A method and system for calculating a blast furnace burden rate

Abstract

The application relates to a blast furnace burden rate calculation method and system and relates to the technical field of blast furnace iron-making process control.The application has the beneficial effects that: the application calculates the unfinished furnace burden batch which crosses the whole point time according to the time proportion segmentation method, eliminates the systematic error caused by the whole point truncation of the traditional method, improves the measurement accuracy of the hourly burden rate, calculates the instantaneous burden rate through the ratio of the real-time descending speed and the average descending speed of the previous batch, shortens the response time from the minute level to the second level, further adopts the sliding window filtering to remove the abnormal jump points, configures corresponding data processing rules for abnormal working conditions such as the hanging burden and the collapsed burden, and ensures the reliability and stability of the burden rate calculation under the abnormal working conditions, and in addition, the application can be realized only by using the existing probe data of the blast furnace, does not need to obtain the easily fluctuating parameters such as the bulk density and the compression rate, and is easy to popularize and apply on the existing blast furnace control system.

Description

Technical Field

[0001] This invention relates to the field of blast furnace ironmaking process control technology, specifically to a method and system for calculating blast furnace charge rate. Background Technology

[0002] In current blast furnace operations, the hourly charge rate is a crucial indicator and judgment parameter. Charge rate refers to the smelting speed of the blast furnace, usually expressed as the number of batches of charge fed into the furnace per hour, reflecting the rate at which the charge is consumed within the furnace (unit: batches / hour). In actual production, operators monitor the charge rate using equipment such as mechanical level gauges / radar level sensors and adjust operations accordingly.

[0003] Under normal furnace conditions, the blast furnace feed rate is affected by various factors, resulting in unstable hourly and instantaneous feed rates. These factors include the amount of molten iron stored in the hearth, tapping speed, tapping interval, oxygen enrichment rate, blast volume, airflow fluctuations, and furnace condition fluctuations. If operators make judgments based on unstable feed rates, it can easily lead to misjudgments and delayed adjustments, which in turn affects the stability and economic indicators of blast furnace production.

[0004] In the description of each batch of blast furnace charge, each batch consists of a certain mass of ore and a certain mass of coke. Ore and coke are charged separately in batches: A charge distributor distributes the charge layer by layer above the blast furnace throat, forming alternating layers of coke and ore. Due to the smelting reactions inside the blast furnace, the lower charge is gradually consumed, and the entire charge layer moves downwards at a certain speed. After the charge is distributed, a mechanical probe hammer is lowered to the top of the charge surface to measure the charge depth. The computer determines the depth of the charge layer based on the depth of the probe's descent. When the probe reaches the set charge depth, it is retracted, and the charge is distributed again. This process is repeated. Generally, the lower edge of the blast furnace throat brick is considered the zero point of the charge line, and the depth from the lower edge of the blast furnace throat brick to the charge surface is the charge depth. When the probe is retracted, there is no material depth signal on the fabric computer. The process of the probe being lowered, contacting the material surface, and being raised, along with changes in the material depth, will generate a material depth record on the fabric computer. Figure 1 As shown.

[0005] Currently, there are two main methods for calculating material batches: (a) The commonly used method for calculating hourly material rate is as follows: the probe is lowered, the probe is gradually moved down, and the probe is retracted when the depth reaches the set material level. This process is counted as half a material. This applies to both ore and coke. The sum of the two is one material. The sum of all half materials in one hour is taken as the material rate for that hour. Therefore, the result can only be a multiple of n (n=0, 1, 2, 3, 4, 5...) of 0.5.

[0006] However, when measuring by hourly intervals, since the calculation is based on the exact hour, the following deviation issues may occur: At the designated time, if the feed line has not reached the required depth but has traveled part of the path, operators often subjectively categorize this unfinished portion into the current or next hour. This method fails to accurately reflect the actual furnace charge consumption within that hour. Figure 1 As shown (7:00 trend line).

[0007] At the exact hour, the material is being fed, but the feed line has already been retracted, and the material level is actually moving downwards. However, this is not counted in the hourly charge consumption statistics, which will cause the actual hourly charge consumption to be underestimated. Figure 1 As shown (8:00 trend line).

[0008] Each hourly interval has two points before and after the start of the hour. The combined problems in (1) and (2) above can sometimes lead to a larger error in the actual material rate within the hourly interval. Figure 1 As shown (7:00-8:00 batch calculation).

[0009] Because coke and ore differ in weight and bulk density, their volumes differ, resulting in varying thicknesses after the charge is laid in the furnace throat. Ore-laid layers are thicker with shallower charge lines, while coke-laid layers are thinner with deeper charge lines (the difference can be 30-50cm). However, both are counted as half a batch, making it impossible to reflect the true rate of charge consumption. Figure 1 As shown (the feed line is narrower after coke is fed, and wider after ore is fed, and the starting depth of the feed line is different).

[0010] (ii) Another method is the theoretical calculation method: This method requires knowing the weight, bulk density, etc. of each batch of material, and combining the blast furnace volume, material line depth, etc. to determine the number of batches of material to be fed per hour through complex calculations.

[0011] However, the theoretical calculation method has the following problems: Because the sources of blast furnace feed are complex and constantly changing, they cannot be stable. For example, there are multiple types of coke from different sources and different types of ore in varying proportions. These factors can easily lead to inconsistencies and large fluctuations in the bulk density, compressibility, and moisture content of the feed. As a result, the calculated volume will differ significantly from the actual volume, and values ​​calculated based on inaccurate data will also be biased.

[0012] When calculating the volume, it is assumed that the entire throat material surface is evenly spread. However, due to the need for blast furnace airflow regulation and other influences, the entire upper material surface of the blast furnace will present a shape of a central funnel plus an edge platform after the material is distributed. Calculating the material surface height by using the volume spread will result in model distortion, and the calculation accuracy is difficult to guarantee.

[0013] In view of this, the purpose of this invention is to provide a simple, practical and accurate calculation method for blast furnace feed rate under normal operating conditions and a method for measuring instantaneous feed rate, so as to solve the existing problems. Summary of the Invention

[0014] The purpose of this invention is to provide a method and system for calculating blast furnace feed rate, addressing the shortcomings and defects of existing technologies.

[0015] To achieve the above objectives, the present invention adopts the following technical solution: a method and system for calculating blast furnace feed rate, comprising the following steps: acquiring data on the feed line depth of the blast furnace probe over time, and recording the moment when the probe completes its retraction action after reaching the set feed line each time; determining the consumption time ΔT of a single batch of feed based on the time difference between the retraction times of two adjacent batches of the same type of feed, i.e., ΔT = T2 - T1, where T1 is the retraction time of the probe for the previous batch of the same type of feed, and T2 is the retraction time of the probe for the current batch of the same type of feed. ΔT is in seconds; for a batch of furnace charge that is not fully consumed before the hour H, after the batch is completely consumed, calculate the time length ΔT1 consumed before the hour H and the time length ΔT2 consumed after the hour H, and ΔT1 + ΔT2 = ΔT. Then, the number of batches contributed by this batch to the hour before the hour is ΔT1 / ΔT, and the number of batches contributed to the hour after the hour is ΔT2 / ΔT; calculate the blast furnace hourly charge rate P within the given hour interval [H, H+1 hours]. h = P complete + P prev + P next Unit: batches / hour, where N complete P represents the number of complete batches of material that were completely consumed within that hour. prev P represents the proportion of the first batch entering that hour that is consumed within that hour's time interval. next P represents the proportion of the last batch consumed within that hour's time interval. prev and P next All calculations were performed using the time-proportional segmentation method.

[0016] Furthermore, it also includes an instantaneous material velocity estimation step: using the real-time descent speed V of the probe during continuous feeding. 现 Compared with the average decline rate V of the previous batch of materials 前 The ratio, combined with the hourly batch number P of the previous batch of material. 前 Calculate the current instantaneous material velocity P 现 =P 前 ×V 现 / V 前 .

[0017] Furthermore, the real-time descent rate V 现V is calculated by dividing the height difference L2-L1 between the two endpoints of the probe trajectory by the corresponding time difference t2-t1, i.e. 现 =(L2-L1)÷(t2-t1); The average descent rate V of the previous batch of material 前 = (V 前焦 +V 前矿 ) ÷ 2, where V 前焦 and V 前矿 These are the average descent rates of the previous coke layer and ore layer, respectively; the hourly batch number P of the previous batch of material. 前 =3600 / ⊿T 前 , ⊿T 前 This is the total consumption time of the previous batch of materials.

[0018] Furthermore, when the probe retracts during the blast furnace charging process and there is no charging curve, the instantaneous charging speed of the blast furnace is defaulted to the charging speed of the first half batch.

[0019] Furthermore, the probe's descent speed V 现 The calculation employs a sliding window filtering algorithm, taking the five most recent valid sampling points for univariate linear regression, and using their slope as V. 现 Abnormal jump points caused by mechanical friction or furnace charge collapse are eliminated.

[0020] Furthermore, it also includes abnormal operating condition handling steps: when the probe descent rate is lower than the first preset rate threshold and the duration exceeds the first preset time threshold, and when the blast furnace air pressure and / or top pressure show abnormal increase characteristics, it is determined to be suspended material, the effective calculation of instantaneous material velocity is suspended and a suspended material alarm signal is output until the probe returns to normal and continuously and stably descends for more than the preset recovery time; when the probe descent depth ΔL within the preset short time window exceeds the preset depth threshold, it is determined to be collapsed material or slippage, and the material consumption corresponding to the abnormal descent segment is smoothed or evenly distributed to multiple normal feeding cycles before and after the abnormal event for calculation.

[0021] Furthermore, the first preset time threshold is 1.5-2 times the normal feeding cycle, the preset recovery time is 1 minute; the preset short time window is less than 1 / 10 of the normal feeding sampling cycle, and the preset depth threshold is 1.5 times the normal material layer thickness.

[0022] Furthermore, the same type of furnace charge is one of coke or ore.

[0023] A blast furnace feed rate calculation system includes: a data acquisition module for acquiring blast furnace probe depth data over time and recording the moment the probe retracts after reaching a set feed line each time; a single batch consumption time determination module for determining the consumption time ΔT of a single batch of feed based on the time difference between the retraction times of two adjacent batches of the same type of feed probe; a cross-hour batch segmentation module for calculating the proportion of consumption time before and after hourly hours H for batches of feed that have not been consumed and cross hourly hours H, and determining the number of feed batches contributed by that batch to the two hours before and after the hourly hours; and an hourly feed rate calculation module for calculating the blast furnace hourly feed rate P within a given hourly interval [H, H+1 hours]. h = P complete + P prev + P next Unit: batches / hour.

[0024] Furthermore, it also includes: an instantaneous material velocity estimation module, used to estimate the real-time descent speed V of the probe during continuous feeding. 现 Compared with the average decline rate V of the previous batch of materials 前 The ratio, combined with the hourly batch number P_previous of the previous batch, is used to calculate the current instantaneous material rate P. 现 =P 前 ×V 现 / V 前 The abnormal operating condition handling module is used to determine whether the probe descent rate is lower than the first preset rate threshold and the duration exceeds the first preset time threshold, and to pause the material rate calculation and output an alarm when combined with abnormal increases in blast furnace air pressure and / or top pressure; and to determine whether the probe descent depth exceeds the preset depth threshold within a preset short time window as material collapse or probe slippage, and to smoothly distribute the corresponding consumption amount to the calculation within multiple normal feeding cycles before and after; the filtering module is used to calculate the probe descent speed V using a sliding window filtering algorithm. 现 Perform a univariate linear regression on the five most recent valid sampling points, and use the slope as V. 现 It eliminates abnormal jump points; the probe redundancy switching module is used to automatically switch to the backup probe or other material level monitoring device data source when the currently used probe data shows signs of jamming or disconnection.

[0025] After adopting the above technical solution, the beneficial effects of the present invention include at least the following: 1. This invention utilizes a time-ratio segmentation method to calculate the contribution of unfinished batches of furnace charge that span the hour. Based on the proportion of time consumed before and after the hour, the number of batches contributing to the two hours before and after the hour is calculated. This eliminates the systematic error (up to 0.5 batches / hour) caused by hourly truncation in traditional methods, reducing calculation deviation to an extremely small level determined by factors such as timing accuracy. Furthermore, by calculating the consumption time of a single batch using the same type of furnace charge (coke or ore) as a benchmark, it avoids statistical distortion caused by simply adding the coke and ore layers due to differences in thickness. This provides blast furnace operators with an accurate basis for judging smelting intensity.

[0026] 2. This invention is the first to propose an instantaneous material velocity calculation model based on the slope of the actual probe trajectory. By comparing the real-time descent speed with the average descent speed of the previous batch, and combining this with the hourly batch number of the previous batch, the current instantaneous material velocity is calculated. The response time is shortened from the traditional "end of each half-batch (approximately 5 minutes)" to "seconds". Operators can observe the changes in material velocity after the implementation of adjustments such as pulverized coal injection rate and oxygen enrichment rate in real time, facilitating precise control of the fuel ratio and reducing adjustment lag.

[0027] 3. This invention uses a sliding window filtering algorithm to eliminate abnormal jump points caused by mechanical friction, furnace charge collapse, etc., avoiding drastic instantaneous speed fluctuations caused by single-point noise. At the same time, it clearly distinguishes abnormal working conditions such as suspended charge, collapsed charge / slipping charge and designs targeted data processing rules. For suspended charge conditions, the material speed calculation is paused and an alarm is triggered. For collapsed charge / slipping charge conditions, the abnormal consumption is smoothly distributed to the calculation within multiple normal cycles before and after, avoiding the contamination of calculation results by abnormal data. This ensures the reliability and stability of material speed calculation under abnormal blast furnace working conditions and provides a reliable input for the intelligent control system.

[0028] 4. This invention does not require obtaining easily fluctuating parameters such as the bulk density, compressibility, and moisture content of the furnace charge, nor does it require establishing a complex material surface volume model. It can achieve accurate calculations using only the material line depth data from the existing blast furnace probes. It has low equipment requirements and is easy to promote and apply to existing blast furnace control systems. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a schematic diagram of the time-depth trend of the blast furnace charging line in this invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0032] Example 1: Calculation of Hourly Material Rate This embodiment uses a certain 1800m 3 Taking a blast furnace as an example, the material line depth is set at 1.5 meters, the ore batch weight is 57 tons, and the coke batch weight is 12 tons.

[0033] (a) Determining the consumption time of a single batch of materials: Using the same type of furnace charge as a benchmark, the time when the probe reaches the set material line and completes the retraction action is recorded. In this embodiment, coke is used as the benchmark, but ore can also be used.

[0034] If the time when the previous batch of coke was collected (i.e., the time when the current batch of coke began to be consumed) is set to 9:50:00, and the time when the current batch of coke was collected is 10:06:00, then the consumption time ΔT for a single batch of furnace charge is: ΔT = T2 - T1 = 10:06:00 - 9:50:00 = 960 seconds This ΔT corresponds to the consumption of one batch of material.

[0035] (ii) Material batch splitting across whole-hour periods: Taking the batch from 9:50:00 to 10:06:00 as an example, this batch spans the hour 10:00:00. After this batch is completely consumed, calculate the time length ΔT1 consumed before the hour 10:00:00 and the time length ΔT2 consumed afterward: ⊿T1 = 10:00:00 - 9:50:00 = 600 seconds (consuming before the hour) ⊿T2 = 10:06:00 - 10:00:00 = 360 seconds (consuming after the hour) ΔT1 + ΔT2 = 600 + 360 = 960 seconds = ΔT The number of batches that this batch contributes to the hour before the hour (9:00-10:00) is: ΔT1 / ΔT = 600 / 960 = 0.625 lots The number of batches contributing material to the hour following the hour (10:00-11:00) is: ΔT2 / ΔT = 360 / 960 = 0.375 lots (III) Calculation of hourly material flow rate: Calculate the hourly blast furnace feed rate P during the 10:00-11:00 hour interval. h .

[0036] The number of complete batches N that were completely consumed within that hour complete Assuming there are 4 batches of materials whose start and end times are all within 10:00-11:00, then NN complete = 4 batches.

[0037] Entering the first batch of P in that hour prev This means that the batch consumed between 9:50:00 and 10:06:00 had a consumption rate of 0.375 batches within the hourly time interval (10:00-11:00).

[0038] Stepping out of the last batch of P that hour next For example, the batch from 10:58:00 to 11:14:00 has consumed 120 seconds by 11:00:00, with a total consumption time of 960 seconds. Therefore, the contribution is 120 / 960 = 0.125 batches.

[0039] The blast furnace feed rate for that hour is: P h = N complete + P prev + P next = 4 + 0.375 + 0.125 = 4.5 batches / hour Example 2: Estimation of Instantaneous Material Rate (a) Calculation of the average descent rate of the previous batch of material: After each half-batch of material is discharged, the probe is lowered and gradually moves downwards with the material layer until the probe reaches the set material depth, at which point the probe is retracted. On the operating computer, the trajectory of the tops of the two material lines of the probe appears as a diagonal line.

[0040] Taking a previous batch of coke as an example, with the height difference between the two endpoints of the probe trajectory as the downward travel distance and the time difference between the endpoints of the two vertical lines as the time taken, the downward movement speed of this half-batch of coke is: V 前焦 = (L2 - L1) ÷ (t2 - t1) Similarly, the downward movement speed of the half-batch of ore from the previous batch is V_front ore.

[0041] The average downward movement speed of the entire batch of material pushed forward by the half-batch is: V 前 = (V 前焦 + V 前矿 ) ÷ 2 By using the time difference ΔT of the previous batch of material (e.g., the consumption time of the previous batch of coke was 750 seconds), estimate the number of batches of material consumed per hour at this material rate during this time period: P 前 = 3600 ÷ ΔT = 3600 ÷ 750 = 4.8 batches / hour (ii) Calculation of the current instantaneous material velocity: When the lower half of the batch moves along the probe trajectory in an oblique line but has not yet reached the set material line, its instantaneous downward movement rate is expressed as: V 现 = (L2 - L1) ÷ (t2 - t1) For example, within 30 seconds from 10:15:30 to 10:16:00, the probe descended from a depth of 1.05 meters to 1.11 meters: V 现 = 0.06 meters ÷ 30 seconds = 0.002 m / s Given that the average descent velocity of the previous batch of material was V_front ≈ 0.003 m / s, and P_front = 4.8 batches / hour, then the current instantaneous material velocity is: P 现 = P 前 × V 现 / V 前 = 4.8 × (0.002 / 0.003) = 3.2 batches / hour (III) Processing during the fabric weaving process: During the blast furnace charging process, the probe is retracted and there is no charging curve. At this time, the instantaneous charging speed of the blast furnace is assumed to be the charging speed of the first half batch as a reference.

[0042] (iv) Sliding window filtering: probe descent speed V 现 The calculation employs a sliding window filtering algorithm, specifically: taking the five most recent valid sampling points (depth, time), performing a univariate linear regression, and using their slope as V. 现 When the depth value of a sampling point deviates from the regression prediction value by more than a preset threshold, that point is determined to be an abnormal jump point and is removed from the V-test. 现 The calculation effectively avoids drastic instantaneous speed fluctuations caused by discontinuous feeding factors such as mechanical friction or local collapse of furnace charge.

[0043] Example 3: Handling Abnormal Operating Conditions (I) Suspended Material Detection and Handling: During a certain operation, the system monitored the probe descent rate and blast furnace air pressure and top pressure data in real time. When it was detected that the probe descent rate was continuously below 0.05 cm / s for 250 consecutive seconds (the normal feeding cycle is 150 seconds, and 250 seconds here is about 1.67 times the normal cycle), and at the same time the air pressure rose from 300 kPa to 350 kPa, and the top pressure showed a corresponding abnormal increase, it was determined to be a case of suspended material.

[0044] Once the system determines that the material is suspended, it pauses the effective calculation of the instantaneous material velocity. The instantaneous material velocity display value remains the valid value of the last frame before the suspension, and a suspension alarm signal is output to prompt the operator to handle the situation.

[0045] After 120 seconds, the probe resumed its descent. The system continuously monitored the probe's descent status. Once the probe had been descending steadily for more than 1 minute, the system automatically exited the suspended material processing mode and resumed normal instantaneous material rate calculation.

[0046] (II) Collapse / Slippage Detection and Handling: During a certain operation, the system detected that the probe descended more than 1.5 times the normal material layer thickness within 2 seconds (the normal material feeding sampling cycle is 30 seconds, and 2 seconds is less than 1 / 10 of the sampling cycle), which was judged as material collapse or probe slippage.

[0047] Once a collapse / slippage is identified, the material consumption corresponding to the abnormal descent segment is not used as a direct basis for calculating the instantaneous material rate. The system smooths or evenly distributes the furnace charge volume (or equivalent number of material batches) represented by the abnormal descent segment across multiple normal feeding cycles before and after the abnormal event for calculation. For example, the abnormal descent amount is evenly distributed across two normal feeding cycles before and after the event to avoid drastic non-physical abrupt changes in the instantaneous material rate.

[0048] (III) Redundancy switching of probes: The system monitors the status of the probe data in real time. When the probe data shows signs of jamming or disconnection, the system automatically switches to the data source of the backup probe or other material level monitoring devices (such as radar level gauges) to ensure the continuity of material rate calculation.

[0049] Example 4: System Module Architecture In specific implementations, the blast furnace feed rate calculation system of the present invention can be integrated into the host computer or PLC (programmable logic controller) of the blast furnace process control system.

[0050] The data acquisition module acquires the material line depth data and timestamp of the probe in real time through industrial communication protocols (such as OPC, Modbus, etc.), and records the moment when the probe completes the retraction action after reaching the set material line each time. The data sampling frequency is not less than 1Hz to ensure the time resolution of instantaneous material speed calculation.

[0051] The single-batch consumption time determination module and the cross-hour batch segmentation module run as timed tasks in the above-mentioned host computer or PLC. After each batch of the same type of furnace material is consumed, a calculation is triggered to update ΔT and the contribution ratio of each hour boundary batch.

[0052] The hourly material rate calculation module is triggered every hour on the hour, summarizing N for that hour. complete P prev and P next Output P h The operation screen will then be displayed.

[0053] The instantaneous material velocity estimation module calculates V in real time using the probe data sampling period as the step size (usually 1 second or less). 现 and P 现 The data is then output to the operation screen for real-time monitoring by the operator.

[0054] The abnormal operating condition handling module and the filtering module are embedded and run in the above-mentioned calculation modules. When an abnormal operating condition is detected, it is handled in accordance with the method described in Example 3.

[0055] The probe redundancy switching module determines the probe's working status through hardware status monitoring (such as probe motor current, encoder signal, etc.) and automatically switches the data source when necessary.

[0056] The specific implementation of each of the above modules can be achieved through software programming, firmware, or hardware logic circuits, all of which fall within the protection scope of this invention.

[0057] The above description is only used to illustrate the technical solution of the present invention and is not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solution of the present invention, as long as they do not depart from the spirit and scope of the technical solution of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A method for calculating blast furnace feed rate, characterized in that, Includes the following steps: Acquire the material depth data of the blast furnace probe over time, and record the moment when the probe completes the retraction action after reaching the set material line each time; Based on the time difference between the retraction times of the probes of two adjacent batches of the same type of furnace charge, the consumption time ΔT of a single batch of furnace charge is determined, i.e., ΔT = T2 - T1, where T1 is the retraction time of the probes of the previous batch of the same type of furnace charge, T2 is the retraction time of the probes of the current batch of the same type of furnace charge, and ΔT is in seconds. For a batch of furnace charge that is not consumed before the hour H, after the batch is completely consumed, calculate the time length ΔT1 consumed before the hour H and the time length ΔT2 consumed after the hour H, and ΔT1+ΔT2=ΔT. Then, the number of batches contributed by this batch to the hour before the hour is ΔT1 / ΔT, and the number of batches contributed to the hour after the hour is ΔT2 / ΔT. Calculate the hourly blast furnace feed rate P within a given hourly interval [H, H+1 hours]. h = P complete + P prev + P next Unit: batches / hour, where N complete P represents the number of complete batches of material that were completely consumed within that hour. prev P represents the proportion of the first batch entering that hour that is consumed within that hour's time interval. next P represents the proportion of the last batch consumed within that hour's time interval. prev and P next All calculations were performed using the time-proportional segmentation method.

2. The method for calculating blast furnace feed rate according to claim 1, characterized in that: It also includes the instantaneous material rate estimation step: The real-time descent speed V of the probe during continuous feeding process 现 Compared with the average decline rate V of the previous batch of materials 前 The ratio, combined with the hourly batch number P of the previous batch of material. 前 Calculate the current instantaneous material velocity P 现 =P 前 ×V 现 / V 前 .

3. The method for calculating blast furnace feed rate according to claim 2, characterized in that: The real-time descent speed V 现 V is calculated by dividing the height difference L2-L1 between the two endpoints of the probe trajectory by the corresponding time difference t2-t1, i.e. 现 =(L2-L1)÷(t2-t1); The average descent rate V of the previous batch of material 前 = (V 前焦 +V 前矿 ) ÷ 2, where V 前焦 and V 前矿 These are the average descent rates of the previous coke layer and ore layer, respectively; the hourly batch number P of the previous batch of material. 前 =3600 / ⊿T 前 , ⊿T 前 This is the total consumption time for the previous batch of materials.

4. The method for calculating blast furnace feed rate according to claim 3, characterized in that: When the probe retracts during the blast furnace charging process and there is no charging curve, the instantaneous charging speed of the blast furnace is assumed to be the charging speed of the first half batch.

5. The method for calculating blast furnace feed rate according to claim 2, characterized in that: The probe descent speed V 现 The calculation employs a sliding window filtering algorithm, taking the five most recent valid sampling points for univariate linear regression, and using their slope as V. 现 Abnormal jump points caused by mechanical friction or furnace charge collapse are eliminated.

6. A method for calculating blast furnace feed rate according to claim 1 or 2, characterized in that: It also includes abnormal operating condition handling procedures: When the probe descent rate is lower than the first preset rate threshold and the duration exceeds the first preset time threshold, and when the blast furnace air pressure and / or top pressure show abnormal increase characteristics, it is determined to be suspended material. The effective calculation of instantaneous material rate is suspended and a suspended material alarm signal is output until the probe returns to normal and the continuous and stable descent exceeds the preset recovery time. When the descent depth ΔL of the probe exceeds the preset depth threshold within a preset short time window, it is determined to be a material collapse or probe slippage. The material consumption corresponding to the abnormal descent segment is then smoothed or evenly distributed over multiple normal feeding cycles before and after the abnormal event for calculation.

7. The method for calculating blast furnace feed rate according to claim 6, characterized in that: The first preset time threshold is 1.5-2 times the normal feeding cycle, the preset recovery time is 1 minute; the preset short time window is less than 1 / 10 of the normal feeding sampling cycle, and the preset depth threshold is 1.5 times the normal material layer thickness.

8. The method for calculating blast furnace feed rate according to claim 1, characterized in that: The same type of furnace charge is one of coke or ore.

9. A blast furnace feed rate calculation system, characterized in that, include: The data acquisition module is used to acquire the material line depth data of the blast furnace probe over time, and record the moment when the probe completes the retraction action after reaching the set material line each time. The single batch consumption time determination module is used to determine the consumption time ΔT of a single batch of furnace charge based on the time difference between the retraction times of two adjacent furnace charge probes of the same type. The cross-hour batch segmentation module is used to calculate the proportion of time length consumed by the batch before and after the hour H for the unfinished batch of furnace charge that crosses the hour H, and determine the number of batches of charge contributed by the batch to the two hours before and after the hour. The hourly material rate calculation module is used to calculate the blast furnace hourly material rate P within a given hourly interval [H, H+1 hours]. h =P complete + P prev + P next Unit: batches / hour.

10. The blast furnace feed rate calculation system according to claim 1, characterized in that, include: The instantaneous material velocity estimation module is used to estimate the real-time descent speed V of the probe during continuous feeding. 现 Compared with the average decline rate V of the previous batch of materials 前 The ratio, combined with the hourly batch number P_previous of the previous batch, is used to calculate the current instantaneous material rate P. 现 =P 前 ×V 现 / V 前 ; The abnormal operating condition processing module is used to determine that when the probe descent rate is lower than the first preset rate threshold and the duration exceeds the first preset time threshold, and when the blast furnace air pressure and / or top pressure show abnormal increase characteristics, it is considered as suspended material and the material rate calculation is paused and an alarm is output; and when the probe descent depth exceeds the preset depth threshold within a preset short time window, it is considered as collapsed material or slippage, and the corresponding consumption is smoothly distributed to the calculation within multiple normal feeding cycles before and after. The filtering module is used to calculate the probe descent speed V using a sliding window filtering algorithm. 现 Perform a univariate linear regression on the five most recent valid sampling points, and use the slope as V. 现 Remove abnormal transition points; The probe redundancy switching module is used to automatically switch to a backup probe or other material level monitoring device data source when the currently used probe data shows signs of jamming or disconnection.

Images