Method for monitoring and adjusting dynamic filtration resistance of a metal mesh filter tube

Abstract

The application discloses a kind of metal mesh filter tube dynamic filtration resistance monitoring and adjusting method, it is related to filter tube filtration control technical field, the method includes: according to the physical property search of fluid to be filtered coding, exports fluid reference viscosity;Receive time sequence flow signal and time sequence temperature signal;Output fluid dynamic viscosity;Interactively obtain the effective filtration area and maximum allowable filtration resistance of metal mesh filter tube;Based on effective filtration area, time sequence differential pressure signal, time sequence flow signal and fluid dynamic viscosity, interactive calculation is carried out, and time sequence filtration resistance is output;Output resistance trend time-varying curve;Carry out backwashing and execute cost optimization, and output resistance adjustment strategy.The application solves the technical problems that the prior art exists metal mesh filter tube filtration resistance monitoring lacks dynamic, filtration efficiency and backwashing energy consumption are difficult to coordinate and balance, achieves the dynamic monitoring of metal mesh filter tube filtration resistance, improves the technical effect of filtration efficiency and backwashing energy consumption balance control.

Description

Technical Field

[0001] This invention relates to the field of filter tube filtration control technology, specifically to a method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube. Background Technology

[0002] In the field of industrial filtration, metal mesh filter tubes are widely used due to their stable structure, high pressure resistance, and ease of cleaning. However, during the filtration process, impurities gradually accumulate on the surface of the filter tubes, leading to a continuous increase in filtration resistance. Excessive resistance reduces filtration flow, increases system energy consumption, and may even cause filter tube blockage or rupture. Therefore, backwashing is necessary to remove impurities and restore filtration performance. In current technologies, resistance monitoring of metal mesh filter tubes mostly relies on manual detection at fixed time intervals or static readings from a single sensor. This makes it difficult to capture the dynamic trend of resistance changes with fluid properties and impurity accumulation rate in real time, and is prone to resistance monitoring lag. At the same time, backwashing strategies often adopt preset fixed cycles or fixed pressure threshold triggering modes, without flexibly adjusting based on the actual volume and time limit of the filtration task, fluid dynamic viscosity, and other parameters. This makes it impossible to achieve synergistic optimization of filtration efficiency and backwashing energy consumption, and fails to meet the industrial requirements for high-precision, low-energy-consumption, and long-life operation of filtration systems.

[0003] Existing technologies suffer from technical problems such as a lack of dynamic monitoring of filtration resistance in metal mesh filter tubes and difficulty in achieving a coordinated balance between filtration efficiency and backwashing energy consumption. Summary of the Invention

[0004] This application provides a method for monitoring and adjusting the dynamic filtration resistance of metal mesh filter tubes, which addresses the technical problems in the prior art where the monitoring of filtration resistance of metal mesh filter tubes lacks dynamism and the filtration efficiency and backwashing energy consumption are difficult to balance.

[0005] In view of the above problems, this application provides a method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube, the method comprising:

[0006] The system performs property retrieval based on the fluid code to be filtered and outputs the fluid reference viscosity. After starting fluid filtration based on predefined operating conditions, it receives time-series flow and temperature signals from flow meters and temperature sensors pre-installed on the main filtration pipeline. It then performs temperature compensation on the fluid reference viscosity based on the time-series temperature signal and outputs the fluid dynamic viscosity. The system interactively obtains the effective filtration area and maximum allowable filtration resistance of the metal mesh filter tube. After receiving the time-series differential pressure signal from the differential pressure transmitter, it performs interactive calculations based on the effective filtration area, time-series differential pressure signal, time-series flow signal, and fluid dynamic viscosity, outputting the time-series filtration resistance. The metal mesh filter tube is deployed on the main filtration pipeline, and the differential pressure transmitter is connected between the inlet and outlet of the metal mesh filter tube. Using the maximum allowable filtration resistance as the prediction termination boundary, it performs resistance time-varying trend fitting based on the time-series filtration resistance and outputs the resistance trend time-varying curve. Finally, using a preset filtration volume time limit as a constraint, it optimizes the backwashing execution cost based on the resistance trend time-varying curve and outputs the resistance adjustment strategy.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The system performs property retrieval based on the fluid code to be filtered, outputting the fluid's baseline viscosity. After fluid filtration is initiated based on predefined operating conditions, it receives time-series flow and temperature signals from flow meters and temperature sensors pre-installed on the main filtration pipeline. Temperature compensation is applied to the fluid's baseline viscosity based on the time-series temperature signal, outputting the fluid's dynamic viscosity. The system interactively obtains the effective filtration area and maximum allowable filtration resistance of the metal mesh filter tube. Based on the effective filtration area, time-series differential pressure signal, time-series flow signal, and fluid dynamic viscosity, it performs interactive calculations, outputting the time-series filtration resistance. Based on the time-series filtration resistance, it performs resistance time-varying trend fitting, outputting a resistance trend time-varying curve. With a preset filtration volume time limit as a constraint, and based on the resistance trend time-varying curve, it optimizes the backwashing execution cost, outputting a resistance adjustment strategy. This achieves dynamic monitoring of the metal mesh filter tube's filtration resistance, improving the balance between filtration efficiency and backwashing energy consumption control. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A flowchart illustrating a method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube, provided in an embodiment of this application;

[0011] Figure 2 This is a flowchart illustrating the time-varying curve of the output resistance trend in a method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube, as provided in an embodiment of this application. Detailed Implementation

[0012] This application provides a method for monitoring and adjusting the dynamic filtration resistance of metal mesh filter tubes, which addresses the technical problems in the prior art where the monitoring of filtration resistance of metal mesh filter tubes lacks dynamism and the filtration efficiency and backwashing energy consumption are difficult to balance.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] Examples, such as Figure 1 As shown, this application provides a method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube, the method comprising:

[0015] Step S100: Perform a property search based on the code of the fluid to be filtered and output the fluid reference viscosity.

[0016] Specifically, the process begins by obtaining a unique code for the fluid to be filtered. This code is pre-associated with the fluid's basic physical properties and can be used as a retrieval index to access a pre-defined physical property database. A precise search is then performed in the physical property database using this code to match the viscosity parameter under standard conditions corresponding to the fluid to be filtered. This parameter is then determined as the fluid's reference viscosity and output. This reference viscosity will serve as the core initial data for subsequent temperature compensation based on time-series temperature signals and for calculating the fluid's dynamic viscosity, providing fundamental physical property support for the precise monitoring and adjustment of the filtration resistance of the entire metal mesh filter tube.

[0017] Step S200: After starting the fluid filtration based on predefined operating conditions, receive the time-series flow signal and time-series temperature signal returned by the installed flow meter and temperature sensor pre-installed on the main filtration pipeline.

[0018] Specifically, the entire fluid filtration system is first started according to a predefined operating condition, which includes parameters such as filtration pressure and initial flow threshold adapted to the filtration requirements of the metal mesh filter tube. This allows the fluid to be filtered to enter the main filtration pipeline along a predefined path. The metal mesh filter tube is deployed in this main pipeline, providing the core carrier for subsequent resistance monitoring. After the filtration system is running stably, it receives signals from two types of sensors pre-installed in the main filtration pipeline in real time. Flowmeters collect real-time flow data of the fluid in the main pipeline at fixed sampling intervals, such as once per second, forming and sending a time-series flow signal. Simultaneously, temperature sensors collect real-time temperature data of the fluid in the main pipeline, forming and sending a time-series temperature signal. These two types of time-series signals will serve as key real-time data sources for calculating the fluid's dynamic viscosity and time-series filtration resistance in subsequent steps.

[0019] Step S300: Perform temperature compensation on the fluid reference viscosity based on the time-series temperature signal, and output the fluid dynamic viscosity.

[0020] Specifically, the received time-series temperature signal is first preprocessed. First, a third-order low-pass filter is performed to eliminate high-frequency interference noise. Then, based on a preset window width, such as a 5-second window, a sliding window averaging operation is performed, ultimately outputting a steady-state temperature value that reflects the true temperature of the fluid under the current filtration conditions. Subsequently, the physical property database is searched again using the fluid code to be filtered to obtain the fluid type, viscosity-temperature coefficient, critical temperature boundary, and the already output fluid reference viscosity. If the steady-state temperature value is within the retrieved critical temperature boundary range, meaning the fluid does not exhibit phase change risk, the corresponding temperature compensation function is called based on the fluid type. The steady-state temperature value and viscosity-temperature coefficient are substituted into this function to perform dynamic viscosity calculation, obtaining and outputting a dynamic viscosity that accurately reflects the actual viscosity characteristics of the fluid at the current temperature. If the steady-state temperature value deviates from the critical temperature boundary, indicating a possible phase change, which would significantly affect the viscosity accuracy, the viscosity compensation calculation process is immediately paused, and a phase change risk warning flag is simultaneously output to prompt staff to promptly investigate any abnormal operating conditions.

[0021] Step S400: Interact to obtain the effective filtration area and maximum allowable filtration resistance of the metal mesh filter tube.

[0022] Specifically, the system obtains key parameters of the metal mesh filter tube through a preset human-machine interface or automated data interaction interface. On the one hand, it receives effective filtration area data input by the operator based on the actual specifications of the metal mesh filter tube currently deployed in the main filtration pipeline, such as filter tube aperture, wire diameter, and effective filtration area size, or directly calls the preset effective filtration area value bound to the metal mesh filter tube model in the filter tube factory parameter database. On the other hand, it determines the maximum allowable filtration resistance through interactive means, based on the material resistance performance of the metal mesh filter tube, the design requirements of the filtration system, and the degree of contamination of the fluid to be filtered. That is, the upper limit of filtration resistance that the filter tube can withstand under the premise of ensuring filtration efficiency and without structural damage or functional failure.

[0023] Step S500: After receiving the time-series differential pressure signal returned by the differential pressure transmitter, perform interactive calculation based on the effective filtration area, time-series differential pressure signal, time-series flow signal and fluid dynamic viscosity, and output time-series filtration resistance. The metal mesh filter tube is deployed in the main filtration pipeline, and the differential pressure transmitter is connected between the inlet and outlet of the metal mesh filter tube.

[0024] Specifically, firstly, the system receives a timing differential pressure signal from a differential pressure transmitter. This transmitter collects data at preset sampling intervals and is deployed across the inlet and outlet of a metal mesh filter tube within the main filter pipeline. This allows for precise capture of the real-time pressure difference across the filter tube caused by impurities in the fluid, ensuring the timing differential pressure signal accurately reflects the initial pressure change in the filter tube's filtration resistance. Next, considering potential differences in sensor response speeds, interpolation compensation is performed on the received timing differential pressure signal and the retained timing flow signal to correct for response delay discrepancies, outputting a timing-corrected differential pressure and a timing-corrected flow rate. Subsequently, combining the interactively obtained effective filtration area of ​​the metal mesh filter tube, the timing surface velocity of the fluid passing through the filter tube—that is, the fluid velocity passing through a unit filtration area per unit time—is calculated based on the timing-corrected flow rate. Simultaneously, the real-time timing differential pressure value is calculated from the timing-corrected differential pressure. Finally, the calculated time-series differential pressure and time-series surface velocity, along with the temperature-compensated fluid dynamic viscosity, are substituted into the preset filtration resistance solution model to perform interactive calculations, generating and outputting in real time the time-series filtration resistance that dynamically reflects the change of the metal mesh filter tube's filtration resistance over time.

[0025] Step S600: Using the maximum allowable filter resistance as the prediction termination boundary, perform time-varying trend fitting based on the time-series filter resistance, and output the time-varying curve of the resistance trend.

[0026] Specifically, first, a dynamic prediction step time is set, such as 1 minute / step, to provide a time interval benchmark and maximum filtration time span for single-step resistance prediction. The maximum filtration time span is retrieved from the preset filtration volume time limit of the entire method to ensure that the prediction range is consistent with the overall filtration task requirements. Next, the terminal resistance value of the output time-series filtration resistance is retrieved, i.e., the latest collected resistance data. Using this terminal resistance value as the prediction starting point, single-step resistance prediction is executed successively according to the set dynamic prediction step time, generating time-series predicted resistance that extends over time. When a certain updated predicted resistance point in the time-series predicted resistance reaches the maximum allowable filtration resistance obtained interactively, i.e., the upper limit of resistance where the filter tube performance is safe and the filtration efficiency meets the standard, it serves as the prediction termination boundary. The prediction boundary timestamp corresponding to this predicted resistance point is immediately recorded. Subsequently... The process involves sequentially connecting the time-series filtered resistance with the generated time-series predicted resistance in chronological order. A cubic spline interpolation algorithm is then used to smooth the connected resistance data, fitting a time-varying resistance trend curve that continuously reflects the resistance's changing patterns. Simultaneously, the starting timestamp of the time-series filtered resistance is retrieved, and a cumulative resistance time window is calculated based on the starting timestamp and the predicted boundary timestamp. This cumulative resistance time window is then used to mark the time-varying resistance trend curve, clearly defining the resistance change time range corresponding to the curve. If the calculated cumulative resistance time window exceeds the preset maximum filtering time span, the excess region of the time-varying resistance trend curve is truncated according to the maximum filtering time span, resulting in a corrected resistance trend curve. This corrected curve is then marked using the maximum filtering time span, yielding the final time-varying resistance trend curve.

[0027] Step S700: With a preset filtration volume time limit as a constraint, optimize the backwashing execution cost based on the time-varying curve of the resistance trend, and output a resistance adjustment strategy.

[0028] Specifically, the preset filtration volume time limit is decomposed to extract the total filtration volume and filtration time limit, which serve as the core constraints for optimizing backwashing costs. Next, a predefined first-scale resistance adjustment cycle, such as 30 minutes, is used as the initial optimization interval. The corresponding first local resistance time-varying curve is segmented on the output resistance trend time-varying curve, focusing on the resistance change pattern within this cycle. Subsequently, the first cumulative filtration flow rate within this cycle is derived by reverse calculation based on the first local resistance time-varying curve. The first end resistance at the end of the curve is retrieved, and backwashing parameters are simulated to output the first backwashing control parameters, including backwashing duration and pressure. The first backwashing control duration is then used to compensate for the first-scale resistance adjustment cycle to obtain the first filtration resistance adjustment cycle. Simultaneously, the first backwashing energy consumption is predicted, and the first backwashing frequency is calculated by combining the total filtration volume and the first cumulative filtration flow rate, thereby determining the first backwashing total energy consumption and the first total filtration duration. Finally, the first total filtration duration is calculated. The first deviation percentage and the first deviation vector from the filtration time limit are used as the starting point for optimization. Gradient compensation adjustment is performed according to the deviation percentage to obtain the second-scale resistance adjustment period. The above backwash benefit evaluation process is repeated to output the second backwash control parameters, the second backwash total energy consumption, and the second total filtration time. If the second total filtration time falls within the filtration time limit, it is stored in the candidate adjustment strategy library. This process is repeated for multiple iterations until the candidate adjustment strategy library accumulates N backwash control strategies. Finally, based on predefined two-dimensional dynamic weights, such as energy consumption weight of 0.6 and time weight of 0.4, the total backwash energy consumption and total filtration time of the N backwash control strategies are weighted and fused to obtain the resistance adjustment performance score of each strategy. After sorting the scores in descending order, the strategy corresponding to the best score is selected as the final resistance adjustment strategy output. This strategy can guide the subsequent online backwashing operation of the metal mesh filter tube to achieve precise adjustment of the filtration resistance.

[0029] In one possible implementation, such as Figure 2 As shown, step S600 further includes:

[0030] Step S610: Set the dynamic prediction step length and the maximum filtration time span, wherein the maximum filtration time span is retrieved from the filtration volume time limit.

[0031] Step S620: Retrieve the end resistance value of the time-series filter resistance, and starting from the end resistance value, perform single-step resistance prediction according to the dynamic prediction step time, and output the time-series predicted resistance.

[0032] Step S630: When the updated predicted resistance point of the time-series predicted resistance satisfies the maximum allowable filtering resistance, record the prediction boundary timestamp.

[0033] Step S640: After connecting the time-series filtering resistance and the time-series predicted resistance, perform smooth trend curve fitting by performing cubic spline interpolation, and output the time-varying resistance trend curve.

[0034] Step S650: Retrieve the starting timestamp of the time-series filtering resistance, and calculate the resistance accumulation time window based on the starting timestamp and the predicted boundary timestamp.

[0035] Step S660: Use the resistance accumulation time window to identify the time-varying curve of the resistance trend.

[0036] Specifically, based on the filtration performance characteristics of the metal mesh filter tube, the degree of contamination of the fluid to be filtered, and the actual filtration efficiency requirements, a reasonable dynamic prediction step time is set, such as 1 minute per step. This time interval needs to balance prediction accuracy and computational efficiency to ensure that the single-step prediction result can accurately reflect the trend of resistance change. At the same time, instead of setting the maximum filtration time span independently, the parameter is directly retrieved from the predefined filtration volume time limit in the entire monitoring and adjustment method, that is, the time constraint corresponding to the completion of the preset total filtration volume, and determined as the maximum filtration time span. This ensures that the time range of subsequent resistance prediction matches the volume target and time requirements of the overall filtration task, and avoids the prediction range from exceeding the actual filtration conditions.

[0037] The system retrieves the calculated time-series filtration resistance data, which is a record of resistance continuously changing over time and generated in real time during the operation of the metal mesh filter tube. The final resistance value is extracted from this data; this is the most recently collected and calculated resistance data for the current filtration stage. This value directly reflects the current actual filtration resistance state of the metal mesh filter tube and is therefore determined as the initial reference point for subsequent resistance prediction. Subsequently, using this final resistance value as the prediction starting point, and combined with the pre-set dynamic prediction step time, a pre-trained neural network prediction model is activated to perform single-step resistance prediction. This neural network model has been trained and optimized using a large amount of historical filtration data, including resistance change samples under different fluid viscosities, flow rates, and filter tube blockage levels. Based on the final resistance value and the previous time-series filtration resistance change patterns, it can successively calculate the predicted resistance value for each time node according to the dynamic prediction step time. After each single-step prediction is completed, the predicted resistance value for the corresponding time point is stored. After completing multiple predictions, all single-step prediction results are integrated in chronological order to ultimately form and output a time-series predicted resistance that reflects the resistance change trend in the future.

[0038] During the generation of time-series predicted resistance, real-time monitoring is maintained for each newly generated updated predicted resistance point. Specifically, whenever a single-step resistance prediction is completed according to the dynamic prediction step time and a new predicted resistance value is obtained, the updated predicted resistance point is immediately compared with the maximum permissible filtration resistance of the metal mesh filter tube obtained through interaction. If an updated predicted resistance point reaches or exceeds the maximum permissible filtration resistance, satisfying the preset prediction termination condition, the time information corresponding to that updated predicted resistance point is immediately captured and recorded, and this is determined as the prediction boundary timestamp. This timestamp clearly marks the time node when the metal mesh filter tube's filtration resistance is expected to reach the safe upper limit.

[0039] Two types of core resistance data are integrated over time. The calculated time-series filtering resistance, reflecting the actual changes in filtration resistance of the metal mesh filter tube, includes continuous resistance data from filtration initiation to the current moment, while the output time-series predicted resistance reflects expected resistance changes in the future, including resistance data from the current moment to the prediction boundary timestamp. These are connected in chronological order to form a complete time-dimensional resistance data sequence covering both past and predicted data, ensuring uninterrupted and non-overlapping data along the time axis. Subsequently, cubic spline interpolation is performed on the integrated resistance data sequence. This operation constructs a smooth cubic polynomial curve between adjacent resistance data points, ensuring that the curve not only accurately passes through each original data point but also guarantees the continuity of the first and second derivatives of adjacent curve segments at the connection points. This effectively eliminates potential fluctuations or dispersion in the original data and avoids abrupt curve changes. Finally, through cubic spline interpolation, a continuous, smooth curve that accurately reflects the time-varying law of filtration resistance of the metal mesh filter tube is fitted and output as the time-varying resistance trend curve.

[0040] From the output time-series filter resistance dataset, retrieve the starting timestamp corresponding to the data set. This timestamp is the initial time when the metal mesh filter tube starts filtering and begins collecting resistance data, directly marking the starting node of actual filter resistance monitoring. Subsequently, call the recorded prediction boundary timestamp, substitute the starting timestamp and the prediction boundary timestamp into the preset time difference calculation formula, and subtract the two time values ​​to obtain the complete time interval from the start of filter resistance monitoring to the expected reaching of the maximum allowable filter resistance. This interval is the resistance accumulation time window.

[0041] First, the calculated resistance accumulation time window is retrieved, and then this time window is linked and bound to the fitted resistance trend time-varying curve. Specifically, the range corresponding to the resistance accumulation time window is clearly marked on the time axis of the resistance trend time-varying curve with prominent markers, such as time interval labels and highlighted lines of specific colors. This allows the curve to intuitively reflect the complete cycle of change in the metal mesh filter tube's filtration resistance from the start of monitoring to the expected reaching of the maximum allowable filtration resistance. For example, if the resistance accumulation time window is "09:00:00-09:30:00", then the corresponding interval is marked on the time axis of the curve, and the curve portion within that time period is highlighted simultaneously, enabling quick identification of the core segments in the curve related to the cumulative change in resistance.

[0042] In one possible implementation, step S700 further includes:

[0043] Step S710: Decompose the filtration volume time limit to obtain the total filtration volume and filtration time limit.

[0044] Step S720: Using a predefined first-scale resistance adjustment period, the first local resistance time-varying curve is obtained by segmenting the resistance trend time-varying curve.

[0045] Step S730: Evaluate the backwashing benefit based on the first local resistance time-varying curve and the total filtration volume, and output the first total backwashing energy consumption and the first total filtration time.

[0046] Step S740: Based on the deviation scale between the first total filtration time and the filtration time limit, take the first scale resistance adjustment cycle as the optimization starting point, perform iterative optimization of filtration resistance adjustment until the resistance adjustment strategy is output.

[0047] Specifically, the pre-set filtration volume time limit parameter is retrieved from the entire monitoring and adjustment process. This parameter is based on the filtration capacity of the metal mesh filter tube, the total required amount of fluid to be filtered, and the actual production task requirements, and integrates two key pieces of information: the target filtration volume to be completed and the time constraint for completing that target. Then, a pre-set parameter analysis algorithm decomposes the filtration volume time limit: on the one hand, it extracts the quantitative indicator representing the total filtration task, determining the total filtration volume to be completed through the metal mesh filter tube; on the other hand, it extracts the quantitative indicator representing the time constraint, determining the maximum allowed time to complete this total filtration volume, i.e., the filtration time limit. Through this decomposition, the ambiguous filtration volume time limit is transformed into two clear and quantifiable parameters.

[0048] The first-scale resistance adjustment period is retrieved. This period is a fixed time interval (e.g., 30 minutes, 1 hour) preset based on the routine maintenance interval of the metal mesh filter tube, the contamination level of the fluid to be filtered, and previous filtration test data. It serves as the basic time unit for segmenting the time-varying resistance trend curve. Subsequently, the output time-varying resistance trend curve is located. This curve, with time on the horizontal axis and filtration resistance on the vertical axis, fully reflects the resistance change of the metal mesh filter tube from filtration startup to reaching the expected maximum allowable filtration resistance. Using the first-scale resistance adjustment period as the time segmentation standard, a starting segmentation point is selected on the time-varying resistance trend curve. This point is typically the current filtration time node or the curve's starting time node. A segment of the curve, starting from the starting segmentation point and lasting for a duration equal to the first-scale resistance adjustment period, is extracted. This segment is the first local resistance time-varying curve.

[0049] Based on the time-varying resistance curve of the first local resistance, and combined with parameters such as fluid dynamic viscosity and effective filtration area of ​​the metal mesh filter tube, the filtration flow rate is inversely estimated. The filtration flow rate at each moment within the local period is calculated using the established resistance-flow correlation model. Then, the flow rate at each moment is integrated over time to output the first cumulative filtration flow rate, clarifying the filtration volume that can be completed within a single adjustment cycle. Next, the first end resistance corresponding to the end of the first local resistance time-varying curve is retrieved. This resistance value represents the actual degree of filter tube blockage at the end of the cycle. Based on this resistance value, the system initiates backwash parameter adjustment simulation. By simulating the removal effect of different backwash pressures, flow rates, and durations on the blockage, the optimal parameter combination is selected, and the first backwash control parameters containing information such as backwash pressure, flow rate, and duration are output. Subsequently, the first backwash is extracted from the first backwash control parameters. The backwash control duration is added to the first-scale resistance adjustment cycle to compensate for the time occupied by backwashing, outputting the first filtration resistance adjustment cycle that includes the entire filtration and backwashing process. Simultaneously, based on the pressure, flow rate, and backwashing equipment power parameters in the first backwash control parameters, the energy consumption required for a single backwash is calculated, outputting the first backwash energy consumption. Then, combining the decomposed total filtration volume and the first cumulative filtration flow rate, the first backwash frequency required to complete all filtration tasks is obtained by calculating the total filtration volume ÷ the first cumulative filtration flow rate. Finally, the total energy consumption for completing all filtration tasks is calculated by multiplying the first backwash energy consumption by the first backwash frequency, and the total filtration time for completing all filtration tasks is calculated by multiplying the first filtration resistance adjustment cycle by the first backwash frequency. This completes the backwash benefit evaluation and outputs these two core parameters.

[0050] The deviation between the calculated first total filtration time and the decomposed filtration time limit is specifically calculated using a preset algorithm to determine the first deviation percentage, i.e., the ratio of the deviation amount to the filtration time limit and the first deviation vector, thereby quantifying the gap between the current filtration time and the task requirements. Next, using the first-scale resistance adjustment cycle as the starting point for optimization, gradient compensation adjustments are performed according to the magnitude of the first deviation percentage and the direction of the first deviation vector. For example, if the first total filtration time exceeds the limit by 15%, adjustments are made along the deviation vector direction of shortening the cycle, outputting the second-scale resistance adjustment cycle. Subsequently, the backwashing benefit evaluation process of steps S720-S730 is repeated for the second-scale resistance adjustment cycle to obtain the second backwashing control parameters, the second backwashing total energy consumption, and the second total filtration time. If the second filtration... If the total duration is within the filtration time limit, the set of parameters is associated and stored in the alternative adjustment strategy library. This logic is repeated for multiple iterations, with each iteration adjusting the resistance adjustment cycle based on the deviation scale of the previous round and evaluating the benefits, until the alternative adjustment strategy library accumulates N backwash control strategies that meet the time constraints. Then, a two-dimensional performance quantification evaluation is performed on the N strategies. Based on predefined two-dimensional dynamic weights, the importance of energy consumption and duration is comprehensively considered, and the total backwash energy consumption and total filtration duration of each strategy are weighted and fused to obtain the corresponding resistance adjustment performance score. Finally, based on the descending order of the scores, the strategy with the highest score among the N strategies is selected and output as the final resistance adjustment strategy, ensuring that this strategy meets both the filtration time requirements and achieves optimal backwash energy consumption.

[0051] In one possible implementation, step S730 further includes:

[0052] Step S731: Based on the first local resistance time-varying curve, reverse the filter flow rate and output the first cumulative filter flow rate.

[0053] Step S732: retrieve the first end resistance of the time-varying curve of the first local resistance, and perform backwash parameter adjustment simulation based on the first end resistance to output the first backwash control parameter.

[0054] Step S733: Retrieve the first backwash control duration from the first backwash control parameters, compensate for the first scale resistance adjustment cycle, and output the first filter resistance adjustment cycle.

[0055] Step S734: Predict the backwash energy consumption of the first backwash control parameter and output the first backwash energy consumption.

[0056] Step S735: Calculate and output the first backwash frequency based on the total filtration volume and the first cumulative filtration flow rate.

[0057] Step S736: Calculate and output the total energy consumption of the first backwash based on the first backwash energy consumption and the first backwash frequency.

[0058] Step S737: Calculate and output the total filtration time based on the first filtration resistance adjustment cycle and the first backwashing frequency.

[0059] Specifically, the first local resistance time-varying curve obtained from the segmentation is acquired. Key parameters previously calculated and stored, including the output fluid dynamic viscosity and the effective filtration area of ​​the metal mesh filter obtained interactively, are then input along with the first local resistance time-varying curve data into a preset flow inverse model. This model is based on the resistance-flow correlation principle in fluid mechanics, such as the Darcy's law derivative formula. It uses the resistance value at each time point in the curve to inversely calculate the instantaneous filtration flow rate at the corresponding moment. Using a modified formula of resistance = (viscosity × flow rate) / (effective filtration area × filter layer permeability), combined with known resistance, viscosity, and effective filtration area, the instantaneous flow rate is calculated. Finally, all instantaneous flow rates within the entire cycle are integrated along the time dimension, and the total filtration volume within the entire first-scale resistance adjustment cycle is accumulated and determined as the first cumulative filtration flow rate, which is then output.

[0060] First, the terminal resistance value is extracted from the time-varying curve of the first local resistance as the core feature, and a feature set is constructed by combining fluid viscosity, filter tube area, etc. Then, a random forest regression model is trained using historical data, including the optimal backwash pressure, flow rate, and duration corresponding to different terminal resistances, so that the model learns the mapping relationship between features and parameters. Finally, the current terminal resistance is input, and the model outputs multiple sets of candidate parameters. By calculating the comprehensive score of the clearance rate and energy consumption corresponding to the parameters, the optimal parameter is selected as the first backwash control parameter output.

[0061] From the output of the first backwash control parameters, key data related to the backwash operation duration are extracted and determined as the first backwash control duration. This duration, based on the simulation of the first end resistance, is the backwash operation time required to ensure effective removal of blockages, including the entire process time from backwash preparation, pressure build-up, continuous flushing, and pressure relief. Subsequently, the first-scale resistance adjustment cycle is retrieved. This cycle initially only covers the filtration stage of the metal mesh filter tube and does not include the time occupied by the backwash operation. To accurately reflect the complete time of a single filtration-backwash cycle, the first backwash control duration and the first-scale resistance adjustment cycle are time-superimposed and compensated. That is, by calculating the first-scale resistance adjustment cycle + the first backwash control duration, the complete cycle duration covering both the filtration and backwash stages is obtained. Finally, this duration is determined as the first filtration resistance adjustment cycle and output.

[0062] First, the output backwash control parameters are collected, and key parameters such as backwash pressure, flow rate, and control duration are extracted as input features. Simultaneously, auxiliary features such as the rated power of the backwash equipment and fluid density are imported to construct the input feature vector of the neural network. Then, this feature vector is input to a pre-trained backwash energy consumption prediction neural network model. This model performs nonlinear transformations on the input features through multiple hidden layers, combining historical training data covering samples of different backwash parameter combinations and corresponding actual energy consumption to learn the mapping relationship between parameters and energy consumption. In the historical samples, the energy consumption label is calculated by integrating the power monitoring data and duration of the actual backwash process, ensuring the accuracy of model training. The model output layer outputs the predicted energy consumption value for a single backwash through a linear activation function. This predicted value is then corrected for operating conditions, such as voltage fluctuations and equipment aging coefficients, to finally determine and output the first backwash energy consumption.

[0063] The total filtration volume obtained from the decomposition and the first cumulative filtration flow rate are used to calculate the number of backwashes required to complete all filtration tasks, i.e., the first backwash frequency, by dividing the total filtration volume by the first cumulative filtration flow rate, thus clarifying the execution frequency of the backwash operation.

[0064] Multiply the obtained first backwash energy consumption by the obtained first backwash frequency (total number of times) to calculate the total backwash energy consumption required to complete all filtration tasks, i.e., the first backwash total energy consumption.

[0065] Multiply the obtained first filtration resistance adjustment cycle (single cycle duration of the entire process) by the obtained first backwash frequency (total number of times) to calculate the total time required to complete all filtration tasks, i.e., the first total filtration duration. This parameter will be used to compare with the filtration time limit to determine whether the current strategy meets the time constraint requirements.

[0066] In one possible implementation, step S740 further includes:

[0067] Step S741: Calculate the first deviation percentage and the first deviation vector of the first total filtering time from the filtering time limit.

[0068] Step S742: Based on the first deviation percentage, perform gradient compensation adjustment of the first deviation vector on the first scale resistance adjustment cycle, and output the second scale resistance adjustment cycle.

[0069] Step S743: Evaluate the backwashing benefit of the second-scale resistance adjustment cycle, and output the second backwashing control parameters, the total energy consumption of the second backwashing, and the total filtration time of the second filtration.

[0070] Step S744: When the second total filtration time falls within the filtration time limit, the second backwash control parameter, the second total backwash energy consumption, and the second total filtration time are associated and stored in the alternative adjustment strategy library.

[0071] Step S745: Perform multiple iterations in a similar manner until the alternative adjustment strategy library is updated to obtain N backwash control strategies.

[0072] Step S746: Perform a two-dimensional performance quantification evaluation on the N backwash control strategies to obtain N resistance regulation performance scores.

[0073] Step S747: Based on the descending order of the N resistance regulation performance scores, locate the resistance regulation strategy among the N backwash control strategies.

[0074] Specifically, the first total filtering time and the decomposed filtering time limit are obtained first, and the degree of deviation between the two is calculated by a preset algorithm: on the one hand, the first deviation percentage is calculated by the formula (first total filtering time - filtering time limit) / filtering time limit × 100%, which quantifies the relative degree of time deviation; on the other hand, the first deviation vector is determined by judging whether the first total filtering time is greater than or less than the filtering time limit. If the time exceeds the limit, the vector is a shortened period; if the time is insufficient, the vector is an extended period, which provides a directional basis for subsequent period adjustment.

[0075] Using the first-scale resistance adjustment cycle as the basis for adjustment, and combining the magnitude of the first deviation percentage and the direction of the first deviation vector, gradient compensation adjustment is performed: for example, if the first deviation percentage is +12%, that is, exceeding the time limit by 12%, and the deviation vector is "shortening the cycle", then the first-scale resistance adjustment cycle is gradient shortened by a deviation ratio of 12%. The adjusted second-scale resistance adjustment cycle is obtained and output by calculating the first-scale resistance adjustment cycle × (1 - first deviation percentage), ensuring that the cycle adjustment magnitude matches the time deviation.

[0076] Repeat steps S720-S737 of the backwash benefit evaluation process, using the second-scale resistance adjustment cycle as the new cycle parameter, and segment the second local resistance time-varying curve on the resistance trend time-varying curve. Then, perform operations such as filter flow back-calculation, backwash parameter adjustment simulation, cycle compensation, energy consumption prediction, and frequency calculation in sequence. Finally, output the second backwash control parameters, the second backwash total energy consumption, and the second filtration total duration corresponding to the second-scale resistance adjustment cycle to form a new backwash strategy scheme.

[0077] The second total filtration time is compared with the filtration time limit. If the second total filtration time is within a reasonable range of the filtration time limit, such as filtration time limit × (1-ε) ≤ second total filtration time ≤ filtration time limit × (1+ε), where ε is a preset error threshold, then the strategy is determined to meet the time constraint. The second backwash control parameters, the second backwash total energy consumption, and the second total filtration time are associated and stored in the alternative adjustment strategy library to retain effective solutions for subsequent multi-strategy screening.

[0078] Following the logical analogy of steps S741-S744, multiple iterations are performed: each iteration is based on the current scale resistance adjustment cycle, and according to the corresponding total filtering time and the deviation scale adjustment cycle of the time limit, new strategy parameters are obtained through benefit evaluation. If the new strategy meets the time constraint, it is stored in the candidate library until N options are accumulated in the candidate adjustment strategy library, where N is a preset quantity threshold, such as 10 or 20 backwash control strategies that meet the requirements, to ensure that there is a sufficient sample size of strategies for subsequent screening.

[0079] The system calls predefined two-dimensional dynamic weight parameters, which are set according to the priority requirements of energy consumption and time for the actual filtering task. For example, when energy consumption control is the core objective, the energy consumption weight can be set to 0.6 and the time weight to 0.4. When the time constraint is more stringent, the time weight can be adjusted to 0.6 and the energy consumption weight to 0.4 to ensure that the evaluation dimensions match the actual needs. Subsequently, the system iterates through the N backwash control strategies stored in the alternative adjustment strategy library, extracts the corresponding total backwash energy consumption and total filtering time for each strategy, and standardizes these two parameters, mapping the energy consumption value and the time value to the interval [0, 1] to eliminate the influence of different dimensions on the evaluation results. For example, the energy consumption standardization formula is: Standardized Energy Consumption = (Current Strategy Energy Consumption - Minimum Energy Consumption) / (Maximum Energy Consumption - Minimum Energy Consumption). Next, according to the formula: Resistance regulation performance score = (1 - Standardized energy consumption) × Energy consumption weight + (1 - Standardized duration) × Duration weight, a comprehensive score is calculated for each strategy. Here, 1 - Standardized energy consumption and 1 - Standardized duration represent the degree of optimization of energy consumption and duration, respectively. The closer the value is to 1, the better the optimization effect. After weighted fusion, the comprehensive performance score of each strategy is obtained. Finally, the score calculation for N backwash control strategies is completed, and the corresponding N resistance regulation performance scores are output.

[0080] The N resistance regulation performance scores are sorted in descending order, and the strategy with the highest score is selected as the optimal solution. The backwash control parameters, total backwash energy consumption, and total filtration time corresponding to this strategy are located from the alternative regulation strategy library. This strategy is determined as the final resistance regulation strategy and output. This strategy can meet the filtration time limit and achieve the optimal balance between energy consumption and time. It can be directly used to guide the dynamic filtration resistance regulation operation of metal mesh filter tubes.

[0081] In one possible implementation, step S300 further includes:

[0082] Step S310: After performing a third-order low-pass filter on the time-series temperature signal, perform a sliding window averaging based on a preset window width to output the steady-state temperature value.

[0083] Step S320: Use the code of the fluid to be filtered to search the physical property database to obtain the fluid type, viscosity-temperature coefficient, critical temperature boundary and the fluid reference viscosity.

[0084] Step S330: If the steady-state temperature value is at the critical temperature boundary, then the temperature compensation function is called according to the fluid type.

[0085] Step S340: By loading the steady-state temperature value and viscosity-temperature coefficient into the prime number temperature compensation function, the dynamic viscosity is solved, and the fluid dynamic viscosity is output.

[0086] Specifically, the system receives a time-series temperature signal from a temperature sensor pre-installed on the main filter pipeline. This signal is subject to high-frequency interference from factors such as equipment vibration and ambient temperature fluctuations during filter system operation, resulting in instantaneous temperature data fluctuations. To eliminate the impact of these interferences on subsequent viscosity compensation calculations, a third-order low-pass filter is first applied to the time-series temperature signal. This filter removes high-frequency noise components while retaining the core trend of temperature change over time. Subsequently, based on a preset window width (which can be set according to the stability requirements of the filtration operation, such as a fixed duration of 5 seconds or 10 seconds), a sliding window averaging operation is performed on the filtered time-series temperature signal. This involves sequentially capturing temperature data and calculating the average value within a preset window starting from the signal's initial moment. By sliding the window point by point, the entire time-series temperature signal is smoothed, ultimately outputting a steady-state temperature value that accurately reflects the actual temperature state within the main filter pipeline with minimal fluctuations.

[0087] The process begins by obtaining a pre-determined code for the fluid to be filtered. This code is a unique identifier for the fluid's physical properties, accurately linking it to its fundamental physicochemical characteristics. This code is then used as a search keyword and input into a pre-built and maintained physical property database. This database stores standardized physical property parameters for various common and special fluids, covering key information such as fluid type, viscosity-temperature coefficient, critical temperature boundary, and reference viscosity. All data has been experimentally verified and calibrated to ensure accuracy and reliability. During the search, the database quickly locates the corresponding fluid's set of physical property parameters through code matching, extracting specific information about the fluid to be filtered: fluid type (e.g., mineral oil, water-glycol solution, compressed air), which determines the basis for selecting the subsequent compensation function; viscosity-temperature coefficient, reflecting the sensitivity of fluid viscosity to temperature changes and serving as a core coefficient for temperature compensation calculations; critical temperature boundary, the temperature range within which fluid viscosity maintains a stable variation, used to determine the effectiveness of subsequent temperature compensation; and the fluid reference viscosity, the viscosity value of the fluid at a standard reference temperature, which serves as the calculation benchmark for temperature compensation.

[0088] The process acquires the output steady-state temperature value, along with two key parameters: the critical temperature boundary and the fluid type, retrieved from the physical property database. The steady-state temperature value is then compared with the critical temperature boundary to determine if it falls within the effective range defined by that boundary. The critical temperature boundary is the temperature range within which the fluid viscosity maintains a stable variation. If the steady-state temperature is within this range, it indicates that the fluid's viscosity characteristics at the current temperature conform to conventional physical laws, fulfilling the prerequisite for temperature compensation calculations. If the steady-state temperature is determined to be within the critical temperature boundary, the process further matches and calls the corresponding temperature compensation function from a pre-defined temperature compensation function library based on the fluid type, such as oil, water-based solution, or gaseous fluid. Different fluid types have different mathematical models for viscosity changes with temperature due to differences in molecular structure and other characteristics. For example, oil fluids are often suited to exponential temperature compensation functions, while water-based solutions are suited to linear or polynomial temperature compensation functions. Precise matching of the fluid type and compensation function ensures that subsequent temperature compensation calculations accurately reflect the current fluid viscosity variation.

[0089] The key parameters obtained in the previous steps, namely the output steady-state temperature value, reflect the actual stable temperature of the fluid under the current filtration conditions and the retrieved viscosity-temperature coefficient. These two parameters are then used as input variables and loaded into a matched prime number temperature compensation function. The prime numbers in this function are specific correction coefficients calibrated based on extensive fluid property experimental data, used to improve the accuracy of viscosity calculations within different temperature ranges. The function form must be compatible with the fluid type; for example, oil fluids correspond to an exponential prime number compensation function, and water-based solutions correspond to a polynomial prime number compensation function. During the function calculation, the steady-state temperature value and viscosity-temperature coefficient are calculated collaboratively, and the fluid's baseline viscosity is compensated for temperature using the prime number correction coefficient to solve for the fluid's true viscosity at the current actual temperature, i.e., the fluid's dynamic viscosity. Finally, the calculation results are validated to determine whether the viscosity value is within a reasonable range under normal physical conditions of the fluid. Once confirmed to be correct, the result is output. This fluid dynamic viscosity accurately reflects the actual physical characteristics of the fluid during the filtration process.

[0090] In one possible implementation, step S330 further includes:

[0091] If the steady-state temperature value deviates from the critical temperature boundary, the viscosity compensation calculation is paused and a phase change risk warning sign is output.

[0092] Specifically, after comparing the steady-state temperature value with the critical temperature boundary, if it is determined that the steady-state temperature value deviates from the critical temperature boundary, that is, the current actual temperature of the fluid exceeds the temperature range in which its viscosity maintains a stable change pattern, an abnormal handling mechanism will be immediately triggered: First, the currently executing fluid viscosity temperature compensation calculation process will be suspended to avoid distortion of viscosity calculation results due to the temperature exceeding the reasonable range, which would affect the accuracy of subsequent filtration resistance calculations and backwashing strategy formulation; then, a phase change risk warning sign will be generated and output. This sign will clearly indicate that the current fluid temperature has deviated from the critical boundary and there is a possibility of phase change. Phase change will cause drastic changes in the physical properties of the fluid, which will not only cause the viscosity compensation model to fail, but may also affect the normal operation of the metal mesh filter tube filtration system, such as clogging the filter tube and damaging the equipment. The warning sign can promptly remind operators to intervene and ensure the safe and stable operation of the filtration system.

[0093] In one possible implementation, step S500 further includes:

[0094] Step S510: Use interpolation compensation to correct the response delay of the time-series differential pressure signal and the time-series flow signal, and output the time-series corrected differential pressure and the time-series corrected flow.

[0095] Step S520: Calculate the time-series surface velocity based on the effective filtration area and the time-series corrected flow rate.

[0096] Step S530: Calculate the timing differential pressure value based on the timing-corrected differential pressure.

[0097] Step S540: Use the time-series differential pressure value, time-series surface velocity and fluid dynamic viscosity to solve for the resistance and output the time-series filtration resistance.

[0098] Specifically, the system first receives the time-series differential pressure signal from the differential pressure transmitter in the filter main pipeline and the time-series flow signal from the flow meter. Due to differences in the installation location and signal transmission path between the differential pressure transmitter and the flow meter, there will be a response delay between the two types of signals, resulting in a mismatch between the differential pressure and flow data at the same time point. To eliminate this error, an interpolation compensation algorithm is used to process the signals: for the time-series differential pressure signal and the time-series flow signal, based on a preset time reference, time alignment correction is performed on the delayed signal data through linear interpolation to fill the time difference in signal transmission. Finally, the system outputs the time-corrected differential pressure (i.e., the time-aligned differential pressure data) and the time-corrected flow (i.e., the time-aligned flow data), ensuring that the differential pressure and flow data used in subsequent calculations remain synchronized in the time dimension.

[0099] The effective filtration area of ​​the metal mesh filter tube obtained from previous interactions is retrieved. Combined with the output time-corrected flow rate, the time-plane velocity corresponding to each time node is obtained by using the calculation formula: time-plane velocity = time-corrected flow rate ÷ effective filtration area.

[0100] The time-corrected differential pressure output is obtained. This data is essentially an electrical signal converted from the pressure difference between the inlet and outlet of the metal mesh filter tube by the differential pressure transmitter, such as a current signal of 4~20mA or a voltage signal of 0~5V, and is not a directly usable physical pressure difference. To obtain pressure parameters with actual physical meaning, a pre-built signal calculation model is invoked. This model is established based on the range, accuracy calibration curve, and signal conversion formula of the differential pressure transmitter. For example, if the transmitter range is 0~100kPa, the corresponding output signal is 4~20mA. The electrical signal value of the time-corrected differential pressure is converted into a physical pressure difference in Pa or kPa, i.e., the time-corrected differential pressure value, using the linear calculation formula: Time-corrected differential pressure value = (current signal value - minimum signal value) ÷ (maximum signal value - minimum signal value) × (maximum range - minimum range) + minimum range. After the calculation is completed, the results are validated to ensure that the timing differential pressure value is within a reasonable physical range, such as not less than 0 and not exceeding the maximum range of the transmitter. Finally, the timing differential pressure value that meets the requirements is output.

[0101] The obtained time-series surface velocity, time-series differential pressure, and fluid dynamic viscosity calculated using temperature compensation are integrated. These three parameters are then substituted into a pre-defined filtration resistance solution model. Based on the resistance calculation formula in fluid mechanics, a modified formula for filtration resistance (time-series differential pressure ÷ (time-series surface velocity × fluid dynamic viscosity)) is derived, and the resistance calculation is performed. By calculating the parameters at each time point, continuous time-series filtration resistance data is obtained and finally output.

[0102] In one possible implementation, step S746 further includes:

[0103] Based on predefined two-dimensional dynamic weights.

[0104] The weighted fusion of the total energy consumption of the second backwash and the total filtration time is performed using the two-dimensional dynamic weights to output the second resistance regulation performance score of the second backwash control strategy.

[0105] Specifically, the first step is to load predefined two-dimensional dynamic weights. These weights will be adjusted according to the actual needs of the filtering task. For example, when the priority of energy consumption control is higher than the time requirement, the energy consumption weight can be set to 0.6 and the time weight to 0.4. When strict adherence to time limits is required, the time weight can be adjusted to 0.6 and the energy consumption weight to 0.4 to ensure that the evaluation dimensions match the task objectives.

[0106] Subsequently, the core performance indicators corresponding to the second backwash control strategy—total energy consumption of the second backwash and total filtration time—were extracted. These two indicators were then standardized to eliminate the influence of different dimensions on the scoring results, mapping both energy consumption and time values ​​to the [0, 1] interval, with values ​​closer to 0 indicating better performance. Next, a two-dimensional dynamic weighting method was used to perform a weighted fusion calculation on the standardized total energy consumption of the second backwash and the total filtration time. Specifically, the formula was: Second Resistance Regulation Performance Score = (1 - Standardized Energy Consumption) × Energy Consumption Weight + (1 - Standardized Time) × Time Weight, where 1 - Standardized Energy Consumption and 1 - Standardized Time represent the degree of optimization of energy consumption and time, respectively. Finally, the second resistance regulation performance score of the second backwash control strategy was obtained through this weighted fusion calculation.

[0107] In one possible implementation, step S600 further includes:

[0108] Step S670: If the resistance accumulation time window exceeds the maximum filtering time span, then the resistance trend time-varying curve is subjected to an over-limit region local stage based on the maximum filtering time span, and a corrected resistance trend curve is output.

[0109] Step S680: Use the maximum filtration time span to identify the time-varying curve of the resistance trend.

[0110] Specifically, the process involves obtaining the resistance accumulation time window and the maximum filtering time span. These two time parameters are then compared. If the resistance accumulation time window exceeds the maximum filtering time span, it indicates that the currently predicted resistance trend duration exceeds the task's allowed time limit, requiring correction of the resistance trend time-varying curve. At this point, using the maximum filtering time span as the boundary, localized processing is performed on the excess region of the resistance trend time-varying curve that exceeds this time span: the effective portion of the curve from the starting timestamp to the timestamp corresponding to the maximum filtering time span is extracted, curve data within the excess time period is removed, and cubic spline interpolation is used to transition the ends of the extracted curve to avoid abrupt data changes. Finally, the corrected resistance trend time-varying curve is output, ensuring that the curve duration matches the maximum allowed time range of the task.

[0111] After the curve correction is completed, the current resistance trend time-varying curve is identified using the maximum filtering time span. The maximum filtering time span value, such as 24 hours or 48 hours, is associated with the metadata of the resistance trend time-varying curve, forming a linked record containing "Curve ID - Maximum Filtering Time Span - Curve Data". This identification operation clearly distinguishes the resistance trend curves under different filtering task time constraints, facilitating rapid retrieval of curve data with corresponding time constraints during subsequent backwashing strategy optimization. It also provides a clear time dimension identifier for curve storage, management, and traceability, ensuring the accuracy and traceability of data throughout the entire filtering resistance monitoring and adjustment process.

[0112] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0113] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0114] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube, characterized in that, The method includes: Based on the code of the fluid to be filtered, perform a property retrieval and output the fluid reference viscosity; After the fluid filtration is started based on predefined operating conditions, the system receives time-series flow and temperature signals from the pre-installed flow meters and temperature sensors on the main filtration pipeline. Temperature compensation is performed on the fluid reference viscosity based on the time-series temperature signal to output the fluid dynamic viscosity. The effective filtration area and maximum allowable filtration resistance of the metal mesh filter tube are obtained interactively. After receiving the time-series differential pressure signal returned by the differential pressure transmitter, the system performs interactive calculations based on the effective filtration area, the time-series differential pressure signal, the time-series flow signal, and the fluid dynamic viscosity, and outputs the time-series filtration resistance. The metal mesh filter tube is deployed in the main filtration pipeline, and the differential pressure transmitter is connected between the inlet and outlet of the metal mesh filter tube. Using the maximum allowable filter resistance as the prediction termination boundary, the time-varying trend of the resistance is fitted based on the time-series filter resistance, and the time-varying curve of the resistance trend is output. With a preset filtration volume time limit as a constraint, and based on the time-varying curve of the resistance trend, the backwashing execution cost is optimized, and a resistance adjustment strategy is output. Using a preset filtration volume time limit as a constraint, and based on the time-varying curve of the resistance trend, the method optimizes the backwashing execution cost and outputs a resistance adjustment strategy. The method includes: Decompose the preset filtration volume time limit to obtain the total filtration volume and filtration time limit; Using a predefined first-scale resistance adjustment period, the first local resistance time-varying curve is obtained by segmenting the resistance trend time-varying curve; The backwashing benefit is evaluated based on the time-varying curve of the first local resistance and the total filtration volume, and the total energy consumption of the first backwashing and the total filtration time of the first filtration are output. Based on the deviation scale between the first total filtration time and the filtration time limit, and taking the first scale resistance adjustment cycle as the optimization starting point, perform iterative optimization of filtration resistance adjustment until the resistance adjustment strategy is output. Based on the deviation scale between the first total filtration time and the filtration time limit, and taking the first scale resistance adjustment cycle as the starting point, iterative optimization of filtration resistance adjustment is performed until the resistance adjustment strategy is output. The method includes: Calculate the first deviation percentage and the first deviation vector between the first total filtering time and the filtering time limit; Based on the first deviation percentage, the gradient compensation adjustment of the first deviation vector is performed on the first scale resistance adjustment cycle, and the second scale resistance adjustment cycle is output. The backwashing benefit is evaluated for the second-scale resistance adjustment cycle, and the second backwashing control parameters, the total energy consumption of the second backwashing, and the total filtration time of the second filtration are output. When the second total filtration time falls within the filtration time limit, the second backwash control parameters, the second total backwash energy consumption, and the second total filtration time are associated and stored in the alternative adjustment strategy library. This process is repeated multiple times until the alternative adjustment strategy library is updated to obtain N backwash control strategies. A two-dimensional performance quantification evaluation was performed on the N backwashing control strategies to obtain N resistance regulation performance scores; Based on the descending order of the N resistance regulation performance scores, the resistance regulation strategy is located among the N backwash control strategies.

2. The method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube as described in claim 1, characterized in that, Using the maximum allowable filtering resistance as the prediction termination boundary, the method involves fitting the time-varying trend of the resistance based on the time-series filtering resistance, and outputting a time-varying curve of the resistance trend. Set the dynamic prediction step time and the maximum filtering time span, wherein the maximum filtering time span is retrieved from the filtering volume time limit; Retrieve the end resistance value of the time-series filtering resistance, and starting from the end resistance value, perform single-step resistance prediction based on the dynamic prediction step time, and output the time-series predicted resistance. When the updated predicted resistance point of the time-series predicted resistance meets the maximum allowable filtering resistance, the prediction boundary timestamp is recorded; After connecting the time-series filtering resistance and the time-series predicted resistance, a smooth trend curve is fitted by performing cubic spline interpolation, and the time-varying trend curve of the resistance is output. Retrieve the start timestamp of the time-series filtering resistance, and calculate the resistance accumulation time window based on the start timestamp and the predicted boundary timestamp; The resistance trend time-varying curve is identified using the resistance accumulation time window.

3. The method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube as described in claim 1, characterized in that, The backwashing benefit is evaluated based on the time-varying curve of the first local resistance and the total filtration volume, and the total energy consumption of the first backwashing and the total filtration time of the first filtration are output. The method includes: Based on the time-varying curve of the first local resistance, the filter flow rate is calculated in reverse to output the first cumulative filter flow rate. The first end resistance of the time-varying curve of the first local resistance is retrieved, and backwash parameter adjustment simulation is performed based on the first end resistance to output the first backwash control parameter. The first backwash control duration is retrieved from the first backwash control parameter, the first scale resistance adjustment cycle is compensated, and the first filter resistance adjustment cycle is output. Predict the backwash energy consumption based on the first backwash control parameters and output the first backwash energy consumption. The first backwash frequency is calculated and output based on the total filtration volume and the first cumulative filtration flow rate; The total energy consumption of the first backwash is calculated and output based on the first backwash energy consumption and the first backwash frequency; The total filtration time is calculated and output based on the first filtration resistance adjustment cycle and the first backwash frequency.

4. The method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube as described in claim 1, characterized in that, The method includes performing temperature compensation on the fluid reference viscosity based on the time-series temperature signal and outputting the fluid dynamic viscosity. After performing a third-order low-pass filter on the time-series temperature signal, a sliding window average is performed based on a preset window width to output the steady-state temperature value. The fluid type, viscosity-temperature coefficient, critical temperature boundary, and reference viscosity of the fluid are obtained by searching the physical property database using the code of the fluid to be filtered. If the steady-state temperature value is at the critical temperature boundary, then the temperature compensation function is called according to the fluid type. The dynamic viscosity of the fluid is output by loading the steady-state temperature value and viscosity-temperature coefficient into the prime number temperature compensation function to perform dynamic viscosity calculation.

5. The method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube as described in claim 4, characterized in that, If the steady-state temperature value deviates from the critical temperature boundary, the viscosity compensation calculation is paused and a phase change risk warning sign is output.

6. The method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube as described in claim 1, characterized in that, Based on the effective filtration area, time-series differential pressure signal, time-series flow rate signal, and fluid dynamic viscosity, an interactive calculation is performed to output the time-series filtration resistance. The method includes: Interpolation compensation is used to correct the response delay of the time-series differential pressure signal and the time-series flow signal, and the time-series corrected differential pressure and time-series corrected flow are output. Calculate the time-series surface velocity based on the effective filtration area and the time-corrected flow rate; The time-series differential pressure value is calculated based on the aforementioned time-series corrected differential pressure. The resistance is calculated using the time-series differential pressure, time-series surface velocity, and fluid dynamic viscosity, and the time-series filtration resistance is output.

7. The method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube as described in claim 1, characterized in that, A two-dimensional performance quantification evaluation is performed on the N backwash control strategies to obtain N resistance regulation performance scores. The method includes: Based on predefined two-dimensional dynamic weights; The weighted fusion of the total energy consumption of the second backwash and the total filtration time is performed using the two-dimensional dynamic weights to output the second resistance regulation performance score of the second backwash control strategy.

8. The method for monitoring and adjusting the dynamic filtration resistance of a metal mesh filter tube as described in claim 2, characterized in that, Also includes: If the resistance accumulation time window exceeds the maximum filtering time span, then the resistance trend time-varying curve is subjected to an over-limit region local stage based on the maximum filtering time span, and a corrected resistance trend curve is output. The time-varying curve of the resistance trend is identified using the maximum filtration time span.

Images