A LF furnace flexible control method and system based on green electricity direct connection

Abstract

This invention discloses a flexible control method and system for an LF furnace based on direct green electricity connection, belonging to the field of new energy power application technology. The invention constructs a virtual energy storage model of molten steel thermal inertia, utilizing the temperature dead zone for overheating energy storage to smooth power gaps; and through the inverse linkage between the short-term voltage stability index (SSI) and the bottom-blown argon flow rate, it utilizes rapid pressure relief to eliminate the gas path pressure capacitance effect and smooth the molten pool surface, achieving active arc stabilization; simultaneously, it combines electrode adjustment variable structure control based on a time window filter to solve the scale mismatch problem of mechanical hydraulic response lag and high-frequency voltage fluctuations. This invention significantly improves the level of new energy absorption without requiring additional hardware energy storage, effectively avoids the risks of electrode arc interruption and carbon increase, and ensures production continuity and metallurgical quality under direct green electricity connection.

Description

Technical Field

[0001] This invention belongs to the field of new energy power application technology, and more specifically, relates to a flexible control method and system for an LF furnace based on direct green electricity connection. Background Technology

[0002] The ladle refining furnace (LF furnace) is an indispensable secondary refining equipment in modern steel production processes, primarily responsible for functions such as steel heating, composition fine-tuning, desulfurization, and inclusion removal. Traditional LF furnace power supply systems typically rely on a rigid power grid dominated by thermal power. However, utilizing renewable energy sources such as wind and solar power (green electricity) to replace traditional thermal power is becoming an inevitable trend.

[0003] Green electricity direct connection refers to a model where renewable energy sources such as wind power, solar power, and biomass power are not directly connected to the public power grid, but instead supplied to individual electricity users through direct connection lines, enabling clear physical traceability of the supplied electricity. However, green electricity inherently possesses randomness, indirectness, and volatility, making it difficult to meet the stable heat energy demands of industry. Therefore, how to bridge the gap between the intermittent supply caused by the volatility of green electricity and the continuous supply to industrial needs is a pressing problem that needs to be solved within the industry.

[0004] A search revealed that patent CN120414624A discloses a coupling system and control method for the gradient utilization of waste heat from electrolysis processes and green energy storage. This application includes a waste heat recovery module for recovering and cascading the use of high-temperature and low-temperature waste heat from electrolysis for power generation or heating; a green energy direct supply module comprising photovoltaic, hybrid energy storage, and electrothermal conversion units for priority direct supply of green energy, absorption of fluctuations, and storage; and a multi-energy coupling control module that communicatively connects each module to the power grid, dynamically optimizing and coordinating the energy distribution among waste heat power, green energy, energy storage, and the power grid based on a preset algorithm.

[0005] For example, patent CN121363886A discloses a graded cogeneration and storage system. This application can simultaneously couple multiple types and sources of energy or heat, converting fluctuating green electricity or waste heat into thermal energy for storage. When energy is needed, the thermal energy stored in the phase change thermal storage medium is released, thereby bridging the gap between intermittent energy and continuous demand.

[0006] In order to mitigate fluctuations in green electricity and ensure a continuous supply of green energy, the above-mentioned applications often require the construction of large-scale physical energy storage facilities, such as supercapacitors, lithium battery arrays, or hydrogen energy storage. This results in extremely high construction and operating costs, which limits the promotion of green electricity in the metallurgical field.

[0007] Furthermore, under the traditional control logic of the metallurgical industry, the regulation of bottom-blown argon flow exhibits significant pneumatic transmission lag, which can easily lead to arc instability or even arc interruption. This not only interferes with the stable operation of the furnace's normal smelting process but also directly affects the quality stability and consistency of the final product. Therefore, designing a green direct-connection system that can be directly adapted to the operating conditions of LF furnaces and effectively ensure the stable operation of the arc is of significant practical importance for promoting the green and high-quality development of the industry. Summary of the Invention

[0008] The problem to be solved

[0009] To address at least some of the problems existing in the prior art, the present invention provides a flexible control method and system for an LF furnace based on direct green electricity connection. The purpose is to solve the problems that the existing direct green electricity connection system requires high energy storage costs and cannot be applied to ladle refining furnaces due to fluctuations in green electricity.

[0010] Technical solution To solve the above problems, the technical solution adopted by the present invention is as follows: The present invention provides a flexible control method for an LF furnace based on direct green electricity connection, comprising the following steps: Step S1: Based on the target tapping temperature, set the allowable temperature dead zone range, and use the temperature dead zone range to construct a virtual energy storage model based on the thermal inertia of molten steel. Step S2: Power the LF furnace using a green direct connection method and collect the power supply voltage signal at the LF furnace power supply terminal in real time to determine the power supply stability. When the power supply is detected to be stable and the green electricity input power is sufficient, the overheat energy storage strategy is executed; the target heating temperature is automatically increased and controlled within the temperature dead zone, and the electrical energy is converted into the sensible heat of the molten steel for storage. When the power supply is detected to be unstable or the green electricity input power is insufficient, the high current heating is suspended, and the pre-stored superheat is used to resist the natural temperature drop during the refining process, so as to keep the temperature of the molten steel within the temperature dead zone. Step S3: When it is determined that the power supply is unstable, perform emergency suppression control operation; Reduce the flow rate of bottom-blown argon gas to the arc-preserving flow rate by adjusting the regulating valve; simultaneously open the pressure relief valve on the pipeline between the regulating valve and the bottom permeable brick to directly release the residual gas in the pipeline to the atmosphere. Step S4: If the power supply becomes unstable, but the duration of the instability does not exceed the threshold, the electrode will not shift. If the duration of power instability exceeds the threshold, perform the normal lifting and lowering operation of the electrodes; Step S5: After the power supply is detected to be stable, the pressure relief valve and the regulating valve are reset. Step S6: Perform supply and demand matching for the next batch.

[0011] In some implementations, the specific steps for determining power supply instability in step S2 are as follows: The short-term voltage stability index (SSI) is calculated based on time sliding window analysis and matched with a preset green electricity transient fluctuation model. When the matching degree exceeds the preset threshold and voltage drop characteristics are detected, the power supply is determined to enter an unstable state. In some embodiments, in step S3, when it is determined that the power supply is in an unstable state, the gas flow rate at the outlet of the permeable brick is rapidly reduced to the arc-preserving flow rate within 200ms by the pressure relief valve, and the minimum voltage threshold required to maintain the electric arc combustion is reduced by using the inertial damping of the liquid surface.

[0012] In some implementations, step S4 specifically involves setting a time window filter threshold t. limit Real-time monitoring of the duration t of power instability; If t < t limit If a voltage dip is detected, the electrode locking mode is activated, and the electrode hydraulic servo valve is forcibly locked in the neutral position to prevent the electrode from tracking voltage fluctuations. The physical elasticity of the arc length is used to adapt to voltage fluctuations. If t≥t limit If the system detects a continuous power interruption, it will release the electrode lockout mode and perform electrode lifting and lowering operations under normal impedance control.

[0013] In some implementations, the time window filter threshold t limit The duration is 0.5 to 3 seconds.

[0014] In some implementations, step S5 specifically involves the following steps: Once the power supply voltage returns to the steady-state range and the short-term voltage stability index (SSI) remains normal for more than the set delay, the pressure relief valve is reset first; then, the electrode position is unlocked; finally, the regulating valve is reset to restore normal impedance closed-loop control; and based on the duration of the emergency suppression mode, the amount of kinetic reaction loss is calculated, and either enhanced stirring or extended refining cycle is performed.

[0015] In some implementations, step S6 specifically involves the following steps: Based on the current fluctuation characteristics and trends of green electricity, estimate the total green electricity supply for the next refining cycle and compare it with the preset standard smelting energy consumption per furnace. When it is determined that the green electricity supply is sufficient to sustain the refining process of the next batch, the refining operation of the next batch shall be carried out. When it is determined that the green electricity supply is insufficient to sustain the refining process of the next batch, a control signal is output to activate the alarm device, prompting production adjustments or cutting off the feed.

[0016] In some implementations, the specific operation of determining whether the green electricity supply is sufficient to sustain the refining process in step S6 is as follows: S61. Calculation of Refining Energy Consumption Requirements Read the production plan data and calculate the theoretical total electrical energy Q required for the next refining batch. need The calculation formula is:

[0017] Among them, C p η is the specific heat capacity of the steel grade; m is the mass of molten steel; ΔT is the target temperature rise; η Q Historical thermal efficiency; S62, Extrapolation of Green Electricity Supply Trends Set the observation window T window Linear regression analysis was performed on the collected active power P(t) to calculate the green electricity supply trend; when the slope K of the power change was monitored... slope <0 and correlation coefficient R 2 When the value is greater than 0.9, it is determined to be a trend of decline, and the next refining cycle T is calculated according to the following analytical formula. refine The green electricity within can supply a total of Q electricity. supply ,

[0018] Wherein, η rel P is the confidence level correction factor. now T represents the current instantaneous power. refine For a complete refining cycle; S63. Supply and Demand Matching and Early Warning Triggering Set the matching coefficient k m The predicted total green electricity supply capacity Q supply With the theoretical total electrical energy Q need Perform a comparison; When inequalities When the system is established, if the green electricity supply is insufficient, the control signal is output; otherwise, if the green electricity supply is sufficient, the control signal is output.

[0019] In some implementations, the criterion for determining the voltage drop characteristic in step S2 is as follows: The effective voltage value was monitored to show a unilateral downward trend for n consecutive power frequency cycles, and the cumulative drop reached the threshold. Alternatively, the Short-Term Voltage Stability Index (SSI) may exceed the preset fluctuation threshold.

[0020] This invention also provides a flexible control system for an LF furnace based on direct green electricity connection, including a green electricity unit and a ladle refining furnace directly connected to the green electricity unit. It also includes, Power monitoring unit, The power monitoring unit is located on the power supply side of the green power unit and is used to collect the voltage signal output by the green power unit. Bottom blow control unit, The bottom blowing control unit includes a permeable brick installed inside the ladle refining furnace, and the permeable brick is connected to a bottom blowing air source through a bottom blowing pipe; The bottom blowing pipe is equipped with a regulating valve and a pressure relief valve; wherein the pressure relief valve is located between the regulating valve and the permeable brick. Electrode adjustment unit, The electrode adjustment unit includes an electrode installed inside the ladle refining furnace, and the electrode is connected to an adjustment mechanism that drives it to perform lifting and lowering operations. and central processing unit, The central processing unit is connected to the power monitoring unit, the bottom blowing control unit, and the electrode adjustment unit, respectively, and is used to receive feedback signals from each unit and send control commands to each unit.

[0021] Beneficial effects Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention provides a flexible control method for an LF furnace based on direct green electricity connection. By setting the allowable temperature dead zone range, when the green electricity input power is sufficient and the power supply is in a stable state, the electrical energy is converted into the sensible heat of the molten steel for storage, without the need to build a large-scale physical energy storage facility, which is beneficial to the control of equipment investment costs. When the green electricity input power is insufficient or the power supply is in an unstable state, the stored superheat is used to resist the natural temperature drop during the refining process and maintain the temperature of the molten steel within the allowable temperature dead zone range. In addition, when the power supply is unstable, the flow rate of bottom blowing argon is reduced to the arc-maintaining flow rate; and the residual gas in the pipeline can be quickly discharged through the pressure relief valve to eliminate the gas pressure capacitance effect and cut off the kinetic energy of the bubble plume; and the inertial damping of the liquid surface is used to quickly attenuate the turbulence and change the molten pool surface from a nonlinear undulating interface to a quasi-static laminar interface, thereby significantly reducing the minimum voltage threshold required to maintain the arc combustion, which can effectively prevent the system from arc interruption and ensure the stable operation of the smelting process.

[0022] (2) The present invention provides a flexible control method for an LF furnace based on direct connection of green electricity. The electrode control strategy based on time window filtering distinguishes between voltage dips and continuous power interruptions of green electricity. During short-term fluctuations, the electrode position is forcibly locked, and the length elasticity of the arc plasma is used to adapt to voltage changes. This not only avoids the accidental carbonization and scrapping caused by the electrode being mistakenly inserted into the molten steel due to tracking random disturbances of green electricity, but also eliminates ineffective oscillations in the mechanical system and extends the equipment life.

[0023] (3) The flexible control method for an LF furnace based on direct green electricity connection of the present invention introduces a dynamic recovery and kinetic compensation strategy. After the green electricity supply stabilizes, the system does not simply reset, but calculates the kinetic loss based on the duration of the emergency wave suppression mode and automatically executes short-term enhanced stirring or extends the refining cycle. This effectively compensates for the reaction process lost due to avoiding green electricity fluctuations, ensuring that key metallurgical indicators such as desulfurization rate and inclusion removal rate are not affected by the quality of green electricity supply, and providing quality assurance for achieving green manufacturing throughout the entire process.

[0024] (4) A flexible control method for an LF furnace based on direct green electricity connection according to the present invention, which calculates the theoretical energy consumption Q of the next furnace cycle. need Green electricity supply trend Q based on linear regression analysis supply By matching supply and demand, irreversible accidents such as molten steel solidifying in the ladle ("cold steel") due to power outages during refining are effectively avoided, and the adaptability of high-energy-consuming industrial loads to the safety of new energy power supply is significantly improved. Attached Figure Description

[0025] Figure 1 This is a flowchart of a flexible control method for an LF furnace based on direct green electricity connection according to the present invention; Figure 2 This is a simplified structural diagram of an LF furnace flexible control system based on direct green electricity connection according to the present invention.

[0026] In the diagram: 100, ladle refining furnace; 200, power monitoring unit; 300, central processing unit; 400. Bottom-blowing control unit; 410. Permeable brick; 420. Bottom blowing pipe; 430. Bottom blowing air source; 440. Regulating valve; 450. Pressure relief valve; 500. Electrode adjustment unit; 510. Electrode; 520. Adjustment mechanism; 600, Green Power Unit; 700, Alarm Unit. Detailed Implementation

[0027] To further understand the content of this invention, a detailed description of the invention will be provided in conjunction with the accompanying drawings.

[0028] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0029] As described in the background section, traditional LF furnace power supply systems typically rely on a rigid, large power grid dominated by thermal power. Their power frequency and voltage are relatively constant, and control strategies primarily focus on optimizing the power supply curve under steady-state conditions to reduce electrode losses and power consumption. When green electricity is directly connected to the LF furnace, voltage flicker or sudden power drops are highly likely to occur on the power supply side (e.g., due to photovoltaic cloud shading or wind gusts). Therefore, existing LF furnaces and green electricity technologies suffer from the following serious drawbacks when facing significant voltage fluctuations: (1) High energy storage costs: To mitigate fluctuations in green electricity, large-scale physical energy storage facilities (such as supercapacitors, lithium battery arrays, or hydrogen energy storage) are often required, resulting in extremely high construction and operating costs and limiting the promotion of green electricity in the metallurgical field.

[0030] (2) Arc instability and arc breakage: Under the existing control logic, there is a significant pneumatic transmission lag in the regulation of bottom-blown argon flow. When the voltage suddenly drops, even if the regulating valve is closed, the residual gas pressure in the pipeline will still maintain strong agitation for several seconds, causing the liquid level to fail to stabilize in time. With the arc rigidity reduced and its ability to resist external physical disturbances (such as airflow, electromagnetic deflection, etc.) and maintain directional stable geometry weakened, the arc is very prone to breakage.

[0031] (3) Electrode malfunction and carbon buildup: Existing electrode adjustment systems typically prioritize high response speeds. When a power supply voltage fluctuation is detected, the hydraulic system attempts to track the voltage by rapidly raising and lowering the electrode. However, because the inertia of mechanical systems (such as mechanohydraulic systems) is much greater than the frequency of voltage fluctuations, phase lag is easily generated, leading to the electrode being mistakenly inserted into the molten steel, causing a severe carbonization accident and resulting in the scrapping of the molten steel.

[0032] Therefore, the present invention aims to provide a green electricity direct connection system suitable for LF furnaces to overcome the above-mentioned defects.

[0033] The present invention will be further described below with reference to specific embodiments.

[0034] This embodiment of a flexible control method for an LF furnace based on direct green electricity connection mainly includes the following steps: Step S1: Based on the target tapping temperature, set the allowable temperature dead zone range, and use the temperature dead zone range to construct a virtual energy storage model based on the thermal inertia of molten steel. Step S2: Power the LF furnace using a green direct connection method and collect the power supply voltage signal at the LF furnace power supply terminal in real time to determine the power supply stability. When the power supply is detected to be stable and the green electricity input power is sufficient, the overheat energy storage strategy is executed; the target heating temperature is automatically increased and controlled within the temperature dead zone, and the electrical energy is converted into the sensible heat of the molten steel for storage. When the power supply is detected to be unstable or the green electricity input power is insufficient, the high current heating is suspended, and the pre-stored superheat is used to resist the natural temperature drop during the refining process, so as to keep the temperature of the molten steel within the temperature dead zone. Step S3: When it is determined that the power supply is unstable, perform emergency suppression control operation; Reduce the flow rate of bottom-blown argon gas to the arc-preserving flow rate by adjusting the regulating valve; simultaneously open the pressure relief valve on the pipeline between the regulating valve and the bottom permeable brick to directly release the residual gas in the pipeline to the atmosphere. Step S4: If the power supply is unstable, but the duration of the instability does not exceed the threshold, the electrode will not perform any displacement action. If the duration of power instability exceeds the threshold, perform the normal lifting and lowering operation of the electrodes; Step S5: After the power supply is detected to be stable, the pressure relief valve and the regulating valve are reset. Step S6: Perform the next batch of refining operations.

[0035] Specifically, in step S1, the temperature dead zone refers to a redundant temperature range that is artificially set above the target tapping temperature required by the process, allowing for fluctuations. Of course, this temperature dead zone range must be smaller than the temperature window allowed by the refining process to ensure normal smelting.

[0036] In this embodiment, to adapt to the intermittent nature of green electricity, traditional rigid constant temperature control is abandoned. The system treats the LF furnace as a huge thermal energy battery and constructs a virtual energy storage model based on the thermal inertia of molten steel, with fluid-electrical reverse linkage as the key support. The discharge logic of this virtual energy storage model is as follows: Charging mode (green power surplus period): When the system detects that the power is sufficient and the voltage is stable, it automatically increases the heating target value to the temperature dead zone range, and converts the excess electrical energy into the sensible heat of the molten steel for physical storage.

[0037] Discharge mode (power gap period): When encountering a power gap caused by photovoltaic cloud shading or a sudden drop in wind power, the system suspends high-current heating and only maintains weak stirring. It also utilizes the pre-stored temperature rise margin to offset the natural temperature drop during the refining process, achieving flexible adaptation to grid fluctuations.

[0038] Specifically, in this embodiment, the target tapping temperature is 1600℃, and the temperature dead zone range is 5-8℃, meaning the upper limit of the temperature dead zone is 1608℃. Simultaneously, the specific heat capacity of the molten steel is set to Cp≈0.8 kJ / (kg·℃), and the mass of the molten steel is m=120000kg.

[0039] In traditional processes, the target tapping temperature is treated as a static point, and full-power heating is immediately applied if it deviates from this point. In this embodiment, however, the LF ladle refining furnace is considered a giant thermal energy cell. Simultaneously, the target tapping temperature is treated as a dynamic range. Within this dynamic range, the system does not force high-precision real-time compensation, but rather uses it as a buffer zone for energy storage.

[0040] Because molten steel has a very high heat capacity and thermal inertia, when there is sufficient green electricity, the system can slightly heat the molten steel to the upper limit of the temperature dead zone, at which point the molten steel essentially becomes a heat energy carrier. When the green electricity drops momentarily, the system can stop heating and use this redundant sensible heat to maintain process requirements without immediately drawing electricity from the outside.

[0041] refer to Figure 1 As shown, in some embodiments, step S2 specifically involves the following steps: The system analyzes the acquired voltage signal U using a time sliding window algorithm. grid Data; at the same time, the system has a built-in green electricity transient fluctuation model, and determines the power supply stability status through dual verification of quantitative indicators and feature matching.

[0042] The quantitative indicator refers to calculating the Short-Term Voltage Stability Index (SSI) in real time and comparing it with a preset threshold (e.g., 0.15). Feature matching refers to comparing the extracted feature vectors in real time with the model database. When the SSI matches the model by more than 85% and a characteristic voltage drop is detected, the power supply is determined to be in an unstable state.

[0043] Specifically, the determination of fall characteristics consists of two parts. (1) Trend determination: The effective voltage value is monitored to show a unilateral downward trend for n consecutive power frequency cycles, and the cumulative drop reaches the threshold. For example, the cumulative drop reaches 10% of the rated voltage for 3 consecutive cycles.

[0044] (2) Pattern recognition: The short-term voltage stability index (SSI) breaks through the preset fluctuation threshold (e.g., 0.15). This index is used to identify non-load-type power supply disturbances by monitoring the instantaneous standard deviation of the waveform.

[0045] When any of the above conditions for determining the voltage drop characteristics are met, the system determines that a characteristic voltage drop has occurred.

[0046] The purpose of setting a preset threshold is that, in the normalized model, if SSI > 0.15, it indicates that the instantaneous fluctuations (noise, harmonics, drops) have exceeded the system's set safety redundancy. Therefore, the system will only execute subsequent rapid pressure relief or electrode locking actions when the SSI (fluctuation intensity) is large enough and the matching degree (recognition accuracy) is high enough, to prevent unnecessary shutdowns caused by erroneous actions.

[0047] Furthermore, the green electricity transient fluctuation model and the SSI-based power state determination logic include, (1) Construction and calibration of green electricity transient fluctuation model Based on the construction of the feature vector space, the model is further built and trained through a combination of offline training and online recognition. The specific process is as follows: First, the feature vector space is constructed, starting with defining three-dimensional feature vectors. The three key parameters are: σ, the fluctuation intensity, which quantifies the dispersion of the waveform from the sinusoidal reference by calculating the standard deviation of the power supply sample values, and is used to identify tiny pulse fluctuations that are insensitive to the effective value (RMS). f, the frequency distribution, which extracts the spectral energy distribution in the 2kHz~15kHz frequency band through Fast Fourier Transform (FFT) to identify the high-frequency carrier characteristics of the inverter. ΔP, the energy trend, is used to define the trend of energy loss, thereby determining whether it is a transient power disturbance or a power trend collapse.

[0048] After the feature vector space is constructed, offline training using offline benchmark database calibration and online recognition using online sample matching are performed to further build and train the model.

[0049] Offline training: The method of calibration using an offline benchmark database is employed.

[0050] Sample classification: Renewable energy output drop signals are collected as positive samples; inductive load start-up / shutdown, electrode short circuit, and excitation inrush current signals are collected as negative samples.

[0051] Regression calibration: The feature vector is optimized by applying a linear regression algorithm. By adjusting the weight factors w1, w2, and w3, a classification discriminant function is established, and a standard feature template for power supply instability is determined.

[0052] Online identification: A method for online sample matching is used. Real-time feature mapping: The system extracts the feature vector V of the current power supply voltage signal in real time. real And project it onto the feature vector space constructed above.

[0053] Distance matching calculation: The feature vector V mentioned above is calculated using Euclidean distance or cosine similarity algorithm.real Geometric distance from positive samples (templates with unstable power supplies) in the benchmark database.

[0054] Confidence conversion: The distance value is converted into a matching index of 0% to 100% through a mapping function. When the matching degree exceeds the 85% threshold, the system determines that the current disturbance conforms to the instability characteristics of the green electricity transient fluctuation model, thereby excluding unrelated load disturbances.

[0055] After the above steps, a preliminary model for transient fluctuations in green electricity can be established.

[0056] (2) SSI feature extraction layer: multi-dimensional coupled decision logic Since the physical units of each dimension of the eigenvector are different, the system uses the normalized weighted summation method to obtain the short-term voltage stability index (SSI).

[0057] Normalization (Norm): The parameters are mapped to the [0, 1] interval by using the Sigmoid function, eliminating the influence of dimensions between different feature vectors.

[0058] Feature discrimination logic: High-frequency identification: High weight is given to the high-frequency flicker unique to green electricity. By weighting the frequency band from 2kHz to 15kHz, load disturbance signals that are concentrated in the low-frequency harmonic band (<2.5kHz) and accompanied by reactive power impact are accurately eliminated.

[0059] Trend determination: Distinguish between transient power disturbances (random power oscillations) and power trend collapses (continuous output losses).

[0060]

[0061] Among them, the basic weight w1 is 0.15-0.25, which is used to reflect the deviation of the sorting; the core identification weight w2 is 0.45-0.55, which is specifically used to identify the inverter characteristics of 2kHz~15kHz, which is the key to distinguishing green electricity; and the decision weight w3 is 0.25-0.35.

[0062] (3) Core architecture of the model: dual time window filter To eliminate the scale mismatch between the mechanical hydraulic system (second-level response) and electrical energy fluctuations (millisecond-level transients), a dual-ring buffer is deployed to execute asynchronous logic: Transient feature extraction stage (narrow window with dual time windows, t < 200 ms): Real-time calculation of instantaneous SSI to capture pulse disturbances. If the preset safety threshold is exceeded, the system immediately triggers rapid pressure relief and electrode locking modes to ensure that electrical safety takes precedence over mechanical action.

[0063] Trend Evolution Assessment Level (wide window with dual time windows, t = 1 s ~ 5 s): Assess the persistence of energy fluctuations. When the wide window energy integral shows that the fluctuations are systematic, determine to enter the virtual energy storage charging and discharging strategy, and smoothly transition to steady-state regulation.

[0064] (4) Model matching degree calculation and judgment execution The system determines whether the power supply is in an unstable state by comparing the real-time feature vector with the reference template, and performs hierarchical linkage based on the SSI determination result.

[0065] Matching degree calculation: Calculate the Euclidean distance between the real-time vector and the standard vector in the model library, and convert it into a matching degree index of 0% to 100%.

[0066] Hierarchical linkage logic: A. When SSI > 0.15 and the matching degree exceeds 85%, and a characteristic voltage drop is detected, the system immediately determines that the power supply is in an unstable state and initiates the following actions in sequence: Execution step S3: Immediately trigger rapid depressurization, and stepwise reduce the bottom-blown argon flow rate to the arc-maintaining flow rate Q. limit Simultaneously execute step S4: start the timer (t=0), activate the electrode lock mode (i.e., t≤t in step S4). limit (Time voltage dip handling), keep the electrodes stationary, and adjust according to the abnormal time t and t limit The comparison determines whether to perform continuous power interruption processing (see steps S3 and S4 below for details).

[0067] B. If only SSI > 0.15 is satisfied, the system is determined to be a local fluctuation / load disturbance: Such interference is usually transient (e.g., the start-up and shutdown of large motors most commonly seen in LF furnace smelting). In this case, the system remains in locked mode until the SSI returns to the safe range. If the SSI abnormality (SSI > 0.15) lasts longer than the threshold and does not have green power characteristics, that is, even if the physical fingerprint of the waveform shows that the interference source is from the internal load of the plant rather than the external green power source, the system can perform protective tripping according to the preset logic instead of performing the safety enhancement logic of continuous power interruption processing in step S4.

[0068] C. If SSI < 0.15, execute the SSI-based dynamic mapping control strategy for power supply status and bottom-blown argon flow rate according to step S3. (See Table 1 in step S3 for details) Step S3: Emergency wave suppression control based on the inverse linkage between bottom-blown argon flow rate and power supply stability Step S3 aims to address the arc instability problem caused by the lag in the bottom blowing system's response. Once step S2 determines that the power supply is unstable, the system immediately triggers the emergency suppression mode and performs the following concurrent actions: Inlet cutoff: The forced adjustment valve for bottom-blown argon flow rate instantly and abruptly closes from the current flow rate (e.g., 150-300 L / min) to the arc-preserving flow rate Q. limit (For example, 50 L / min).

[0069] Rapid pressure relief in pipelines: Simultaneously open the pressure relief valve on the pipeline between the regulating valve and the bottom permeable brick.

[0070] The arc-preserving flow rate refers to the minimum critical flow rate for arc stabilization that matches the Short-Term Voltage Stability Index (SSI). Traditional LF furnace bottom blowing is typically divided into strong blowing (desulfurization) and weak blowing (inclusion removal) based on the process stage. If the grid voltage becomes unstable, traditional systems usually trip directly for protection rather than adjusting the bottom-blown argon flow rate to maintain the arc. However, in this embodiment, due to the possibility of voltage drops during direct connection to green electricity, a dedicated arc-preserving flow rate Q needs to be set. limit This serves as a buffer to accommodate the strong fluctuations in green electricity.

[0071] In this embodiment, the core lies in establishing the short-term voltage stability index (SSI) and the bottom-blown argon flow rate Q. Ar The dynamic hierarchical mapping model is based on the following principle: A decrease in the short-term voltage stability index (SSI value) leads to a "peak-trough" superposition effect between the physical fluctuations of the molten pool surface and the voltage fluctuations, drastically amplifying the risk of arc interruption. This results in the negative impact of the physical fluctuations of the molten pool surface on arc stability being amplified further. Therefore, compensating for voltage instability by forcibly suppressing surface fluctuations is crucial for ensuring continuous production under green power direct connection conditions.

[0072] In this embodiment, the standard bottom-blown argon flow rate Q in the conventional operation of the LF furnace is used. std Based on this, a minimum critical flow rate state for maintaining the electric arc by adjusting the bottom-blown argon flow rate, namely the arc-maintaining flow rate Q, was proposed. limit Based on this, as the Short-Term Voltage Stability Index (SSI) increases, the power supply stability decreases, transitioning from a highly stable state (i.e., normal state) to an unstable state. The following bottom-blowing argon flow mapping control strategy, based on the SSI range division, is implemented: Process stability zone (SSI < 0.05): The power supply is considered to be in a high stability state. At this time, the conventional refining process is executed, and the bottom-blown argon flow rate is the standard refining flow rate Q. std (Typically 150-300L / min) to ensure desulfurization and component uniformity.

[0073] Collaborative early warning zone (0.05≤SSI≤0.15): If a moderate disturbance is detected in the power supply, the system will initiate a gradual current reduction mechanism, linearly mapping the flow rate to 50%-70% of the process reference value based on the short-term voltage stability index (SSI) offset. This mechanism aims to reduce molten pool fluctuations in advance and prevent short-circuit tripping caused by liquid surface impacting the electrodes.

[0074] Arc-preservation trigger zone (SSI > 0.15): This indicates the power supply is in a state of deep instability or voltage dip. The system triggers a rapid pressure relief command, using the exhaust valve to eliminate the pipeline pressure capacitance effect and instantly switching the flow rate to the dynamic arc-preservation position flow rate Q. limit .

[0075] Furthermore, the arc-position flow rate Q limit The dynamic determination process includes, Physical model construction: The gas-liquid-slag three-phase coupled interface was simulated using fluid dynamics (CFD). The relationships between the arc length L, the liquid surface rise height ΔH, and the bottom-blown argon flow rate Q were established. Ar The correlation equation:

[0076] Among them, L effective This represents the actual physical distance between the electrode tip and the fluctuating liquid surface, taking into account the liquid surface bulge ΔH caused by bottom blowing and the arc characteristic shift due to the short-term voltage stability index (SSI). Under green electricity conditions, the plasma stability of the arc decreases due to the reduced power supply rigidity. effective It must be maintained within a very narrow dynamic range: if it is too long, it will cause arc interruption; if it is too short (or collision due to liquid bulging), it will cause short circuit tripping.

[0077] L set To preset the reference arc length, the system sets the ideal arc length based on the current transformer voltage level under stable power supply (SSI < 0.05) and static liquid surface conditions. ΔH represents the height of the liquid surface rise induced by bottom blowing. During the ascent of the bottom-blown argon plume, the kinetic energy transferred to the molten steel causes a physical displacement of the liquid surface below the electrode. ΔH is the bottom-blown argon flow rate Q. Ar The nonlinear function, when the bottom-blown argon flow rate Q Ar Step to arc-preserving flow rate Q limit At that time, the decrease in ΔH can instantly compensate for the insufficient arc stiffness caused by the voltage drop, preventing the liquid surface from contacting the electrode.

[0078] δ(SSI) is the arc characteristic offset correction term, which reflects the drift of the arc current-voltage characteristic curve due to the decrease in power supply stability (increased SSI).

[0079] In this embodiment, the fluid energy domain is manipulated (controlling Q) Ar By changing ΔH, the loss in the electrical energy domain (SSI anomaly) is actively compensated, thereby achieving the goal of maintaining the arc position flow rate Q. limit The goal is to stabilize the fluctuations in the liquid surface.

[0080] In normal mode, the power supply is stable, and bottom blowing performs strong stirring (with a large ΔH). At this time, the arc relies on stable power output to maintain Leffectiveness. However, in the suppression mode of this embodiment, when SSI > 0.15, the power supply cannot provide sufficient energy to maintain a long arc. In this case, the system rapidly reduces ΔH by quickly depressurizing.

[0081] As the height of the liquid surface rises rapidly by ΔH, the liquid surface actively drops while the electrode remains stationary. This provides additional physical space under low-voltage conditions, forcibly maintaining the geometric continuity of the electric arc and avoiding frequent short circuits caused by liquid splashing.

[0082] Based on this, by calculating the minimum current density required to maintain continuous refining of the electric arc under different voltage drop amplitudes, the maximum allowable fluctuation threshold of the liquid level can be deduced, thereby locking Q. limit The range of values ​​for .

[0083] The boundary definition of the arc-preserving flow rate needs to be combined with experimental measurements to define the upper and lower limits of the bottom-blown argon flow rate and to comprehensively calibrate the arc-preserving flow rate boundary to ensure that during power instability, bottom blowing can both stabilize the liquid surface and maintain basic metallurgical kinetic conditions.

[0084] The upper limit of the arc-preserving flow rate must ensure that the total harmonic distortion (THD) of the arc remains low to avoid electrode carbonization or mechanical damage caused by drastic fluctuations. Determining this upper limit requires combining cold-state physical simulation and hot-state harmonic analysis. Cold-state physical simulation: An LF boiler water model was constructed using similarity criteria, with a focus on observing different bottom-blown argon flow rates Q. Ar The height of the liquid surface bulge ΔH and the surface wave frequency were measured. The experimental criterion was that when ΔH exceeded the critical margin of the arc length (usually 20%-30% of the normal arc length), it was considered to be an excessive fluctuation. This experiment established the mapping function between flow rate and liquid level fluctuation amplitude, and determined the upper limit of the bottom-blown argon flow rate to prevent physical interference between the melt and the electrode.

[0085] Hot harmonic analysis: During actual furnace operation, real-time monitoring is performed using an energy quality analyzer. The upper limit of the arc-maintaining flow rate must ensure that the total harmonic distortion (THD) of the arc remains low. If the flow rate is too high, violent liquid surface fluctuations will cause the arc to be frequently elongated or short-circuited, generating a large number of 3rd and 5th characteristic harmonics. By finding the inflection point of the THD curve, the maximum safe flow rate that will not cause electrode carbonization and mechanical vibration is determined.

[0086] The lower limit of the arc-maintaining flow rate must ensure that the molten steel below the electrode has basic thermal circulation capacity to prevent uneven metallurgical quality caused by localized overheating. The determination of this lower limit of the arc-maintaining flow rate needs to be combined with hot-state heat transfer efficiency experiments, specifically... Hot-state heat transfer efficiency experiment: The lower limit of the flow rate at the arc position must ensure that the molten steel below the electrode has basic thermal circulation capacity. The experiment uses an online temperature probe to monitor the temperature difference ΔT between the edge and center regions of the ladle. p .

[0087] If the flow rate falls below this lower limit, the circulation of the molten steel at the bottom will stagnate, leading to overheating of the molten steel in the arc-heated center below the electrode while the surrounding area experiences a slow temperature rise. This will not only cause metallurgical composition segregation but may also damage the furnace lining due to localized overheating. The experimentally determined lower limit must satisfy the difference (ΔT) between the temperature rise in the core area below the electrode and the temperature rise in the surrounding molten steel. p It no longer increases over time.

[0088] Finally, the comprehensive calibration of the arc-preserving flow rate boundary still needs to be combined with the hot bottom-blowing step reduction experiment (based on the standard bottom-blowing argon flow rate Q). std Gradually reduce to minimum sustaining flow rate Q min Among them, the minimum sustaining flow rate Q min To ensure that the bottom-blown permeable bricks do not leak steel and to maintain the limit flow rate of the basic thermal circulation at the bottom of the ladle.

[0089] During the refining intervals when the LF furnace is operating normally, small-amplitude power supply fluctuations are artificially simulated (such as adjusting transformer taps or simulating green electricity fluctuations) to perform a hot-state bottom-blowing argon flow rate reduction experiment (from the standard bottom-blowing argon flow rate Q). std Gradually reduce to minimum sustaining flow rate Q min ).

[0090] Simultaneously, the arc stability current waveform, arc sound spectrum, and voltage fluctuation compensation rate are recorded for each flow rate level. When the flow rate Q... Ar When the arc current falls within a specific range, the standard deviation is minimized and the total harmonic distortion (THD) is lowest. This range is defined as the initial arc-preserving flow range, serving as the default target value for the control system when SSI > 0.15.

[0091] In this embodiment, the arc-maintaining flow rate is not a fixed value, but a response function of a closed-loop feedback logic based on the Short-Term Voltage Stability Index (SSI). It stabilizes the liquid level and compensates for voltage fluctuations by dynamically adjusting the bottom-blown argon flow rate. Table 1 below provides a dynamic mapping between the power supply state and the bottom-blown argon flow rate based on SSI: Table 1: Dynamic Mapping Comparison of Power State and Bottom-blown Argon Flow Rate Based on SSI

[0092] In this embodiment, the objective of step S4 is to resolve the mismatch between the response lag of the mechanical hydraulic system and the high-frequency fluctuation of the electrical system. Specifically, this includes... Variable structure control logic: The system sets the time window filter threshold t. limit (This threshold can be adjusted between 0.5 and 5 seconds based on the characteristic frequency of power grid fluctuations and the response speed of the mechanical hydraulic system. In this embodiment, taking into account the gust characteristics of the wind farm, the threshold t is set as follows.) limit (Set to 2 seconds.) Meanwhile, the system calculates the duration t of the monitored power instability and the preset time window filter threshold t. limit The comparison is performed to determine whether to perform electrode position locking or regular lifting operation.

[0093] When power instability is detected, a timer is started to record the duration t. When the system detects a short-term drop in the effective power value, and the duration t < t limit At this point, it is determined that the voltage has temporarily dropped, and the locking mode of the mechanical hydraulic system is activated, forcibly locking the electrode hydraulic servo valve to the neutral position (i.e. keeping the electrode in place), and blocking the lifting and lowering commands output by the conventional PID regulation.

[0094] This stage utilizes the stable working conditions formed after bottom blowing and leveling the liquid surface in step S3. It relies on the elastic support of the electric arc (i.e., under constant electrode distance, the electric arc compensates for small voltage deviations through its own electromagnetic properties) to maintain combustion, so as to avoid the mechanical system blindly following the millisecond-level high-frequency flicker and protect the mechanical system from fatigue damage and hydraulic shock caused by ineffective oscillation.

[0095] When the power supply abnormality duration t≥t limit At this point, a continuous power interruption is detected, and the safety boost logic is executed. The mechanical system automatically releases the lock on the electrode, switches back to the conventional impedance control logic, and performs normal electrode lifting and lowering operations.

[0096] Meanwhile, after confirming the persistence of the fluctuations, the system achieves real-time reconfiguration of the electrode position through a variable structure for electrode adjustment. This ensures that the electrode can be lifted in time when the power drops significantly, preventing carbon buildup or mechanical collisions caused by the electrode hitting the bottom, thereby protecting production continuity and metallurgical quality during trend fluctuations.

[0097] The locking mode in this embodiment utilizes the physical elasticity (volt-ampere characteristic elasticity) of the arc plasma column maintaining conductivity within a certain stretching range to adapt to transient voltage fluctuations. At this time, the electrode displacement remains constant (i.e., dead zone locking state), which fundamentally eliminates the risk of electrode mis-insertion into the molten pool (carburization) caused by mechanical hysteresis and reduces ineffective oscillation wear of the hydraulic valve assembly.

[0098] Additionally, it is worth mentioning that in this embodiment, the time window filter threshold t limit Essentially, it is the inertial filtering time of mechanical motion, and its value must balance the following two contradictions: Time window filter threshold t limit The lower limit is limited by mechanical wear: if t limit If the duration is too short (e.g., <0.5s), the electrode adjustment system will attempt to follow the instantaneous high-frequency flicker of the green electricity, causing the electrode lead screw and motor to frequently reverse, resulting in mechanical fatigue and thermal overload.

[0099] Time window filter threshold t limit The upper limit is limited by arc physics: if t limit If the arc length is too long (e.g., >3s), the arc length will exceed the current-voltage characteristic compensation limit of the power supply during a deep drop, resulting in arc breakage.

[0100] Therefore, in this embodiment, the decision threshold t of the time window filter limit The preferred setting is 2.0s. This value is chosen based on the following reasons: Matching mechanical response characteristics: The physical response time of the electrode regulator used in this system simulation from receiving the signal to the motor reaching its rated speed is 0.6s. If t limit If the setting is too small, the electrode will frequently generate invalid oscillations under the interference of millisecond-level pulses of green electricity.

[0101] Combined with the fluid dynamics smoothing effect: Since step S3 has switched the bottom-blowing argon flow rate to the arc-maintaining flow rate, the fluctuation amplitude ΔH of the molten pool surface has decreased. According to the static characteristic curve of the arc, under a stable liquid surface, the arc's ability to resist voltage drops is significantly improved. Experimental data shows that, under the premise that the bottom-blowing linkage is effective, voltage fluctuations within 2 seconds can be offset by the volt-ampere characteristic compensation margin, without triggering physical electrode raising or lowering.

[0102] Ensuring timely response: Setting a 2-second upper limit is to prevent the system from missing the optimal time to lift the electrode due to excessive filtering when a trend of power failure occurs, thereby avoiding the risks of electrode carbonization and mechanical impact.

[0103] Step S5: Dynamic recovery and kinetic compensation after power restoration This step smoothly restarts the system after power is restored and compensates for losses from the metallurgical reaction. Specifically, it includes: 1. Hysteresis Reset Logic The system determines that the alarm is cleared once the power supply voltage returns to the steady-state range and the Short-Term Voltage Stability Index (SSI) remains normal for more than a set delay (e.g., 5 seconds). To avoid system overload, the following reset actions are executed in sequence: (1) Close the pressure relief valve (T = 0 ms): Close the pressure relief valve and rebuild the pipeline pressure; (2) Unlock (T + 100 ms): Unlock the electrode position and restore closed-loop impedance control; (3) Restore bottom blowing (T + 200 ms): Smoothly restore the bottom blowing argon flow rate regulating valve to the normal operating position (e.g., 300 L / min).

[0104] 2. Two-layer delay logic By separating the originally highly coupled electrical, pneumatic, and mechanical actions along the time axis, the transient complexity of the system is reduced. This ensures a high degree of coordination between physical stirring and energy injection in the time step during the steel refining process.

[0105] First-level delay: global observation period, to prevent frequent start-stop cycles caused by transient voltage fluctuations.

[0106] Delay logic: During the initial recovery phase of green electricity (wind / solar), a very short secondary voltage drop may occur. The system uses voltage U... grid The system employs dual verification with the Short-Term Voltage Stability Index (SSI) and enforces a steady-state verification period. Only when the signal remains stable throughout the window is the system considered to have completed a valid steady-state identification, thus preventing losses due to misjudgment.

[0107] Second-level delay: Millisecond-level sequential logic Delay logic: Once recovery is determined, the system enters the second-level delay—action decoupling delay. Sequential operations are used in 100ms increments, primarily to achieve temporal decoupling of fluid pressure reconstruction, electrical energy injection recovery, and enhanced kinetic stirring at the physical layer.

[0108] 3. Dynamic compensation The system automatically records the duration t of the weak stirring state in emergency wave suppression mode. low After reset, the system adjusts according to the compensation coefficient k. comp Automatically execute a compensation period of T comp =k comp ·t low Enhanced stirring operations (e.g., temporarily increasing the flow rate to 350 L / min) accelerate mass transfer at the slag-metal interface, compensate for the loss of desulfurization and inclusion removal kinetic processes due to fluctuations in the liquid surface, and ensure that the final steel quality is not affected by fluctuations in green electricity.

[0109] When the duration of the emergency suppression mode is t low Time t exceeding the maximum safety threshold max， If the system determines that there is a serious power supply failure, it will exit the flexible control process and execute emergency protective safety tripping and automatic electrode lifting safety protection actions.

[0110] Considering that: the limit time for the overheated energy storage caused by the temperature dead zone to resist the natural temperature drop is 0.5-0.7 min / ℃; under weak stirring, the LF furnace generally needs to pay close attention to the LF slag line after 5 minutes to prevent the risk of crusting; the green electricity fluctuation characteristics mean that the time for it to recover to stability is generally within 3 minutes, and at most no more than 5 minutes.

[0111] Therefore, considering the above three key time points, the maximum safety threshold time t max Set the time to 300 to 600 seconds.

[0112] In addition, in this embodiment, to prevent energy shortages in the refining process of molten steel due to the trend of insufficient green electricity (such as solar sunset decay and continuous weakening of wind power), an energy supply and demand matching algorithm is provided. Specifically, this algorithm includes refining energy consumption demand calculation, green electricity supply trend extrapolation, and supply and demand matching and alarm triggering. That is, step S6: Execute the next heat's energy supply and demand prediction and interlocking control. Based on the current fluctuation characteristics and trends of green electricity, the system estimates the total amount of green electricity to be supplied in the next refining cycle and compares it with the preset standard smelting energy consumption per furnace. When it is determined that the green electricity supply is insufficient to maintain the refining process of the next furnace, the system outputs a control signal to activate the alarm device, prompting production adjustments or cutting off the feed.

[0113] The specific steps are as follows: Step S61: Calculation of refining energy consumption requirements (Q) need ) Before the current heat cycle ends or before a new heat cycle begins, read the production plan data. Simultaneously, calculate the following parameters: steel mass m, steel specific heat capacity C. p Target temperature rise ΔT.

[0114] The system incorporates historical thermal efficiency η Q Calculate the theoretical total electrical energy Q required for the next refining cycle. need :

[0115] Step S62: Extrapolation of Green Electricity Supply Trends (Q supply ) Extrapolating from historical data, the total electricity Q that green electricity can supply within a future refining cycle (e.g., 40 minutes) is estimated. supply The system sets up an observation window T. window (e.g., the past 15 minutes), for the collected active power P (t) Perform feature analysis and trend projection.

[0116] 1. Feature Differentiation Logic: The system's built-in algorithm distinguishes between non-stationary random disturbances and deterministic trend drops. When voltage and power data satisfy the change slope K...slope <0 and correlation coefficient R 2 If the value is greater than 0.9, it is determined that the current state is in a deterministic downward trend (i.e., showing a monotonically decreasing trend, such as solar sunset, wind farm cut-off, etc.).

[0117] The technical value of introducing R²: If R² is low, it means that the power fluctuation is seriously affected by random factors (such as the instantaneous shading caused by passing clouds), and the power may rebound. It cannot be directly extrapolated based on the slope at this time. If R² > 0.9, it means that the power shows a very strong linear downward trend. Only in this case can the extrapolation calculation have a highly reliable reference value.

[0118] 2. Extrapolation calculation steps (integral prediction method) Core logic: It is assumed that the current energy decay trend (slope) will continue in the next cycle.

[0119] Step 1 (Establishing the Predictive Model): Construct a predictive power function for the future time domain based on linear regression. The calculation formula can be expressed as:

[0120] Where: Δt is the time prediction amount shifted forward from the current time t1 (Δt>0), P now K represents the current instantaneous power. slope The slope (negative value) is obtained from linear regression.

[0121] Step 2 (Solving for total supply): For the predicted power function P... (Δt) In the next refining cycle T refine Perform time-domain integration within (e.g., 40 minutes) to obtain the integral value Q of the total supply. supply1 The calculation formula can be expressed as:

[0122] Wherein: T refine For a pre-set complete refining cycle (e.g., 40 minutes), P (Δt) This is the predicted power function.

[0123] An analytical formula applicable to PLC execution is derived, and a confidence correction coefficient η is introduced. rel (For example, 0.85-0.9) A conservative estimate is made of the integral result to finally obtain Q used to determine supply and demand. supply The value, expressed by the formula, is:

[0124] It is worth mentioning that, due to K slope It is a negative value; the second term actually deducts the amount of electricity lost due to energy decay.

[0125] S63: Supply and Demand Matching and Alarm Triggering The predicted total green electricity supply is compared with the smelting energy consumption demand in real time, and a global matching coefficient k is set. m .

[0126] Decision logic:

[0127] If the above inequality does not hold, the current green electricity trend is determined to be sufficient to support the complete smelting of the next heat; if the above inequality holds, the current green electricity trend is determined to be insufficient to support the complete smelting of the next heat. In this case, an "insufficient green electricity" alarm is issued on-site. Simultaneously, production is locked, and a "prohibit steel intake" or "standby heat preservation" command is sent to the upper-level dispatch system. By intercepting the ladle entering the station at the physical layer, the temperature of the molten steel in the LF furnace is prevented from dropping to the solidification point due to prolonged power shortage, fundamentally eliminating the occurrence of "cold steel" and carbon increase accidents.

[0128] To achieve the above control method, this embodiment also proposes a flexible control system for an LF furnace based on direct green electricity connection.

[0129] refer to Figure 2 As shown, this embodiment of a flexible control system for an LF furnace based on direct green electricity connection includes a power monitoring unit 200, a central processing unit 300, a bottom blowing control unit 400, an electrode adjustment unit 500, and a green electricity unit 600. Among them, A power monitoring unit 200 is installed on the power supply side of the green power unit 600 to collect the voltage signal output by the green power unit 600. Preferably, the power monitoring unit 200 is a power quality analyzer, which acquires the power value U in real time with a sampling period of 10ms. grid Data such as frequency are used to calculate the short-term voltage stability index (SSI).

[0130] The bottom blowing control unit 400 includes a permeable brick 410 disposed inside the ladle refining furnace 100, and the permeable brick 410 is connected to a bottom blowing air source 430 via a bottom blowing pipe 420. The bottom blowing pipe 420 is provided with a regulating valve 440 and a pressure relief valve 450; wherein, the pressure relief valve 450 is located between the regulating valve 440 and the permeable brick 410.

[0131] The electrode adjustment unit 500 includes an electrode 510 disposed within the ladle refining furnace 100, and the electrode 510 is connected to an adjustment mechanism 520 that drives it to perform lifting and lowering operations. Preferably, the adjustment mechanism 520 is an electrode hydraulic servo system, which has two working modes: conventional impedance control and logic locking.

[0132] The central processing unit 300 is connected to the power monitoring unit 200, the bottom blowing control unit 400, and the electrode adjustment unit 500, respectively, and is used to receive feedback signals from each module and send control commands to each module.

[0133] Preferably, the central processing unit 300 uses a high-performance PLC as its control core and is connected to other units via a high-speed industrial fieldbus. Simultaneously, the central processing unit 300 is also connected to an alarm unit 700, used to issue audible and visual alarms or send stop signals to the production scheduling system when green electricity supply is insufficient.

[0134] The present invention and its embodiments have been described above illustratively. This description is not restrictive, and the figures shown are only one embodiment of the present invention; the actual structure is not limited thereto. Therefore, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the present invention, such designs should fall within the protection scope of the present invention.

Claims

1. A flexible control method for an LF furnace based on direct green electricity connection, characterized in that: Includes the following steps, Step S1: Based on the target tapping temperature, set the allowable temperature dead zone range, and use the temperature dead zone range to construct a virtual energy storage model based on the thermal inertia of molten steel. Step S2: Power the LF furnace using a green direct connection method and collect the power supply voltage signal at the LF furnace power supply terminal in real time to determine the power supply stability. When the power supply is detected to be stable and the green electricity input power is sufficient, the overheat energy storage strategy is executed; the target heating temperature is automatically increased and controlled within the temperature dead zone, and the electrical energy is converted into the sensible heat of the molten steel for storage. When the power supply is detected to be unstable or the green electricity input power is insufficient, the high current heating is suspended, and the pre-stored superheat is used to resist the natural temperature drop during the refining process, so as to keep the temperature of the molten steel within the temperature dead zone. Step S3: When it is determined that the power supply is unstable, perform emergency suppression control operation; Reduce the flow rate of bottom-blown argon gas to the arc-preserving flow rate by adjusting the regulating valve; simultaneously open the pressure relief valve on the pipeline between the regulating valve and the bottom permeable brick to directly release the residual gas in the pipeline to the atmosphere. Step S4: If the power supply becomes unstable, but the duration of the instability does not exceed the threshold, the electrode will not shift. If the duration of power instability exceeds the threshold, perform the normal lifting and lowering operation of the electrodes; Step S5: After the power supply is detected to be stable, the pressure relief valve and the regulating valve are reset. Step S6: Perform the next batch of refining supply and demand matching.

2. The flexible control method for an LF furnace based on direct green electricity connection according to claim 1, characterized in that: In step S2, the specific steps for determining power supply instability are as follows: The short-term voltage stability index (SSI) is calculated based on time sliding window analysis and matched with a preset green electricity transient fluctuation model. When the matching degree exceeds the preset threshold and voltage drop characteristics are detected, the power supply is determined to enter an unstable state.

3. The flexible control method for an LF furnace based on direct green electricity connection according to claim 2, characterized in that: In step S3, when the power supply is determined to be in an unstable state, the gas flow rate at the outlet of the permeable brick is rapidly reduced to the arc-preserving flow rate within 200ms by the pressure relief valve, and the minimum voltage threshold required to maintain the electric arc combustion is reduced by the inertial damping of the liquid surface.

4. A flexible control method for an LF furnace based on direct green electricity connection as described in claims 1-3, characterized in that: The specific operation of step S4 is to set the time window filter threshold t. limit Real-time monitoring of the duration t of power instability; If t < t limit If a voltage dip is detected, the electrode locking mode is activated, and the electrode hydraulic servo valve is forcibly locked in the neutral position to prevent the electrode from tracking voltage fluctuations. The physical elasticity of the arc length is used to adapt to voltage fluctuations. If t≥t limit If the system detects a continuous power interruption, it will release the electrode lockout mode and perform electrode lifting and lowering operations under normal impedance control.

5. The flexible control method for an LF furnace based on direct green electricity connection according to claim 4, characterized in that: The time window filter threshold t limit The duration is 0.5 to 3 seconds.

6. The flexible control method for an LF furnace based on direct green electricity connection according to claim 4, characterized in that: The specific operation of step S5 is as follows: Once the power supply voltage returns to the steady-state range and the short-term voltage stability index (SSI) remains normal for more than the set delay, the pressure relief valve is reset first; then, the electrode position is unlocked; finally, the regulating valve is reset to restore normal impedance closed-loop control; and based on the duration of the emergency suppression mode, the amount of kinetic reaction loss is calculated, and either enhanced stirring or extended refining cycle is performed.

7. The flexible control method for an LF furnace based on direct green electricity connection according to claim 1, characterized in that: The specific operation of step S6 is as follows: Based on the current fluctuation characteristics and trends of green electricity, estimate the total green electricity supply for the next refining cycle and compare it with the preset standard smelting energy consumption per furnace. When it is determined that the green electricity supply is sufficient to sustain the refining process of the next batch, the refining operation of the next batch shall be carried out. When it is determined that the green electricity supply is insufficient to sustain the refining process of the next batch, a control signal is output to activate the alarm device, prompting production adjustments or cutting off the feed.

8. The flexible control method for an LF furnace based on direct green electricity connection according to claim 7, characterized in that: In step S6, the specific operation for determining whether the green electricity supply is sufficient to sustain the refining process of the next batch is as follows: S61. Calculation of Refining Energy Consumption Requirements Read the production plan data and calculate the theoretical total electrical energy Q required for the next refining batch. need The calculation formula is: Among them, C p η is the specific heat capacity of the steel grade; m is the mass of molten steel; ΔT is the target temperature rise; η Q Historical thermal efficiency; S62, Extrapolation of Green Electricity Supply Trends Set the observation window T window Linear regression analysis was performed on the collected active power P(t) to calculate the green electricity supply trend; when the slope K of the power change was monitored... slope <0 and correlation coefficient R 2 When the value is greater than 0.9, it is determined to be a deterministic downward trend, and the next refining cycle T is calculated according to the following analytical formula. refine The green electricity within can supply a total of Q electricity. supply , Wherein, η rel P is the confidence level correction factor. now T represents the current instantaneous power. refine For a complete refining cycle; S63. Supply and Demand Matching and Early Warning Triggering Set the matching coefficient k m The predicted total green electricity supply capacity Q supply With the theoretical total electrical energy Q need Perform a comparison; When inequalities When the system is established, if the green electricity supply is insufficient, the control signal is output; otherwise, if the green electricity supply is sufficient, the control signal is output.

9. A flexible control method for an LF furnace based on direct green electricity connection according to claim 2, characterized in that: In step S2, the criterion for judging the voltage drop characteristics is as follows: The effective voltage value was monitored to show a unilateral downward trend for n consecutive power frequency cycles, and the cumulative drop reached the threshold. Alternatively, the Short-Term Voltage Stability Index (SSI) may exceed the preset fluctuation threshold.

10. A flexible control system for an LF furnace based on direct green electricity connection, comprising a green electricity unit (600) and a ladle refining furnace (100) directly connected to the green electricity unit (600), characterized in that: It also includes, Power monitoring unit (200). The power monitoring unit (200) is located on the power supply side of the green power unit (600) and is used to collect the voltage signal output by the green power unit (600); Bottom blowing control unit (400). The bottom blowing control unit (400) includes a permeable brick (410) installed in the ladle refining furnace (100), and the permeable brick (410) is connected to a bottom blowing air source (430) through a bottom blowing pipe (420). The bottom blowing pipe (420) is equipped with a regulating valve (440) and a pressure relief valve (450); wherein the pressure relief valve (450) is located between the regulating valve (440) and the permeable brick (410); Electrode adjustment unit (500). The electrode adjustment unit (500) includes an electrode (510) disposed in the ladle refining furnace (100), and the electrode (510) is connected to an adjustment mechanism (520) that drives it to perform lifting and lowering operations. And a central processing unit (300). The central processing unit (300) is connected to the power monitoring unit (200), the bottom blowing control unit (400), and the electrode adjustment unit (500) respectively, and is used to receive feedback signals from each unit and send control commands to each unit.

Images