Apparatus and Method for Redundant Precoat Filter Regeneration Using Adaptive Resonance Tracking Control

The implementation of FFT-based resonance analysis and dual control architecture with software redundancy addresses the inefficiencies and reliability issues of conventional bumping methods, enhancing energy efficiency and operational continuity in precoat filtration systems.

KR103046424B1Active Publication Date: 2026-11-11유한회사한국기계설비기술
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020260037591
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2026-03-01
Publication Date
2026-11-11
Estimated Expiration
2046-03-01

AI Technical Summary

Technical Problem

Conventional mechanical bumping methods for precoat filtration systems suffer from fixed frequency operation, lack of resonance frequency detection, inefficient energy use, and single-point failure vulnerabilities, leading to suboptimal performance and reliability issues.

Method used

Implementing FFT-based resonance analysis and adaptive resonance tracking to tune bumping frequency to the natural frequency of the precoat layer, coupled with dual control architecture and software redundancy for continuous operation, ensuring efficient detachment and reliability.

Benefits of technology

Reduces energy consumption by 30-35%, improves delamination rate to 90-95%, and ensures uninterrupted operation through adaptive frequency tracking and redundancy mechanisms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112026025060027-PAT00001_ABST
    Figure 112026025060027-PAT00001_ABST
Patent Text Reader

Abstract

The present invention relates to an adaptive resonance tracking control regeneration device and method for a precoat filter. The natural frequency of the precoat layer is detected by FFT analysis of an acceleration sensor mounted on a tube sheet, and after tuning the bumping frequency, a quasi-resonant state is maintained by tracking the shift in natural frequency caused by the reduction of precoat mass during bumping in real time and re-tuning. Automatic recovery in the event of a software failure is achieved through software redundancy (control process + watchdog process) within a field embedded PC, and remote fallback control is performed even in the event of a hardware failure by a software mirror of the central control system, thereby ensuring uninterrupted regeneration operation. Compared to conventional methods, this achieves an improvement in delamination rate of approximately 7%p, a reduction in energy consumption of approximately 35%, and a reduction in time of approximately 42%.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to an apparatus and method for regenerating filtration performance by removing a precoat layer formed on the surface of a filtration element in a precoat filter through mechanical bumping. Specifically, the present invention relates to a technology that significantly improves bumping energy efficiency and removal performance with minimal equipment investment by maintaining existing single-axis vertical bumping hardware while tuning the bumping frequency to the natural frequency of the precoat layer in real time through FFT-based resonance analysis and adaptive resonance tracking control. Furthermore, the present invention relates to a technology that ensures the reliability of uninterrupted regeneration operation by continuing bumping without interruption even in the event of a failure on one side through software redundancy of a field embedded PC and remote fallback control by a central control system. Background Technology

[0002] Precoat filtration is a filtration method in which a filtration aid such as diatomaceous earth, perlite, or cellulose is formed as a precoat layer on the outer surface of a filtration element, and this precoat layer is utilized as an actual filtration medium to capture fine suspended solids in raw water. Since filtration performance deteriorates when the precoat layer becomes saturated with contaminants, a regeneration process is essential to periodically detach the precoat layer from the surface of the element and re-form a new precoat layer.

[0003] The most common conventional regeneration method is mechanical bumping, which utilizes hydraulic or pneumatic cylinders and operating rods to reciprocate the entire bundle of tube sheets and filtration elements vertically, and the principle is to detach the precoat filtration medium from the surface of the elements through the inertial force generated during this movement.

[0004] However, conventional mechanical bumping methods have the following fundamental control technical limitations.

[0005] First, conventional bumping operates with a preset fixed amplitude and fixed frequency. Although the physical properties of the precoat layer vary with each regeneration cycle depending on the type of precoat, coating thickness, degree of contamination, and consolidation state, fixed-parameter bumping fails to adapt to these variable conditions at all.

[0006] Secondly, a tube sheet-element system with a pre-coat layer possesses unique mechanical resonance characteristics. By matching the bumping frequency to this natural frequency, effective detachment is possible with less energy through resonance amplification. However, conventional technology lacks the ability to detect the natural frequency or to tune the bumping frequency to it.

[0007] Thirdly, conventional bumping systems lack sensor feedback mechanisms to objectively determine whether detachment is complete. Since the number of bumps and the duration are determined based on the driver's experience and visual observation, energy waste and mechanical wear of the elements due to unnecessarily long bumping are inevitable.

[0008] Fourth, since conventional bumping systems are operated by a single controller, there is a Single Point of Failure problem in which the bumping control function is completely lost in the event of a controller failure. Under the requirements for continuous operation of water treatment plants, this vulnerability poses a serious risk that reduces the plant's operating rate. Prior art literature

[0009] U.S. Patent No. 10,406,458 (Registered Sep. 10, 2019), "Regenerative Media Filtration"; U.S. Patent No. 3,422,958 (Registered Jan. 21, 1969), "Precoat Filter"; German Patent Published No. 30 26 769 (Published Feb. 4, 1982), "Precoat-Verfahren zur Reinigung von Filterelementen in Anschwemmfiltern"

[0010] Svarovsky, L., "Solid-Liquid Separation", 4th Edition, Butterworth-Heinemann, 2000Harris, CM & Piersol, AG, "Harris' Shock and Vibration Handbook", 6th Edition, McGraw-Hill, 2010Cooley, JW & Tukey, JW, "An Algorithm for the Machine Calculation of Complex Fourier Series", Mathematics of Computation, Vol. 19, no. 90, 1965 The problem to be solved

[0011] The present invention was devised to overcome the limitations of the aforementioned conventional mechanical bumping technology and aims to solve the following technical problems.

[0012] The first technical task is to detect the natural frequency of a tube sheet-element system with a pre-coat layer attached by FFT analysis of the acceleration sensor output, and to realize high-efficiency bumping by utilizing the resonance amplification phenomenon by tuning the bumping frequency to the said natural frequency.

[0013] The second technical challenge is to realize adaptive resonance tracking, which re-detects the natural frequency at every predetermined time interval as the bumping progresses due to the reduction in precoat mass in real time, and re-tunes the bumping frequency to maintain a quasi-resonant state throughout the entire bumping process.

[0014] The third technical task is to objectively and automatically determine the completion of detachment by cross-verifying the bumping efficiency indicators calculated from the outputs of the differential pressure sensor and the acceleration sensor.

[0015] The fourth technical challenge is to realize a practical configuration in which all the aforementioned control functions are executed within the industrial site controller, enabling retrofitting to existing facilities at minimal cost.

[0016] The fifth technical challenge is to realize uninterrupted regenerative operation, in which the other side takes over control functions and continues bumping without interruption even in the event of a failure on one side, through software redundancy within the field embedded PC and remote fallback control by a central control system. means of solving the problem

[0017] To solve the above problem, according to a first aspect of the present invention, a precoat filter regeneration device is provided. The precoat filter regeneration device of the present invention comprises a filtration vessel, a vertical driving unit, a sensor unit, and a control unit.

[0018] The filtration vessel is internally divided into an upper chamber and a lower chamber by a tube sheet, a plurality of filtration elements are vertically suspended and mounted on the lower surface of the tube sheet, and a precoat layer made of a filtration aid is formed on the outer surface of each filtration element.

[0019] The vertical drive unit includes a hydraulic or pneumatic cylinder and an operating rod, and its amplitude and frequency are variably controlled by a control signal from the control unit to reciprocate the tube sheet in the vertical direction.

[0020] The sensor unit includes an acceleration sensor mounted on the tube sheet, a differential pressure sensor for measuring the differential pressure between the inlet and outlet sides of the filter container, and a position sensor for measuring the position of the operating rod.

[0021] The control unit may be composed of an industrial controller selected from a PLC (Programmable Logic Controller), an embedded PC, or an embedded controller. Within the control unit, an FFT resonance analysis module, an adaptive resonance tracking module, and a detachment determination module are implemented in software.

[0022] The FFT resonance analysis module calculates the natural frequency of the tube sheet-element system with the pre-coat layer attached by real-time FFT analysis of the acceleration sensor output. Separation of the bumping excitation frequency components is performed by a frequency masking method or a notch filter method.

[0023] The adaptive resonance tracking module re-performs the FFT using a sliding window method at intervals of 2 to 5 seconds during bumping to recalculate the current natural frequency, and re-tunes the bumping frequency to a range of 0.9 to 1.1 times the re-detected natural frequency so that a quasi-resonant state (a state in which an amplification ratio of 1 / √2, or about -3dB or more relative to the maximum value of the resonant frequency response is maintained) is maintained throughout the entire bumping process.

[0024] The detachment determination module determines the completion of detachment using a cross-verification method based on dual criteria. The first criterion is the convergence of the differential pressure change rate, and the second criterion is the saturation of the Bumping Efficiency Index (BEI). The BEI is calculated using the formula BEI = (a_rms)² / E_input.

[0025] According to a second aspect of the present invention, the control unit may be configured with a hierarchical dual control architecture of a field embedded PC and a central control system.

[0026] The field embedded PC (hereinafter referred to as the "local controller") is installed near the filtration vessel and independently performs all real-time control operations, such as sensor data acquisition, FFT resonance analysis, adaptive resonance tracking, valve timing generation, and detachment determination. The local controller operates on a Real-Time Operating System (RTOS) or an operating system with a Linux RT patch applied, and the control loop period is 5ms to 20ms.

[0027] Software redundancy is implemented within the local controller. The control process performs core control logic such as FFT analysis, resonance tracking, valve timing generation, and clearance determination. The watchdog process runs independently of the control process and monitors the normal operation of the control process. The watchdog process sends a survival check request to the control process at intervals of 50 ms (preferably 20 ms to 100 ms) and declares a software failure of the control process if there is a failure to respond three consecutive times (preferably 2 to 5 times). Upon declaration of failure, the watchdog process automatically restarts the control process and performs safe mode bumping based on the last optimal bumping frequency stored in non-volatile memory during the transition period (approximately 1 to 2 seconds) until the restart is complete.

[0028] The central control system (SCADA, DCS, or upper-level control server) is located in the plant's central control room and is connected to local controllers via an industrial network (Ethernet / IP, PROFINET, or Modbus TCP). Under normal conditions, the central control system performs a monitoring role, receiving sensor data, control status, and FFT analysis results from local controllers, and displaying and logging them. In the central control system, a software mirror of the same control algorithm as the local controller is executed to perform shadow computation using the received sensor data as input, and compares and verifies the results with those of the local controller.

[0029] If the local controller becomes completely incapacitated due to a hardware failure (power interruption, processor failure, etc.), the central control system detects the failure via a network timeout (preferably 1 to 3 seconds) and initiates remote fallback control. The central control system transmits remote control commands to the solenoid valve via an industrial network to take over bumping. Since the software mirror running in the central control system performs FFT analysis and adaptive tracking, the adaptive resonance tracking function is maintained. However, valve timing precision may be slightly reduced compared to local control due to network delay (several milliseconds to tens of milliseconds); in this case, the bumping amplitude is limited to a safe range. When the recovery of the local controller is detected, it automatically returns to normal mode.

[0030] In addition, the central control system performs integrated management functions for multiple filtration vessels. By integrating and analyzing sensor data and control history collected from multiple local controllers, it performs optimization of bumping schedules between vessels, self-learning parameter optimization based on accumulated data from multiple cycles, and predictive maintenance through element fatigue estimation. Effects of the invention

[0031] According to the present invention, by adaptive resonance tracking, bumping energy consumption is reduced by about 30 to 35% compared to conventional fixed frequency bumping, the delamination rate is improved from about 85% to about 90 to 93%, and the regeneration time is shortened by about 35 to 42%.

[0032] In addition, an objective and reproducible determination of detachment completion is realized through dual automatic judgment based on cross-verification of differential pressure and bumping efficiency indicators.

[0033] Furthermore, automatic recovery in the event of a software failure is performed through software redundancy (control process + watchdog process) within the field embedded PC, and bumping continues without interruption even in the event of a hardware failure through remote fallback control by the central control system. This resolves the single point of failure problem and ensures the reliability of continuous operation of the water treatment plant.

[0034] When integrated management of multiple containers by a central control system is applied, the detachment rate is further improved to about 95% through self-learning optimization based on accumulated data from multiple cycles, and the timing of element replacement can be predicted in advance through predictive maintenance. Brief explanation of the drawing

[0035] FIG. 1 is a system block diagram showing the overall configuration of a PLC-based adaptive resonance tracking control precoat filter regeneration device according to a first embodiment of the present invention. Figure 2 is a schematic plan view of the upper surface of the tube sheet showing the installation location of the acceleration sensor. Figure 3 is a conceptual diagram explaining the principle of adaptive resonance tracking. Figure 4 is a block diagram illustrating the operation of the FFT resonance analysis module. Figure 5 is an integrated control flowchart of the control unit. Figure 6 is a double cross-validation Tally determination flowchart. Figure 7 is a graph comparing performance with the conventional method. FIG. 8 is a system block diagram of a hierarchical dual control architecture of an embedded PC and a central control system according to a second embodiment of the present invention. Figure 9 is a software redundancy and remote fallback control flowchart. Specific details for implementing the invention

[0036] Preferred embodiments of the present invention will be described in detail below with reference to the attached drawings. However, the concept of the present invention is not limited to the presented embodiments, and those skilled in the art who understand the concept of the present invention will be able to easily propose other embodiments within the scope of the same concept.

[0037] The term "FFT analysis" as used in this specification should be interpreted in a broad sense to encompass "frequency response analysis" or "signal transformation techniques," including not only the Fast Fourier Transform but also the Short-Time Fourier Transform (STFT), Wavelet Transform, and autocorrelation analysis in the time domain. Furthermore, "cylinder" encompasses not only hydraulic / pneumatic cylinders but also "vertical drive units that perform linear reciprocating motion," such as motor-based electric linear actuators.

[0038] [First Embodiment: PLC-based Adaptive Resonance Tracking Control Device]

[0039] FIG. 1 is a system block diagram showing the overall configuration of an adaptive resonance tracking control precoat filter regeneration device according to a first embodiment of the present invention, and FIG. 2 is a schematic plan view of the upper surface of a tube sheet showing the installation location of an acceleration sensor.

[0040] Referring to FIGS. 1 and 2, a device (100) according to a first embodiment of the present invention comprises a filtration vessel (10), a vertical driving unit (30), a sensor unit (50), and a control unit (70). The filtration vessel (10) is divided into an upper chamber (14) and a lower chamber (15) by a tube sheet (21) inside. A plurality of filtration elements (22) are vertically suspended and mounted on the lower surface of the tube sheet (21), and a precoat layer (23) is coated on the outer surface of each filtration element (22).

[0041] The vertical drive unit (30) is an actuator that performs linear reciprocating drive and includes a hydraulic or pneumatic cylinder (31) and an operating rod (32) that reciprocates the tube seat (21) in a vertical direction. A solenoid valve (34) for controlling the drive frequency and amplitude is connected to the cylinder (31).

[0042] The sensor unit (50) includes a first acceleration sensor (51a) and a second acceleration sensor (51b) mounted on the upper surface of the tube sheet (21) as shown in FIG. 2. The first acceleration sensor (51a) is positioned at the center near the connection point of the operating rod (32) to measure the main excitation force, and the second acceleration sensor (51b) is positioned at the edge to monitor the eccentric vibration of the tube sheet. Additionally, the sensor unit may include a differential pressure sensor (52) that measures the differential pressure between the inlet and outlet sides of the filtration vessel (10) and a position sensor (33) that measures the position of the operating rod. The sensor cable is connected to an external control unit (70) through only a minimum number of cable fittings (e.g., M16 to M20) for ease of retrofitting.

[0043] The control unit (70) may be composed of an industrial PLC, DDC, or FPGA-based controller, and the FFT resonance analysis module (71), adaptive resonance tracking module (72), and detachment determination module (73) are implemented in software.

[0044] [Principles of Resonance Search and Adaptive Resonance Tracking]

[0045] Figure 3 is a conceptual diagram explaining the principle of adaptive resonance tracking, and Figure 4 is a block diagram explaining the operation of the FFT resonance analysis module.

[0046] Unlike conventional fixed-frequency bumping, the present invention reduces the total mass (m) of the system as the precoat layer (23) detaches from the element (22) as bumping proceeds, and consequently reduces the natural frequency (f) of the system. n It was inspired by the physical phenomenon of ∝ √(k / m)) gradually rising.

[0047] Referring to FIG. 4, the control unit (70) applies a micro-sweep excitation (e.g., 1–50 Hz, amplitude 0.5–1 mm) during the initial sweep, extracts data sampled at approximately 200 Hz from the acceleration sensor (51a) using a 2-second sliding window method, and performs a 1024-point FFT transformation. Subsequently, after separating the excitation component using frequency masking or a notch filter, the peak spectrum is detected to determine the natural frequency (f) of the current state. n derives ).

[0048] As illustrated in FIG. 3, the adaptive resonance tracking module (72) reanalyzes the frequency response of the acceleration sensor output at intervals of 2 to 5 seconds during bumping to re-detect the natural frequency (solid line graph) moving in real time. Then, it immediately re-tunes the bumping frequency (f_bump, dotted line step graph) to within the range of 0.9 to 1.1 times the re-detected natural frequency. Through this, a quasi-resonance state is maintained throughout the entire bumping process, ensuring an amplification ratio of at least -3dB relative to the maximum value of the resonance response.

[0049] [Integrated Control Method and Tally Determination] (Double Cross-Verification)

[0050] FIG. 5 is an integrated control flowchart of the control unit, and FIG. 6 is a double cross-verification flowchart for preventing material incompleteness and objectively determining the completion of detachment.

[0051] Referring to FIG. 5, the bumping process proceeds in the steps of residual raw water discharge (S1), initial resonance search through a micro-amplitude sweep (S2), resonance tuning bumping initiation (S3), resonance tracking and re-tuning every 2 to 5 seconds (S4), detachment determination (S5), and bumping stop (S7).

[0052] In particular, the detachment determination module (73) of the present invention adopts a cross-verification method based on dual criteria as shown in FIG. 6.

[0053] The first criterion is whether the output change rate (|dΔP / dt|) of the differential pressure sensor (52) converges to a predetermined threshold value (e.g., 0.1 kPa / s).

[0054] The second criterion is whether the rate of change (|dBEI / dt|) of the Bumping Efficiency Index (BEI), calculated based on the output of the acceleration sensor (51a) and the input energy of the vertical drive unit (30), has converged to a predetermined threshold value or lower and saturated.

[0055] Here, the above bumping efficiency index (BEI) is calculated using the following [Equation 1].

[0056] [Mathematical Formula 1] BEI = (a_rms)² / E_input

[0057] (Here, a_rms is the RMS value of the acceleration response within the sliding window interval.)

[0058] The E_input (estimated input energy value) corresponding to the above denominator may be calculated or estimated through any one of the following various implementation methods so that a person skilled in the art can easily implement the present invention.

[0059] 1. Pressure-based calculation: E_input can be calculated in real time as the product of the fluid pressure (P) measured by the pressure sensor mounted on the cylinder (31) and the load displacement amount (Δx) measured by the position sensor (33) (Work = P × A × Δx).

[0060] 2. Control signal-based estimation (sensorless method): Without attaching additional sensors, a system constant proportional to the ON time duty cycle of the control signal output by the control unit (70) to the solenoid valve (34) can be used as a substitute for the E_input value.

[0061] The detachment determination module (73) confirms objective detachment completion only when both the first criterion (differential pressure convergence) and the second criterion (BEI saturation) are satisfied (AND condition), and immediately turns off the valve to prevent unnecessary over-bumping.

[0062] [Second Embodiment: Hierarchical Dual Control Architecture and Remote Fallback]

[0063] FIG. 8 is a system block diagram of a hierarchical dual control architecture of an embedded PC and a central control system according to a second embodiment of the present invention, and FIG. 9 is a flowchart of software redundancy and remote fallback control. This embodiment is suitable for nuclear power plant condensate desalination facilities or large water purification plants where zero-downtime operation is essential.

[0064] Referring to FIG. 8, the control unit is divided into a local controller (80, embedded PC) located at the filtration vessel site and a central control system (90, SCADA / DCS) connected via a communication network.

[0065] Software redundancy is implemented within the local controller (80), so that a control process (86) that controls the actual equipment and a watchdog process (87) that independently monitors it are run in parallel. The central control system (90) is equipped with a software mirror (93) that performs the same operation as the local control algorithm to perform so-called shadow computation.

[0066] Referring to Fig. 9, the uninterrupted fallback control procedure is explained.

[0067] During normal operation, the watchdog process (87) checks the survival of the control process (86) every 50ms (IPC-based). If three consecutive response failures occur on the control process side due to a software bug or delay, a Level 1 local fallback is triggered. The watchdog process (87) forcibly terminates and restarts the control process, and during the transition period (about 1 to 2 seconds) required for the restart, the 'last optimal natural frequency (f)' stored in the non-volatile storage (82) n It reads the value of )' and continues bumping in safe mode. Therefore, valve operation does not stop even in the event of a software failure.

[0068] If a critical hardware failure occurs, such as the power to the local controller (80) being cut off, a two-stage remote fallback is triggered. When the central control system (90) detects a network timeout (1 to 3 seconds), it activates the remote fallback control module (94). From this point on, the frequency tracking result calculated by the software mirror (93) within the central control system is transmitted directly to the field solenoid valve (34) via an industrial network (Ethernet / IP, etc.). In this remote fallback mode, the valve switching amplitude is automatically limited to a physical safety range to account for network latency. Subsequently, when the local controller is repaired and communication resumes, it immediately returns to normal operation mode.

[0069] Through this redundancy and fallback mechanism, the present invention maximizes the operational reliability of the plant by fundamentally eliminating single point of failure (SPOF). Furthermore, the central control system (90) can learn accumulated sensor data from multiple vessels through the learning module (95) and prediction module (96) shown in FIG. 8 to gradually optimize bumping parameters and diagnose the replacement timing of elements in advance.

[0070] [Examples and Comparative Examples]

[0071] Figure 7 is a graph comparing the performance of embodiments of the present invention and a conventional method (comparative example). To demonstrate the significant effects of the adaptive resonance tracking control and redundancy system of the present invention, experiments were conducted by configuring comparative examples and embodiments as follows.

[0072] - Comparative Example 1 (Conventional fixed frequency bumping): As a result of performing fixed bumping for 90 seconds with an amplitude of 30 mm and a frequency of 2 Hz, the detachment rate was 85.2%, the energy consumed was 4.8 kJ, and the standard deviation of the detachment rate in 10 repeated experiments was 4.2%.

[0073] - Comparative Example 2 (Resonance tuning, no tracking): Initial natural frequency (f n This is a case where the bumping frequency was tuned once to find (= 8.1Hz), but frequency re-tracking due to mass change was not performed. There was some performance improvement with a delamination rate of 88.3% (+3.1%p improvement), energy of 3.8kJ (-21% reduction), and time of 68 seconds (-24% reduction).

[0074] - Example 1 (Adaptive resonance tracking applied - PLC alone): f during bumping n As a result of tracking the transition from 8.1 Hz to 11.4 Hz in real time and performing a total of 5 re-tunings, the de-tuning rate was significantly improved to 91.5%, energy consumption to 3.1 kJ (-35% reduction), and time to 52 seconds (-42% reduction).

[0075] - Example 2 (Adaptive Resonance Tracking + Dual Automatic Judgment): As a result of adding differential pressure and BEI dual cross-verification automatic judgment to Example 1, as shown in Fig. 7, (a) a detachment rate of 92.3%, (b) energy of 2.8 kJ (-42% reduction), and (c) time of 43 seconds (-52% reduction) were achieved, and the standard deviation was reduced to 1.5%, significantly improving reproducibility.

[0076] - Example 3 (Embedded PC + Central Control Redundancy + Self-Learning): As a result of the central control system (90) learning the accumulated data of 40 cycles of multiple containers, from the 35th cycle onwards, optimized performance (standard deviation 0.8%) was achieved with a delamination rate of 95.1%, energy of 2.7 kJ (-44% reduction), and time of 40 seconds (-56% reduction).

[0077] - Example 4 (Verification of Fallback Operation): (a) When a local software failure (forced termination of the control process) occurred, the watchdog process (87) detected it within 150ms and completed an automatic restart in 1.2 seconds, and safe mode bumping was maintained during the transition period, so no bumping interruption occurred. (b) When a local hardware failure (power cut off to the local controller) occurred, the central control system (90) detected the failure after 2.1 seconds and initiated a remote fallback via the software mirror (93) to maintain adaptive tracking, and successfully defended against a high fallback rate of 90.3% (level of Example 1) even in a fallback mode with communication delay. Subsequently, when the local controller was restarted, it returned to normal operation within 8 seconds.

[0078] [Variation Example]

[0079] The present invention is not limited to the embodiments described above and can be implemented with various modifications within the scope of the technical concept of the present invention.

[0080] For example, the control unit (70) is not limited to a PLC and may be composed of an embedded PC, DDC, FPGA-based controller, or a combination thereof. Also, the vertical drive unit (30) is not limited to a hydraulic cylinder and may be a pneumatic cylinder, an electric linear actuator, etc.

[0081] Signal analysis techniques for calculating natural frequencies may also include the Short-Time Fourier Transform (STFT), wavelet transform, and autocorrelation function analysis, in addition to the Fast Fourier Transform (FFT).

[0082] In addition, in a software redundancy architecture, the monitoring cycle and the number of response failure judgments of the watchdog process can be adjusted according to the real-time requirements of the plant, and various protocols compliant with industry standards, such as Ethernet / IP, PROFINET, Modbus TCP, and OPC UA, can also be used as communication protocols with the central control system. Explanation of the symbols

[0083] 10: Filter container 14: Upper chamber 15: Lower chamber 21 : Tube sheet 22 : Filter element 23 : Precoat layer 30: Vertical drive unit 31 : Cylinder 32 : Operating Load 33: Position sensor 34: Solenoid valve 50 : Sensor section 51a: First accelerometer 51b: Second accelerometer 52: Differential pressure sensor 70 : Control unit (PLC, 1st embodiment) 71: FFT Resonance Analysis Module 72: Adaptive Resonance Tracking Module 73: Talisman Judgment Module 80 : Local controller (embedded PC, 2nd embodiment) 81: Application processor 82: Non-volatile storage 83: Analog input module 84: Digital output module 85: Industrial network interface 86 : Control Process 87 : Watchdog Process 90 : Central Control System (SCADA / DCS) 91: HMI Display Module 92: Data Logging Module 93 : Software Mirror 94: Remote fallback control module 95: Self-learning optimization engine 96 : Predictive Maintenance Engine 100: Adaptive Resonance Tracking Control Redundant Precoat Filter Regeneration Device

Claims

Claim 1 A device for regenerating filtration performance by detaching a precoat filter medium from a precoat filter in which a precoat filter medium is coated on the outer surface of a filtration element contained within a filtration vessel, comprising: a vertical drive unit including a cylinder and an operating rod for reciprocating a tube sheet on which the filtration element is mounted in a vertical direction; and a sensor unit including an acceleration sensor mounted on the tube sheet for measuring an acceleration response during bumping, and a differential pressure sensor for measuring the differential pressure between the inlet and outlet sides of the filtration vessel. The control unit comprises: calculating the natural frequency of the system with the precoat layer attached by FFT analysis of the acceleration sensor output; tuning the bumping frequency of the vertical drive unit to the natural frequency; re-detecting the natural frequency, which gradually rises due to the reduction in precoat mass during bumping, by FFT re-analysis using a sliding window method at intervals of every 2 to 5 seconds; and performing adaptive resonance tracking control to maintain a quasi-resonant state throughout the entire bumping process by re-tuning the bumping frequency to a range of 0.9 to 1.1 times the re-detected natural frequency; cross-verifying a first criterion in which the rate of change of the output of the differential pressure sensor converges to a predetermined threshold value or less and a second criterion in which the rate of change of the bumping efficiency index (BEI) calculated from the acceleration sensor output converges to a predetermined threshold value or less; and confirming the completion of detachment and automatically stopping bumping when both criteria are satisfied, wherein the BEI is defined as the value obtained by dividing the input energy estimate value having the square of the acceleration RMS value within the sliding window section. Redundant precoat filter regeneration device using adaptive resonance tracking control. Claim 2 An apparatus according to claim 1, wherein the control unit applies a frequency sweep vibration of a small amplitude at the start of playback and determines the peak frequency in the FFT spectrum as the natural frequency. Claim 3 delete Claim 4 The device according to claim 1, wherein the control unit comprises: a field embedded PC in which a control process that performs FFT analysis, adaptive resonance tracking, and a fall-out determination, and a watchdog process that monitors the normal operation of the control process are executed independently; and a central control system having a software mirror connected to the field embedded PC via an industrial network and performing shadow operations of the same control algorithm on sensor data received from the field embedded PC; wherein the watchdog process automatically restarts the control process upon detection of a software failure of the control process and performs safe mode bumping by a bumping frequency that was last retuned immediately before detection of the software failure by the adaptive resonance tracking control, which is stored in non-volatile memory until the restart is completed, and the central control system performs remote fallback control based on the operation result of the software mirror upon detection of a hardware failure of the field embedded PC. Claim 5 delete Claim 6 A precoat filter regeneration method for detaching a precoat filter medium coated on the outer surface of a filtration element contained within a filtration vessel by vertical bumping, comprising: a resonance search step of applying a frequency sweep vibration of a small amplitude and detecting the natural frequency of a system with the precoat layer attached by FFT analysis of an acceleration sensor output; a resonance tuning bumping step of performing vertical bumping by tuning the bumping frequency to the detected natural frequency; and a resonance tracking step of re-detecting the natural frequency, which gradually rises according to the reduction in precoat mass during bumping, by FFT re-analysis using a sliding window method at intervals of every 2 to 5 seconds, and re-tuning the bumping frequency to a range of 0.9 to 1.1 times the re-detected natural frequency. A precoat filter regeneration method comprising: a first criterion in which the rate of change of differential pressure between the inlet and outlet sides of a filtration vessel converges to a predetermined threshold value or less, and a second criterion in which the rate of change of the bumping efficiency index (BEI) calculated from the acceleration sensor output converges to a predetermined threshold value or less, and a feedback control step in which, when both criteria are satisfied, the completion of detachment is confirmed and bumping is automatically stopped; wherein the BEI is defined as a value divided by an estimated input energy value having the square of the acceleration RMS value within the sliding window section. Claim 7 delete

Citation Information

Patent Citations

  • Device, method and computer program for determining facility state by using virtual model

    KR1020210112598A

  • Dust removing system and operating method thereof

    KR1020240115652A

  • Precoat filter

    US3422958A

  • Helical spring apparatus and process for detaching filter cake

    US4804481A