Conveying belt alarm system and method based on infrared signal and dust concentration detection
Through the data fusion processing of infrared signal and dust concentration detection system, combined with Kalman filter and Bayesian network, the threshold is dynamically adjusted, which solves the problem of insufficient detection accuracy and reliability in the existing technology, and achieves efficient and accurate delivery The belt tear detection reduces the false alarm rate and ensures the safe operation of the conveyor belt system.
Patent Information
- Application Number
- CN202410955662.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-12-06
AI Technical Summary
The existing conveyor belt tear detection system is difficult to meet the needs of modern production for efficient, accurate and real-time detection due to problems such as large environmental interference and complex signal processing, resulting in insufficient detection accuracy and reliability.
The infrared signal and dust concentration detection system are adopted, and the infrared detector and dust concentration detector are arranged through a multi-point distributed array, combined with the Kalman filter and Bayesian network for data fusion processing, and the threshold is dynamically adjusted to generate the final fusion data. Improve detection accuracy and reliability.
Significantly improve the accuracy and real-timeness of conveyor belt tear detection, reduce the impact of environmental noise and sensor error, reduce false alarm rate, improve detection sensitivity and adaptability, ensure timely detection of tear failures, and provide delivery The safe operation with the system provides strong technical guarantees.
Smart Images

Figure CN119091593A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of conveyor belt alarms, in particular to a conveyor belt alarm system and method based on infrared signals and dust concentration detection. Background Art
[0002] With the continuous improvement of industrial automation, conveyor belt systems have been widely used in mining, metallurgy, ports and other fields. As the core component of continuous transportation equipment, the working state of the conveyor belt directly affects the efficiency and safety of the entire production system. Conveyor belt tearing is one of the common and serious faults in the operation of the conveyor belt. It may cause material leakage and production stagnation at the least, and may cause equipment damage and personal injury at the worst. Traditional conveyor belt tearing detection methods mainly rely on manual inspections and simple mechanical sensors. These methods often have problems such as low detection sensitivity, slow response speed and high false alarm rate, which are difficult to meet the needs of modern production for efficient, accurate and real-time detection. With the development of sensing technology and intelligent algorithms, new detection methods are gradually applied to conveyor belt tearing detection, including detection systems based on infrared signals and dust concentration sensors. However, the existing detection systems still have certain limitations in practical applications, such as large environmental interference and complex signal processing. Therefore, how to improve the accuracy and reliability of conveyor belt tearing detection systems has become an important topic in current research and application. Summary of the invention
[0003] In view of the above problems existing in the prior art, the present invention is proposed.
[0004] Therefore, the problem to be solved by the present invention is to improve the accuracy and reliability of the conveyor belt tear detection system.
[0005] To solve the above technical problems, the present invention provides the following technical solutions: a conveyor belt alarm system based on infrared signals and dust concentration detection, comprising: infrared detectors, which are arranged at both longitudinal ends of the conveyor belt in a multi-point distributed array, below the conveyor belt bearing surface between each group of roller supports, for transmitting and receiving infrared signals; a dust concentration detector, which is installed between the conveyor belt bearing surface and the return surface, for detecting dust concentration; a data processing unit, which receives and processes data from the infrared detector and the dust concentration detector, and fuses the data using a fusion algorithm to generate preliminary fused data; a threshold dynamic adjustment module, which dynamically adjusts the threshold of tear detection based on the fused data; and an alarm unit, which sends an alarm signal when the fused data exceeds the dynamically adjusted threshold.
[0006] As a preferred solution of the conveyor belt alarm system based on infrared signals and dust concentration detection described in the present invention, the data processing unit includes: a Kalman filter for receiving preliminary fusion data, performing state estimation and updating, and generating final fusion data; a Bayesian network module for processing the data dependency relationship between the infrared detector and the dust concentration detector, and calculating preliminary fusion data.
[0007] As a preferred solution of the conveyor belt alarm system based on infrared signal and dust concentration detection described in the present invention, the prediction unit predicts the state of the next time step based on the initial state estimation and process noise; the update unit updates the state estimation and covariance in combination with the observed noise and preliminary fusion data.
[0008] As a preferred solution of the conveyor belt alarm system based on infrared signal and dust concentration detection of the present invention, the prediction unit calculates the state of the next time step by the following formula:
[0009]
[0010] P k|k-1 =AP k-1|k-1 A T +Q
[0011] in is the predicted state at time step k, A is the state transfer matrix, is the state estimation of the time step, A is the control input matrix, B is the control input, u k -1 is the prediction covariance at time step k, and Q is the process noise covariance.
[0012] As a preferred solution of the conveyor belt alarm system based on infrared signal and dust concentration detection of the present invention, the updating unit updates the state estimation and covariance by the following formula:
[0013] K k =P k|k-1 H T (HP k|k-1 H T +R) -1
[0014]
[0015] P k|k =(IK k H)P k|k-1
[0016] Where K k is the Kalman gain, R is the observation noise covariance, H is the observation matrix, z k is the observed value, and I is the unit matrix.
[0017] As a preferred solution of the conveyor belt alarm system based on infrared signal and dust concentration detection described in the present invention, the Bayesian network module includes: a conditional probability calculation unit, which calculates the conditional probability of the infrared signal and the dust concentration according to historical data; a joint probability calculation unit, which calculates the joint probability distribution based on the conditional probability and the current sensor data to generate preliminary fusion data; the conditional probability of the conditional probability calculation unit is determined by experimental data, when the difference between the infrared signal and the dust concentration is small, the conditional probability is high, and when the difference is large, the conditional probability is low; the Bayesian network calculation formula is as follows:
[0018]
[0019] Among them, P(S|I, C) is the conditional probability that the system state is tearing, and P(I, C|S) is the joint conditional probability of infrared signal and dust concentration when the system state is tearing, which is determined by experimental data; P(S) is the prior probability that the system state is tearing, and P(I, C) is the joint probability of infrared signal and dust concentration.
[0020] As a preferred solution of the conveyor belt alarm system based on infrared signal and dust concentration detection of the present invention, the preliminary fusion data is calculated by the following formula:
[0021] F(t)=α·X infrared (t)+(1-α)·X dust (t)
[0022] Among them, α is the weight coefficient obtained by Bayesian network reasoning, reflecting the importance of infrared signals and dust concentration to the system state judgment, and X infrared (t) and X dust (t) are infrared signals and dust concentration data respectively.
[0023] As a preferred solution of the conveyor belt alarm system based on infrared signal and dust concentration detection described in the present invention, the threshold dynamic adjustment module includes a threshold calculation unit, which calculates the dynamic threshold based on the state estimation and covariance of the Kalman filter; and a threshold comparison unit, which compares the fused data with the dynamic threshold to determine whether to issue an alarm signal.
[0024] As a preferred solution of the conveyor belt alarm system based on infrared signal and dust concentration detection of the present invention, the threshold calculation unit calculates the dynamic threshold using the following formula:
[0025]
[0026] Threshold t is the dynamic threshold at the current moment, is the state estimate at the current moment, P t is the covariance at the current moment.
[0027] The present invention also provides a conveyor belt tear alarm method, which includes a conveyor belt alarm system based on infrared signals and dust concentration detection, and also includes the following steps: collecting data from infrared detectors and dust concentration detectors; using a Bayesian network to calculate the joint probability distribution of sensor data to generate preliminary fusion data; using a Kalman filter to perform state estimation and update the preliminary fusion data to generate final fusion data; based on the final fusion data, calculating a dynamic threshold; comparing the fusion data with the dynamic threshold, and when the fusion data exceeds the dynamic threshold, issuing an alarm signal.
[0028] The beneficial effects of the present invention are as follows: the present invention significantly improves the accuracy and real-time performance of conveyor belt tear detection by adopting data fusion processing of infrared signals and dust concentration sensors. The combination of Kalman filter and Bayesian network effectively reduces the influence of environmental noise and sensor error on the detection results, thereby reducing the false alarm rate. At the same time, the introduction of dynamic adjustment threshold enables the system to automatically optimize the alarm judgment criteria according to real-time data changes, further improving the detection sensitivity and adaptability. Compared with the traditional fixed threshold detection method, the present invention has higher reliability and stability in dealing with complex working conditions and changing environments. Through comparison of experimental data, it is proved that the scheme of the present invention reduces false alarms while ensuring the timely detection of tear faults, providing a strong technical guarantee for the safe operation of the conveyor belt system. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:
[0030] Figure 1 Schematic diagram of the conveyor belt alarm system based on infrared signals and dust concentration detection.
[0031] Figure 2 The figure is a flow chart of a conveyor belt alarm method based on infrared signals and dust concentration detection.
[0032] Figure 3 This is a schematic diagram of the infrared signal installation location of the conveyor belt alarm system based on infrared signals and dust concentration detection. DETAILED DESCRIPTION
[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.
[0034] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0035] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is mutually exclusive with other embodiments, either individually or selectively.
[0036] Example 1
[0037] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and this embodiment provides a conveyor belt alarm system based on infrared signals and dust concentration detection. The conveyor belt alarm system based on infrared signals and dust concentration detection includes an infrared detector 100, a dust concentration detector 200, a data processing unit 300, a threshold dynamic adjustment module 400 and an alarm unit 500.
[0038] Specifically, the infrared detectors 100 are arranged at both longitudinal ends of the conveyor belt in a multi-point distributed array, below the conveyor belt bearing surface between each group of roller supports, for transmitting and receiving infrared signals; the dust concentration detector 200 is installed between the conveyor belt bearing surface and the return surface, for detecting dust concentration; the data processing unit 300 receives and processes the data of the infrared detector and the dust concentration detector, fuses the data using a fusion algorithm, and generates preliminary fusion data; the threshold dynamic adjustment module 400 dynamically adjusts the threshold of tear detection based on the fusion data; the alarm unit 500 sends an alarm signal when the fusion data exceeds the dynamically adjusted threshold.
[0039] Specifically, the data processing unit 300 includes: a Kalman filter 301 for receiving preliminary fusion data, performing state estimation and updating, and generating final fusion data; a Bayesian network module 302 for processing the data dependency between the infrared detector and the dust concentration detector, and calculating preliminary fusion data.
[0040] Specifically, the Kalman filter includes: a prediction unit 301a, which predicts the state of the next time step based on the initial state estimate and process noise; an update unit 301b, which updates the state estimate and covariance in combination with the observation noise and the preliminary fusion data.
[0041] Specifically, the prediction unit 301a calculates the state of the next time step by the following formula:
[0042]
[0043] P k|k-1 =AP k-1|k-1 A T +Q
[0044] in is the predicted state at time step k, A is the state transfer matrix, is the state estimation of the time step, A is the control input matrix, B is the control input, u k -1 is the prediction covariance at time step k, Q is the process noise covariance;
[0045] The state transfer matrix A is used to describe the dynamic changes of the system state, and combined with the time step, it provides a prediction of the next state. The control input matrix B describes the impact of the control input on the system state. In this embodiment, I have no external control input that affects the system state, so the control input matrix B is a zero matrix. The process noise covariance matrix Q represents the degree of random disturbance within the system and reflects the accuracy of the model.
[0046] Specifically, the updating unit 301b updates the state estimation and covariance by the following formula:
[0047] K k =P k|k-1 H T (HP k|k-1 H T +R) -1
[0048]
[0049] P k|k =(IK k H)P k|k-1
[0050] Where K k is the Kalman gain, R is the observation noise covariance, H is the observation matrix, z k is the observed value, I is the unit matrix;
[0051] The observation matrix H describes how to obtain observation values from the system state. In this embodiment, we directly observe the infrared signal and dust concentration, so the observation matrix is a unit matrix I. The Kalman filter estimates and adjusts the system state through two steps of prediction and update, thereby improving the accuracy and reliability of the system.
[0052] Specifically, the Bayesian network module includes: a conditional probability calculation unit 302a, which calculates the conditional probability of the infrared signal and the dust concentration according to the historical data; a joint probability calculation unit 302b, which calculates the joint probability distribution based on the conditional probability and the current sensor data to generate preliminary fusion data; the conditional probability of the conditional probability calculation unit 302a is determined by the experimental data, when the difference between the infrared signal and the dust concentration is small, the conditional probability is high, when the difference is large, the conditional probability is low, wherein, when the difference between the infrared signal and the dust concentration is less than 1.5, the conditional probability that the system state is torn is greater than 0.7; the Bayesian network calculation formula is as follows:
[0053]
[0054] Among them, P(S|I, C) is the conditional probability that the system state is tearing, and P(I, C|S) is the joint conditional probability of infrared signal and dust concentration when the system state is tearing, which is determined by experimental data; P(S) is the prior probability that the system state is tearing, and P(I, C) is the joint probability of infrared signal and dust concentration.
[0055] Specifically, the preliminary fusion data is calculated by the following formula:
[0056] F(t)=α·X infrared (t)+(1-α)·X dust (t)
[0057] Among them, α is the weight coefficient obtained by Bayesian network reasoning, reflecting the importance of infrared signals and dust concentration to the system state judgment, X infrared (t) and X dust (t) are infrared signals and dust concentration data, respectively. In this embodiment, a large number of experiments have shown that α is approximately equal to 0.7.
[0058] Specifically, the threshold dynamic adjustment module 400 includes a threshold calculation unit 401, which calculates the dynamic threshold based on the state estimation and covariance of the Kalman filter; and a threshold comparison unit 402, which compares the fused data with the dynamic threshold to determine whether to issue an alarm signal.
[0059] Specifically, the threshold calculation unit 401 calculates the dynamic threshold using the following formula:
[0060]
[0061] Threshold t is the dynamic threshold at the current moment, is the state estimate at the current moment, P t is the covariance at the current moment.
[0062] Example 2
[0063] In this embodiment, the calculation process of each parameter and the experimental data and initial values used are described in detail. First, the state transfer matrix, process noise covariance matrix and observation noise covariance matrix are calculated through experimental data, and then the Kalman filter and Bayesian network are applied for data fusion, and finally the alarm judgment is made through the dynamic threshold adjustment module.
[0064] The experimental data used are as follows:
[0065]
[0066]
[0067] Using the above experimental data, calculate the initial state estimate And the initial covariance matrix P0:
[0068] 1. Calculate the mean of infrared signal and dust concentration:
[0069]
[0070] 2. Calculate the initial state estimate:
[0071]
[0072] 3. Calculate the initial covariance matrix:
[0073]
[0074] Among them, Var(X IR ) and Var(X D ) are the variances of infrared signal and dust concentration, Cov(X IR ,X D ) is the covariance between the two.
[0075]
[0076]
[0077] Calculated
[0078] Var(X IR )=0.0556
[0079] Var(XD )=0.5556
[0080] Cov(X IR , X D )=0.1333
[0081] therefore
[0082]
[0083] Calculate the state transfer matrix A, process noise covariance matrix Q, and measurement noise covariance matrix R:
[0084] 1. State transfer matrix A: Estimate A using the least squares method
[0085]
[0086] Bring in data for calculation:
[0087]
[0088]
[0089]
[0090] Through the above calculation process, we get the state transfer matrix A. This matrix describes the change relationship between the system state between time steps, including infrared signals and dust concentration detection data. This matrix not only reflects the dependency between sensor data, but also can help the Kalman filter to estimate and update the state more accurately.
[0091] 2. Process noise covariance matrix Q: Based on the prediction error, calculate the error covariance matrix Q
[0092]
[0093] Calculated
[0094]
[0095] 2. Measurement noise covariance matrix R: Estimate R through sensor measurement data
[0096]
[0097] The observation matrix H describes how to obtain observation values from the system state. In this embodiment, we directly observe the infrared signal and the dust concentration, so the observation matrix H is an identity matrix I.
[0098] Calculated
[0099]
[0100] Example 3
[0101] This embodiment uses the coefficients calculated above to illustrate the superiority of the present invention through comparison of experimental data.
[0102]
[0103] This table shows the infrared signal, dust concentration, whether the fixed threshold alarms, false alarms, preliminary fusion data, final fusion data, dynamic thresholds, and comparisons of dynamic threshold alarms and false alarms at each time step. From the above data, it can be seen that the use of the final fusion data and dynamic threshold method can effectively reduce false alarms.
[0104] Example 4
[0105] This embodiment provides a conveyor belt tear alarm method, the steps include:
[0106] S1: Collect data from infrared detectors and dust concentration detectors.
[0107] S2: Use the Bayesian network to calculate the joint probability distribution of sensor data and generate preliminary fusion data.
[0108] S3: Use the Kalman filter to perform state estimation and update on the preliminary fusion data to generate the final fusion data.
[0109] S4: Calculate the dynamic threshold based on the final fused data.
[0110] S5: Compare the fused data with the dynamic threshold, and send out an alarm signal when the fused data exceeds the dynamic threshold.
[0111] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
Claims
1. The conveyor belt alarm system based on infrared signal and dust concentration detection is characterized by: include, Infrared detectors (100) are arranged at both ends of the conveyor belt in the longitudinal direction, in a multi-point distributed array arrangement, below the conveyor belt bearing surface between each set of roller brackets, for transmitting and receiving infrared signals; A dust concentration detector (200) is installed between the carrying surface and the return surface of the conveyor belt and is used to detect the dust concentration; A data processing unit (300) receives and processes data from the infrared detector and the dust concentration detector, fuses the data using a fusion algorithm, and generates preliminary fusion data; A threshold dynamic adjustment module (400) dynamically adjusts the threshold of tear detection based on the fused data; The alarm unit (500) sends out an alarm signal when the fused data exceeds the dynamically adjusted threshold value.
2. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 1, characterized in that: The data processing unit (300) comprises: a Kalman filter (301) for receiving preliminary fusion data, performing state estimation and updating, and generating final fusion data; The Bayesian network module (302) is used to process the data dependency between the infrared detector and the dust concentration detector and calculate preliminary fusion data.
3. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 2, characterized in that: The Kalman filter comprises: a prediction unit (301a) for predicting the state of the next time step based on the initial state estimate and process noise; The updating unit (301b) combines the observation noise and the preliminary fusion data to update the state estimate and covariance.
4. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 3, characterized in that: The prediction unit (301a) calculates the state of the next time step by the following formula: P k|k-1 =AP k-1|k-1 From T +Q in is the predicted state at time step k, A is the state transfer matrix, is the state estimation of the time step, A is the control input matrix, B is the control input, u k-1 is the prediction covariance at time step k, and Q is the process noise covariance.
5. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 4, characterized in that: The updating unit (301b) updates the state estimation and covariance by the following formula: K k =P k|k-1 H T (HP k|k-1 H T +R) -1 P k|k =(I-K k H)P k|k-1 Where K k is the Kalman gain, R is the observation noise covariance, H is the observation matrix, z k is the observed value, and I is the unit matrix.
6. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 5, characterized in that: The Bayesian network module includes: A conditional probability calculation unit (302a), which calculates the conditional probability of the infrared signal and the dust concentration according to the historical data; A joint probability calculation unit (302b), based on the conditional probability and the current sensor data, calculates the joint probability distribution and generates preliminary fusion data; The conditional probability of the conditional probability calculation unit (302a) is determined by experimental data, wherein when the difference between the infrared signal and the dust concentration is less than 1.5, the conditional probability that the system state is tearing is greater than 0.7, and when the difference between the infrared signal and the dust concentration is greater than 1.5, the conditional probability that the system state is tearing is less than 0.3; The Bayesian network calculation formula is as follows: Among them, P(S|I, C) is the conditional probability that the system state is tearing, and P(I, C|S) is the joint conditional probability of infrared signal and dust concentration when the system state is tearing, which is determined by experimental data; P(S) is the prior probability that the system state is tearing, and P(I, C) is the joint probability of infrared signal and dust concentration.
7. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 6, characterized in that: The preliminary fusion data is calculated by the following formula: F(t)=α·X infrared (t)+(1-α)·X dust (t) Among them, α is the weight coefficient obtained by Bayesian network reasoning, reflecting the importance of infrared signals and dust concentration to the system state judgment, and X infrared (t) and X dust (t) are infrared signals and dust concentration data respectively.
8. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 7, characterized in that: The threshold dynamic adjustment module (400) comprises a threshold calculation unit (401) for calculating a dynamic threshold based on a state estimation and covariance of a Kalman filter; The threshold comparison unit (402) compares the fused data with the dynamic threshold to determine whether to issue an alarm signal.
9. The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in claim 8, characterized in that: The threshold calculation unit (401) calculates the dynamic threshold using the following formula: Threshold t is the dynamic threshold at the current moment, is the state estimate at the current moment, P t is the covariance at the current moment.
10. The conveyor belt alarm method based on infrared signal and dust concentration detection is characterized by: The conveyor belt alarm system based on infrared signal and dust concentration detection as claimed in any one of claims 1 to 9 further comprises: The following steps are included: Collect data from infrared detectors and dust concentration monitors; Use Bayesian networks to calculate the joint probability distribution of sensor data and generate preliminary fusion data; Use Kalman filter to estimate and update the state of preliminary fusion data to generate final fusion data; Based on the final fused data, calculate the dynamic threshold; The fused data is compared with the dynamic threshold, and an alarm signal is issued when the fused data exceeds the dynamic threshold.
