TWI819911B - Method for predicting gas level of coke oven gas reservoir - Google Patents

Abstract

The present disclosure provides a method for predicting gas level of coke oven gas reservoir, including implementing a coke oven gas pressure to gas level of coke oven gas reservoir analysis, so as to derive a relationship between the coke oven gas pressure and a gas level variation of coke oven gas reservoir; implementing a natural gas input to gas level of coke oven gas reservoir analysis, so as to derive a relationship between the natural gas input and the gas level variation of coke oven gas reservoir; and implementing an analysis of a gas level variation prediction of coke oven gas reservoir after an input of the natural gas according to the relationship between the coke oven gas pressure and a gas level variation of coke oven gas reservoir, the natural gas input, coke oven gas pressure in a predefined previous time interval, and a current gas level of coke oven gas reservoir, so as to predict the gas level of coke oven gas reservoir of next minute.

Description

Method for predicting the location of coke oven gas storage tanks

The present disclosure relates to a method for predicting storage tank slots, and in particular, to a method for predicting coke oven gas storage tank slots.

In the petrochemical industry, the coke oven gas produced by the coking process can be stored in coke oven gas storage tanks and used to supply to various production plants as power generation energy when needed.

Under normal circumstances, a single coke oven gas storage tank is often connected to multiple production plants to supply energy through connecting equipment such as pipeline networks. If there is a shortage of coke oven gas, the pressure in the coke oven gas storage tank and pipe network will easily be too low, causing operational risks. Specifically, when the pipe network pressure is lower than 400mmH ₂ O, the equipment in each factory may trip due to insufficient power. At this time, a common practice is to manually inject natural gas and raise the coke oven gas storage tank to avoid equipment tripping in the plant. However, it is actually difficult to predict how the amount of natural gas injected will affect the location of the coke oven gas storage tank. People often can only adjust the amount of natural gas injected based on experience. If the operator is inexperienced, he may not be able to adjust the amount of natural gas injection in time, which may cause the equipment in the plant to shut down. Therefore, how to accurately predict the changes in the coke oven gas storage tank position has become a crucial issue in the coke oven manufacturing process.

Therefore, there is an urgent need for a method that can accurately predict the coke oven gas storage tank slots. It can predict the coke oven gas storage tank slot changes within a certain period of time based on the past parameters of the coke oven gas storage tanks for online staff to browse. Natural gas is injected immediately after the prediction result to prevent the coke oven gas storage tank from being too low.

Therefore, one purpose of an embodiment of the present disclosure is to provide a method for predicting the coke oven gas storage tank slot, which includes measuring the coke oven gas pressure vs. the coke oven gas storage tank slot and the natural gas input amount vs. the coke oven gas storage tank slot. Analyze and obtain the relationship between coke oven gas pressure and natural gas input volume on the change of coke oven gas storage tank slots, and predict the coke oven gas storage tank slots by combining the two relationship expressions and other parameters to accurately grasp the coke oven gas storage Changes in slot slots.

According to the above purpose of the present disclosure, a method for predicting the coke oven gas storage tank slot is proposed, which includes analyzing the coke oven gas pressure versus the coke oven gas storage tank slot to obtain the relationship between the coke oven gas pressure and the coke oven gas storage tank slot location. ; Carry out an analysis of the natural gas input volume and the coke oven gas storage tank slots to obtain the relationship between the natural gas input volume and the coke oven gas storage tank slot changes; and the relationship between the coke oven gas pressure and the coke oven gas storage tank slot changes, and the natural gas input volume and coke oven gas storage tank slot changes. The relationship between gas storage tank slot changes, natural gas input volume, coke oven gas pressure before the preset time period, and current coke oven gas storage tank slots are used to predict and analyze the changes in coke oven gas storage tank slots after natural gas injection, and predict Coke oven gas storage tank slot for the next minute.

According to some embodiments of the present disclosure, the above-mentioned analysis of coke oven gas pressure on coke oven gas storage tank locations includes performing several coke oven gas pressure detections on several coke oven gas storage tanks to obtain the data of the coke oven gas storage tanks. Coke oven gas pressure; perform detection of several coke oven gas storage tank slots on the coke oven gas storage tank to obtain changes in the coke oven gas storage tank slots; and detect the coke oven gas pressure and The coke oven gas storage tank slot is subjected to the first regression analysis process to obtain the relationship between the coke oven gas pressure and the coke oven gas storage tank slot position.

According to some embodiments of the present disclosure, the above-mentioned relationship between the coke oven gas pressure and the coke oven gas storage tank position change is: Y1=(0.021297X1-21.55493)， Among them, Y1 is the percentage change of the coke oven gas storage tank position per minute, and X1 is the average coke oven gas tank pressure per minute.

According to some embodiments of the present disclosure, the above-mentioned analysis of the natural gas input amount to the coke oven gas storage tank slot includes calculating the coke oven gas supply amount of the coking process to the coke oven gas storage tank based on several coking process records per hour; based on Several production process records are used to calculate the coke oven gas consumption of the coke oven gas storage tank per hour; based on the coke oven gas supply and coke oven gas consumption, the hourly coke oven gas consumption of the coke oven gas storage tank is calculated. gas injection amount; perform a second regression analysis process on the natural gas injection amount and the coke oven gas injection amount to obtain the relationship between the natural gas injection amount and the coke oven gas injection amount; and obtain the natural gas input based on the relationship between the natural gas injection amount and the coke oven gas injection amount. The relationship between quantity and coke oven gas storage tank position changes.

According to some embodiments of the present disclosure, the above relationship between the natural gas injection amount and the coke oven gas injection amount is: 1.836822X2=X3， Among them, X2 is the natural gas injection amount per minute, and X3 is the coke oven gas injection amount per minute.

According to some embodiments of the present disclosure, the relationship between the above-mentioned natural gas input volume and the change of the coke oven gas storage tank slot is: Y1=0.000051X2， Among them, Y1 is the percentage change of the coke oven gas storage tank slot per minute, and X2 is the natural gas injection volume per minute.

According to some embodiments of the present disclosure, the above-mentioned natural gas input volume analysis of coke oven gas storage tank slots further includes performing a data cleaning operation to remove several invalid data or a plurality of low-relevance data in the coking process records and production process records.

According to some embodiments of the present disclosure, the above-mentioned coking process records include coking process parameters of each stage of coking.

According to some embodiments of the present disclosure, the above-mentioned production process records include several hot rolling process records, several steel plate production process records, several steel blank production process records, several bar production process records, and several wire rod production process records. , coke oven gas storage tank change records, and several coke oven gas usage records.

According to some embodiments of the present disclosure, the above-mentioned method of predicting coke oven gas storage tank slots further includes repeatedly performing a prediction analysis of coke oven gas storage tank slot changes after natural gas injection to predict the long-term performance of the current coke oven gas storage tank slots. change.

It can be seen from the above embodiments of the present disclosure that the present disclosure mainly obtains the relationship between the coke oven gas pressure and the change of the coke oven gas storage tank slot and the relationship between the natural gas input amount and the change of the coke oven gas storage tank slot through regression analysis, and then combines the relationship between the coke oven gas pressure and the coke oven gas storage tank slot. The relationship between gas pressure and the change of coke oven gas storage tank slots and the relationship between natural gas input volume and coke oven gas storage tank slot changes are used to predict the long-term changes in coke oven gas storage tank slots after natural gas injection. The disclosed method for predicting coke oven gas storage tank slots has a high degree of accuracy in prediction, and can instantly present the prediction results to online staff. Therefore, it can effectively avoid the occurrence of excessive coke oven gas storage tank slot levels in various factories. The device tripped. In addition, the analysis of natural gas input volume on coke oven gas storage tank slots also includes data cleaning operations to remove invalid data and low-relevance data, thereby improving the accuracy of the analysis.

Embodiments of the present disclosure are discussed in detail below. It is to be appreciated, however, that the embodiments provide many applicable concepts that can be embodied in a wide variety of specific contexts. The embodiments discussed and disclosed are for illustration only and are not intended to limit the scope of the present disclosure.

Please refer to FIG. 1 , which is a flow chart illustrating a method for predicting coke oven gas storage tank slots according to some embodiments of the present disclosure. In some embodiments, method 100 includes step 200 , step 300 , step 400 , and step 500 . It should be noted that the method 100 is only an example and is not intended to limit the present disclosure to the clearly disclosed patent protection scope. Additional steps may be provided before, during, and after method 100 , and some of the steps described may be moved, removed, replaced, or deleted for additional implementations of method 100 .

When performing the method 100 of predicting the coke oven gas storage tank slot, step 200 can be performed first to analyze the coke oven gas pressure versus the coke oven gas storage tank slot to obtain the relationship between the coke oven gas pressure and the coke oven gas storage tank slot location. . Please also refer to FIG. 2 , which is a flow chart illustrating coke oven gas pressure versus coke oven gas storage tank slot analysis according to some embodiments of the present disclosure. In some embodiments, performing coke oven gas pressure versus coke oven gas storage tank location analysis includes step 210 , step 220 , and step 230 . Likewise, step 200 is merely illustrative and is not intended to limit the present disclosure to the clearly disclosed scope of patent protection. Additional steps may be provided before, during, and after step 200, and some of the steps described may be moved, removed, replaced, or deleted for additional implementations of step 200.

When analyzing coke oven gas pressure versus coke oven gas storage tank slots, step 210 can be performed first to perform several coke oven gas pressure detections on several coke oven gas storage tanks to obtain several coke oven gas storage tank locations. Coke oven gas pressure. Specifically, in some embodiments, online staff can measure the coke oven gas pressure in each coke oven gas storage tank through a coke oven gas pressure detector pre-connected to each coke oven gas storage tank, and continuously record the coke oven gas pressure. Then, the average coke oven gas pressure within a specific time interval (for example, but not limited to, 15 minutes) is obtained. In some examples, the coke oven gas pressure detector may be, but is not limited to, a pointer pressure gauge, a digital contact pressure gauge, a diaphragm pressure gauge, etc. The coke oven gas pressure detector is not limited to the type of pressure gauge, but it only needs to be able to accurately measure the coke oven gas pressure in each coke oven gas storage tank.

When the coke oven gas pressure is analyzed for the coke oven gas storage tank slots, step 220 can be performed next to perform several coke oven gas storage tank slot detections on the coke oven gas storage tanks to obtain several coke oven gas storage tank slots. Coke oven gas storage tank location changes. Specifically, in some embodiments, online operators can measure the material level in each coke oven gas storage tank through a coke oven gas tank level detector pre-connected to each coke oven gas storage tank, and after continuous recording The average coke oven gas tank position change within a specific time interval (eg, but not limited to, 15 minutes) is obtained. In some examples, the coke oven gas tank level detector may be, but is not limited to, a capacitive level meter, an ultrasonic level meter, a radar wave level meter, etc. The coke oven gas tank level detector is not limited to the type of material level meter, but it only needs to be able to accurately measure the material level in each coke oven gas tank.

Next, step 230 is performed to perform a first regression analysis process on the coke oven gas pressure and the coke oven gas storage tank location to obtain the relationship between the coke oven gas pressure and the coke oven gas storage tank location. For example, after measuring the average coke oven gas pressure of each coke oven gas storage tank and the average coke oven gas tank position change every 15 minutes, the operator can use various analysis programming languages or software, for example, but Not limited to, R language, Python, Excel, and SAS are used to perform regression analysis to obtain the relationship between coke oven gas pressure and coke oven gas storage tank slot changes. In some examples, the relationship between coke oven gas pressure and coke oven gas storage tank location is: Y1=(0.021297X1-21.55493)， Among them, Y1 is the percentage change of the coke oven gas storage tank position per minute, and X1 is the average coke oven gas tank pressure per minute.

The method of predicting the location of the coke oven gas storage tank then proceeds to step 300. Please refer to FIG. 1 and FIG. 3 at the same time. FIG. 3 is a flow chart illustrating the analysis of the natural gas input amount and the coke oven gas storage tank slot according to some embodiments of the present disclosure. When performing step 300, the natural gas input amount and the coke oven gas storage tank slots can be analyzed first to obtain the relationship between the natural gas input amount and the coke oven gas storage tank slot positions. The analysis of the natural gas input amount to the coke oven gas storage tank slot may include step 310 , step 320 , step 330 , step 340 , step 350 , and step 360 . It should be emphasized that step 300 is only illustrative and is not intended to limit the present disclosure to the clearly disclosed scope. Additional steps may be provided before, during, and after step 300, and some of the steps described may be moved, removed, replaced, or deleted for additional implementations of step 300.

Please refer to Figure 3. When analyzing the natural gas input volume and the coke oven gas storage tank slot, step 310 can be performed first to calculate the coke oven gas supplied to the coke oven gas storage tank by the coking process per hour based on several coking process records. Amount provided. Specifically, in some examples, the coking process records include the coking process parameters of each stage of coking. Therefore, when the coking process is performed, the staff will record the coke oven gas output generated per hour in each coking process, and record it in the corresponding The coking process of the coking process is being recorded. In some examples, several coking process records can be stored in the form of digital data in a relational database (such as MySQL, SQL Server, and Oracle) or a non-relational database (such as MongoDB), so workers can Use programming languages such as SQL or Python to query or update the contents of the database in real time.

When the natural gas input amount is analyzed for the coke oven gas storage tank slot, step 320 can be performed to calculate the hourly coke oven gas consumption of the coke oven gas storage tank based on several production process records. Specifically, during each production process, the coke oven can be used as electric energy for each production process. Online staff can record the coke oven gas consumption per hour in each production process and produce the corresponding Production process records. In some examples, the production process records include, but are not limited to, several hot rolling process records, several steel plate production process records, several steel blank production process records, several bar steel production process records, and several wire rod production process records. , coke oven gas storage tank change records, and several coke oven gas usage records. In some examples, several production process records can be stored in the form of digital data in a relational database (such as MySQL, SQL Server, and Oracle) or a non-relational database (such as MongoDB), so workers can Use programming languages such as SQL or Python to query or update the contents of the database in real time.

Next, step 330 is performed to calculate the coke oven gas injection amount per hour into the coke oven gas storage tank based on the coke oven gas supply amount and the coke oven gas consumption amount.

When the natural gas input amount is analyzed for the coke oven gas storage tank slot, step 340 is performed to perform a data cleaning operation to remove several invalid data or several low-relevance data in the coking process records and production process records. Specifically, in some examples, the coking process records and the production process records may contain several fields that store incorrect information or no information. These data may easily lead to inaccurate analysis results when conducting regression analysis in the future, or even make analysis impossible. Therefore, online staff can perform data cleaning to remove incomplete, inaccurate, invalid, or low-relevance data in the database to improve the accuracy of future regression analysis results.

When analyzing the natural gas input amount versus the coke oven gas storage tank location, after completing step 340, step 350 is then performed to perform a second regression analysis process on the natural gas injection amount and the coke oven gas injection amount to obtain the natural gas injection amount versus the coke oven gas Injection volume relationship. Specifically, after obtaining the coke oven gas injection volume per hour into the coke oven gas storage tank, online staff can extract the hourly natural gas injection volume into the coke oven gas storage tank from the previously stored historical data, and Use various analysis software or programming languages to perform regression analysis on the coke oven gas injection volume and natural gas injection volume. In some examples, the relationship between natural gas injection volume and coke oven gas injection volume is: 1.836822X2=X3， Among them, X2 is the natural gas injection amount per minute, and X3 is the coke oven gas injection amount per minute.

When analyzing the natural gas input amount versus the coke oven gas storage tank slot, step 360 is finally performed to obtain the relationship between the natural gas input amount and the coke oven gas storage tank slot change based on the relationship between the natural gas injection amount and the coke oven gas injection amount. Specifically, on the premise that it is known that the capacity of the coke oven gas storage tank when it is full is 60,000NM3, online staff can measure the volume of coke oven gas injected into the coke oven gas storage tank every hour to estimate the coke oven gas volume per minute. The position of the gas storage tank changes due to the injection of coke oven gas. Therefore, through the relationship between the amount of natural gas injection and the amount of coke oven gas, the relationship between the amount of natural gas injection and the change in the slot position of the coke oven gas storage tank can be easily obtained. In some examples, the relationship between natural gas input volume and coke oven gas storage tank slot changes is: Y1=0.000051X2， Among them, Y1 is the percentage change of the coke oven gas storage tank slot per minute, and X2 is the natural gas injection volume per minute.

Please refer to Figure 1 again. When performing the method 100 of predicting the coke oven gas storage tank slot, step 400 can be performed to determine the relationship between the coke oven gas pressure and the coke oven gas storage tank slot changes and the natural gas input amount to the coke oven gas storage tank. The relationship between slot changes, natural gas input volume, coke oven gas pressure before the preset time period, and current coke oven gas storage tank slots are used to predict and analyze the changes in coke oven gas storage tank slots after natural gas injection, and predict the next Minutes of coke oven gas storage tank slot. Specifically, after obtaining the relationship between the coke oven gas pressure and the change in the coke oven gas storage tank slot and the relationship between the natural gas input amount and the change in the coke oven gas storage tank slot, online staff can use machine learning and other methods to calculate the relationship between the coke oven gas pressure and the coke oven gas storage tank slot. A prediction model is built based on the relationship between the change of coke oven gas storage tank slots and the relationship between the natural gas input volume and the change of coke oven gas storage tank slots, and combines the natural gas input volume, the coke volume before the preset time period (for example, 3 minutes) The oven gas pressure and the current coke oven gas storage tank slot are input into the built model to predict the changes in the coke oven gas storage tank slot in the next minute, and then obtain the coke oven gas storage tank slot in the next minute.

When performing method 100 of predicting coke oven gas storage tank slots, after completing step 400, step 500 can be followed to repeat the prediction analysis of coke oven gas storage tank slot changes after natural gas injection to predict the current coke oven gas storage tank. Long-term changes in slots. Specifically, after using step 400 to obtain the coke oven gas storage tank slot for the next minute, the online staff can substitute the coke oven gas storage tank slot for the next minute and other parameters into the model again to obtain 2 Coke oven gas storage tank slot after minutes. By analogy, the long-term (for example, 1 hour) changes in the current coke oven gas storage tank position can be predicted.

It can be seen from the above embodiments of the present disclosure that the present disclosure mainly obtains the relationship between the coke oven gas pressure and the change of the coke oven gas storage tank slot and the relationship between the natural gas input amount and the change of the coke oven gas storage tank slot through regression analysis, and then combines the relationship between the coke oven gas pressure and the coke oven gas storage tank slot. The relationship between gas pressure and the change of coke oven gas storage tank slots and the relationship between natural gas input volume and coke oven gas storage tank slot changes are used to predict the long-term changes in coke oven gas storage tank slots after natural gas injection. The disclosed method for predicting coke oven gas storage tank slots has a high degree of accuracy in prediction, and can instantly present the prediction results to online staff. Therefore, it can effectively avoid the occurrence of excessive coke oven gas storage tank slot levels in various factories. The device tripped. In addition, the analysis of natural gas input volume on coke oven gas storage tank slots also includes data cleaning operations to remove invalid data and low-relevance data, thereby improving the accuracy of the analysis.

Although the disclosure has been disclosed above through embodiments, they are not intended to limit the disclosure. Anyone with ordinary knowledge in the technical field may make slight changes and modifications without departing from the spirit and scope of the disclosure. Therefore, The scope of protection of this disclosure shall be determined by the scope of the appended patent application.

100:Method 200: steps 210: Step 220:Step 230:Step 300: steps 310: Steps 320: Steps 330: Steps 340: Steps 350: Steps 360: steps 400: steps 500: steps

In order to make the above and other objects, features, advantages and embodiments of the present invention more apparent and understandable, the detailed description of the accompanying drawings is as follows: Figure 1 is a flowchart illustrating a method for predicting coke oven gas storage tank slots according to some embodiments of the present disclosure; Figure 2 is a flow chart illustrating coke oven gas pressure versus coke oven gas storage tank location analysis according to some embodiments of the present disclosure; and Figure 3 is a flow chart illustrating natural gas input volume analysis of coke oven gas storage tank slots according to some embodiments of the present disclosure.

Domestic storage information (please note in order of storage institution, date and number) without Overseas storage information (please note in order of storage country, institution, date, and number) without

100:Method

200: steps

300: steps

400: steps

500: steps

Claims (10)

A method for predicting the location of coke oven gas storage tanks, including: Carry out an analysis of the coke oven gas pressure and the coke oven gas storage tank slot to obtain the change relationship between the coke oven gas pressure and the coke oven gas storage tank slot; Conduct an analysis of a natural gas input amount versus a coke oven gas storage tank slot to obtain a relationship between a natural gas input amount and a change in the coke oven gas storage tank slot location; and According to the relationship between the coke oven gas pressure and the change in the coke oven gas storage tank slot, the natural gas input amount and the change in the coke oven gas storage tank slot, a natural gas input amount, a coke oven gas pressure before a preset time period, and a For the current coke oven gas storage tank slot, perform a prediction analysis on the change of the coke oven gas storage tank slot after natural gas injection, and predict the coke oven gas storage tank slot position in the next minute. The method for predicting coke oven gas storage tank slots as described in claim 1, wherein the coke oven gas pressure analysis of coke oven gas storage tank slots includes: Perform coke oven gas pressure detection on a plurality of coke oven gas storage tanks respectively to obtain a plurality of coke oven gas pressures of the coke oven gas storage tanks; Conducting a plurality of coke oven gas storage tank slot detections on these coke oven gas storage tanks respectively to obtain changes in the plurality of coke oven gas storage tank slots of these coke oven gas storage tanks; and A first regression analysis process is performed on the coke oven gas pressures and the coke oven gas storage tank slots to obtain a relationship between the changes in coke oven gas pressure and coke oven gas storage tank slots. The method of predicting the coke oven gas storage tank slot as described in claim 2, wherein the relationship between the coke oven gas pressure and the change in the coke oven gas storage tank slot is: Y1=(0.021297X1-21.55493)， Among them, Y1 is the percentage change of the coke oven gas storage tank position per minute, and X1 is the average coke oven gas tank pressure per minute. The method for predicting coke oven gas storage tank slots as described in request 1, wherein the natural gas input volume analysis of coke oven gas storage tank slots includes: Based on multiple coking process records, calculate the amount of coke oven gas supplied to a coke oven gas storage tank for these coking processes per hour; Calculate the coke oven gas consumption of the coke oven gas storage tank per hour based on multiple production process records; Calculate the coke oven gas injection amount per hour of the coke oven gas storage tank based on the coke oven gas supply amount and the coke oven gas consumption amount; Perform a second regression analysis process on a natural gas injection amount and the coke oven gas injection amount to obtain a relationship between a natural gas injection amount and the coke oven gas injection amount; and Based on the relationship between the natural gas injection amount and the coke oven gas injection amount, a relationship between the natural gas input amount and the change in the coke oven gas storage tank location is obtained. The method for predicting coke oven gas storage tank slots as described in claim 4, wherein the relationship between the natural gas injection amount and the coke oven gas injection amount is: 1.836822X2=X3， Among them, X2 is the natural gas injection amount per minute, and X3 is the coke oven gas injection amount per minute. The method for predicting coke oven gas storage tank slots as described in claim 4, wherein the relationship between the natural gas input volume and the change in coke oven gas storage tank slots is: Y1=0.000051X2， Among them, Y1 is the percentage change of the coke oven gas storage tank slot per minute, and X2 is the natural gas injection volume per minute. The method for predicting coke oven gas storage tank slots as described in claim 4, wherein the natural gas input volume analysis of coke oven gas storage tank slots further includes: A data cleaning operation is performed to remove a plurality of invalid data or a plurality of low-relevance data in the coking process records and the production process records. The method for predicting coke oven gas storage tank slots as described in claim 4, wherein the coking process records include coking process parameters for each stage of coking. The method for predicting coke oven gas storage tank slots as described in request item 4, wherein the production process records include a plurality of hot rolling process records, a plurality of steel plate production process records, a plurality of steel blank production process records, a plurality of strips Steel production process records, multiple wire rod production process records, a coke oven gas storage tank change record, and multiple coke oven gas usage records. The method for predicting coke oven gas storage tank slots as described in claim 1, wherein the method for predicting coke oven gas storage tank slots further includes: Repeat the prediction analysis of the coke oven gas storage tank location changes after natural gas injection to predict long-term changes in the current coke oven gas storage tank location.

Images