Optimized filtering method and system for petroleum filtering device

Abstract

The invention discloses an adjusting and optimizing filtering method and system for a petroleum filtering device, and relates to the technical field of petroleum filtering, and the method comprises the following steps: initially checking and establishing a petroleum feature set; and inputting the petroleum feature set into a filter screen setting model to generate a filter screen configuration result. And key adjustment parameters of the petroleum filtering device are established. Big data is called through a petroleum feature set, a fitness function is constructed, data selection is performed based on the big data and in-plant control data, and a solution space and an initial solution set are established. And evaluating the initial solution set based on a fitness function, performing iterative screening, and establishing an optimal solution identifier and an inferior solution identifier. And performing initial solution set reproduction and solution space variation through optimal solution and inferior solution identification, and executing initial solution set iteration. And generating an optimization result according to the iteration result, and performing tuning filtering treatment according to the filter screen configuration result and the optimization result. Therefore, the technical effects of improving the adaptability of the petroleum filtering device and improving the filtering efficiency are achieved.

Description

Technical Field

[0001] The invention relates to the technical field of petroleum filtration, and in particular to a tuning filtration method and system for a petroleum filtration device. Background Art

[0002] Oil filtration devices play a key role in oil extraction, transportation and processing. They are used to remove impurities and pollutants in oil to ensure the purity and quality of oil products. Existing oil filtration device configuration methods usually use fixed filter media and a single filter mode, which is difficult to cope with the complex changes in oil composition and the filtering requirements under different working conditions. There are technical problems such as low filtration efficiency and poor adaptability. Summary of the invention

[0003] The present invention provides a tuning filtration method and system for a petroleum filtration device, so as to solve the technical problems of low filtration efficiency and poor adaptability in the prior art, and achieve the technical effect of improving the adaptability of the petroleum filtration device and enhancing the filtration efficiency.

[0004] In a first aspect, the present invention provides a method for optimizing filtration of a petroleum filtration device, wherein the method comprises: The petroleum to be filtered is subjected to a preliminary petroleum inspection, and a petroleum feature set is established, wherein the petroleum feature set includes petroleum impurity features and petroleum viscosity features.

[0005] The oil feature set is input into a filter setting model to generate a filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications under each grade.

[0006] Establish key adjustment parameters of the petroleum filtering device, wherein the key adjustment parameters include flow rate parameters, pressure parameters, and temperature parameters.

[0007] The big data is called through the petroleum feature set to construct a fitness function, and data selection is performed based on the big data and in-plant control data to establish a solution space and an initial solution set, wherein the solution space is a space for searching and optimizing, and the initial solution set is a solution set constructed based on a preset population size and data selection.

[0008] The solutions in the initial solution set are evaluated based on the fitness function, and iterative screening is performed based on the evaluation results to establish superior solution identifiers and inferior solution identifiers.

[0009] The initial solution set is multiplied and mutated in the solution space through the superior solution identifier and the inferior solution identifier, and the initial solution set iteration is performed.

[0010] Generate an optimization result according to the iteration result, and perform tuning and filtering processing based on the filter configuration result and the optimization result.

[0011] In a second aspect, the present invention also provides a tuning and filtering system for a petroleum filtering device, wherein the system comprises: The petroleum initial inspection module is used to perform an initial inspection on the petroleum to be filtered and establish a petroleum feature set, wherein the petroleum feature set includes petroleum impurity features and petroleum viscosity features.

[0012] A filter configuration module is used to input the petroleum feature set into a filter setting model to generate a filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications under each grade.

[0013] The adjustment parameter module is used to establish key adjustment parameters of the petroleum filtering device, wherein the key adjustment parameters include flow rate parameters, pressure parameters, and temperature parameters.

[0014] An evaluation initialization module is used to call big data through the petroleum feature set, construct a fitness function, and perform data selection based on the big data and in-plant control data to establish a solution space and an initial solution set, wherein the solution space is a space for searching and optimizing, and the initial solution set is a solution set constructed based on a preset population size and data selection.

[0015] A screening identification module is used to evaluate solutions in the initial solution set based on the fitness function, perform iterative screening based on the evaluation results, and establish superior solution identification and inferior solution identification.

[0016] A mutation iteration module is used to perform initial solution set propagation and mutation in the solution space through the superior solution identifier and the inferior solution identifier, and perform initial solution set iteration.

[0017] The tuning and filtering module is used to generate an optimization result according to the iteration result, and perform tuning and filtering processing based on the filter configuration result and the optimization result.

[0018] The invention discloses a tuning and filtering method and system for a petroleum filtering device, comprising: performing preliminary detection on the petroleum to be filtered, establishing a petroleum feature set, including petroleum impurity features and petroleum viscosity features; inputting the petroleum feature set into a filter screen setting model, generating a filter screen configuration result, the result including the number of filter screen classifications and the specifications of each filter screen; establishing key adjustment parameters of the petroleum filtering device, including flow rate parameters, pressure parameters and temperature parameters; calling big data through the petroleum feature set, constructing a fitness function, and performing data selection based on the big data and in-plant control data to establish a solution space and an initial solution set, wherein the solution space is a search optimization space, and the initial solution set is a solution set constructed based on a preset population size and data selection; evaluating solutions in the initial solution set based on the fitness function, performing iterative screening based on the evaluation results, and establishing a superior solution identifier and a inferior solution identifier; performing initial solution set propagation and variation in the solution space through the superior solution identifier and the inferior solution identifier, and performing initial solution set iteration; generating an optimization result according to the iteration result, and performing tuning and filtering processing with the filter screen configuration result and the optimization result. The tuning and filtering method and system for a petroleum filtering device disclosed in the present invention solve the technical problems of low filtering efficiency and poor adaptability, and achieve the technical effect of improving the adaptability of the petroleum filtering device and enhancing the filtering efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The present invention is a schematic flow chart of a method for optimizing filtration of a petroleum filtration device.

[0020] Figure 2 The figure is a schematic diagram of the structure of the tuning and filtering system for the petroleum filtering device of the present invention.

[0021] Explanation of the accompanying reference numerals: oil initial inspection module 11, filter configuration module 12, parameter adjustment module 13, evaluation initialization module 14, screening identification module 15, variation iteration module 16, tuning and filtering module 17. DETAILED DESCRIPTION

[0022] The technical solution provided in the embodiments of the present invention is to solve the technical problems of low filtration efficiency and poor adaptability in the prior art. The overall idea adopted is as follows: First, the oil to be filtered is preliminarily tested to establish an oil feature set, which includes the impurity characteristics and viscosity characteristics of the oil. Then, these oil features are input into the filter setting model to generate the filter configuration results, which will tell us the number of filter grades required and the specifications of the filters in each grade. Then, the key adjustment parameters of the oil filtration device are established, including flow rate, pressure and temperature. Through the oil feature set, we can call big data to build a fitness function, and perform data selection based on these big data and the control data in the factory to establish the solution space and the initial solution set. The solution space is the area for searching and optimizing, and the initial solution set is a set of solutions constructed according to the preset population size and data selection. Next, the solutions in the initial solution set are evaluated based on the fitness function, and iterative screening is performed according to the evaluation results to establish the superior solution identification and inferior solution identification. Through these superior solution and inferior solution identification, the initial solution set is propagated, mutated in the solution space, and the iteration of the initial solution set is performed. Finally, the optimization result is generated according to the iterative results, and the filter configuration result and the optimization result are used to perform tuning and filtering processing.

[0023] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them. Embodiment 1

[0024] Figure 1 The present invention is a schematic flow chart of a method for optimizing filtration of a petroleum filtration device, wherein the method comprises:

[0025] The petroleum to be filtered is subjected to a preliminary petroleum inspection, and a petroleum feature set is established, wherein the petroleum feature set includes petroleum impurity features and petroleum viscosity features.

[0026] Optionally, first obtain a petroleum sample of the target filtering scene, and detect impurities in the petroleum through analytical means (such as particle counting, specific gravity analysis, spectral analysis, etc.) to obtain petroleum impurity characteristics. Impurity characteristics include characteristics of impurities such as solid particles, water, and salt in the petroleum.

[0027] Optionally, the characteristics of petroleum impurities include particle size distribution, number concentration, morphological characteristics, etc. of solid particles in petroleum. Specifically, particle size distribution can reflect the size range of solid particles. In most cases, the distribution of solid particles in petroleum is uneven, and local high-concentration particle size areas may be formed. Particle size distribution is measured by laser particle size analyzer, sieving method or microscope and other equipment; number concentration is the content of solid particles in petroleum, which directly affects the quality and filtration effect of petroleum, and is measured by number concentration density gradient centrifugation, buoyancy sedimentation method or automatic counter; morphological characteristics of particles (such as shape, surface texture, etc.) can provide information about the source of particles and possible impacts, which is helpful to improve possible defects in the mining process. Morphological characteristics are obtained by observing solid particles with scanning electron microscope (SEM) or transmission electron microscope (TEM).

[0028] Optionally, a viscometer or other device is used to measure the viscosity of the above oil sample at different temperatures to obtain oil viscosity characteristics. Specifically, the oil viscosity characteristics reflect the fluidity and viscosity of the oil, and affect the flow state of the oil in the oil filtration device.

[0029] Furthermore, the collected impurity feature and viscosity feature data are organized into a set to generate an oil feature set. The oil feature set contains key feature parameters such as impurity content and viscosity in oil, which is used to provide a decision basis for subsequent filtering and processing.

[0030] The oil feature set is input into a filter setting model to generate a filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications under each grade.

[0031] Optionally, the previously established petroleum feature set (including impurity features and viscosity features, etc.) is input into the filter setting model to perform a decision analysis on the filter configuration and generate a filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications at each grade. The above parameters in the filter configuration result determine the efficiency and performance of the filtering device.

[0032] Optionally, the filter setting model can be a regression model or a machine learning model constructed based on a statistical analysis method, and the filter configuration result is generated according to the input petroleum characteristic data. Exemplarily, based on a machine learning method, multiple historical filter configurations with good filtering effect and filtering quality and corresponding historical petroleum characteristic data are obtained as training data, and the filter setting model is constructed and trained, wherein the filter setting model includes: a feature data input layer, which is used to receive information such as the oil viscosity coefficient, the size, shape, number, concentration, chemical composition and distribution characteristics of solid particles in the oil; a data preprocessing layer, which is used to standardize the acquired feature information and extract important features; a feature analysis layer, which is used to analyze the size distribution, shape distribution and number distribution of particles, determine the multi-peak distribution characteristics of particles, analyze the concentration of particles of different sizes, and determine the number of particles in each size range; a configuration generation layer, which is constructed based on decision trees, random forests, support vector machines, etc., and outputs the analysis results output by the above feature analysis layer based on the complex mapping relationship obtained through training and learning, and outputs the number of filter grades and the filter specifications under each grade. Through the above method, the filter configuration can be predicted efficiently and accurately based on the characteristic data of oil. On the one hand, it improves the adaptability of the filter configuration to the oil in the target scene, and on the other hand, it improves the automation and intelligence of the filter configuration process.

[0033] Specifically, the number of filter screen grades depends on the size and content of impurities in the oil, as well as the expected filtration accuracy. For example, if the particle distribution is wide (such as from a few nanometers to hundreds of microns), at least three to four levels of filter screens are required to filter particles of different sizes, thereby improving the filtration efficiency and filtration effect of the filtration device. The wider the particle size distribution of solid particle impurities in the oil, the more dispersed the distribution, and the higher the filtration accuracy required, the greater the number of filter screen grades.

[0034] Specifically, the filter specification refers to the pore size of the filter. The filter specification matches the particle size distribution of impurities in the petroleum. The filter specification is determined according to the particle size range that each level of the filter needs to filter to ensure an effective filtering effect. Exemplarily, if the number of filter levels is 4, the filter specifications can be configured as follows: The first-level filter has a pore size of 100 microns, which is used to filter larger particles. The second-level filter has a pore size of 50 microns, which is used to filter medium-sized particles. The third-level filter has a pore size of 10 microns, which is used to filter smaller particles. The fourth-level filter has a pore size of 1 micron, which is used to filter the smallest particles.

[0035] Establish key adjustment parameters of the petroleum filtering device, wherein the key adjustment parameters include flow rate parameters, pressure parameters, and temperature parameters.

[0036] Optionally, the petroleum filtering device includes key components such as a filter, a pump, and a control valve. By adjusting and controlling key parameters of the above key components accordingly, the operating state of the petroleum filtering device can be changed to optimize the filtering effect.

[0037] Optionally, the key adjustment parameters refer to the key parameters of the key components in the oil filter device, and the key adjustment parameters include flow rate parameters, pressure parameters, temperature parameters, etc. Among them, the flow rate parameter is the flow rate of the oil in the filter device, taking into account the flow area of ​​the oil filter device, usually in liters per minute (L / min) or cubic meters per hour (m³ / h). The pressure parameter refers to the pressure in the filter device, including the feed pressure, filter cake pressure, etc., usually in Pascal (Pa) or Bar (Bar). The temperature parameter refers to the temperature in the filter device, including the feed temperature, filter cake temperature, etc.

[0038] Specifically, the flow rate parameter is monitored in real time by a flow meter and adjusted by a flow control valve. A flow rate that is too high may cause damage to the filter medium or particles to penetrate the filter layer. A flow rate that is too low may result in low filtration efficiency and increased energy consumption. The pressure parameter is monitored in real time by a pressure sensor. Appropriate pressure can ensure the normal operation of the filter medium and prevent excessive compression or breakage. Excessive pressure will reduce the service life of the filter medium or cause the filter to rupture, while too low pressure may result in poor filtration. The temperature parameter is adjusted using a preheater or cooler. Maintaining the appropriate temperature can ensure that the viscosity of the oil is within a range that is conducive to efficient filtration. However, too high a temperature may cause the material performance of the filter medium to deteriorate or even damage the filter, while too low a temperature may increase the viscosity of the oil and reduce the filtration efficiency. In addition, the regulation of temperature parameters also needs to consider the regulation of energy consumption.

[0039] The big data is called through the petroleum feature set to construct a fitness function, and data selection is performed based on the big data and in-plant control data to establish a solution space and an initial solution set, wherein the solution space is a space for searching and optimizing, and the initial solution set is a solution set constructed based on a preset population size and data selection.

[0040] Optionally, a fitness function is used to evaluate the quality of the solution. By calling the big data in the petroleum feature set, a fitness function can be constructed. Specifically, the fitness function is defined based on the key adjustment parameters of the filter device or the filtering performance index. The fitness function can be a proxy model or a theoretical model based on experimental data or historical data.

[0041] Exemplarily, a fitness function is constructed. First, a distance-based fitness evaluation is performed based on the deviation between the target filtering performance and the historical filtering performance, the deviation between the target filtering performance and the historical filtering performance is calculated, and the fitness is defined as the inverse of the deviation to obtain multiple fitnesses; then, based on the correlation between the historical filtering performance and the historical control records, a mapping relationship between multiple fitnesses and multiple groups of corresponding key adjustment parameters in multiple historical control records is established, that is, by analyzing multiple fitnesses and multiple groups of corresponding key adjustment parameters, a function is obtained to describe the relationship between fitness and key adjustment parameters. This can be achieved through data fitting, such as multivariate regression, neural network or other machine learning methods; finally, the function is stored as a fitness function to predict the fitness of a given group of key adjustment parameters, thereby facilitating the acquisition of the best key adjustment parameters.

[0042] Optionally, the fitness function needs to be adaptively adjusted according to the specific requirements of the target scenario. For example, while optimizing the filtering performance, other objectives such as minimizing energy consumption or cost need to be considered.

[0043] Furthermore, based on the interactive petroleum filtration big data of the petroleum feature set, big data petroleum filtration control records that are similar to the petroleum feature set are collected; then, the in-plant control data and the big data petroleum filtration control records are merged, the value range of each key adjustment parameter is defined, and a solution space is generated, which is a set of all possible solutions; then, multiple groups of control data or control records whose filtration performance meets the preset filtration performance are selected as the initial solution set, which is a subset in the solution space and is used to initialize the optimization algorithm. Some solutions with better filtration performance are selected as the initial solution set through the threshold screening method, which helps to improve the optimization efficiency of subsequent optimization.

[0044] Optionally, the preset filtering performance can be the target filtering performance, the current filtering performance or other higher standard filtering performance, and the selection of the preset filtering performance is based on the optimization preference of the target scenario. Specifically, if the target scenario has no rigid filtering performance requirements, the current filtering performance can be used as the preset filtering performance; if the target scenario is highly volatile, a filtering performance with a higher standard than the target filtering performance can be selected as the preset filtering performance, thereby ensuring that the key adjustment parameters finally obtained can still meet the target filtering performance when operating under the most unfavorable conditions of the target scenario.

[0045] Optionally, if the size of the obtained initial solution set does not meet the preset population size, the population size of the initial solution set is adjusted, specifically, including supplementing the selection of a smaller initial solution set in the above solution space or eliminating the larger initial solution set based on filtering performance.

[0046] The solutions in the initial solution set are evaluated based on the fitness function, and iterative screening is performed based on the evaluation results to establish superior solution identifiers and inferior solution identifiers.

[0047] Specifically, first, the fitness function evaluates the fitness of the solutions in the initial solution set, obtains a fitness value set, and outputs the evaluation result; then, it screens according to the fitness value of the evaluation result, selects the best solution and the worst solution, and optimizes the initial solution set through iteration, wherein the screening process can be achieved through screening strategies such as roulette selection and tournament selection.

[0048] Exemplarily, iterative screening is performed based on the evaluation results to establish superior solution identification and inferior solution identification, including: sorting the solutions according to the fitness of each solution; selecting a part of solutions with the highest fitness values ​​(such as the top 5% of solutions in terms of fitness) as superior solutions, and a part of solutions with the lowest fitness values ​​(such as the bottom 5% of solutions in terms of fitness) as inferior solutions; iterative screening of the initial solution set based on the identification results until the established superior solution identification and inferior solution identification meet the identification quantity requirement.

[0049] The initial solution set is multiplied and mutated in the solution space through the superior solution identifier and the inferior solution identifier, and the initial solution set iteration is performed.

[0050] In some embodiments, the performing of initial solution set propagation and mutation in the solution space by using the superior solution identifier and the inferior solution identifier to perform initial solution set iteration further includes: Get the solution set with the optimal solution mark, perform data interactive crossover on the corresponding solution set, and generate the crossover result.

[0051] The crossover result is fine-tuned and mutated, a new solution is established according to the fine-tuned and mutated perturbation result, and the new solution is updated to the initial solution set to complete the iteration of the initial solution set.

[0052] Optionally, perform initial solution set propagation and mutation in solution space. First, select the optimal solution from the initial solution set, and perform data crossover within the solution set marked with the optimal solution, that is, randomly exchange the key adjustment parameters of the solution in the solution set marked with the optimal solution to generate a new solution. The crossover result contains some features of the optimal solution and has higher potential performance.

[0053] Furthermore, the crossover results are fine-tuned and perturbed, new solutions are established based on the perturbation results, and these new solutions are updated to the initial solution set to supplement the original initial solution set and complete the iteration of the initial solution set. Among them, fine-tuning and perturbation refers to making small random or regular changes to the crossover results to increase the diversity of solutions.

[0054] Through the above steps, the best solution is identified and the worst solution is identified, the initial solution set is multiplied and mutated, and an iterative process is performed, thereby gradually optimizing the quality of the solution and improving the filtering efficiency of the filtering device.

[0055] In some implementations, updating the newly added solution to the initial solution set to complete the initial solution set iteration further includes: The solution set of the inferior solution marker is subjected to random perturbation mutation based on the solution space to construct a mutant solution set.

[0056] The mutated solution set and the newly added solution set are synchronously updated to the initial solution set, the updated initial solution set is screened by the fitness function, and a round of iteration of the initial solution set is completed according to the screening result.

[0057] Optionally, random perturbation is introduced into the iteration process to increase the diversity of the search space and prevent premature convergence. First, the solution set identified as a poor solution in the previous iteration is randomly perturbed to search for new possible optimization directions in the solution space. In other words, the poor solution currently performs poorly, but may have potential in some directions.

[0058] Furthermore, new possible solutions (including variant solutions and newly added solutions) are introduced into the initial solution set, and the updated initial solution set is evaluated and screened through the fitness function. The best solutions will be retained, while the worst solutions will be eliminated, thus gradually approaching the global optimal solution.

[0059] Optionally, the above process is repeated in multiple rounds of iterations to achieve multiple rounds of iterations of the initial solution set. This helps to find the optimal solution in the search space while maintaining the diversity of the search space and preventing premature convergence to a local optimal solution.

[0060] In some implementations, the performing solution screening on the updated initial solution set by using the fitness function, and completing a round of iteration of the initial solution set according to the screening result, further includes: A preset retention threshold of inferior solutions is set, and a minimum number of inferior solutions to be retained is configured based on the preset retention threshold and the preset population size.

[0061] When the updated initial solution set is screened by the fitness function, if the number of inferior solutions triggers the minimum number of inferior solutions to be retained, the elimination of inferior solutions is stopped and the last-place elimination of superior solutions is performed until the number of solutions in the initial solution set meets the preset population size.

[0062] Optionally, according to the optimization goal, a preset retention threshold of inferior solutions is set, which determines the minimum number of inferior solutions to be retained during the iteration process, thereby avoiding premature convergence and maintaining the diversity and exploratory nature of the population. The preset retention threshold is represented as a ratio or coefficient, representing the proportion of the minimum number of inferior solutions to be retained to the updated initial solution set.

[0063] Specifically, the updated initial solution set is evaluated using the fitness function, and the solutions are sorted according to the evaluation results. When the number of inferior solutions reaches or exceeds the preset minimum retention number, the elimination of inferior solutions is stopped and the elimination of superior solutions is performed instead. The elimination of superior solutions helps to maintain the competitiveness of the population and prevent over-reliance on certain superior solutions.

[0064] Exemplarily, the last-ranked elimination of optimal solutions is performed, starting from the solution with the highest fitness, and the last-ranked solutions are eliminated in sequence until the number of solutions in the initial solution set meets the preset population size, thereby ensuring the quality and diversity of the solution set while keeping the size of the population unchanged.

[0065] In general, the above steps help maintain diversity and competitiveness during the search process, while controlling the size of the initial solution set to increase the probability of finding the optimal solution or a near-optimal solution.

[0066] In some implementations, setting a preset retention threshold for inferior solutions further includes: A supervision network is configured, and the optimization iteration process is monitored and analyzed through the supervision network to generate an iteration cycle prediction result.

[0067] A threshold-retention period decreasing function of the inferior solution is established, and the preset retention threshold is adaptively updated through the threshold-retention period decreasing function and the iterative period prediction result to complete the setting of the preset retention threshold.

[0068] Specifically, first, a supervision network is designed and trained to monitor and analyze the optimization iterative process, obtain the stage of the iterative process, and generate an iterative cycle prediction result. The iterative cycle prediction result includes the early, middle, and late stages of the iterative cycle. Preferably, the iterative cycle prediction result is characterized by the iterated proportion of the iterative cycle.

[0069] Optionally, the threshold-retention period decreasing function is a function that describes how the threshold decreases with the retention period, and the preset retention threshold can be adaptively updated according to the prediction results of the iteration period. In this way, the preset retention threshold can be automatically adjusted as the iteration process progresses, which can help retain more inferior solutions that may have potential, while also preventing excessive invalid exploration.

[0070] Generate an optimization result according to the iteration result, and perform tuning and filtering processing based on the filter configuration result and the optimization result.

[0071] Optionally, after the iteration is completed, the optimal solution in the iteration result is selected as the optimization result, and the optimization result is the optimal key adjustment parameter. Then, according to the optimal solution of the optimization result, the parameters of the filter device configured with the filter screen are adjusted to achieve the best filtering effect, ensuring that the filter device achieves the expected filtering effect and quality standards in actual application.

[0072] In some embodiments, the method further comprises: A monitoring sensor is established based on the filter configuration result, and data monitoring is performed based on the monitoring sensor to establish a monitoring data set.

[0073] The monitoring data set is used to perform time series filtering verification and establish a filter clogging warning.

[0074] Based on the filter clogging warning, an automatic backwashing operation is performed to perform self-cleaning of the filter.

[0075] Optionally, according to the filter configuration results, select the appropriate sensor type and location, install monitoring equipment in the oil filtration device and the target scene, and use the installed monitoring sensors to monitor key parameters in the filtration device in real time, such as flow, pressure, temperature, etc. The collected data is sorted and stored, and output as a monitoring data set, providing real-time information for subsequent analysis and processing.

[0076] Furthermore, the monitoring data set is analyzed using methods such as time series analysis to verify the performance of the filter device to find abnormal conditions in the filtering process, such as filter blockage, filter damage, etc., and a filter blockage warning is established based on the results of the time series filtering verification. The filter blockage warning is characterized by a filter blockage warning instruction, which includes the filter level of the filter blockage, the severity of the blockage, etc.

[0077] Optionally, when the filter blockage warning is triggered, the backwash operation is automatically started to remove impurities on the filter and restore the performance of the filter. The filter self-cleaning can extend the service life of the filter and reduce maintenance costs. Through this process, real-time monitoring and automatic control of the filter can be achieved. In some implementations, the automatic backwashing operation is performed based on the filter blockage warning to perform filter self-cleaning, and further includes: A verification timing window is established, and the window verification after the filter self-cleaning is performed through the verification timing window.

[0078] Abnormal data is extracted through the window verification result, and characteristic analysis is performed based on the abnormal data and the historical characteristic data of the petroleum filtering device, and the device abnormality is reported according to the characteristic analysis result.

[0079] Among them, the verification timing window defines the time interval and verification collection time for verifying the filter device after the filter self-cleaning, ensuring that there is enough time to evaluate the effect of the filter self-cleaning after the backwashing operation.

[0080] Optionally, within the set verification timing window, the data of the filter device is collected by monitoring sensors, and by comparing the data difference before and after the filter self-cleaning, it is determined whether the filter self-cleaning operation is effective and whether the filter device has resumed normal operation.

[0081] Optionally, use statistical analysis, machine learning and other data analysis methods to extract and analyze features of abnormal data, determine the root cause of the abnormality, and provide a basis for subsequent maintenance and troubleshooting. For example, if the filtration pressure or filtration rate does not return to the expected level after cleaning, there may be abnormalities such as insufficient backwashing or backwashing equipment failure.

[0082] By automatically responding to filter blockage warnings and verifying the effectiveness of filter self-cleaning operations, detailed information on abnormal problems can be analyzed and obtained, which helps to quickly locate and solve problems, thereby improving the reliability and production efficiency of the filtration device.

[0083] In summary, the tuning and filtering method for a petroleum filtering device provided by the present invention has the following technical effects: Through the initial inspection of the oil to be filtered, the oil feature set is established, which includes oil impurity features and oil viscosity features; the oil feature set is input into the filter setting model to generate the filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications under each grade; the key adjustment parameters of the oil filter device are established, wherein the key adjustment parameters include flow rate parameters, pressure parameters, and temperature parameters; the big data is called through the oil feature set, the fitness function is constructed, and data selection is performed based on the big data and the in-plant control data to establish the solution space and the initial solution set, wherein the solution space is the space for searching and optimizing, and the initial solution set is the solution set constructed according to the preset population size and data selection; the solution evaluation in the initial solution set is performed based on the fitness function, and iterative screening is performed based on the evaluation results to establish the optimal solution mark and the inferior solution mark; the initial solution set is propagated and the variation in the solution space is performed through the optimal solution mark and the inferior solution mark, and the initial solution set is iterated; the optimization result is generated according to the iterative result, and the filter configuration result and the optimization result are used to tune the filtering process. Thus, the technical effect of improving the adaptability of the oil filter device and improving the filtration efficiency is achieved. Embodiment 2

[0084] Figure 2 Schematic diagram of the structure of the tuning filter system of the present invention for petroleum filtration device. For example, Figure 1 The flow chart of the method for optimizing the filtration of the petroleum filtration device of the present invention can be shown as follows: Figure 2 The structure shown is implemented.

[0085] Based on the same concept as the tuning and filtering method for the petroleum filtering device in the above embodiment, the present invention also provides a tuning and filtering system for the petroleum filtering device, comprising: The oil preliminary inspection module 11 is used to perform preliminary inspection on the oil to be filtered and establish an oil feature set, wherein the oil feature set includes oil impurity features and oil viscosity features.

[0086] The filter configuration module 12 is used to input the oil feature set into the filter setting model to generate a filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications under each grade.

[0087] The adjustment parameter module 13 is used to establish key adjustment parameters of the petroleum filtering device, wherein the key adjustment parameters include flow rate parameters, pressure parameters, and temperature parameters.

[0088] The evaluation initialization module 14 is used to call big data through the petroleum feature set, construct a fitness function, and perform data selection based on the big data and in-plant control data to establish a solution space and an initial solution set, wherein the solution space is a space for searching and optimizing, and the initial solution set is a solution set constructed based on a preset population size and data selection.

[0089] The screening and identification module 15 is used to evaluate the solutions in the initial solution set based on the fitness function, perform iterative screening based on the evaluation results, and establish the identification of the best solutions and the identification of the worst solutions.

[0090] The mutation and iteration module 16 is used to perform the initial solution set propagation and mutation in the solution space through the superior solution identifier and the inferior solution identifier, and perform the initial solution set iteration.

[0091] The tuning and filtering module 17 is used to generate an optimization result according to the iteration result, and perform tuning and filtering processing based on the filter configuration result and the optimization result.

[0092] The mutation iteration module 16 includes: The data interactive cross unit is used to obtain a solution set with an optimal solution identifier, perform data interactive cross on the corresponding solution set, and generate a cross result.

[0093] A fine-tuning unit is used to perform fine-tuning and variation disturbance on the crossover result, establish a new solution according to the fine-tuning and variation disturbance result, and update the new solution to the initial solution set to complete the iteration of the initial solution set.

[0094] In some embodiments, the fine-tuning unit in the mutation iteration module 16 includes: The random perturbation mutation unit is used to perform random perturbation mutation on the solution set of the inferior solution marker based on the solution space to construct a mutant solution set.

[0095] The solution screening unit is updated to synchronously update the variant solution set and the newly added solution set to the initial solution set, screen the updated initial solution set through the fitness function, and complete a round of iteration of the initial solution set according to the screening result.

[0096] In some implementations, updating the solution screening unit in the mutation iteration module 16 includes: The retention amount configuration unit is used to set a preset retention threshold of inferior solutions and configure a minimum retention amount of inferior solutions based on the preset retention threshold and the preset population size.

[0097] The screening and retention unit is used to stop the elimination of inferior solutions and perform the last-place elimination of superior solutions when the number of inferior solutions triggers the minimum retention number of inferior solutions when screening the updated initial solution set through the fitness function, until the number of solutions in the initial solution set meets the preset population size.

[0098] In some implementations, the retention amount configuration unit in the mutation iteration module 16 includes: The cycle prediction unit is used to configure a supervision network, monitor and analyze the optimization iteration process through the supervision network, and generate an iteration cycle prediction result.

[0099] The retention threshold updating unit is used to establish a threshold-retention period decreasing function of the inferior solution, and adaptively update the preset retention threshold through the threshold-retention period decreasing function and the iterative period prediction result to complete the setting of the preset retention threshold.

[0100] Furthermore, the system also includes: The monitoring and collecting unit is used to establish a monitoring sensor based on the filter configuration result, and perform data monitoring based on the monitoring sensor to establish a monitoring data set.

[0101] The clogging warning unit is used to perform time series filtering verification through the monitoring data set and establish a filter clogging warning.

[0102] The self-cleaning unit is used to perform automatic backwashing operation based on the filter clogging warning to perform self-cleaning of the filter.

[0103] In some embodiments, the self-cleaning unit further comprises: The self-cleaning verification unit is used to establish a verification timing window, and to perform window verification after the filter self-cleaning through the verification timing window.

[0104] The abnormality analysis unit is used to extract abnormal data through the window verification result, and perform characteristic analysis based on the abnormal data and the historical characteristic data of the petroleum filtering device, and report the device abnormality according to the characteristic analysis result.

[0105] It should be understood that the embodiments mentioned in this specification focus on their differences from other embodiments. The specific embodiments in the aforementioned embodiment one are also applicable to the tuning and filtering system for the petroleum filtering device described in embodiment two. For the sake of brevity of the specification, they will not be further expanded here.

[0106] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for optimizing filtration of a petroleum filtration device, characterized in that: The method comprises: Performing a preliminary petroleum inspection on the petroleum to be filtered and establishing a petroleum feature set, wherein the petroleum feature set includes petroleum impurity features and petroleum viscosity features; Inputting the oil feature set into a filter setting model to generate a filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications under each grade; Establishing key adjustment parameters of the petroleum filtration device, wherein the key adjustment parameters include flow rate parameters, pressure parameters, and temperature parameters; The big data is called by the petroleum feature set to construct a fitness function, and data selection is performed based on the big data and the in-plant control data to establish a solution space and an initial solution set, wherein the solution space is a space for searching and optimizing, and the initial solution set is a solution set constructed according to a preset population size and data selection; Evaluate the solutions in the initial solution set based on the fitness function, perform iterative screening based on the evaluation results, and establish a superior solution identifier and a inferior solution identifier; The initial solution set is multiplied and mutated in the solution space by using the superior solution identifier and the inferior solution identifier, and the initial solution set is iterated; Generate an optimization result according to the iteration result, and perform tuning and filtering processing based on the filter configuration result and the optimization result.

2. The method according to claim 1, characterized in that The method of performing initial solution set propagation and mutation in the solution space by using the superior solution identifier and the inferior solution identifier to perform initial solution set iteration also includes: Obtain the solution set with the optimal solution mark, perform data interactive crossover on the corresponding solution set, and generate crossover results; The crossover result is fine-tuned and mutated, a new solution is established according to the fine-tuned and mutated perturbation result, and the new solution is updated to the initial solution set to complete the iteration of the initial solution set.

3. The method according to claim 2, characterized in that The updating of the newly added solution to the initial solution set to complete the iteration of the initial solution set further includes: Performing random perturbation mutation on the solution set of the inferior solution marker based on the solution space to construct a mutation solution set; The mutated solution set and the newly added solution set are synchronously updated to the initial solution set, the updated initial solution set is screened by the fitness function, and a round of iteration of the initial solution set is completed according to the screening result.

4. The method according to claim 3, characterized in that The method further comprises: screening the updated initial solution set by using the fitness function, and completing a round of iteration of the initial solution set according to the screening result; Setting a preset retention threshold for inferior solutions, and configuring a minimum number of inferior solutions to be retained based on the preset retention threshold and the preset population size; When the updated initial solution set is screened by the fitness function, if the number of inferior solutions triggers the minimum number of inferior solutions to be retained, the elimination of inferior solutions is stopped and the last-place elimination of superior solutions is performed until the number of solutions in the initial solution set meets the preset population size.

5. The method according to claim 4, characterized in that The setting of the preset retention threshold of the inferior solution also includes: A supervision network is configured to monitor and analyze the optimization iteration process through the supervision network to generate an iteration cycle prediction result; A threshold-retention period decreasing function of the inferior solution is established, and the preset retention threshold is adaptively updated through the threshold-retention period decreasing function and the iterative period prediction result to complete the setting of the preset retention threshold.

6. The method according to claim 1, characterized in that The method further comprises: Establishing a monitoring sensor based on the filter configuration result, and performing data monitoring based on the monitoring sensor to establish a monitoring data set; Perform time series filtering verification through the monitoring data set to establish filter blockage warning; Based on the filter clogging warning, an automatic backwashing operation is performed to perform self-cleaning of the filter.

7. The method according to claim 6, characterized in that The automatic backwashing operation is performed based on the filter blockage warning to perform filter self-cleaning, and further includes: Establishing a verification timing window, and performing window verification after filter self-cleaning through the verification timing window; Abnormal data is extracted through the window verification result, and characteristic analysis is performed based on the abnormal data and the historical characteristic data of the petroleum filtering device, and the device abnormality is reported according to the characteristic analysis result.

8. A tuning filter system for a petroleum filter, characterized in that: The system is used to execute the tuning and filtering method for a petroleum filtering device according to any one of claims 1 to 7, and the system comprises: An oil preliminary inspection module, which is used to perform preliminary inspection on the oil to be filtered and establish an oil feature set, wherein the oil feature set includes oil impurity features and oil viscosity features; A filter configuration module, the filter configuration module is used to input the oil feature set into the filter setting model to generate a filter configuration result, wherein the filter configuration result includes the number of filter grades and the filter specifications under each grade; An adjustment parameter module, the adjustment parameter module is used to establish key adjustment parameters of the petroleum filtering device, wherein the key adjustment parameters include flow rate parameters, pressure parameters, and temperature parameters; An evaluation initialization module, the evaluation initialization module is used to call big data through the petroleum feature set, construct a fitness function, and perform data selection based on the big data and in-plant control data to establish a solution space and an initial solution set, wherein the solution space is a space for searching and optimizing, and the initial solution set is a solution set constructed according to a preset population size and data selection; A screening identification module, wherein the screening identification module is used to evaluate solutions in the initial solution set based on the fitness function, perform iterative screening based on the evaluation results, and establish a superior solution identification and a inferior solution identification; A mutation iteration module, the mutation iteration module is used to perform initial solution set propagation and mutation in the solution space through the superior solution identifier and the inferior solution identifier, and perform initial solution set iteration; The tuning and filtering module is used to generate an optimization result according to the iteration result, and perform tuning and filtering processing based on the filter configuration result and the optimization result.

Images