A molten steel level detection method and device based on temperature measurement sampling robot top pressure sensing

Abstract

This invention belongs to the field of iron and steel metallurgy technology, and specifically relates to a method and device for detecting the molten steel surface based on a pressure sensor at the top of a temperature-sensing sampling robot. The method involves installing a three-dimensional pressure sensor between the robot's front flange and the temperature-sensing sampling gun rod to collect pressure signals Px, Py, and Pz along three axes in real time, as well as the robot's position signal Lx perpendicular to the molten steel surface. The robot drives the gun rod to insert into the molten steel surface. The processing unit calculates the comprehensive pressure P = a·Px + b·Py + c·Pz, constructs the pressure curve P(t) and the position curve L(t), and discretizes them. By calculating the first and second order differences of the pressure sequence, the insertion point of the probe contacting the molten steel surface is determined when the product of adjacent second order differences is less than zero. Then, based on the insertion point position and a preset depth, the robot is controlled to complete precise temperature measurement and sampling. This invention utilizes the inflection point of pressure change to automatically identify the contact time with the molten steel, eliminating errors from human experience and achieving precise control of the insertion depth and standardized operation.

Description

Technical Field

[0001] This invention belongs to the field of iron and steel metallurgy technology, and specifically relates to a method and device for detecting the steel liquid level based on the top pressure sensor of a temperature sampling robot. Background Technology

[0002] In the RH refining process of the iron and steel metallurgy industry, accurate temperature measurement and composition sampling of molten steel are crucial for ensuring steel quality and optimizing production processes. Temperature sampling is typically performed by an industrial robot carrying a temperature sampling probe, which is inserted into the molten steel in the ladle to a designated depth to obtain representative data on the temperature and chemical composition of the molten steel. The accuracy of this operation directly affects the adjustment of subsequent process parameters and the quality control of the final product.

[0003] However, in actual production, the molten steel level in the ladle is not constant. Influenced by various factors such as the volume of molten steel in the ladle, the thickness of the slag, fluctuations in molten steel during the refining process, and the deformation of the ladle itself, the molten steel level varies between different heats and even at different times within the same heat. Traditional temperature measurement and sampling methods rely heavily on the operator's on-site experience to judge the molten steel level and control the robot's insertion depth. Operators estimate the molten steel level by observing visual information such as the brightness of the molten steel surface and the state of the slag layer, combined with experience. This method is highly subjective. The uncertainty of manual judgment can easily lead to probe insertion that is too shallow or too deep: if inserted too shallowly, the probe may not be fully immersed in the molten steel or may only contact the surface slag layer, resulting in lower temperature readings or failed sampling; if inserted too deeply, the probe may touch the refractory material at the bottom of the ladle or entrain excessive slag, affecting data accuracy and potentially damaging the probe or even endangering equipment safety. Furthermore, consistency in manual operation is difficult to guarantee. Differences in judgment standards exist between different operators and between different shifts for the same operator, hindering the implementation of standardized operations.

[0004] Therefore, there is an urgent need for a method that can automatically, in real-time, and accurately detect the contact point between molten steel and the surface to eliminate errors caused by human experience, automate and standardize temperature sampling operations, and improve the reliability and efficiency of molten steel testing data. Summary of the Invention

[0005] To address the aforementioned problems, the present invention aims to provide a method and device for detecting the molten steel surface based on the pressure sensor at the top of a temperature sampling robot. This method uses a three-dimensional pressure sensor installed between the robot flange and the gun rod to collect triaxial pressure signals in real time, and uses the second-order difference sign change to accurately identify the inflection point of the probe contacting the molten steel surface, eliminating the subjective error caused by traditional manual visual inspection experience. The insertion point detection accuracy can reach the millisecond level.

[0006] The technical solution of the present invention lies in: a method for detecting the molten steel surface based on the top pressure sensing of a temperature measuring and sampling robot, including the following steps: Step 1: Install a three-dimensional pressure sensor between the front flange of the robot's robotic arm and the gun barrel of the temperature measuring and sampling gun. The three-dimensional pressure sensor real-time collects the pressure signals Px, Py, and Pz in three axial directions, and simultaneously obtains the position signal Lx of the robot in the direction perpendicular to the molten steel horizontal plane; Step 2: The robot drives the gun barrel to insert into the molten steel surface at a fixed speed, and the processing unit real-time receives the pressure signal and the position signal; Step 3: Conduct comprehensive processing on the pressure signal, and calculate the comprehensive pressure P = a·Px + b·Py + c·Pz, where a, b, and c are weight coefficients, and the value range is 0 to 1; Step 4: Construct the curve P(t) of the comprehensive pressure changing with time and the curve L(t) of the position changing with time, and convert the continuous signals into discrete point sequences (ti, Pi) and (ti, Li); Step 5: Take the first-order difference of the pressure discrete sequence △Pi = P{i+1} - Pi, and then take the second-order difference △²Pi = △P{i+1} - △Pi = P{i+2}-2P{i+1}+Pi. When the product of adjacent second-order differences satisfies △²Pi · △²P{i+1} < 0, determine that the corresponding moment t{i+1} is the insertion point where the gun barrel probe touches the molten steel surface; Step 6: Obtain the corresponding insertion point height position from the position curve L(t) according to the insertion point moment t{i+1}, calculate the end position of the robot based on the preset target value of the insertion depth, and control the robot to insert the probe to the specified depth to complete the temperature measuring and sampling operation.

[0007] In the said Step 2, the robot drives the gun barrel to insert into the molten steel surface at a fixed angle, and the insertion speed is constant.

[0008] The weight coefficients a, b, and c in the said Step 3 are adjusted according to the actual working conditions, and a < b < c. The weight coefficient c of the axial component perpendicular to the molten steel horizontal plane approaches 1, and the weight coefficient a of the axial component parallel to the molten steel horizontal plane approaches 0.

[0009] Before calculating the comprehensive pressure P in the said Step 3, first perform filtering processing on the axial pressure signals Px, Py, and Pz by using the sliding window averaging method.

[0010] In the said Step 2, the processing unit starts to operate after the robot runs to a fixed height position above the molten steel surface, and starts to receive the pressure signal and the position signal.

[0011] When converting the continuous signals into discrete point sequences in the said Step 4, perform equidistant sampling at a fixed sampling frequency.

[0012] A steel liquid level detection device based on top pressure sensing of a temperature sampling robot is used in a steel liquid level detection method based on top pressure sensing of a temperature sampling robot. The device includes a robotic arm, a three-dimensional pressure sensor, and a processing unit. The front flange of the robotic arm is provided with a connection adapter, and the three-dimensional pressure sensor is installed between the connection adapter and the barrel of the temperature sampling gun.

[0013] The technical advantages of this invention are as follows: 1. This invention uses a three-dimensional pressure sensor installed between the robot flange and the gun barrel to collect triaxial pressure signals in real time, and accurately identifies the inflection point of the probe contacting the molten steel surface using second-order differential sign changes, eliminating subjective errors caused by traditional manual visual inspection. The insertion point detection accuracy can reach the millisecond level. 2. Based on the identified insertion point height and the preset insertion depth target value, this invention automatically calculates the robot's endpoint position and controls execution, ensuring that the temperature sampling probe is inserted into the molten steel at the same specified depth each time. This avoids measurement failure due to shallow insertion or damage to the probe / equipment due to excessive insertion, achieving fully automated and standardized operation with highly consistent results across different furnace batches and shifts. 3. This invention only requires the addition of a three-dimensional pressure sensor between the flange and the gun barrel of an existing temperature sampling robot, along with a processing unit and algorithm. It does not require large-scale on-site modifications, resulting in low cost and convenient deployment.

[0014] The following will provide further explanation in conjunction with the accompanying drawings. Attached Figure Description

[0015] Figure 1 This is a flowchart of a method for detecting the surface of molten steel based on the top pressure sensor of a temperature sampling robot, according to the present invention.

[0016] Figure 2 This is a structural schematic diagram of a temperature sampling robot probe safety detection device according to the present invention.

[0017] Figure 3 This is the connection intention of the three-dimensional pressure sensor of the present invention. Detailed Implementation

[0018] Example 1 like Figure 1 As shown, a method for detecting the level of molten steel based on pressure sensing at the top of a temperature sampling robot includes the following steps: A method for detecting the molten steel level based on pressure sensing at the top of a temperature sampling robot includes the following steps: Step 1: Install a three-dimensional pressure sensor between the flange at the front end of the robot's robotic arm and the barrel of the temperature sampling gun. The three-dimensional pressure sensor collects pressure signals Px, Py, and Pz in three axes in real time, and simultaneously acquires the position signal Lx of the robot along the direction perpendicular to the horizontal plane of the molten steel. Step 2: The robot drives the lance to insert into the molten steel surface at a fixed speed, and the processing unit receives the pressure signal and position signal in real time; Step 3: Conduct comprehensive processing on the pressure signal, and calculate the comprehensive pressure P = a·Px + b·Py + c·Pz, where a, b, and c are weighting coefficients, and the value range is 0 to 1; Step 4: Construct the curve P(t) of the comprehensive pressure changing with time and the curve L(t) of the position changing with time, and convert the continuous signals into discrete point sequences (ti, Pi) and (ti, Li); Step 5: Take the first-order difference of the pressure discrete sequence △Pi = P{i + 1} - Pi, and then take the second-order difference △²Pi = △P{i + 1} - △Pi = P{i + 2} - 2P{i + 1} + Pi. When the product of adjacent second-order differences satisfies △²Pi · △²P{i + 1} < 0, determine the corresponding time ti + 1 as the insertion point where the lance probe touches the molten steel surface; Step 6: Obtain the corresponding insertion point height position from the position curve L(t) according to the insertion point time ti + 1, calculate the end position of the robot based on the preset target value of the insertion depth, and control the robot to insert the probe to the specified depth to complete the temperature measurement and sampling operation.

[0019] In the above Step 2, the robot drives the lance to insert into the molten steel surface at a fixed angle and with a constant insertion speed.

[0020] The weighting coefficients a, b, and c in the above Step 3 are adjusted according to the actual working conditions, and a < b < c. The weighting coefficient c of the axial component perpendicular to the molten steel horizontal plane approaches 1, and the weighting coefficient a of the axial component parallel to the molten steel horizontal plane approaches 0.

[0021] Before calculating the comprehensive pressure P in the above Step 3, the axial pressure signals Px, Py, and Pz are first filtered by the sliding window averaging method.

[0022] In the above Step 2, the processing unit starts to receive the pressure signal and position signal after the robot runs to a fixed height position above the molten steel surface.

[0023] When converting the continuous signals into discrete point sequences in the above Step 4, equidistant sampling is performed at a fixed sampling frequency.

[0024] Embodiment 2 As Figure 2 、 Figure 3As shown, a steel liquid level detection device based on the top pressure sensor of a temperature sampling robot is used for the steel liquid level detection method based on the top pressure sensor of a temperature sampling robot. It includes a robotic arm, a three-dimensional pressure sensor and a processing unit. The front flange of the robotic arm is provided with a connection adapter, and the three-dimensional pressure sensor is installed between the connection adapter and the barrel of the temperature sampling gun.

[0025] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting the level of molten steel based on pressure sensing at the top of a temperature-sensing robot, characterized in that, It includes the following steps: Step 1: Install a three-dimensional pressure sensor between the front flange of the robotic arm and the gun barrel of the temperature measuring and sampling gun. The three-dimensional pressure sensor real-time collects pressure signals Px, Py, and Pz in three axial directions, and simultaneously obtains the position signal Lx of the robot along the direction perpendicular to the molten steel horizontal plane; Step 2: The robot drives the gun barrel to insert into the molten steel surface at a fixed speed, and the processing unit real-time receives the pressure signal and the position signal; Step 3: Conduct comprehensive processing on the pressure signal, calculate the comprehensive pressure P = a·Px + b·Py + c·Pz, where a, b, and c are weight coefficients, and the value range is 0 to 1; Step 4: Construct the curve P(t) of the comprehensive pressure changing with time and the curve L(t) of the position changing with time, and convert the continuous signals into discrete point sequences (ti, Pi) and (ti, Li); Step 5: Take the first-order difference of the pressure discrete sequence △Pi = P{i + 1} - Pi, and then take the second-order difference △²Pi = △P{i + 1} - △Pi = P{i + 2} - 2P{i + 1} + Pi. When the product of adjacent second-order differences satisfies △²Pi · △²P{i + 1} < 0, determine the corresponding moment t{i + 1} as the insertion point where the gun barrel probe touches the molten steel surface; Step 6: Obtain the corresponding insertion point height position from the position curve L(t) according to the insertion point moment t{i + 1}, calculate the end position of the robot based on the preset insertion depth target value, and control the robot to insert the probe to the specified depth to complete the temperature measuring and sampling operation.

2. The method for detecting the molten steel level based on the top pressure sensor of a temperature sampling robot according to claim 1, characterized in that, In the said Step II, the robot drives the gun barrel to insert into the molten steel surface at a fixed angle and with a constant insertion speed.

3. The method for detecting the molten steel level based on the top pressure sensor of a temperature sampling robot according to claim 1, characterized in that, The weight coefficients a, b, and c in the said Step III are adjusted according to the actual working conditions, and a < b < c. The weight coefficient c of the axial component perpendicular to the molten steel horizontal plane approaches 1, and the weight coefficient a of the axial component parallel to the molten steel horizontal plane approaches 0.

4. The method for detecting the molten steel level based on the top pressure sensor of a temperature sampling robot according to claim 1, characterized in that, Before calculating the comprehensive pressure P in the said Step III, first perform filtering processing on the axial pressure signals Px, Py, and Pz by using the sliding window averaging method.

5. The method for detecting the molten steel level based on the top pressure sensor of a temperature sampling robot according to claim 1, characterized in that, In the said Step II, the processing unit is put into operation after the robot runs to a fixed height position above the molten steel surface and starts to receive the pressure signal and the position signal.

6. The method for detecting the molten steel level based on the top pressure sensor of a temperature sampling robot according to claim 1, characterized in that, When converting the continuous signal into a discrete point sequence in the said Step IV, perform equidistant sampling at a fixed sampling frequency.

7. A steel molten surface detection device based on pressure sensing at the top of a temperature sampling robot, used in the steel molten surface detection method based on pressure sensing at the top of a temperature sampling robot as described in claim 1, characterized in that, It includes a robotic arm, a three-dimensional pressure sensor, and a processing unit. A connection adapter is provided on the front flange of the robotic arm, and a three-dimensional pressure sensor is installed between the connection adapter and the gun barrel of the temperature measuring and sampling gun.

Images