Multi-gun top-blown furnace foundation independent support and spiral micrometer system pre-embedding construction method

Abstract

The invention relates to the technical field of non-ferrous metal smelting equipment foundation construction and deformation monitoring, and discloses a multi-gun top-blown furnace foundation independent support and spiral micrometer system embedding construction method which comprises the steps that a foundation support independent of a furnace body foundation is designed, and thermal deformation conduction is reduced; a spiral micrometer system is pre-embedded before support concrete pouring, and micron-sized deformation monitoring is achieved; a wireless transmission module is adopted to upload data to a control center in real time; and the deformation trend is predicted by combining a recurrent neural network model, and early warning and automatic calibration are realized. According to the invention, the problem that the mounting precision is reduced due to thermal deformation of a traditional bracket is solved, high-precision, real-time and long-term automatic monitoring is realized, the operation stability and safety of the multi-gun top-blowing furnace are remarkably improved, and the bracket is suitable for mounting and monitoring integrated construction of high-temperature industrial equipment foundation engineering.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of non-ferrous metal smelting equipment foundation construction and structural deformation monitoring, and particularly relates to a construction and long-term health monitoring method for a foundation support structure of a multi-lance top-blown furnace, in particular to a construction method and system in which a high-precision screw micrometer system is embedded in an independent foundation support, and intelligent prediction and closed-loop calibration functions are integrated. BACKGROUND

[0002] A multi-lance top-blown furnace is a core device in modern non-ferrous metal continuous smelting processes. Its operational stability is highly dependent on the accuracy of the spatial pose of the lance group, and the installation reference of the lance is directly determined by the geometric precision of the furnace body foundation support. Under long-term high-temperature and high-load smelting conditions, the furnace body foundation will bear huge thermal and mechanical stresses, causing the foundation structure to creep and deform, and then causing the support platform installed thereon to displace and settle, ultimately affecting the set values of the lance height and angle, reducing oxygen utilization, increasing the risk of splashing, and even causing equipment safety accidents.

[0003] Currently, the deformation monitoring of industrial equipment foundations mainly has the following technical bottlenecks: first, the monitoring means is lagging. It generally relies on manual periodic offline measurement using a total station, with a long measurement period (usually weeks or months), which cannot capture dynamic deformation in an instant or short period, and cannot achieve early warning. Second, the monitoring is disconnected from the structure. Existing automatic monitoring sensors (such as inclinometers and strain gauges) are mostly installed on the surface or shallowly embedded, which are easily disturbed by construction, aged by high temperature, or damaged by machinery, with low survival rate, and the measurement reference has weak relevance to the core of the structure. Third, there is a lack of intelligent decision-making closed loop. Even if deformation data is obtained, it is only limited to display and record, and cannot form an effective prediction and feedback closed loop between the data and equipment control (such as lance adjustment), so it cannot achieve automatic regulation and control of "perception-decision-execution".

[0004] Regarding patent literature, existing technologies offer some related ideas, but none of them solve the aforementioned comprehensive problems. For example, patent literature CN115773701A provides a calibration system and method for a micrometer screw gauge, achieving high-precision displacement measurement and calibration. However, its application is limited to measuring instruments in laboratory environments and does not address the issues of long-term pre-embedded installation and data fusion in harsh industrial environments. Patent literature CN120426779A utilizes infrared monitoring and recurrent neural networks to achieve intelligent control of the spray gun height, but its focus is on the feedback of process parameters within the furnace, without addressing the precise monitoring and compensation of the deformation of the foundation support itself supporting the spray gun. Patent literature CN118008321A proposes a scheme for pre-embedding displacement monitoring rods in tunnel construction, achieving concealed continuous monitoring. However, its monitoring object is the convergence of surrounding rock, and the monitoring accuracy (typically at the millimeter level) and sensor principle cannot meet the needs of monitoring the micron-level deformation of industrial equipment foundations, nor does it integrate intelligent prediction and active calibration functions.

[0005] Therefore, there is an urgent need for an innovative construction method that can fundamentally solve the problem of long-term deformation monitoring and accuracy maintenance of multi-gun top-blown furnace foundations. This method requires the deep integration of high-precision monitoring sensors into the foundation structure to obtain real stress and deformation data from the source. Combined with modern data intelligence technology, it can achieve full-process automation from deformation perception and trend prediction to active calibration, thereby ensuring the long-term stability and efficiency of the smelting process.

[0006] As a core piece of equipment in non-ferrous metal smelting, the installation accuracy of the foundation support of a multi-lance top-blown furnace directly affects the furnace's operational stability and the spray lance's performance. Traditional supports are an integral structure with the furnace foundation, which, under long-term exposure to high-temperature radiation and thermal stress, is prone to thermal deformation, leading to a decrease in the flatness of the support platform and consequently affecting the spray lance height accuracy. Existing monitoring methods largely rely on manual, periodic total station measurements, which are inefficient, lack real-time performance, and cannot provide early warning or automatic calibration. While some technologies involve high-temperature monitoring or pre-embedded measurements, a complete construction method has yet to be developed that integrates a high-precision spiral micrometer system embedded in an independent support with intelligent prediction and closed-loop calibration. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a construction method and system for pre-embedding an independent support frame and a spiral micrometer system for a multi-nozzle top-blown furnace foundation. This method, through innovative structural design and construction techniques, deeply embeds a high-precision spiral micrometer system within a foundation support frame independent of the furnace heat source. Combined with wireless data transmission, intelligent algorithm prediction, and closed-loop calibration control, it achieves real-time, online, high-precision monitoring and active compensation for three-dimensional micro-deformation at key points of the support frame, ultimately ensuring the long-term stability of the multi-nozzle top-blown furnace nozzle installation reference.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] A construction method for pre-embedding an independent support frame and a spiral micrometer system for a multi-gun top-blown furnace foundation includes the following steps:

[0010] S1. Independent Foundation Support Design and Construction: On the ground of the smelting workshop, in the outer area of ​​the multi-gun top-blown furnace foundation, an independent reinforced concrete foundation support is designed and poured. This support is completely separated from the furnace foundation structure, with a gap reserved between them and filled with a layer of high-performance heat insulation material to block the direct heat conduction from the furnace body's high temperature to the support, thereby reducing heat-induced deformation at its source.

[0011] S2. Pre-installation of the micrometer spiral system: During the binding of the steel reinforcement cage of the independent foundation support, fixed bases are pre-installed at key stress nodes (such as the top of the column, the mid-span of the main beam, and the support). Multiple micrometer spiral monitoring units are connected end-to-end via rigid connecting rods to form a monitoring chain distributed along the main stress path of the support. Each monitoring unit has a built-in high-precision micrometer spiral, photoelectric or magnetic grating displacement sensor, signal processing circuit, and independent digital address. The monitoring chain is led out through a high-temperature shielded cable and connected to an active gateway fixed in a protective box on the side of the support.

[0012] S3. Concrete Pouring and Curing: A low-heat-of-hydration, high-durability concrete formula will be used for integral pouring. During pouring, specific protective measures must be taken for the embedded monitoring units and cables to prevent impact or displacement. After pouring, standard curing will be performed, and the monitoring system will be powered on for initial adjustment during the initial curing phase, recording the initial "zero point" reading.

[0013] S4. Furnace Installation and Initial Baseline Establishment: After the foundation support concrete reaches its design strength, the multi-gun top-blown furnace body is hoisted onto the precision-machined installation platform at the top of the support. Using a high-precision total station, with the workshop's overall control network as the reference, the three-dimensional coordinates of preset marker points on the furnace body and support are measured and adjusted to meet the design installation accuracy requirements. These coordinates serve as the initial spatial reference for all subsequent deformation monitoring.

[0014] S5. Monitoring System Startup and Data Transmission: After formal commissioning, the pre-embedded spiral micrometer monitoring system will be activated. The system will automatically collect relative displacement data of each monitoring point according to a preset sampling frequency (e.g., once per minute). The active gateway is responsible for collecting the data and sending encrypted data packets to the data server of the remote central control platform via an industrial wireless network (e.g., 4G / 5G or dedicated frequency band).

[0015] S6. Intelligent Data Processing and Deformation Prediction: The data processing module of the central control platform cleans, filters, and calculates the coordinates of the uploaded raw data to obtain the deformation components of each monitoring point in a two-dimensional plane perpendicular to the monitoring chain direction. Principal component analysis is used to reduce the dimensionality of the multi-point data and extract the core deformation patterns. The processed time-series data is input into a pre-trained recurrent neural network model (such as LSTM). The model learns the correlation between historical deformation and operating parameters such as temperature and load, and predicts the deformation development trend within a certain time window (such as the next hour).

[0016] S7. Intelligent Decision-Making and Active Calibration: When the future deformation output by the predictive model exceeds a preset safety threshold, the decision module of the central control platform automatically generates a calibration command. The command is sent to the corresponding micrometer unit through the control network. The micromotor driven encoder built into the unit starts working, and according to whether the command is "completion" or "reduction", it precisely adjusts the extension length of the micrometer screw, thereby applying a small reverse mechanical compensation to the local structure where the monitoring point is located, actively counteracting the predicted deformation trend.

[0017] S8. Periodic Absolute Benchmark Verification: To eliminate potential long-term cumulative errors and benchmark drift in the monitoring system, a periodic verification mechanism is established. At fixed intervals (e.g., monthly), a high-precision total station is used to measure the absolute coordinates of the prism target positioned on the outer surface of the protective casing. The measurement results are compared and integrated with the relative deformation data of the monitoring system during the same period, and the benchmark and model parameters of the monitoring system are calibrated and corrected.

[0018] S9. Closed-Loop System Operation and Optimization: Steps S5 to S8 above form a complete closed loop of "data acquisition-transmission-processing-prediction-decision-execution-verification". The system runs continuously, accumulating data and optimizing the prediction model parameters through machine learning algorithms, forming an intelligent basic health protection system with self-learning and self-optimization capabilities.

[0019] Furthermore, the surface of the top mounting platform of the independent base support needs to be precision milled, and its overall flatness error is required to be no more than ±0.3mm, so as to ensure the initial accuracy of the furnace body installation.

[0020] Furthermore, the protection level of the micrometer monitoring unit is no less than IP68, and the internal structure uses a vacuum insulation cavity and phase change heat dissipation material to ensure that its core sensor can withstand an ambient temperature of up to 120°C for a long time inside the pre-embedded concrete without performance degradation.

[0021] Furthermore, the active gateway has a built-in rechargeable battery and solar charging panel, enabling it to operate continuously for 72 hours in the event of a temporary power outage in the workshop, ensuring the continuity of monitoring.

[0022] Furthermore, the training data for the recurrent neural network model comes from historical monitoring data of this furnace, monitoring data of similar furnaces, and finite element thermodynamic simulation data, giving it good generalization ability for operating conditions.

[0023] An embedded intelligent monitoring system for implementing the above method includes:

[0024] Sensing layer: It consists of several micrometer monitoring units embedded in the concrete foundation support, which are connected in series and are responsible for collecting micro displacement signals.

[0025] Transport layer: including field active gateways, industrial wireless networks and workshop intranets, responsible for reliable data aggregation and remote transmission.

[0026] Platform layer: This is the central control platform, which includes data servers, intelligent algorithm engines (PCA dimensionality reduction, LSTM prediction models), and human-computer interaction interfaces. It is responsible for data storage, processing, prediction, decision-making, and visualization.

[0027] Execution layer: refers to the micro-drive calibration mechanism built into the micrometer unit, which is responsible for executing the compensation instructions issued by the platform.

[0028] Verification layer: refers to the total station measurement system, which serves as an absolute spatial reference and is periodically calibrated.

[0029] The beneficial effects of this invention are as follows:

[0030] 1. Source-based thermal deformation prevention: Through the design of independent supports and insulation layers, the main heat source is physically isolated, reducing the thermal deformation of the support body by more than 60%, which greatly improves the stability of the foundation.

[0031] 2. High monitoring accuracy and reliability: The high-precision micrometer is deeply embedded in the key stress points of the structure, making the measurement direct and accurate, avoiding the error and damage risk caused by surface sensor installation, and the monitoring accuracy reaches ±0.005mm.

[0032] 3. Predictive maintenance is achieved: By combining intelligent algorithms, passive recording is transformed into proactive prediction. It can issue early warnings and initiate compensation before deformation causes observable equipment problems, turning post-event processing into pre-event intervention.

[0033] 4. A smart closed loop has been formed: By creatively combining structural health monitoring with active mechanical compensation, a complete "perception-analysis-decision-execution" closed loop has been formed, which significantly reduces the frequency of manual intervention and improves the level of system automation.

[0034] 5. Controllable construction and operation and maintenance costs: Although the initial pre-embedded construction is somewhat complex, a single construction can ensure the foundation accuracy of the equipment throughout its entire life cycle, avoiding frequent production stoppages and adjustments caused by foundation inaccuracies and the resulting huge production losses, thus significantly reducing the total life cycle cost. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0036] Figure 1 This is a flowchart illustrating the closed-loop construction process for furnace installation and intelligent monitoring in this invention. Detailed Implementation

[0037] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. The following embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0038] Example 1: Foundation Construction of a Large Copper Smelting Multi-lance Top-blown Furnace

[0039] This embodiment is applied to a multi-lance top-blown smelting furnace with a daily processing capacity of 2,000 tons of copper concentrate.

[0040] Step S1 Implementation: The furnace foundation is a large block reinforced concrete foundation. Two independent reinforced concrete supports are symmetrically designed on its east and west sides. Each support is 12m long, 3m wide, and 2.5m high. The bottom of the support is piled, and the top is a 1m thick slab beam structure with a grid of mounting bolts on the upper surface. The side wall of the support is 800mm away from the outer wall of the furnace foundation. This gap is filled with a 100mm thick nano-aerogel insulation blanket (thermal conductivity ≤0.018W / (m·K)) and then sealed with refractory fiber cotton.

[0041] Step S2 Implementation: Four monitoring chains are pre-installed along the length of the main beam reinforcement of each independent support. Each monitoring chain consists of five spiral micrometer monitoring units connected in series, spaced 2.5m apart, located at the two end supports and three quarter points. The monitoring unit housing is a 316L stainless steel cylinder, 50mm in diameter and 300mm in length, welded and fixed to the upper main reinforcement of the main beam using special clamps. The unit integrates a grating ruler displacement sensor (0.001mm resolution) and a stepper motor-driven calibration mechanism. All units are connected in series via armored mineral-insulated cables, with the cable ends leading to a wall-mounted stainless steel protective box (IP66), housing an active gateway and battery. A forced centering base is welded to the outer surface of the protective box for mounting the total station prism.

[0042] Step S3 Implementation: Use C40 micro-expansion concrete, and pour it from one end of the support to the other to avoid direct impact on the monitoring unit. Carefully vibrate the unit using an immersion vibrator. After pouring, cover with geotextile and water for 28 days. On the 7th day of curing, power on the monitoring system and record the initial readings of all monitoring units under no-load conditions, setting them to "zero".

[0043] Step S4 Implementation: Using a 1000-ton crawler crane, the top-blown furnace body is hoisted in sections onto independent supports on both sides, and precisely positioned using adjusting bolts on the top surface of the supports. A high-precision total station with a 0.5-second accuracy is used to measure the prism coordinates corresponding to 36 feature points on the furnace body and 24 monitoring points on the supports. Least squares fitting is then used to adjust the furnace body to the designed orientation, with the error controlled within ±1mm. These coordinates are stored as a reference in the central database.

[0044] Step S5 Implementation: After smelting production begins, the monitoring system is set to collect data every 5 minutes. The active gateway sends encrypted data packets to the cloud server of the plant dispatch center via the workshop's 5G private network.

[0045] Step S6 Implementation: The data processing module on the cloud server runs automatically. First, outlier data with significant jumps is removed, and then smoothed using a Kalman filter. Data from symmetrical measurement points on the east and west sides are merged and processed to calculate the deformation curves of the support in the horizontal direction (perpendicular to the furnace axis) and vertical direction. PCA analysis reveals that the first two principal components contribute 85% of the deformation variance, corresponding to the overall uneven settlement and lateral bending due to heat, respectively. The deformation principal component time series of the past 24 hours, the thermocouple temperatures of each section of the furnace, and the air flow rate are used as inputs to a trained LSTM network (three layers, 128 neurons per layer). The network outputs the predicted values ​​of the deformation principal components for the next two hours.

[0046] Step S7 Implementation: Set the safety thresholds to 0.15mm for lateral deformation and 0.2mm for vertical settlement. A prediction indicates that the lateral deformation in the middle of the eastern support will reach 0.18mm after 1.5 hours. The decision module immediately generates instructions and sends them to the two monitoring units corresponding to the middle of the eastern support. The stepper motor in the unit starts, driving the micrometer screw to extend in the opposite direction by 0.03mm, generating a small pre-pressure on the surrounding concrete through its rigid shell to counteract part of the predicted deformation.

[0047] Step S8 Implementation: During the monthly shutdown and maintenance period on the first Monday of each month, surveyors use a total station to measure the absolute coordinates of the prisms on all eight protective boxes on both sides of the support. The measurement results are compared with the cumulative deformation values ​​calculated by the monitoring system. A cumulative deviation of 0.05mm in the vertical direction was found in the system's calculated values. This deviation was used to correct the calculation benchmark of the monitoring system and added as feedback data to the training set of the LSTM model to optimize the next prediction.

[0048] Step S9 Implementation: The above process continues to run. After one year of system operation, statistics show that the number of unplanned spray gun adjustments due to foundation deformation decreased by 90%, and the uniformity of temperature distribution in the furnace improved by 15%, proving the effectiveness of the closed-loop system.

[0049] Example 2: Redundancy Design of System Architecture

[0050] To cope with extreme situations, the system of this invention can be designed with redundancy. Each active gateway simultaneously connects to two wireless networks with different routes (such as a 5G private network and a Wi-Fi 6 mesh network). The central control platform adopts dual-machine hot standby. When the predictive model fails, the system can automatically switch to a simple rule control mode based on fixed thresholds to ensure that basic compensation functions are not interrupted.

[0051] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A construction method of a multi-gun top-blown furnace foundation independent support and a pre-embedded screw micrometer system, characterized in that, The method comprises the following steps: S1, an independent foundation support is designed and constructed outside the furnace body foundation periphery, a heat insulation layer is arranged between the independent foundation support and the furnace body foundation to reduce thermal conduction deformation; S2, a screw micrometer system is pre-buried in the steel reinforcement framework of the independent foundation support, the screw micrometer system comprises a screw micrometer, a wireless transmission module and a power supply module, the screw micrometer is installed at a key stress point of the steel reinforcement framework through a fixing support; S3, concrete is poured, the screw micrometer system is completely pre-buried in the concrete, and only a data interface and a wireless antenna are exposed; S4, a main structure of a multi-gun top-blown furnace is installed on the independent foundation support, initial position calibration is performed through a total station, and initial coordinate data are recorded; S5, the screw micrometer system is started, support deformation data are collected in real time, and the deformation data are sent to a central control platform through the wireless transmission module; S6, the central control platform pre-processes and reduces dimensions of the received deformation data, and a recurrent neural network model is used to predict a deformation trend; S7, when a predicted deformation value exceeds a preset threshold value, a calibration instruction is automatically generated by the system, and the screw micrometer is dynamically calibrated through an encoder; S8, key measuring points are regularly rechecked and calibrated through the total station, and system cumulative errors are corrected; S9, a "monitoring-prediction-calibration-rechecking" closed-loop control system is formed, and long-term stable operation is realized.

2. The method of claim 1, wherein, The independent foundation support adopts a reinforced concrete structure, a plurality of steel reinforcement meshes are arranged inside, reinforcing ribs are additionally arranged at key stress points, an installation platform is arranged at the top of the support, the surface of the platform is finely processed, and the flatness error is not more than ±0.5 mm.

3. The method of claim 1, wherein, The screw micrometer system comprises a plurality of screw micrometer units, each unit is provided with an independent IP address, is connected in series through an RS-485 bus and is connected to an active gateway, the active gateway is internally provided with a power management and wireless communication module and supports timing wake-up and data packaging upload.

4. The method of claim 3, wherein, The screw micrometer unit comprises a protective shell, a micrometer screw, a sensing element and a signal processing circuit, the protective shell is made of a high-temperature-resistant alloy material and is filled with heat-insulating ceramic fibers inside, and long-term stable operation below 150 DEG C is ensured.

5. The method of claim 1, wherein, The pre-processing in step S6 comprises data cleaning, abnormal value elimination and sliding average filtering; the dimension reduction adopts a principal component analysis method to extract main characteristic vectors of the deformation data; and the recurrent neural network model is an LSTM network, the input is time-series deformation data, and the output is deformation prediction values in future 3-5 sampling periods.

6. The method of claim 1, wherein, The calibration instruction in step S7 comprises two modes of complete calibration and reduced calibration: when the predicted deformation value is greater than the real-time measured value, the complete calibration is performed, and the micrometer screw extension length is increased; and when the predicted deformation value is less than the real-time measured value, the reduced calibration is performed, and the micrometer screw extension length is reduced.

7. The method of claim 1, wherein, The central control platform is also provided with a visual interface, real-time deformation data of each measuring point, a prediction curve and an alarm state are displayed, historical data query and report export are supported.

8. The method of claim 1, wherein, The heat insulation layer is an aerogel composite material, the thickness is 50-100 mm, and the thermal conductivity is not greater than 0.02 W / (m·K).

9. The method of claim 1, wherein, The total station review calibration in step S8 is performed once every 30 days, and the calibration data is automatically uploaded to the central control platform to correct the prediction deviation of the neural network model.

10. A construction system for embedding a plurality of gun top blown furnace foundation independent supports and a screw micrometer system, for implementing the method according to any one of claims 1 to 9, characterized in that, Comprise: Independent foundation support, pre-buried screw micrometer system, active gateway, sampling and control host, central control platform; the screw micrometer system is connected to the active gateway through a wired connection, the active gateway communicates with the sampling and control host through a wireless network, and the sampling and control host is connected with the central control platform through a 4G / 5G network, forming a three-level data transmission architecture.

Images