A method for processing production data of gasoline and diesel antistatic agents

Abstract

The present invention relates to the field of data processing technology, and in particular to a method for processing production data of gasoline and diesel antistatic agents, the method comprising: clustering data points composed of production data of gasoline and diesel antistatic agents in a historical time period, predicting the probability of the data points corresponding to the time to be predicted belonging to each category according to the distance between the data points corresponding to each reference time at the time to be predicted and the center of the category to which the data points corresponding to each reference time belong, and all data points in the category to which the data points corresponding to each reference time belong, weighting the centers of multiple categories with the largest probability according to the probability of the data points corresponding to the time to be predicted belonging to each category, using the weighted results as the predicted data points at the time to be predicted, and judging whether overheating or overpressure will occur at the time to be predicted according to the temperature and pressure of the predicted data points. The present invention improves the accuracy of the final prediction result, thereby improving the stability of the production reaction conditions of the gasoline and diesel antistatic agent and the safety of production.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically, to a method for processing production data of gasoline and diesel antistatic agents. Background Art

[0002] In the process of producing gasoline and diesel antistatic agents, overheating and overpressure may lead to explosion; in order to ensure the stability of reaction conditions and the safety of production, temperature and pressure need to be monitored.

[0003] In the related technology, for example, the Chinese patent document with the authorization announcement number CN114610096B has developed a reactor jacket temperature control system based on temperature control technology, which belongs to the field of temperature control technology. The purpose is to solve the problem of reactor jacket temperature control accuracy, receive the cooling signal sent by the server, obtain the liquid temperature value, liquid flow rate and liquid pressure when the liquid temperature exceeds E in real time, and integrate the liquid temperature value, working environment, liquid flow rate and liquid pressure as input data; obtain a prediction model; input the input data into the prediction model to obtain the optimal cooling temperature and flow rate when cooling, including a monitoring alarm module, a cooling module, a heating module, a server and a storage module; the monitoring alarm module includes a temperature detection alarm unit, a pressure detection alarm unit and a line detection alarm unit. By adopting the jet enthalpy increase technology for cooling, the single-stage refrigeration temperature can be as low as -50°C, and the control accuracy is high, which meets the production process requirements for wide range and high precision.

[0004] The relevant technology relies on a single artificial intelligence model for prediction, which lacks accuracy when faced with dynamically changing temperature and pressure data. Summary of the invention

[0005] In order to solve the technical problem that the above-mentioned related technologies have insufficient prediction accuracy when facing dynamically changing temperature and pressure data, the present invention provides a method for processing gasoline and diesel antistatic agent production data, comprising: clustering data points composed of gasoline and diesel antistatic agent production data in a historical time period to obtain multiple categories, wherein the production data includes temperature and pressure; obtaining multiple reference times for the time to be predicted; predicting that the data point corresponding to the time to be predicted belongs to the category Probability , , The time difference from the time to be predicted is equal to The data point corresponding to the reference time is for Integer in the range, , Represent the minimum time difference and the maximum time difference respectively, For data points Category, For data points and categories The distance from the center; For Category The The time corresponding to the data point is For the moment The corresponding data points, For data points Category, For Category The number of all data points in ; is the judgment function, responding to the category Yes Category hour, ,otherwise ; is a natural exponential function; the centers of multiple categories with the largest probabilities are weighted according to the probability that the data points corresponding to the time to be predicted belong to each category, and the weighted results are used as the predicted data points at the time to be predicted; based on the temperature and pressure of the predicted data points, it is determined whether overheating or overpressure will occur at the time to be predicted.

[0006] The present invention can subdivide the data points into multiple categories by clustering the data points corresponding to the historical moments, so as to better capture the performance under different production conditions; by analyzing the data points corresponding to multiple reference moments of the moment to be predicted, the time dependency in the production process can be captured; by dynamically analyzing the data points in the category to which the data points corresponding to the reference moments under different time differences belong, and weighting the data points corresponding to the reference moments by the distance between the centers of the categories to which they belong, the time series characteristics can be better utilized, so that the prediction results are closer to the actual production situation; by predicting the probability that the data points corresponding to the moment to be predicted belong to each category, the attribution of the data points corresponding to the moment to be predicted can be quantified, and finally by weighting the centers of the multiple categories with the largest probabilities, the accuracy of the final prediction results can be improved.

[0007] Preferably, the historical time period includes multiple historical production cycles, the historical production cycle is earlier than the production cycle at the current moment, and the production cycle includes a mixing stage, a reaction stage, a separation stage, a purification stage and a preparation stage.

[0008] Preferably, the production data also includes flow rate and pH value. , determine the stage at that moment , the stage The mean of the production data of all the collection locations in the composition time The corresponding data points.

[0009] Preferably, clustering the data points consisting of the production data of gasoline and diesel antistatic agents within the historical time period to obtain multiple categories includes: clustering the data points corresponding to all historical moments within the historical time period by the OPTICS clustering algorithm to obtain multiple categories.

[0010] The present invention can subdivide the data points into multiple categories by clustering the data points corresponding to historical moments, so as to better capture the performance under different production conditions.

[0011] Preferably, the time to be predicted is later than the current time, and a plurality of reference times of the time to be predicted are within the production cycle of the current time and are not later than the current time.

[0012] Preferably, obtain the data points Categories include: Calculate the data points The distance from the center of each category, taking the category with the smallest distance as the data point Category.

[0013] Preferably, the method weights the centers of the multiple categories with the largest probabilities according to the probability that the data points corresponding to the time to be predicted belong to each category, and uses the weighted results as the predicted data points of the time to be predicted, including: sorting all categories in descending order of the probability that the data points corresponding to the time to be predicted belong to each category; categories, as reference categories for the time to be predicted; is a preset number; according to the probability that the data point corresponding to the time to be predicted belongs to each reference category, the temperature and pressure of the center of each reference category are weighted to obtain the predicted data point at the time to be predicted.

[0014] The present invention can quantify the attribution of data points corresponding to the time to be predicted by predicting the probability that the data points corresponding to the time to be predicted belong to each category, and finally improve the accuracy of the final prediction result by weighting the centers of multiple categories with the highest probabilities.

[0015] Preferably, the temperature and pressure of the predicted data point satisfy the expression: ; ; In the formula, , are the temperature and pressure of the predicted data point, , The reference categories The temperature and pressure at the center, The data point corresponding to the time to be predicted belongs to the reference category The probability of It represents the sum of the probabilities that the data point corresponding to the time to be predicted belongs to each reference category.

[0016] Preferably, judging whether overheating or overpressure will occur at the time to be predicted based on the temperature and pressure of the predicted data point includes: in response to the temperature of the predicted data point being greater than a preset temperature threshold, overheating occurs at the time to be predicted; in response to the pressure of the predicted data point being greater than a preset pressure threshold, overpressure occurs at the time to be predicted.

[0017] The present invention determines whether overheating or overpressure occurs based on the temperature and pressure of the predicted data point, so that the production process can be dynamically adjusted with the support of real-time data, thereby improving the stability of the production reaction conditions of the gasoline and diesel antistatic agent and the safety of production.

[0018] Preferably, the method also includes: in response to overheating or overpressure occurring at the predicted time, taking corresponding measures to reduce the temperature and pressure of the production process of gasoline and diesel antistatic agents; the measures include adjusting the cooling system, slowing down the addition rate of reactants, increasing the ventilation speed and ventilation volume of the reaction equipment, and opening a safety valve or an exhaust valve.

[0019] Through the comprehensive application of the above measures, the temperature and pressure in the production process of gasoline and diesel antistatic agents can be effectively reduced, thereby greatly reducing the possibility of explosion and ensuring the safety of the production process.

[0020] The beneficial effects of the present invention are:

[0021] The present invention can subdivide the data points into multiple categories by clustering the data points corresponding to the historical moments, so as to better capture the performance under different production conditions; by analyzing the data points corresponding to multiple reference moments of the moment to be predicted, the time dependency in the production process can be captured; by dynamically analyzing the data points in the category to which the data points corresponding to the reference moments under different time differences belong, and weighting the data points corresponding to the reference moments by the distance between the center of the category to which they belong, the time series characteristics can be better utilized, so that the prediction results are closer to the actual production situation; by predicting the probability that the data points corresponding to the moment to be predicted belong to each category, the attribution of the data points corresponding to the moment to be predicted can be quantified, and finally by weighting the centers of multiple categories with the largest probabilities, the accuracy of the final prediction results can be improved;

[0022] Furthermore, the present invention determines whether overheating or overpressure occurs based on the temperature and pressure of the predicted data points, so that the production process can be dynamically adjusted with the support of real-time data, thereby improving the stability of the production reaction conditions of gasoline and diesel antistatic agents and the safety of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below through the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0024] Figure 1 It is a flow chart schematically showing a method for processing production data of a gasoline and diesel antistatic agent in the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0026] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0027] The embodiment of the present invention discloses a method for processing production data of a gasoline and diesel antistatic agent, referring to Figure 1 , including steps S1 to S6:

[0028] S1. Collect the production data of gasoline and diesel antistatic agents and obtain the data points corresponding to each moment.

[0029] Specifically, the entire production cycle of gasoline and diesel antistatic agents includes a mixing stage, a reaction stage, a separation stage, a purification stage, and a preparation stage, and the mixing stage, the reaction stage, the separation stage, the purification stage, and the preparation stage are performed sequentially; in the mixing stage, the prepared raw materials are placed in a mixing device according to certain process parameters and uniformly mixed by physical or chemical methods; in the reaction stage, a chemical reaction is performed under specific conditions to generate the main components of the antistatic agent; in the separation stage, the reaction mixture is separated to remove excess raw materials or by-products; in the purification stage, purification is performed by distillation, filtration, and the like; in the preparation stage, the antistatic agent is prepared according to the specific application requirements of the product to achieve the required physical and chemical properties.

[0030] It should be noted that in the process of producing gasoline and diesel antistatic agents, monitoring temperature and pressure can prevent explosions caused by overheating or overpressure, which is crucial to ensuring the stability of reaction conditions and production safety.

[0031] Specifically, a plurality of collection positions are provided in each stage, and a plurality of collectors are installed at each collection position; production data of each collection position is obtained through various collectors, and the production data includes temperature, pressure, flow rate and pH value, the collected production data is normalized, and subsequent production data are all normalized results, and the normalization method is maximum and minimum value normalization; wherein, the temperature of each collection position is collected through a temperature collector, the pressure of each collection position is collected through a pressure collector, the flow rate of each collection position is collected through a flow collector, and the pH value of each collection position is collected through a pH collector; wherein, the temperature collector is a thermocouple or a resistance temperature detector (RTD); the pressure collector is a piezoelectric sensor or a pressure transmitter; the flow collector is an electromagnetic flowmeter or a turbine flowmeter; and the pH collector is a pH meter or a pH sensor.

[0032] Among them, the maximum temperature in the production process of gasoline and diesel antistatic agent is between 150°C and 200°C, the maximum pressure is usually between 0.5MPa (megapascals) and 1.5MPa, and the maximum flow rate may be between 100L / h and 500L / h, depending on the chemical nature of the reaction, the type of catalyst used and the design of the reactor; in this embodiment, when normalizing the production data, the minimum temperature is 0°C, the maximum temperature is 160°C, the minimum pressure is 0MPa, the maximum pressure is 1MPa, the minimum flow rate is 0L / h, and the maximum flow rate is 400L / h; and the pH value is usually between 0 and 14.

[0033] Furthermore, for any moment , determine the stage at that moment , the stage The mean of the production data of all the collection locations in the composition time The corresponding data points.

[0034] S2. Cluster the data points corresponding to all historical moments in the historical time period to obtain multiple categories.

[0035] Specifically, the historical time period includes multiple historical production cycles, which are earlier than the production cycle at the current moment; in this embodiment, the week before the natural day of the current moment is taken as the historical time period, and therefore, the historical production cycle included in the historical time period is earlier than the production cycle at the current moment; in other embodiments, the implementers can set it according to the actual application scenarios and needs.

[0036] Furthermore, since the number of categories is uncertain, the present invention selects the OPTICS (Ordering points to identify the clustering structure) clustering algorithm. Through the OPTICS clustering algorithm, the data points corresponding to all historical moments in the historical time period are clustered to obtain multiple categories; the OPTICS clustering algorithm is a well-known technology and will not be described in detail here.

[0037] It should be noted that the present invention can subdivide the data points into multiple categories by clustering the data points corresponding to historical moments, so as to better capture the performance under different production conditions.

[0038] S3. Obtain a time to be predicted and multiple reference times of the time to be predicted.

[0039] It should be noted that by analyzing the data points corresponding to multiple reference times of the predicted time, it is possible to fully utilize the distribution of production data at the reference times, thereby significantly improving the accuracy of the prediction.

[0040] Specifically, the time to be predicted is later than the current time, and the reference time of the time to be predicted is within the production cycle of the current time and is not later than the current time; therefore, the time difference with the time to be predicted is equal to The time is taken as the reference time of the time to be predicted, and the time difference between the reference time and the time to be predicted refers to the difference between the time to be predicted and the reference time; wherein, for Therefore, the number of all reference moments to be predicted is equal to , , Respectively represent the minimum time difference and the maximum time difference; is equal to the preset first value, is equal to a preset second value, and the preset first value is less than the preset second value.

[0041] Among them, the specific values ​​of the preset first value and the preset second value can be set according to the actual application scenario and requirements, and the preset first value is an integer in the range of [1,5], and the preset second value is an integer in the range of [15,20]. The present invention sets the preset first value to 1 and sets the preset second value to 20.

[0042] S4. Calculate the probability that the data point corresponding to the predicted moment belongs to each category based on the distance between the data point corresponding to each reference moment and the center of the category to which it belongs, and all data points in the category to which the data point corresponding to each reference moment belongs.

[0043] Specifically, for the time difference from the time to be predicted equal to The reference time, get the data point corresponding to the reference time , including: calculating the data points The distance from the center of each category, taking the category with the smallest distance as the data point Category.

[0044] Furthermore, based on the distance between the data point corresponding to each reference time and the center of the category to which it belongs, and all the data points in the category to which the data point corresponding to each reference time belongs, the probability of the data point corresponding to the predicted time belonging to each category is predicted, and the data point corresponding to the predicted time belongs to the category Probability Satisfies the expression:

[0045] ;

[0046] In the formula, The time difference from the time to be predicted is equal to The data point corresponding to the reference time is for Integer in the range, , Represent the minimum time difference and the maximum time difference respectively, For data points Category, For data points and categories The distance from the center; For Category The The time corresponding to the data point is For the moment The corresponding data points, For data points Category, For Category The number of all data points in ; is the judgment function, is a natural exponential function.

[0047] Among them, in response to the category Yes Category When , otherwise the judgment function .

[0048] It should be noted that the time difference between the reference time and the time to be predicted is equal to , then for the category The time difference of the data points in is equal to The corresponding data point , can be used as a reference category to predict the probability that the data point corresponding to the predicted time belongs to each category; and the data point and categories The smaller the distance to the center of the data point, the The more you obey the category The data distribution of The greater the reference degree for predicting the probability that the data point corresponding to the time to be predicted belongs to each category.

[0049] It should be further explained that by analyzing the data points corresponding to multiple reference moments of the moment to be predicted, the time dependence of the production process can be captured; by dynamically analyzing the data points in the category to which the data points corresponding to the reference moments at different time differences belong, and weighting the data points by the distance between the data points corresponding to the reference moments and the center of the category to which they belong, the time series characteristics can be better utilized, making the prediction results closer to the actual production situation.

[0050] S5. Weight the centers of multiple categories with the largest probabilities according to the probability that the data point corresponding to the time to be predicted belongs to each category, and use the weighted result as the predicted data point of the time to be predicted.

[0051] Specifically, all categories are sorted in descending order according to the probability of the data points corresponding to the predicted time belonging to each category; categories, as reference categories for the time to be predicted; is the preset number, then all reference categories at the time to be predicted are equal to .

[0052] The specific value of the preset number can be set according to the actual application scenario and requirements, and the preset number is Integer in the range, is the total number of all categories, and the present invention sets the preset number to .

[0053] Furthermore, according to the probability that the data point corresponding to the time to be predicted belongs to each reference category, the temperature and pressure of the center of each reference category are weighted to obtain the predicted data point of the time to be predicted.

[0054] The temperature and pressure of the predicted data point satisfy the expression:

[0055] ;

[0056] ;

[0057] In the formula, , are the temperature and pressure of the predicted data point, , The reference categories The temperature and pressure at the center, The data point corresponding to the time to be predicted belongs to the reference category The probability of It represents the sum of the probabilities that the data point corresponding to the time to be predicted belongs to each reference category.

[0058] It should be noted that the probability that the data point corresponding to the predicted time belongs to each reference category is obtained by summing the probabilities of the data point corresponding to the predicted time belonging to the reference category. The probability of is normalized and the normalized result is As a weight, the temperature and pressure at the time to be predicted are predicted.

[0059] It should be further explained that by predicting the probability that the data points corresponding to the time to be predicted belong to each category, the attribution of the data points corresponding to the time to be predicted can be quantified, and finally by weighting the centers of multiple categories with the largest probabilities, the accuracy of the final prediction result can be improved.

[0060] S6. Determine whether overheating or overpressure will occur at the time to be predicted based on the temperature and pressure of the predicted data point.

[0061] It should be noted that in the process of producing gasoline and diesel antistatic agents, overheating and overpressure may lead to explosion. Therefore, the present invention determines whether overheating or overpressure will occur at the time to be predicted by predicting the temperature and pressure at the time to be predicted. When it is predicted that the temperature may be too high or the pressure may be too high at the time to be predicted, corresponding measures are taken to reduce the possibility of explosion and ensure the stability of the production reaction conditions of the gasoline and diesel antistatic agent and the safety of production.

[0062] Specifically, in response to the temperature of the predicted data point being greater than a preset temperature threshold, overheating occurs at the time to be predicted; in response to the pressure of the predicted data point being greater than a preset pressure threshold, overpressure occurs at the time to be predicted.

[0063] The specific values ​​of the temperature threshold and the pressure threshold can be set according to the actual application scenario and requirements, and the temperature threshold and the pressure threshold are both greater than 1. In the present invention, the temperature threshold and the pressure threshold are both set to .

[0064] It should be noted that the present invention determines whether overheating or overpressure occurs based on the temperature and pressure of the predicted data points, so that the production process can be dynamically adjusted with the support of real-time data, thereby improving the stability of the production reaction conditions of gasoline and diesel antistatic agents and the safety of production.

[0065] Further, in response to overheating or overpressure occurring at the predicted time, corresponding measures are taken to reduce the temperature and pressure of the production process of the gasoline and diesel antistatic agent; the measures include:

[0066] 1. Adjust the cooling system, including: enhancing the function of the cooling system, increasing the flow of cooling water or lowering the temperature of cooling water; lowering the temperature by quickly taking away the reaction heat.

[0067] 2. Slow down the rate of addition of reactants, thereby reducing the rate of increase of temperature and pressure.

[0068] 3. Increase the ventilation speed and volume of the reaction equipment to discharge excess heat and gas, reduce local temperature, and prevent gas accumulation, thereby reducing the rate of pressure rise.

[0069] 4. Open the safety valve or exhaust valve to release the excessive pressure in the system in time.

[0070] Through the comprehensive application of the above measures, the temperature and pressure in the production process of gasoline and diesel antistatic agents can be effectively reduced, thereby greatly reducing the possibility of explosion and ensuring the safety of the production process.

[0071] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0072] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for processing production data of gasoline and diesel antistatic agents, characterized in that: include: Clustering data points consisting of production data of gasoline and diesel antistatic agents in a historical period to obtain multiple categories, wherein the production data includes temperature and pressure; Get multiple reference moments of the moment to be predicted; predict that the data point corresponding to the moment to be predicted belongs to the category Probability , , The time difference from the time to be predicted is equal to The data point corresponding to the reference time is for Integer in the range, , Represent the minimum time difference and the maximum time difference respectively, For data points Category, For data points and categories The distance from the center; For Category The The time corresponding to the data point is For the moment The corresponding data points, For data points Category, For Category The number of all data points in ; is the judgment function, responding to the category Yes Category hour, ,otherwise ; According to the probability that the data point corresponding to the time to be predicted belongs to each category, the centers of the multiple categories with the largest probability are weighted, and the weighted results are used as the predicted data points of the time to be predicted, including: sorting all categories in descending order according to the probability that the data point corresponding to the time to be predicted belongs to each category; Sort by categories, as reference categories for the time to be predicted; is a preset number; according to the probability that the corresponding data point at the time to be predicted belongs to each reference category, the temperature and pressure of the center of each reference category are weighted to obtain the predicted data point at the time to be predicted; according to the temperature and pressure of the predicted data point, it is determined whether overheating or overpressure will occur at the time to be predicted.

2. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: The historical time period includes multiple historical production cycles, which are earlier than the production cycle at the current moment. The production cycle includes a mixing stage, a reaction stage, a separation stage, a purification stage and a preparation stage.

3. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: The production data also includes flow rate and pH value. , determine the stage at that moment , the stage The mean of the production data of all the collection locations in the composition time The corresponding data points.

4. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: The data points composed of the production data of gasoline and diesel antistatic agents in the historical time period are clustered to obtain multiple categories, including: The OPTICS clustering algorithm is used to cluster the data points corresponding to all historical moments in the historical time period to obtain multiple categories.

5. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: The time to be predicted is later than the current time, and a plurality of reference times of the time to be predicted are within the production cycle of the current time and are not later than the current time.

6. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: Get data points Categories include: Calculate the data points The distance from the center of each category, taking the category with the smallest distance as the data point Category.

7. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: The temperature and pressure of the predicted data point satisfy the expression: ； ； In the formula, , are the temperature and pressure of the predicted data point, , The reference categories The temperature and pressure at the center, The data point corresponding to the time to be predicted belongs to the reference category The probability of It represents the sum of the probabilities that the data point corresponding to the time to be predicted belongs to each reference category.

8. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: The step of judging whether overheating or overpressure will occur at the time to be predicted based on the temperature and pressure of the predicted data point includes: In response to the temperature of the predicted data point being greater than a preset temperature threshold, overheating occurs at the time to be predicted; in response to the pressure of the predicted data point being greater than a preset pressure threshold, overpressure occurs at the time to be predicted.

9. A gasoline and diesel antistatic agent production data processing method according to claim 1, characterized in that: The method further comprises: In response to overheating or overpressure occurring at the predicted time, taking corresponding measures to reduce the temperature and pressure of the production process of the gasoline and diesel antistatic agent; The measures include adjusting the cooling system, slowing down the rate of adding reactants, increasing the ventilation speed and volume of the reaction equipment, and opening the safety valve or exhaust valve.

Images