Intelligent detection method for roller bar of roller kiln
By installing sensors in the roller kiln roller rod and using intelligent algorithms to detect the status of the roller rod, the problem of inaccurate monitoring of roller rod usage status in the prior art is solved, real-time monitoring of roller rod wear and fracture is achieved, and production safety and efficiency are improved.
Patent Information
- Application Number
- CN202510171188.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-17
- Publication Date
- 2025-04-18
AI Technical Summary
The prior art is difficult to accurately monitor the use status of roller kiln roller rods in a timely manner, resulting in frequent problems such as roller rod breakage, causing safety hazards and low production efficiency.
Using intelligent detection method, by installing torque sensors and proximity sensors in the roller kiln roller rods, the torque and rotation signals of the roller rods are detected in real time, and combined with intelligent algorithms to calculate the wear rate and status of the roller rods, reminding operators to replace the roller rods in time.
It realizes comprehensive monitoring of the use status of roller rods, improves the accuracy and timeliness of detection, reduces safety hazards and production accidents caused by roller rod problems, and improves overall production efficiency and safety.
Smart Images
Figure CN119845028A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of roller kiln roller detection, and in particular relates to an intelligent detection method for roller kiln rollers. Background Art
[0002] The roller kiln is a type of tunnel kiln. Its full name is roller tunnel kiln, also known as roller bottom kiln. The working principle of this kiln is relatively simple and efficient: first, the various products to be fired are placed directly or indirectly on the rollers, and then the rollers are continuously rotated so that the fired products can move forward steadily in a predetermined order. The ends of each roller are connected by a precision sprocket and driven by the chain for orderly operation. In order to ensure the smoothness and safety of the transmission process, the chains are usually divided into several groups for multi-stage transmission. In the low-temperature area of the kiln, the rollers are generally made of high-performance materials such as heat-resistant nickel-chromium alloy steel to cope with the appropriate working temperature. In the high-temperature area, high-temperature resistant alumina or silicon carbide ceramic rods are mainly used as roller materials. In terms of heating source, the roller kiln usually uses gas or electric heating elements, such as silicon carbon rods, silicon molybdenum rods, resistance wires, etc., to provide stable and uniform heat.
[0003] Roller kiln plays a vital role in the production and application of many industries such as ceramic tiles and powder materials. As a continuous kiln, it has the significant advantage of low energy consumption. However, this also puts forward high requirements on the material selection and installation technology of rollers. The roller structure of the roller kiln is carefully composed of rollers arranged in parallel and crossing the cross-section of the kiln working channel. The products can be placed directly on these rollers, or they can be placed on the pads first, and then the rollers are driven to rotate by the efficient transmission system, so that the products are smoothly transported into the kiln as the rollers rotate, and the key firing process is completed in the kiln.
[0004] In recent years, roller kilns have been used more and more widely. The operators of the kilns need to observe the rotation of the rollers carefully with their naked eyes through regular inspections. However, even so, it is difficult to accurately judge the status of the rollers. Once a roller breaks, it will be difficult to quickly and accurately detect the specific location of the broken roller, which can easily cause a kiln blockage accident and bring huge safety hazards to the production process.
[0005] Under existing technical conditions, the means of detecting rollers in roller kilns are relatively limited. They can only detect whether there are broken rollers, but cannot conduct a comprehensive inspection of the actual usage status of the rollers. This limitation often leads to sudden roller breakage during the sintering process, and at this time, it is no longer possible to replace the rollers in time during sintering, which in turn leads to a series of problems such as displacement, collapse, and jamming of the sintered workpiece, which not only greatly reduces the sintering effect, but may even cause more serious safety accidents. Summary of the invention
[0006] In order to effectively solve the technical problem that the current technical field cannot accurately monitor the use status of rollers in real time, the purpose of the present invention is to provide an intelligent detection method for rollers of a roller kiln.
[0007] In order to achieve the above object, the present invention adopts the following technical solution: An intelligent detection method for a roller bar of a roller kiln, wherein the roller bar of the roller kiln is a stepped shaft, that is, the roller bar of the roller kiln has a roller bar central shaft and a roller driving shaft and a roller driven shaft located at both ends of the roller bar central shaft; the diameters of the roller driving shaft and the roller driven shaft are both smaller than the diameter of the roller bar central shaft; the end of the roller driving shaft is equipped with a torque sensor for collecting signals from the roller driving shaft, and the torque sensor is connected to a transmission mechanism through a transmission connector; the torque sensor detects the output load of the connected transmission structure in real time, which can be converted into the weight of the connected roller bar by calculation, and the roller bar wear rate can be calculated; the roller bar wear rate calculation formula is: wear rate (%) = ×100; in Indicates the no-load torque of the roller in the initial state, that is, the torque value when the roller is just installed; Indicates the no-load torque of the roller in the current state, that is, the torque value after a period of use; A roller driven shaft detection block is installed on the roller driven shaft, and a proximity sensor is provided corresponding to the roller driven shaft detection block, and the proximity sensor is used to detect whether the roller driven shaft rotates with the roller driving end; The specific steps of the intelligent detection method are as follows: Step 1: No-load detection: Before loading the kiln, check whether all roller driven shafts 1 fixed to the roller driven shaft fixing parts and the roller driving shaft fixed to the roller driving shaft fixing block are installed correctly and reliably, and then perform no-load self-test operation. By measuring the value of the torque sensor connected to the transmission connection and the transmission structure, the torque sensor is reliably connected to the roller driving shaft, and the current roller weight is indirectly calculated through the intelligent algorithm. At the same time, the proximity sensor detects that the roller driven shaft detection block fixed on the roller driven shaft can obtain the roller driven shaft rotation signal to determine whether the roller driven shaft rotates with the transmission structure, thereby indirectly determining whether the roller driven shaft or the roller middle shaft or the roller driving shaft is broken or deformed; Step 2: Intelligent reminder: If the detected roller is a low-temperature metal roller, the roller driven shaft rotation signal is valid periodically, and the current roller no-load weight is basically consistent with the previous roller no-load weight value, it can be judged that the current roller state is normal; if the roller driven shaft rotation signal is non-periodic and discontinuous, and the difference between the current roller no-load weight and the previous roller no-load weight value is more than 5%, it can be judged that the current roller state is abnormal and may be deformed, and it should be reminded to check and replace the roller in time; If the detected roller is a high-temperature non-metallic roller, the roller driven shaft rotation signal is valid periodically, and the current roller no-load weight is basically consistent with the previous roller no-load weight value, it can be judged that the current roller state is normal; if the roller driven shaft rotation signal is valid periodically, and the difference between the current roller no-load weight and the new roller no-load weight is more than 20%, it can be judged that the current roller state is abnormal and there may be a risk of breakage, and you should be reminded to check and replace the roller in time; if the roller driven shaft rotation signal is continuous or non-periodic valid, and the difference between the current roller no-load weight 11 and the previous roller no-load weight is more than 5%, it can be judged that the current roller state is abnormal and there is a break, and you should be reminded to check and replace the roller in time; Step 3: Sintering detection: After the roller kiln is loaded and started to operate, the non-metallic rollers in the high-temperature zone are subjected to cyclic inspection, and the value of the torque sensor and the detection signal of the proximity sensor are read in real time to calculate the current roller load weight and determine the driven shaft rotation signal. The intelligent algorithm compares and calculates the current roller load weight, the weight of the previous cycle monitoring roller, and the load weights of 10 adjacent groups of rollers in real time. If the difference between the current roller load weight and the previous cycle monitoring roller weight and the average load weight of 10 adjacent groups of rollers continues to be more than 5%, and the driven shaft rotation signal is abnormal, it can be determined that the current roller is broken, and it should be reminded to check and replace the roller in time after the end of this operation.
[0008] The description of the intelligent algorithm is as follows: 1) Algorithm input: input torque sensor data, proximity sensor data and historical data; the torque sensor data: real-time detection of the roller torque value, used to calculate the roller weight; the proximity sensor data: detection of the roller driven shaft rotation signal, to determine whether the roller rotates normally; the historical data includes the last no-load weight, the no-load weight of the new roller, and the average load weight of adjacent rollers; 2) Algorithm output: roller status: normal, deformed, broken; early warning prompt: whether the roller needs to be checked or replaced.
[0009] 3) Algorithm steps: Data collection: The torque value of the current roller is obtained through the torque sensor and converted into the current roller weight; the rotation signal of the roller driven shaft is obtained through the proximity sensor to determine whether the roller rotates normally; No-load detection: Before loading the kiln, perform no-load detection to calculate the current no-load weight of the roller. Compare the difference between the current no-load weight and the last no-load weight to determine whether the roller is deformed or worn; Sintering detection: During the operation of the kiln, the load weight and rotation signal of the roller are monitored in real time. Compare the difference between the current load weight and the historical data (such as the weight monitored in the previous cycle and the average load weight of adjacent rollers); if the weight difference continues to exceed the threshold (such as 5%) and the rotation signal is abnormal, it is judged that the roller may be broken; Status judgment and early warning: judge the status of the roller (normal, deformed, broken) according to the weight change and rotation signal; if the status is abnormal, issue an early warning to remind the operator to check or replace the roller; 4) Algorithm core logic: Ignoring the effect of temperature, the relationship between weight and torque can be simplified to: , Where, T represents the torque of the detection roller, τ represents the allowable shear stress of the material (MPa), ρ represents the material density (kg / m³), L represents the roller length (m), and g represents the acceleration due to gravity (9.8 m / s²). ; Weight change analysis: 100%; among which, is the current roller weight, It is the historical weight (such as the last empty weight or the average weight of adjacent rollers). If ΔW exceeds the set threshold (such as 5% or 20%, etc.), the roller state is judged to be abnormal.
[0010] Rotation signal analysis: If the rotation signal of the roller driven shaft is non-periodic or discontinuous, it is judged that the roller may be broken or deformed.
[0011] The proximity sensor can be a photoelectric, mechanical or proximity switch, a displacement sensor, a 2D / 3D machine vision sensor, etc.
[0012] The torque sensor is a static or dynamic torque sensor.
[0013] The roller driven shaft refers to the end of the roller connected to the side wall of the kiln. The roller and the side wall are connected by a bearing and its support, which plays a role in axially fixing the roller.
[0014] By adopting the above technical solution, the present invention can realize comprehensive monitoring of the use status of the roller, including key information such as the degree of wear, bending condition and whether the roller is broken. This real-time monitoring method greatly improves the accuracy and timeliness of detection, allowing operators to obtain early warning prompts in time before the roller reaches the replacement standard, thereby ensuring that the roller is replaced before the sintering process begins; in addition, the intelligent algorithm provided by the patent of this invention can also scientifically evaluate the use status of the roller based on the collected data, providing strong data support for decision-making in the production process. This not only helps to reduce the probability of unsuccessful sintering or other production accidents caused by roller problems, but also can further improve the overall production efficiency and safety, laying a solid foundation for the steady development of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a schematic diagram of a roller mechanism.
[0016] Figure 2 The invention is a flow chart of an intelligent detection method for roller bars of a roller kiln.
[0017] In the figure: 1. roller driven shaft, 2. roller driven shaft detection block, 3. proximity sensor, 4. roller driven shaft fixing part, 5. roller middle shaft, 6. roller driving shaft fixing block, 7. roller driving shaft, 8. torque sensor, 9. transmission connecting part, 10. transmission structure, 11. current roller no-load weight, 12. roller driven shaft rotation signal, 13. last roller no-load weight, 14. new roller no-load weight, 15. current roller load weight, 16. upper cycle monitoring roller weight, 17. average load weight of 10 adjacent roller groups. DETAILED DESCRIPTION
[0018] The present invention is described in detail with reference to the accompanying drawings and specific embodiments: like Figure 1 As shown, an intelligent detection method for rollers of a roller kiln is provided. The rollers of the roller kiln are stepped shafts, that is, the rollers of the roller kiln have a roller center shaft 5 and a roller driving shaft 7 and a roller driven shaft 1 located at both ends of the roller center shaft; the diameters of the roller driving shaft and the roller driven shaft are both smaller than the diameter of the roller center shaft; a torque sensor for collecting signals from the roller driving shaft is installed at the end of the roller driving shaft, and the torque sensor is connected to the transmission mechanism through a transmission connector; the torque sensor detects the output load of the connected transmission structure in real time, which can be converted into the weight of the connected roller by calculation, and the roller wear rate can be calculated; the roller wear rate calculation formula is: wear rate (%) = ×100; in Indicates the no-load torque of the roller in the initial state, that is, the torque value when the roller is just installed; Indicates the no-load torque of the roller in the current state, that is, the torque value after a period of use; A roller driven shaft detection block is installed on the roller driven shaft, and a proximity sensor is provided corresponding to the roller driven shaft detection block, and the proximity sensor is used to detect whether the roller driven shaft rotates with the roller driving end; The specific steps of the intelligent detection method are as follows: Step 1: No-load detection: Before loading the kiln, check whether all roller driven shafts 1 fixed to the roller driven shaft fixing member 4 and the roller driving shaft 7 fixed to the roller driving shaft fixing block 6 are installed correctly and reliably, and then perform no-load self-check operation. By measuring the value of the torque sensor 8 connected to the transmission connection member 9 and the transmission structure 10, the torque sensor 8 is reliably connected to the roller driving shaft 7. The current roller weight 11 can be indirectly calculated through the intelligent algorithm. At the same time, the proximity sensor 3 detects that the roller driven shaft detection block 2 fixed on the roller driven shaft 3 can obtain the roller driven shaft rotation signal 12 to determine whether the roller driven shaft 1 follows the transmission structure 10 for rotational movement, thereby indirectly determining whether the roller driven shaft 1 or the roller middle shaft 5 or the roller driving shaft 7 is broken or deformed; Step 2: Intelligent reminder: If the detected roller is a low-temperature metal roller, the roller driven shaft rotation signal 12 is periodically valid, and the current roller no-load weight 11 is basically consistent with the previous roller no-load weight 13, it can be judged that the current roller state is normal; if the roller driven shaft rotation signal 12 is non-periodic and discontinuous, and the difference between the current roller no-load weight 11 and the previous roller no-load weight 13 is more than 5%, it can be judged that the current roller state is abnormal and may be deformed, and you should be reminded to check and replace the roller in time. If the detected roller is a high-position non-metallic roller, the roller driven shaft rotation signal 12 is periodically valid, and the current roller no-load weight 11 is basically consistent with the previous roller no-load weight 13, it can be judged that the current roller state is normal; if the roller driven shaft rotation signal 12 is periodically valid, and the difference between the current roller no-load weight 11 and the new roller no-load weight 14 is more than 20%, it can be judged that the current roller state is abnormal and there may be a risk of breakage, and you should be reminded to check and replace the roller in time; if the roller driven shaft rotation signal 12 is continuous or non-periodically valid, and the difference between the current roller no-load weight 11 and the previous roller no-load weight 13 is more than 5%, it can be judged that the current roller state is abnormal and there is a break, and you should be reminded to check and replace the roller in time.
[0019] Step 3: Sintering detection: After the roller kiln is loaded and started to operate, the non-metallic rollers in the high-temperature zone are subjected to cyclic inspection, and the value of the torque sensor 8 and the detection signal of the proximity sensor 3 are read in real time to calculate the current roller load weight 15 and judge the driven shaft rotation signal 12. The intelligent algorithm compares and calculates the current roller load weight 15, the upper cycle monitoring roller weight 16, and the adjacent 10 groups of roller load weights 17 in real time. If the difference between the current roller load weight 15 and the upper cycle monitoring roller weight 16 and the average load weight 17 of the adjacent 10 groups of rollers is continuously above 5%, and the driven shaft rotation signal 12 is in an abnormal state, it can be determined that the current roller is broken, and it should be reminded to check and replace the roller in time after the end of this operation.
[0020] The description of the intelligent algorithm is as follows: 1) Algorithm input: input torque sensor data, proximity sensor data and historical data; the torque sensor data: real-time detection of the roller torque value, used to calculate the roller weight; the proximity sensor data: detection of the roller driven shaft rotation signal, to determine whether the roller rotates normally; the historical data includes the last no-load weight, the no-load weight of the new roller, and the average load weight of adjacent rollers; 2) Algorithm output: roller status: normal, deformed, broken; early warning prompt: whether the roller needs to be checked or replaced.
[0021] 3) Algorithm steps: Data collection: The torque value of the current roller is obtained through the torque sensor and converted into the current roller weight; the rotation signal of the roller driven shaft is obtained through the proximity sensor to determine whether the roller rotates normally; No-load detection: Before loading the kiln, perform no-load detection to calculate the current no-load weight of the roller. Compare the difference between the current no-load weight and the last no-load weight to determine whether the roller is deformed or worn; Sintering detection: During the operation of the kiln, the load weight and rotation signal of the roller are monitored in real time. Compare the difference between the current load weight and the historical data (such as the weight monitored in the previous cycle and the average load weight of adjacent rollers); if the weight difference continues to exceed the threshold (such as 5%) and the rotation signal is abnormal, it is judged that the roller may be broken; Status judgment and early warning: judge the status of the roller (normal, deformed, broken) according to the weight change and rotation signal; if the status is abnormal, issue an early warning to remind the operator to check or replace the roller; 4) Algorithm core logic: Ignoring the effect of temperature, the relationship between weight and torque can be simplified to: , Where, T represents the torque of the detection roller, τ represents the allowable shear stress of the material (MPa), ρ represents the material density (kg / m³), L represents the roller length (m), and g represents the acceleration due to gravity (9.8 m / s²). ; Weight change analysis: 100%; among which, is the current roller weight, It is the historical weight (such as the last empty weight or the average weight of adjacent rollers). If ΔW exceeds the set threshold (such as 5% or 20%, etc.), the roller state is judged to be abnormal.
[0022] Rotation signal analysis: If the rotation signal of the roller driven shaft is non-periodic or discontinuous, it is judged that the roller may be broken or deformed.
Claims
1. An intelligent detection method for rollers of a roller kiln, characterized in that: The roller of the roller kiln is a stepped shaft, that is, the roller of the roller kiln has a roller center shaft and a roller driving shaft and a roller driven shaft located at both ends of the roller center shaft; the diameters of the roller driving shaft and the roller driven shaft are both smaller than the diameter of the roller center shaft; the end of the roller driving shaft is equipped with a torque sensor for collecting signals from the roller driving shaft, and the torque sensor is connected to the transmission mechanism through a transmission connector; the torque sensor detects the output load of the connected transmission structure in real time, which can be converted into the weight of the connected roller by calculation, and the roller wear rate can be calculated; the roller wear rate calculation formula is: wear rate (%) = ×100; in Indicates the no-load torque of the roller in the initial state, that is, the torque value when the roller is just installed; Indicates the no-load torque of the roller in the current state, that is, the torque value after a period of use; A roller driven shaft detection block is installed on the roller driven shaft, and a proximity sensor is provided corresponding to the roller driven shaft detection block, and the proximity sensor is used to detect whether the roller driven shaft rotates with the roller driving end; The specific steps of the intelligent detection method are as follows: Step 1: No-load detection: Before loading the kiln, check whether all roller driven shafts 1 fixed to the roller driven shaft fixing parts and the roller driving shaft fixed to the roller driving shaft fixing block are installed correctly and reliably, and then perform no-load self-test operation. By measuring the value of the torque sensor connected to the transmission connection and the transmission structure, the torque sensor is reliably connected to the roller driving shaft, and the current roller weight is indirectly calculated through the intelligent algorithm. At the same time, the proximity sensor detects that the roller driven shaft detection block fixed on the roller driven shaft can obtain the roller driven shaft rotation signal to determine whether the roller driven shaft rotates with the transmission structure, thereby indirectly determining whether the roller driven shaft or the roller middle shaft or the roller driving shaft is broken or deformed; Step 2: Intelligent reminder: If the detected roller is a low-temperature metal roller, the roller driven shaft rotation signal is valid periodically, and the current roller no-load weight is basically consistent with the previous roller no-load weight value, it can be judged that the current roller state is normal; if the roller driven shaft rotation signal is non-periodic and discontinuous, and the difference between the current roller no-load weight and the previous roller no-load weight value is more than 5%, it can be judged that the current roller state is abnormal and may be deformed, and it should be reminded to check and replace the roller in time; If the detected roller is a high-temperature non-metallic roller, the roller driven shaft rotation signal is valid periodically, and the current roller no-load weight is basically consistent with the previous roller no-load weight value, it can be judged that the current roller state is normal; if the roller driven shaft rotation signal is valid periodically, and the difference between the current roller no-load weight and the new roller no-load weight is more than 20%, it can be judged that the current roller state is abnormal and there may be a risk of breakage, and you should be reminded to check and replace the roller in time; if the roller driven shaft rotation signal is continuous or non-periodic valid, and the difference between the current roller no-load weight 11 and the previous roller no-load weight is more than 5%, it can be judged that the current roller state is abnormal and there is a break, and you should be reminded to check and replace the roller in time; Step 3: Sintering detection: After the roller kiln is loaded and started to operate, the non-metallic rollers in the high-temperature zone are subjected to cyclic inspection, and the value of the torque sensor and the detection signal of the proximity sensor are read in real time to calculate the current roller load weight and determine the driven shaft rotation signal. The intelligent algorithm compares and calculates the current roller load weight, the weight of the previous cycle monitoring roller, and the load weights of 10 adjacent groups of rollers in real time. If the difference between the current roller load weight and the previous cycle monitoring roller weight and the average load weight of 10 adjacent groups of rollers continues to be more than 5%, and the driven shaft rotation signal is abnormal, it can be determined that the current roller is broken, and it should be reminded to check and replace the roller in time after the end of this operation.
2. The intelligent detection method for roller kiln rollers according to claim 1, characterized in that: The description of the intelligent algorithm is as follows: 1) Algorithm input: input torque sensor data, proximity sensor data and historical data; the torque sensor data: real-time detection of the roller torque value, used to calculate the roller weight; the proximity sensor data: detection of the roller driven shaft rotation signal, to determine whether the roller rotates normally; the historical data includes the last no-load weight, the no-load weight of the new roller, and the average load weight of adjacent rollers; 2) Algorithm output: roller status: normal, deformed, broken; early warning prompt: whether the roller needs to be checked or replaced; 3) Algorithm steps: Data collection: The torque value of the current roller is obtained through the torque sensor and converted into the current roller weight; the rotation signal of the roller driven shaft is obtained through the proximity sensor to determine whether the roller rotates normally; No-load detection: Before loading the kiln, perform no-load detection to calculate the current no-load weight of the roller; compare the difference between the current no-load weight and the last no-load weight to determine whether the roller is deformed or worn; Sintering detection: During the operation of the kiln, the load weight and rotation signal of the roller are monitored in real time. Compare the difference between the current load weight and the historical data (such as the weight monitored in the previous cycle and the average load weight of adjacent rollers); if the weight difference continues to exceed the threshold (such as 5%) and the rotation signal is abnormal, it is judged that the roller may be broken; Status judgment and early warning: judge the status of the roller (normal, deformed, broken) according to the weight change and rotation signal; if the status is abnormal, issue an early warning to remind the operator to check or replace the roller; 4) Algorithm core logic: Ignoring the effect of temperature, the relationship between weight and torque can be simplified to: , Where, T represents the torque of the detection roller, τ represents the allowable shear stress of the material (MPa), ρ represents the material density (kg / m³), L represents the roller length (m), and g represents the acceleration due to gravity (9.8 m / s²); Weight change analysis: 100%; among which, is the current roller weight, is the historical weight (such as the last empty weight or the average weight of adjacent rollers); if ΔW exceeds the set threshold (such as 5% or 20%, etc.), the roller state is judged to be abnormal; Rotation signal analysis: If the rotation signal of the roller driven shaft is non-periodic or discontinuous, it is judged that the roller may be broken or deformed.
3. The intelligent detection method for roller kiln roller rods according to claim 1, characterized in that: The proximity sensor can be a photoelectric, mechanical or proximity switch, a displacement sensor, a 2D / 3D machine vision sensor, etc.
4. The intelligent detection method for roller kiln roller rods according to claim 1, characterized in that: The torque sensor is a static or dynamic torque sensor.
5. The intelligent detection method for roller kiln roller rods according to claim 1, characterized in that: The roller driven shaft refers to the end of the roller connected to the side wall of the kiln. The roller and the side wall are connected by a bearing and its support, which plays a role in axially fixing the roller.