there is active roller mechanical power P A (W):
[0050] P A =(C1+C3+C2q m )v b
[0051] In the formula, C1 = gf(L h + L0)q, C2 = gf(L h +L0)+gH,C3=0.2gd 2 L h γ.
[0052] Furthermore, in step S4, a normal distribution fitting is performed on the efficiency values of the belt conveyor under different working conditions to obtain the efficiency distribution characteristics of the belt conveyor under different working conditions.
[0053] Furthermore, in step S4, the data collected at different times are longitudinally analyzed under the same working conditions, and the efficiency change trends of the belt conveyor under the same working conditions over a long period of time are compared, and a basis for health level classification is formulated. According to the efficiency, the health level of the belt conveyor system is divided into four levels: excellent, good, general, and early potential failure.
[0054] Furthermore, in step S5, in the evaluation of the health status of the belt conveyor system: under stable operating conditions except for the startup and shutdown processes, operating condition identification, efficiency value calculation and health evaluation are performed every 10 seconds, and the evaluation results are statistically analyzed on a daily basis to obtain the overall health status of the system.
[0055] The beneficial effects of the present invention are as follows: the present invention adopts the TICC algorithm and combines real-time operation data to realize the evaluation of the health status of the belt conveyor system. The present invention does not need to rely on an accurate physical model, and evaluates from the perspective of system energy rather than the fault status of a single subsystem, making it easier to migrate to field equipment; it does not rely on the setting of a single threshold, reducing the impact of false alarms, missed alarms, and delayed start-up of protective devices. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 It is a flow chart of the online assessment method of the health status of a mining belt conveyor;
[0057] Figure 2 Flowchart of the multidimensional time series clustering algorithm TICC;
[0058] Figure 3 is the histogram of system efficiency under various working conditions;
[0059] Figure 4 It is a comparison of the efficiency of belt conveyor systems under the same working conditions over a long time scale; DETAILED DESCRIPTION
[0060] The following will provide embodiments of the present invention in conjunction with the accompanying drawings, and further explain the technical solution of the present invention clearly and completely through the embodiments. Obviously, the embodiments described are only part of the embodiments of the present invention, rather than all the embodiments.
[0061] like Figure 1 As shown, an embodiment provides an online assessment method for the health status of a mining belt conveyor, the method mainly comprising the following steps:
[0062] S1. Obtaining belt conveyor belt speed, instantaneous coal flow rate, and driving motor current as operating condition characteristic parameters;
[0063] The operating condition identification problem discussed in this step refers to the secondary identification of the operating condition caused by coal flow fluctuations, belt speed changes, etc. in the steady-state operating condition, except for start-up and shutdown. The operating condition identification of the belt conveyor system first depends on the reasonable selection of the operating condition characteristic parameters. The operating parameters that have a direct or indirect impact on the health status of the belt conveyor system, that is, the equipment efficiency value, and can be provided by the site should be selected.
[0064] Get the belt conveyor speed v b , instantaneous coal flow q m And the current i driving the motor fc As characteristic parameters for working condition identification.
[0065] S2, using the acquired characteristic parameters, using the multidimensional time series clustering algorithm TICC to identify the operating conditions of the belt conveyor system;
[0066] The process of multidimensional time series clustering algorithm is as follows Figure 2 As shown, it can be divided into the following steps:
[0067] The historical operation data of the equipment with a time length of T is:
[0068] X=[x1,x2,x3,…x t ,…x T ]
[0069] In the formula, T is the number of samples in the time series dataset, x t The n-dimensional operating data vector measured by the n sensors installed on the equipment at time t:
[0070]
[0071] Taking each moment t as the benchmark, we extract a data subsequence X with a window length of k. t , recorded as:
[0072] X t =[x t ,xt+1 ,x t+2 ,…x t+k-1 ]
[0073] Assuming that the data set can be divided into M categories, the set of p sensor signals contained in the mth cluster is recorded as:
[0074] W m =[W m1 ,W m2 ,…W mp ]
[0075] Assume that each cluster can be defined using the inverse Gaussian covariance matrix as Θ m =R nk*nk , then the TICC solution model is:
[0076]
[0077] In the formula, represents a set of symmetric Teoplitz matrices; represents the L1 norm penalty of the Hadamard product, using the incentive sparse quasi-covariance; α represents the temporal consistency coefficient; is used to check the last subsequence X t-1 The logarithmic likelihood of belonging to the mth category can be expressed as:
[0078]
[0079] In the formula, β m represents the empirical mean of category m.
[0080] Taking the belt conveyor operation data of a coal mine on a certain day as an example, the system steady-state operation condition data was screened out according to the equipment start and stop records saved by the belt centralized control system. After the bad data was eliminated, a total of 15,000 continuous sampling records were selected. This data segment was input into the TICC algorithm for cluster analysis, and the number of clusters was set to 4.
[0081] S3. Compare the efficiency of belt conveyor systems under different working conditions horizontally, and fit the probability density of system operating efficiency corresponding to each working condition by normal distribution;
[0082] The process for solving the efficiency of a belt conveyor system is as follows:
[0083] From an energy perspective, the efficiency of the belt conveyor system can be expressed as:
[0084]
[0085] In the formula, E inRefers to the total energy input of the belt conveyor system, that is, the electrical energy consumed. This part can be obtained from the voltage and current data of the belt conveyor system:
[0086]
[0087] From the perspective of energy conversion, the operating power P provided by the main roller of the belt conveyor A It is mainly consumed by the following four parts: ① No-load operation power consumption P nl (W):
[0088] P nl =gf(L h +L0)qv b
[0089] In the formula, g is the acceleration due to gravity, which is 9.8 m / s 2 , f is the simulated friction factor, L h is the horizontal length of the belt conveyor roller center distance (m), L0 is the length compensation coefficient (m), which is used to compensate for the friction force component that is not related to the belt length, v b is the conveyor belt speed (m / s), q is the unit mass of the moving part of the equipment (kg / m), and the calculation formula is as follows:
[0090]
[0091] In the formula, is the unit mass of the rotating part of the supporting roller (kg / m), is the unit mass of the rotating part of the return roller (kg / m), q belt is the mass of the conveyor belt per unit length (kg / m).
[0092] ②Power consumed in horizontal material transport P l (W):
[0093] P l =gf(L h +L0)q m v b
[0094] Where H is the lifting or lower height of the material (m).
[0095] ③Power consumed in vertical material transport P h (W):
[0096] P h =gHq m v b
[0097] Where H is the lifting or lowering height of the material (m).
[0098] ④ The power consumed by the friction between the accessories and the conveyor belt when the belt conveyor is running P ACS (W):
[0099] P ACS =0.2gd 2 L h γv b
[0100] Where d is the material cross-sectional depth (m), γ is the material bulk density (kg / m 3 ).
[0101] In summary, based on the perspective of energy conversion, the total effective energy E consumed by the belt conveyor for continuous material transportation for t seconds is out (J) is:
[0102]
[0103] Further simplified to:
[0104]
[0105] At the same time, there is active roller mechanical power P A (W):
[0106] P A =(C1+C3+C2q m )v b
[0107] In the formula, C1 = gf(L h + L0)q, C2 = gf(L h +L0)+gH,C3=0.2gd 2 L h γ.
[0108] Using the above method, the efficiency of the belt conveyor system under different working conditions in S2 is calculated. The results are as follows: Figure 3 The efficiency values of the belt conveyor under different working conditions are fitted with normal distribution, and the normal distribution fitting parameters of the belt conveyor system efficiency under different working conditions are shown in Table 1. From the data in the table, it can be seen that there are obvious differences in efficiency expectations under different working conditions, which verifies the importance of working condition identification for health status assessment, and it is necessary to judge the health level under the same working conditions.
[0109] Table 1 Normal distribution fitting parameters of belt conveyor system efficiency under different working conditions
[0110]
[0111] S4. Conduct longitudinal analysis on the sampling data under the same working conditions to obtain the trend of the belt conveyor's health status change, and divide the health status evaluation level based on the actual on-site experience;
[0112] The three sampling data collected at different times were analyzed and the system efficiency under different working conditions was compared. The results are as follows: Figure 4 As shown. Figure 4 It can be seen that the efficiency of the belt conveyor system under the same working conditions decays to a certain extent over time, reflecting the decrease in the health of the system. Combining the expected efficiency value of the first sampling in Table 1 as the benchmark value and the actual experience on site, the health of the belt conveyor is divided into 4 levels, as shown in Table 2.
[0113] Table 4-2 Basis for classification of health levels under different working conditions
[0114] Table 4-2 The basis of health classification under different working conditions
[0115]
[0116]
[0117] Among them, "Excellent" and "Good" indicate that the efficiency value of the system under the current working condition is at a high level and the system is running stably; "Fair" indicates that the efficiency value of the system under the current working condition is relatively low. If it remains in this state for a long time, appropriate inspection or maintenance work should be carried out; "Early latent failure" indicates that the efficiency value of the system under the current working condition is obviously low. If it remains in this state for a long time and is not handled in time, the system may be at risk of failure.
[0118] S5. Establish a belt conveyor health status evaluation model and combine it with real-time operation data to evaluate the health status of the belt conveyor system.
[0119] The evaluation rules of the belt conveyor system are as follows: in a stable operating state except for the start-up and shutdown processes, the operating condition is judged, the efficiency value is calculated, and the health assessment is performed every 10 seconds. The evaluation results are statistically analyzed on a daily basis to obtain the overall health status of the system.
[0120] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A method for online assessment of the health status of a mining belt conveyor, characterized in that: The following steps are involved: S1. Obtaining characteristic parameters of operating conditions, including: belt conveyor speed, instantaneous coal flow rate and driving motor current; S2. Using the acquired operating condition characteristic parameters, the multidimensional time series clustering algorithm TICC method is used to identify the operating condition of the belt conveyor system and divide it into n operating conditions; S3. Compare the efficiency of belt conveyor systems under different working conditions horizontally, and fit the probability density of system operating efficiency corresponding to each working condition by normal distribution; S4. Conduct longitudinal analysis on the sampling data under the same working conditions to obtain the trend of the belt conveyor's health status change, and divide the health status evaluation level based on the actual experience on site; S5. Establish a belt conveyor health status evaluation model and combine it with real-time operation data to evaluate the health status of the belt conveyor system.
2. The method for online assessment of the health status of a mining belt conveyor according to claim 1 is characterized in that: In step S2, the specific process of the TICC algorithm used includes: The historical operation data of the equipment with a time length of T is: X=[x1,x2,x3,…x t ,…x T ] In the formula, T is the number of samples in the time series dataset, x t The n-dimensional operating data vector measured by the n sensors installed on the equipment at time t: Taking each moment t as the benchmark, we extract a data subsequence X with a window length of k. t , recorded as: X t =[x t ,x t+1 ,x t+2 ,…x t+k-1 ] Assuming that the data set is divided into M categories, the set of p sensor signals contained in the mth cluster is recorded as: IN m =[W m1 ,IN m2 ,…IN mp ] Assume that each cluster uses the inverse Gaussian covariance matrix defined as Θ m =R nk*nk , the TICC solution model is: In the formula, represents a set of symmetric Teoplitz matrices; represents the L1 norm penalty of the Hadamard product, using the incentive sparse quasi-covariance; α represents the temporal consistency coefficient; is used to check the last subsequence X t-1 The logarithmic likelihood of belonging to the mth category is expressed as: In the formula, β m represents the empirical mean of category m.
3. The method for online assessment of the health status of a mining belt conveyor according to claim 1 or 2, characterized in that: In step S3, the process of obtaining the efficiency of the belt conveyor system is as follows: From the energy perspective, the efficiency η of the belt conveyor system is expressed as: In the formula, E out Refers to the energy output of the belt conveyor, that is, the work done during the operation of the equipment; E in Refers to the total energy input of the belt conveyor system, that is, the electrical energy consumed. This part is obtained from the voltage and current data of the belt conveyor system: In the formula, U refers to the line voltage of the belt conveyor system, I refers to the line current of the belt conveyor system, It refers to the phase angle between the line voltage and the line current, and t refers to the time range corresponding to the energy input to be calculated; From the perspective of energy conversion, the operating power P provided by the main roller of the belt conveyor A It is mainly consumed by the following four parts: ① No-load operation power consumption P nl (W): P nl =gf(L h +L0)qv b In the formula, g is the acceleration due to gravity, which is 9.8 m / s 2 , f is the simulated friction factor, L h is the horizontal length of the belt conveyor roller center distance (m), L0 is the length compensation coefficient (m), which is used to compensate for the friction force component that is not related to the belt length, v b is the conveyor belt speed (m / s), q is the unit mass of the moving part of the equipment (kg / m), and the calculation formula is as follows: In the formula, is the unit mass of the rotating part of the supporting roller (kg / m), is the unit mass of the rotating part of the return roller (kg / m), q belt is the mass of the conveyor belt per unit length (kg / m); ②Power consumed in horizontal material transport P l (W): P l =gf(L h +L0)q m v b Where H is the height of the material being lifted or below (m), q m is the mass of material per unit length on the conveyor belt (kg / m); ③Power consumed in vertical material transport P h (W): P h =gHq m v b Where H is the lifting or lowering height of the material (m); ④ The power consumed by the friction between the accessories and the conveyor belt when the belt conveyor is running P ACS (W): P ACS =0.2gd 2 50 h γv b Where d is the material cross-sectional depth (m), γ is the material bulk density (kg / m 3 ); In summary, based on the perspective of energy conversion, the total effective energy E consumed by the belt conveyor for continuous material transportation for t seconds is out (J) is: Further simplified to: At the same time, there is active roller mechanical power P A (W): P A =(C1+C3+C2q m )v b In the formula, C1 = gf(L h + L0)q, C2 = gf(L h +L0)+gH,C3=0.2gd 2 L h γ.
4. The method for online assessment of the health status of a mining belt conveyor according to claim 1, characterized in that: In the step S4, a normal distribution fitting is performed on the efficiency values of the belt conveyor under different working conditions to obtain the efficiency distribution characteristics of the belt conveyor under different working conditions.
5. The method for online assessment of the health status of a mining belt conveyor according to claim 1, characterized in that: In step S4, the data collected at different times are longitudinally analyzed under the same working conditions, and the efficiency change trends of the belt conveyor over a long time scale under the same working conditions are compared. A basis for dividing the health level is formulated, and the health level of the belt conveyor system is divided into four levels: excellent, good, general, and early potential failure according to the efficiency.
6. The method for online assessment of the health status of a mining belt conveyor according to claim 1, characterized in that: In step S5, in the assessment of the health status of the belt conveyor system: under stable operating conditions except for the start-up and shutdown processes, the operating condition is judged, the efficiency value is calculated and the health assessment is performed every 10 seconds, the assessment results are statistically calculated on a daily basis, and the daily cumulative assessment results are statistically analyzed to obtain the overall health status of the system.
