A device life prediction method and system based on working state change curve

Abstract

The present application relates to the technical field of device life prediction, and more particularly to a device life prediction method and system based on working state change curve, the present application proposes the following scheme, by acquiring the working state change curve of the device in the working process, the resampling of the flow curve is realized based on the cumulative volume mapping, the hysteresis back scanning is carried out in the neighborhood of the candidate boundary point, the final boundary point is determined, and the updated equivalent window is constructed. Further, according to the volume difference between the plurality of updated equivalent windows, correction is carried out, when the difference is within the tolerance range, it is directly used, otherwise a correction factor is introduced to correct the updated window, and the self-cleaning recovery characteristics of the device during the cycle switching are combined to compensate, so that the updated equivalent window more in line with the actual situation is obtained. The attenuation and recovery process of the device is accurately characterized, and reliable basis is provided for device life prediction and operation and maintenance strategy.

Description

Technical Field

[0001] This invention relates to the field of device life prediction technology, and in particular to a device life prediction method and system based on operating state change curves. Background Technology

[0002] In existing technologies, device performance evaluation primarily relies on flow monitoring and empirical threshold judgments during operation. However, these methods generally suffer from insufficient accuracy, failing to accurately reflect performance changes at different operating stages, leading to unreliable judgments regarding cleaning and replacement timing. Furthermore, existing flow curve processing methods are susceptible to data fluctuations, resulting in significant errors in identifying the device's degradation process and affecting the calculation of the updated equivalent window. Moreover, existing technologies fail to adequately characterize the device's self-cleaning function during cycle switching, causing discrepancies between window update results and actual operating conditions. These deficiencies collectively contribute to inaccurate device lifetime prediction, thereby impacting system stability and maintenance efficiency.

[0003] To address the above issues, this application presents a method and system for predicting device lifespan based on operating state change curves. Summary of the Invention

[0004] The technical problem this invention aims to solve is to address the shortcomings of existing technologies by providing a method and system for predicting device lifetime based on operating state change curves. This method acquires the operating state change curves of the device during operation, resamples the flow rate curve based on cumulative volume mapping, and performs hysteresis scan verification within the neighborhood of candidate boundary points to determine the final boundary point and construct an updated equivalent window. Furthermore, corrections are made based on the volume differences between multiple updated equivalent windows. When the difference is within the tolerance range, it is directly adopted; otherwise, a correction factor is introduced to modify the updated window, and compensation is made by combining the self-cleaning recovery characteristics of the device during cycle switching, thereby obtaining an updated equivalent window that more closely reflects reality.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A device lifetime prediction method based on operating state change curves is applied to an MCU in an offline fine filtration system. The MCU stores a device lifetime table and a current equivalent window, where the current equivalent window is an updated equivalent window from the previous operating cycle of the device. The updated equivalent window is a characterization index obtained by mapping the state information of the device's operating object through equivalent processing. The method is executed by the MCU after the current operating cycle ends, and specifically includes:

[0007] Obtain the status information of the current equivalent window and the working object of the current work cycle;

[0008] According to the state information, an updated equivalent window is determined, and the updated equivalent window is divided into a first equivalent window and a second equivalent window by monitoring a working state change curve of the device in a current working period, wherein a size of the first equivalent window is the same as that of the current equivalent window, and a size of the second equivalent window is the same as that of the updated equivalent window;

[0009] The updated equivalent window is corrected according to the first equivalent window and the second equivalent window, and replaces the current equivalent window;

[0010] The corrected updated equivalent window is compared with the device life table to obtain a device life.

[0011] According to the state information, an updated equivalent window is determined, and the updated equivalent window is divided into a first equivalent window and a second equivalent window by monitoring a working state change curve of the device in a current working period, wherein a size of the first equivalent window is the same as that of the current equivalent window, and a size of the second equivalent window is the same as that of the updated equivalent window;

[0012] According to state information of a working object of the device, an impurity accumulation rate of a current working period is obtained, wherein the state information at least includes one of turbidity value, sand content, particle size distribution and dissolved solid content;

[0013] According to the impurity accumulation rate and the current equivalent window, a predicted attenuation amount is calculated, and the predicted attenuation amount is applied to the current equivalent window to obtain an updated equivalent window.

[0014] According to the impurity accumulation rate and the current equivalent window, a predicted attenuation amount is calculated, and the predicted attenuation amount is applied to the current equivalent window to obtain an updated equivalent window.

[0015] The current equivalent window is divided into a plurality of continuous sub-windows, wherein lengths of the continuous sub-windows are calculated according to a change trend of the impurity accumulation rate;

[0016] For each continuous sub-window, an attenuation contribution value is calculated according to a corresponding impurity accumulation rate, and the attenuation contribution values are accumulated to obtain a first attenuation curve;

[0017] The first attenuation curve is aligned with a predicted working state change curve in the current working period to obtain a lag compensation factor;

[0018] The first attenuation curve is compensated according to the lag compensation factor to obtain a second attenuation curve;

[0019] According to a descending slope of the second attenuation curve and a cumulative attenuation area, a predicted attenuation amount is determined.

[0020] The state information further includes an inlet water flow, and the updated equivalent window is divided into the first equivalent window and the second equivalent window by monitoring a working state change curve of the device in a current working period, and the method comprises the following steps:

[0021] acquire a working state change curve of the device in a current working period, and compare the working state change curve with a reference cleaning curve to obtain a candidate division point, wherein the reference cleaning curve is calculated according to a current equivalent window and a water inflow;

[0022] For each candidate division point, the working state change curve is resampled in a neighborhood of the candidate division point, and hysteresis review is performed according to the resampling result to obtain a final division point;

[0023] According to the position of the final division point on the working state change curve, the updated equivalent window is segmented to obtain a first equivalent window and a second equivalent window, and the first equivalent window is assigned a value according to the current equivalent window.

[0024] The working state change curve is compared with the reference cleaning curve to obtain a candidate division point, including:

[0025] A first pointer and a second pointer are configured in a time axis corresponding to the working state change curve and the reference cleaning curve, wherein the first pointer is used to traverse the working state change curve, and the second pointer is used to traverse the reference cleaning curve;

[0026] The first pointer and the second pointer are advanced in a preset candidate time window, and a time stretching factor and a stability index in the corresponding candidate time window are calculated according to an advancing mode, wherein the stability index includes a variance and a range of the candidate time window;

[0027] When the time stretching factor is greater than or equal to a preset time stretching threshold, and the offset value of the stability index from a tolerance band is greater than a preset deviation threshold, the average time of the corresponding candidate time window is taken as a candidate division point, wherein the tolerance band is calculated according to a minimum cleaning volume of the reference cleaning curve.

[0028] The advancing mode includes:

[0029] The first pointer and the second pointer are advanced synchronously;

[0030] The first pointer is fast forwarded and the second pointer is slow forwarded;

[0031] The first pointer is slow forwarded and the second pointer is fast forwarded;

[0032] The fast forwarding and the slow forwarding correspond to the advancing frequency of the corresponding pointer, and the advancing frequency is calculated according to the average slope of the curve of the working state change curve and the reference cleaning curve in the corresponding candidate time window.

[0033] The working state change curve is resampled in the neighborhood of the candidate boundary point, and hysteresis review is performed according to the resampling result, including:

[0034] The cumulative volume is calculated according to the working state change curve, the working state change curve corresponding to the neighborhood of the candidate boundary point is resampled according to the cumulative volume, and a candidate pair corresponding to the uplink segment and the return segment is obtained, wherein the volume span of the candidate pair is greater than or equal to the minimum cleaning volume;

[0035] The uplink segment and the return segment are aligned and interpolated on the volume axis, the flow difference of the uplink segment and the return segment under the consistent volume grid is calculated, and the corresponding pseudo-hysteresis area is obtained by integration;

[0036] The pseudo-hysteresis areas of each candidate boundary point are sorted in descending order, and the first candidate boundary point is taken as the final boundary point for output, wherein the descending sorting further includes weighting assignment of the pseudo-hysteresis area according to the area of the neighborhood of the candidate boundary point.

[0037] The first equivalent window and the second equivalent window are modified according to the update equivalent window, including:

[0038] The volume difference between the first equivalent window and the second equivalent window is calculated;

[0039] If the volume difference is less than or equal to a preset tolerance threshold, the second equivalent window is taken as the modified update equivalent window;

[0040] If the volume difference is greater than the preset tolerance threshold, a correction factor is calculated according to the volume difference, and the update equivalent window is weighted and assigned according to the correction factor to obtain the modified update equivalent window.

[0041] The modified update equivalent window is compared with the device life table to obtain the device life, including:

[0042] The modified update equivalent window is input into the device life table to retrieve the corresponding life reference value;

[0043] The remaining life of the device is determined according to the life reference value, and the device life is output.

[0044] A device life prediction system based on working state change curve, the system is applied to MCU of offline fine filtration system, the system is applied to MCU of offline fine filtration system, the MCU stores device life table and current equivalent window, wherein the current equivalent window is the updated equivalent window of the previous operation cycle of the device, the updated equivalent window is a characteristic index mapped by equivalent processing of the state information of the working object of the device, the system comprises a data acquisition module, an operation processing module and a life evaluation module, wherein:

[0045] The data acquisition module is used for acquiring the state information of the water sample to be filtered in the current working cycle and the working state change curve of the device in the current working cycle, and transmitting the data to the operation processing module.

[0046] The operation processing module is arranged in the MCU and is used for maintaining the updated equivalent window and the device life table, and updating, correcting, segmenting the updated equivalent window and calculating the prediction attenuation amount.

[0047] The life evaluation module is used for comparing the corrected updated equivalent window with the device life table to obtain the device life.

[0048] Compared with the prior art, the beneficial effects of the present application are:

[0049] The present application can effectively reduce the interference caused by instantaneous fluctuation and uneven sampling in the identification process of the candidate boundary point by introducing the resampling method based on the volume axis and the hysteresis back-scan verification mechanism, thereby improving the stability and accuracy of the boundary point positioning. By comparing the differences between the uplink segment and the return segment under the consistent volume grid, and taking the pseudo-hysteresis area as the basis for judgment, not only can the distinguishability of the device attenuation and recovery characteristics be enhanced, but also the boundary point selection result can be ensured to be consistent with the actual operation process. BRIEF DESCRIPTION OF DRAWINGS

[0050] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the accompanying drawings:

[0051] Figure 1 It is a device life attenuation schematic diagram of the embodiment of the present application;

[0052] Figure 2 It is an updated equivalent window segmentation schematic diagram of the embodiment of the present application;

[0053] Figure 3 It is an exemplary application scenario diagram of the embodiment of the present application;

[0054] Figure 4 It is a flowchart of a device life prediction method based on working state change curve of the embodiment of the present application;

[0055] Figure 5 The segmentation process schematic diagram for updating the equivalent window of the embodiment of the application;

[0056] Figure 6 The principle schematic diagram for candidate boundary point screening of the embodiment of the application;

[0057] Figure 7 The correction principle schematic diagram for updating the equivalent window of the embodiment of the application. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the application will be apparently and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments.

[0059] In this document, reference to“an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase“in an embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. Those skilled in the art will appreciate from the present description that embodiments described herein can be combined with other embodiments in various ways.

[0060] The application is applicable to device products with periodic water discharge characteristics and multi-factor water quality disturbance conditions. The life prediction process of the device products is limited by the differences in the running environment water quality and the dynamic fluctuations of the device water discharge performance. In the device life evaluation process, there are prediction deviation problems caused by the nonlinear cumulative effect of water quality and the random disturbance of water discharge flow.

[0061] The application scenarios include but are not limited to:

[0062] Household or commercial water purification systems with multi-stage series devices and periodic backwashing functions;

[0063] Offline water treatment devices using high-precision devices for pretreatment or fine treatment in industrial pipelines;

[0064] Reverse osmosis pre-protective devices running in complex water quality environments, whose water discharge performance is affected by the double coupling of dissolved solid content and particulate matter concentration.

[0065] The selection of the application scenarios is based on the common difficulties of device life prediction. The application scenarios at least have one of the following characteristics:

[0066] The attenuation process of the device is accumulated at a non-uniform rate, showing multi-section decline or lag compensation behavior;

[0067] The water outlet flow curve is affected by external water inlet pressure and instantaneous water use behavior, resulting in short-term fluctuations or false abnormal points;

[0068] The traditional life estimation method based on cumulative flow threshold is over-conservative or delayed in alarming;

[0069] The device attenuation is a non-structural geometric mutation, only showing a small disturbance attenuation of the equivalent cleaning ability.

[0070] It should be noted that the device life prediction method proposed in the present application does not rely on the fixed geometry of the device or a single water quality indicator as a prerequisite basis, but is oriented towards working scenarios with complex water quality disturbances and nonlinear changes in water outlet performance.

[0071] It is worth noting that the update equivalent window update and hysteresis back-scan review strategy of the present application is not designed for a specific type of device or a specific water treatment system, but can also be applied to scenarios where dynamic fluctuations in water quality parameters result in cumulative deviations in the prediction model, and scenarios where the device performance shows time-dependent decay rules under periodic working conditions.

[0072] In some optional specific embodiments, the device in the present application can be understood as a filter element.

[0073] Taking the filter element as an example, it can be understood that the present application mainly faces filter elements for oil filtration, i.e., the medium filtered by the filter element is oil, and oil filtration filter elements are widely used in hydraulic systems, lubrication systems, and various industrial equipment. Their main function is to remove particulate impurities and gum deposits from oil to ensure the normal operation of system components and extend the service life of equipment. Unlike water filtration, oil generally has higher viscosity and more complex pollutant composition, including metal abrasives, carbon soot particles, oxide deposits, and additive degradation products, etc. These pollutants can cause more obvious plugging effects inside the filter element.

[0074] In the following content, the water sample to be filtered and water body information can be understood as oil to be filtered and oil information.

[0075] Reference Figure 1 , Figure 1 The device life attenuation schematic diagram provided for the embodiments of the present application.

[0076] Figure 1It is shown that a certain device includes three filtration periods of t1, t2 and t3, and the water production performance of the device gradually deteriorates as the working time elapses, and the flow rate decay rate in different periods is different: in the initial period t1, the internal filter medium of the device is relatively clean, and the decay rate is low; after entering the middle period t2, the device gradually accumulates impurities, and the water flow decay rate accelerates; when entering the end period t3, the device has reached a saturation state, and the water flow rapidly decreases, and the life enters a critical stage.

[0077] It should be noted that, Figure 1 The decay curve shown is only a schematic illustration, mainly used to show the basic principles and stage characteristics of the device life decay, and does not represent the actual working state of the device. In actual application, the device life curve will be affected by many factors such as water quality, working conditions, pressure conditions and device materials, and its decay pattern may have complex characteristics such as multi-section, non-linearity or local fluctuations.

[0078] It can be understood that in this application, the change of water flow can be regarded as a direct representation of the working state of the device, and by equivalent processing of the water flow, it can be mapped to an update equivalent window. In other words, the update equivalent window is a unified measurement method for reflecting the filtering performance and decay degree of the device in different periods. When the device is in the initial period, the value of the update equivalent window is larger, corresponding to the stable and ideal reference of the water flow; as the device gradually accumulates pollutants, the update equivalent window is correspondingly reduced, until it significantly decreases in the critical period, corresponding to the sharp decrease of the water flow.

[0079] It should be noted that the aforementioned update equivalent window is not a real physical quantity, but a representation index obtained by modeling conversion based on the water flow curve, which is used to simplify the description and calculation of the device life decay process. The method of the present application is to establish a mapping relationship between the current equivalent window and the device life, so as to realize unified life prediction in different devices and different use environments. It can be understood that the MCU always maintains a current equivalent window in this application, which can be understood as the update equivalent window of the previous running period. The current equivalent window is updated according to the update equivalent window of the current running period in each running period to achieve the purpose of life prediction of the present application.

[0080] It is easy to understand that in the water filtration work of the device, the water flow of a certain period is not a linear change state, so the corresponding update equivalent window is not a constant value in the corresponding period. In general, it can be considered that the update equivalent window is maintained at a relatively high level in the initial stage of the period, and as the device gradually accumulates pollutants, the water flow slowly decreases, and the update equivalent window also gradually decreases.

[0081] Therefore, the updating equivalent window is not a static fixed amount, but a dynamic amount that evolves over time and cumulative usage. The present application accurately extracts features reflecting the device attenuation state by characterizing this dynamic change process, combined with candidate division point identification, local resampling, and hysteresis review processing steps, to realize reliable prediction of device life.

[0082] Reference Figure 2 , Figure 2 The updating equivalent window segmentation schematic diagram provided by the embodiment of the present application is shown.

[0083] Figure 2 On the basis of Figure 1 , the performance attenuation law of the device in different working periods and the corresponding relationship between the updating equivalent window are further shown.

[0084] It can be understood that in the actual water filtration process of the device, the water flow is not an ideal straight line attenuation, but is affected by water quality fluctuations, pumping pressure changes, and valve start-stop and other factors, showing nonlinear changes. Based on the modeling method proposed in the present application, this complex flow change can be understood as a reduction process of the updating equivalent window:

[0085] In the initial stage of each period, the pore structure has not been significantly blocked, and the updating equivalent window is similar to the updating equivalent window of the previous period, showing a high and stable water flow;

[0086] With the extension of the use time, impurities gradually deposit inside the device, reducing the effective filtration area, resulting in a gradual reduction of the updating equivalent window, showing a decrease in water flow.

[0087] It should be noted that Figure 2 The above is only an exemplary explanation of the foregoing principles, which is used to intuitively show the trend of the updating equivalent window gradually reducing over time, and is not limited to the specific change curve in real applications.

[0088] Reference Figure 3 , Figure 3 An exemplary application scenario provided by the embodiment of the present application is shown.

[0089] Figure 3 An application scenario is shown, which includes a water filtration device 100, a device 101, a water inlet 102, a water outlet 103, and a processor 104, wherein the solid line represents the water flow channel, and the dashed line represents data transmission, specifically:

[0090] The water inlet 102 is used to receive raw water to be filtered;

[0091] The device 101 is arranged inside the water filtration device 100, and is used to filter the raw water;

[0092] The water outlet 103 is in communication with the device 101, and is used to output filtered water after the device 101 filters the water;

[0093] The processor 104 collects working parameters related to the state of the device 101 through a sensor, such as water flow, working time, and the like, and performs device life prediction based on the working parameters.

[0094] It can be understood that, Figure 3 The application scenarios shown are only exemplary descriptions of embodiments of the present application, and are used to help understand the application mode of the device life prediction method and system proposed by the present application.

[0095] In actual applications, the specific structure and functional modules of the water filtering device 100 can be adjusted according to different product forms or use requirements, for example:

[0096] The processor 104 can be built into the water filtering device 100, or can be independently arranged in an external control unit and communicate with the water filtering device 100 through wired or wireless means;

[0097] The working parameters related to the state of the device 101 are not limited to water flow and working time, but can also include water quality indicators, pressure changes, and valve start-stop times;

[0098] The water outlet 103 can be further connected to a water storage tank, a water supply pipeline, or an end-use water appliance to meet different application scenarios.

[0099] Therefore, Figure 3 The application scenarios shown do not constitute a limitation on the protection scope of the present application, and the method and system of the present application are also applicable to other types of water filtering equipment and device structures.

[0100] Next, with reference to the accompanying drawings, a device life prediction method based on a working state change curve provided by an embodiment of the present application will be described, Figure 4 The method is applied to an MCU of an offline precision filtering system, the MCU maintains a current equivalent window and a device life table corresponding to the current equivalent window in the MCU, the current equivalent window is an updated equivalent window of a previous running period of the device, the updated equivalent window is a characteristic index mapped by equivalent processing of state information of a device working object, and the method is executed by the MCU after a current working period ends, and the specific steps are as follows:

[0101] S1: Obtain the current equivalent window of the current working period and the state information of the device working object;

[0102] In the embodiment, the device working object can be understood as a water sample to be filtered, and the state information of the water sample to be filtered can include indexes such as turbidity, hardness, TDS (total dissolved solids), and can also be working parameters obtained through a flow sensor and time accumulation. The specific types can be flexibly configured according to application scenarios, as long as they can support the evaluation of the working state of the device to the minimum extent, and the present application does not make more limitations.

[0103] By collecting the above information at the beginning of the working period, boundary conditions can be provided for subsequent updating of the update equivalent window, thereby avoiding life prediction errors caused by water quality differences.

[0104] S2: According to the state information, determine the update equivalent window, and after the end of the current working period, divide the update equivalent window into a first equivalent window and a second equivalent window through monitoring the working state change curve of the device in the current working period.

[0105] In the embodiment, the working state change curve can be understood as a change curve of the water outlet flow, which is recorded in real time by a sensor and processed after the end of the period. By calculating the cumulative water volume, the working state change curve can be resampled at the candidate division point, and the effectiveness of the division point can be verified by using a hysteresis sweep. If the curve shows different attenuation characteristics before and after the division point, the updated update equivalent window is divided into a first equivalent window and a second equivalent window.

[0106] S3: According to the first equivalent window and the second equivalent window, correct the update equivalent window and replace the current equivalent window.

[0107] In the embodiment, the first equivalent window and the second equivalent window respectively represent the flow performance in different stages of the period. After superimposing and comparing them by the MCU, the correction result is obtained by interpolation, weighted average or difference correction. The corrected update equivalent window replaces the current equivalent window and is stored as the starting state of the next working period. In this way, errors caused by short-term abnormal fluctuations can be dynamically eliminated, thereby ensuring the continuity and stability of life prediction.

[0108] S4: Compare the current equivalent window with the device life table to obtain the device life.

[0109] It should be noted that the current equivalent window at this time is the corrected update equivalent window.

[0110] In the embodiment, the device life table can be obtained by laboratory accelerated aging test, actual work big data statistics or experience model training, and stored in the non-volatile memory of the MCU; after the current equivalent window is obtained, the MCU can quickly obtain the quantitative value of the remaining life of the device by table lookup or interpolation operation, avoid complex calculation burden, ensure that the prediction result can be output in real time in the embedded environment, and facilitate the user to timely master the device state.

[0111] It can be understood that the MCU refers to a micro control unit, which is essentially an embedded control chip integrating a processor core, a memory and a peripheral interface. In the embodiment, the MCU is used to execute each step of the device life prediction method, and maintain the correspondence between the current equivalent window and the device life table. The MCU can be any model of 8-bit, 16-bit or 32-bit, or a special chip with low power consumption characteristics, and the specific type can be selected according to the application scenario.

[0112] Before expanding the specific technical content corresponding to the step, the embodiment of the application needs to emphasize again that the core problem faced by the device life prediction of the application is not simply the fitting of the life curve, but how to obtain an accurate judgment of the device attenuation trend only with limited real-time working data in the case of lacking massive historical data support in the offline filtering environment.

[0113] It can be understood that in typical household or distributed water purification equipment, the MCU is often subject to storage and computing power, and does not have the condition of long-term accumulation of large-scale samples. If the traditional life prediction method based on big data regression or long-term accumulation is relied on, the model will be difficult to converge, and the environmental adaptability will be insufficient, resulting in large deviation of the prediction result.

[0114] The life prediction logic proposed in the embodiment is not based on historical full data, but updates the definition of the equivalent window to normalize the complex flow decay process in a single cycle. In this way, the performance of each cycle of the device is mapped to an equivalent value corresponding to the device life table, so that the comparison and update across cycles can also be realized in the case of limited data.

[0115] It is easy to understand that even if the MCU only saves limited recent cycle information, the continuity of prediction can still be maintained through the inheritance and correction of window values, without the need for complete historical curves. In other words, the method of the application avoids the hardware constraint of insufficient data storage through structured modeling, and improves the usability of the prediction method in offline working scenarios.

[0116] Further, the application does not emphasize static assumptions about water quality or the environment, but through the division and update of periodic windows, the combined effects of flow disturbance, water quality fluctuation and pressure change are projected onto the reduced trend of the updated equivalent window. This mapping not only reduces the influence of external interference on life prediction, but also provides quantifiable and iterative indicators for the construction of device life table. In engineering practice, the life table of the device does not have to be an accurate global function, but through continuous comparison of different periodic windows, it gradually approaches the real decay law, and in the face of different water quality and different use habits, it still maintains high adaptability and robustness.

[0117] Next, the principle of the method of the application about the updated equivalent window is further expanded.

[0118] In one example, according to the state information, the updated equivalent window is determined, comprising:

[0119] S2.1: According to the state information of the water sample to be tested, the impurity accumulation rate of the current working period is obtained;

[0120] Specifically, in the device life prediction process, relying only on the downward trend of the outlet flow is often insufficient to reflect the real speed of device clogging, because the nature of impurities carried by different water bodies is quite different. Even under the same flow decline amplitude, the actual load of the device may be completely different.

[0121] For example, when the particle size of the particulate matter in the water body is small and the concentration is high, the filter pores are more likely to be micro-clogged in a short time, and if the main impurity in the water body is large particle suspended matter, these impurities may be quickly trapped on the surface of the device and form a blocking layer in the short term, but under the action of a certain scouring, they can be partially released. Therefore, when updating the equivalent window, it is necessary to first quantify the impurity accumulation rate of the current working period according to the state information, so as to serve as the basis for the decay amount calculation.

[0122] In the embodiment, the state information at least includes one of turbidity value, sand content, particle size distribution and dissolved solids content, which can comprehensively reflect the impurity characteristics in raw water. The turbidity value can reflect the overall influence of suspended particles on optical properties, which can be regarded as an index comprehensively reflecting the particle concentration in water; the sand content directly determines the total amount of inorganic particles that need to be intercepted by the device in unit time; the particle size distribution can reveal the corresponding relationship between the fineness of particles in water and the risk of blockage, and the smaller the particle size, the easier it is to penetrate the pores of the filter material; the dissolved solids content is used to reflect the crystallizable salts or chemical impurities in water, which will affect the permeability of the device in the form of deposition or scaling in the long-term work. Through the combination evaluation of the above information, the comprehensive load level of the device in the current working period, i.e. the impurity accumulation rate, can be obtained, so that the subsequent life prediction no longer depends on the flow signal alone, but integrates the quality characteristics of the water body, improving the accuracy and adaptability of life prediction.

[0123] Further, after collecting the foregoing state information, the MCU will perform structured processing on the data. For example, the turbidity value can be collected by an optical turbidity sensor and mapped to a standardized numerical interval, the sand content can be calculated by combining the pressure difference change and the particle monitor, the particle size distribution can be obtained by a laser particle size meter or statistical derivation based on flow rate disturbance, and the dissolved solids content can be indirectly calculated by a conductivity sensor. Due to the limitation of cost and volume in the offline precision filtration environment, the MCU will not save all the original data for a long time, but will map the above original data indicators to a periodic impurity accumulation rate through a preset algorithm, and update at the end of the current working period and in combination with the update equivalent window.

[0124] S2.2: According to the impurity accumulation rate and the current equivalent window, a predicted attenuation amount is calculated, and the predicted attenuation amount is applied to the current equivalent window to obtain an updated equivalent window;

[0125] Specifically, the updated equivalent window can be understood as a mapped value of the effective water purification capacity of the device in the current working period, and the impurity accumulation rate represents the actual load amount of the water body to the device in the period. If the original flow curve is directly used for calculation, it will often be affected by instantaneous fluctuations or use habit differences, resulting in unstable prediction results.

[0126] In the embodiment, by establishing the corresponding relationship between the impurity accumulation rate and the updated equivalent window, a theoretical predicted attenuation amount is first calculated, and then the predicted attenuation amount is applied to the current equivalent window to obtain the updated updated equivalent window. Combining the characteristics of the water body and the performance of the device avoids the prediction deviation that may be caused by single-dimensional data, so that the updating process not only considers the actual filtration performance of the device, but also takes into account the influence of water body differences on the attenuation rate.

[0127] Further, the calculation process of the predicted attenuation amount is not fixed value replacement, but adopts a periodic correction manner. Specifically, the MCU will analyze the working state change curve of the device in each period after the end of each period, and combine the high and low of the impurity accumulation rate to calculate an attenuation ratio corresponding to the device life table.

[0128] For example, when it is detected that the impurity accumulation rate is significantly higher than the standard water body condition, even if the outflow decreases by a limited amount, the MCU will correspondingly amplify the attenuation ratio to reflect the potential blockage of the device pores that may occur; on the contrary, when the impurity accumulation rate is low, the predicted attenuation amount is reduced to avoid overestimating the shortening of the device life. The updated equivalent window obtained finally will replace the original window value for the life prediction iteration of the next period.

[0129] In one example, calculating a predicted attenuation amount according to the impurity accumulation rate and the current equivalent window comprises:

[0130] S2.2.1: dividing the current equivalent window into a plurality of continuous sub-windows, wherein the length of the continuous sub-windows is calculated according to the change trend of the impurity accumulation rate;

[0131] Specifically, the current equivalent window is divided into a plurality of continuous sub-windows because the impurity accumulation process is not linear change, but presents a phased difference with the change of water quality and working conditions. If the whole window is still taken as the analysis object, the situation that the local pollution rate is too fast may be covered up, or the buffer segment appearing in the work cannot be reflected. By subdividing the updated equivalent window into continuous sub-windows, each sub-window can correspond to the impurity accumulation state in a specific time period, so that the subsequent prediction is more refined and dynamically adaptive.

[0132] In the present embodiment, the length of the sub-window is not fixedly set, but is adaptively calculated in combination with the change trend of the impurity accumulation rate. When it is detected that the content of particulate matter or dissolved solids in the water body increases, the division of the sub-window will be more intensive to capture the sudden performance attenuation of the device; on the contrary, when the water quality is stable, the sub-window is appropriately lengthened to reduce unnecessary operation burden caused by data dispersion.

[0133] S2.2.2: for each continuous sub-window, calculating an attenuation contribution value according to the corresponding impurity accumulation rate, and accumulating the attenuation contribution values to obtain a first attenuation curve;

[0134] Specifically, the impurity accumulation rate of each sub-window can be regarded as the load intensity of the device in this stage. After the load intensity accumulates to a certain degree, it will directly reflect the downward trend of the outflow or the increase of the pressure difference.

[0135] In the present embodiment, the decay contribution value in the time period is obtained by combining the impurity accumulation rate with the sub-window duration, and then the first decay curve is obtained by accumulating one by one. The performance decline trajectory of the device can be completely depicted in the time dimension, rather than speculating the overall decay condition by relying on a single point of water quality parameter. At the same time, since each sub-window retains its independent impurity characteristics, this accumulation method can avoid dilution of extreme water quality in global prediction, thereby more accurately reflecting the decay law of the device in actual work.

[0136] S2.2.3: aligning the first decay curve with the predicted working state change curve in the current working period to obtain a lag compensation factor;

[0137] Specifically, the first decay curve is aligned with the predicted working state change curve in the current working period because the performance decline of the device does not immediately appear when the contaminant enters the filter material, but there is a certain lag, for example, the pore gradually fills up before showing a significant increase in resistance.

[0138] S2.2.4: compensating the first decay curve according to the lag compensation factor to obtain a second decay curve;

[0139] In the present embodiment, the compensation process is not only a simple time shift, but also includes adjustment in the decay amplitude, that is, considering the energy consumption difference caused by lag, by adjusting the descending slope and intercept of the curve, so that the curve is consistent with the actual flow curve in the overall form. The second decay curve processed in this way not only coincides with the working condition in the time dimension, but also can truly map the decay trend of the device in the decay amplitude, ensuring the accuracy and reliability of the prediction model.

[0140] S2.2.5: determining a predicted decay amount according to the descending slope of the second decay curve and the cumulative decay area;

[0141] In the present embodiment, by calculating the descending slope of the second decay curve, the rate of performance decline of the device per unit time can be obtained, so as to evaluate the decay intensity under the current water quality condition; by accumulating the decay area, the total purification capacity loss of the device in a period can be quantified.

[0142] Next, the principle part of the method of the present application about updating the equivalent window segmentation is further expanded.

[0143] It can be understood that the segmentation of the updated equivalent window is proposed based on the unevenness of the performance decay of the device in the working period. The pollutant load borne by the device during use is often not a stable input, but dynamically fluctuates with the turbidity, sand content, particle size distribution and dissolved solid content of the water body. These fluctuations make the device performance decline have a phased feature. If the entire working period is directly taken as a single calculation window, the impurity accumulation rate in different stages will be averaged, causing a significant deviation between the prediction result and the actual decay performance.

[0144] In the present embodiment, the aforementioned calculated updated equivalent window can be understood as the residual filtration capacity of the device after the filtration behavior in the present period in theory.

[0145] It is easy to understand that the residual filtration capacity is a quantitative index obtained by comprehensively considering the impurity load borne by the device in the present period, the change of the pore structure of the filtration medium and the pressure drop generated when the water flow passes through the device. Since the device does not directly appear phenomena such as pore blockage, reduction of effective filtration area and increase of internal flow channel resistance during filtration, if this situation is not dynamically corrected, it will often cause distortion of life prediction, and may cause premature failure or over-conservative replacement of the device. Therefore, the updated updated equivalent window is essentially a mathematical characterization of the true residual filtration potential of the device under the current water quality environment, which can more accurately reflect the working state of the device.

[0146] Reference Figure 5 , Figure 5 The segmentation process diagram of the updated equivalent window provided by the embodiment of the present application is shown.

[0147] In one example, the state information further includes an inlet flow rate, the updated equivalent window is divided into a first equivalent window and a second equivalent window by monitoring the working state change curve of the device in the current working period, comprising:

[0148] S2.3: Obtain the working state change curve of the device in the current working period, and compare the working state change curve with a reference cleaning curve to obtain a candidate dividing point, wherein the reference cleaning curve is calculated according to the current equivalent window and the inlet flow rate;

[0149] Specifically, the working state of the device is different in different stages, especially between the middle and late stages of use, the pores of the filtration medium gradually shrink, causing the working state change curve to deviate from the reference cleaning curve of the ideal state. If the reference cleaning curve is not introduced as a reference, it is difficult to accurately determine the time when the performance of the device occurs inflection point, so that it is difficult to reasonably segment the updated equivalent window.

[0150] In the embodiment, the water outflow data of the current working period is collected in real time, and a change curve is drawn, and then a reference cleaning curve is generated according to the set water inflow and the theoretical water outflow performance of the device under the initial update equivalent window. The reference cleaning curve can reflect the flow attenuation trend of the device under ideal cleaning conditions. When the actual working state change curve deviates from the reference cleaning curve, the deviation points are marked as candidate boundary points. The time or working interval corresponding to the candidate boundary points often means that the filtering resistance of the device changes, which becomes a reference for subsequent division of the first and second equivalent windows.

[0151] Further, in the process of comparing the working state change curve and the reference cleaning curve, the embodiment introduces a double-pointer screening method to determine the candidate boundary points. Single curve comparison is often disturbed by local fluctuations. If the point with the largest difference on the adjacent curve is directly taken as the candidate boundary point, false boundary points are likely to occur. Through the double-pointer form, the region with persistent deviation characteristics can be more stably screened in the global range.

[0152] Reference Figure 6 , Figure 6 The principle schematic diagram of the candidate boundary point screening provided by the embodiment is shown.

[0153] Figure 6 The deviation of the working state change curve (dashed line) from the reference cleaning curve (solid line) in a working period is shown, where the horizontal axis represents time and the vertical axis represents the instantaneous water outflow at the corresponding time.

[0154] It can be understood that, due to the fluctuation of the pollution rate of the device in actual use, the working state change curve presents a stepped or sudden drop trend compared with the ideal reference cleaning curve.

[0155] Figure 6 Further, seven candidate time windows are shown, and the first pointer and the second pointer are set to advance along the two curves. Through the differential advancement of the double pointers, the local residual between the two curves can be calculated segment by segment, and the candidate boundary points are marked at the positions where the residual change rate reaches the preset threshold. Figure 6 In the embodiment, two candidate boundary points, candidate boundary point one and candidate boundary point two, are included.

[0156] In one example, the working state change curve is compared with the reference cleaning curve to obtain the candidate boundary points, including:

[0157] S2.3.1: configuring a first pointer and a second pointer in a time axis corresponding to the working state change curve and the reference cleaning curve, wherein the first pointer is used to traverse the working state change curve, and the second pointer is used to traverse the reference cleaning curve;

[0158] Specifically, when establishing the time axis correspondence between the working state change curve and the reference cleaning curve, the first pointer and the second pointer are introduced to enable dynamic comparison of the two curves in the same time reference system. If only relying on overall statistical characteristics or comparison of discrete points, it is easy to lose the capture of curve details due to local fluctuations in the device working process. Through the configuration of the double pointers, point-by-point traversal and synchronous / asynchronous advancement of the two curves can be ensured, so that the flow difference in any time period can be detected in time.

[0159] S2.3.2: advancing the first pointer and the second pointer in a preset candidate time window, and calculating a time stretching factor and a stability index in the corresponding candidate time window according to the advancing mode, wherein the stability index includes variance and range of the candidate time window;

[0160] In this embodiment, by setting the candidate time window on the time axis, the risk of misjudgment caused by excessive sensitivity to single-point data can be avoided.

[0161] It can be understood that the role of the candidate time window is to limit the advancing process of the pointers within a certain interval, so as to comprehensively evaluate the curve difference within the interval. The time stretching factor is determined by comparing the change speed and change amplitude of the actual water flow curve and the reference cleaning curve within the candidate time window, and its significance lies in depicting whether there is a significant time misalignment or delay between the actual curve and the ideal curve. If the time stretching factor is large, it means that the deviation between the actual curve and the ideal curve in the window is persistent rather than instantaneous, and such persistent deviation often corresponds to the real attenuation of the device. The stability index measures the fluctuation degree of the data by calculating the variance and range of the candidate time window, so as to judge whether the curve performance in the interval is smooth transition or sharp fluctuation. Through the combination of the time stretching factor and the stability index, the persistence and fluctuation characteristics of the offset can be considered within the candidate time window, so as to ensure that the selected candidate dividing point has stable physical meaning, rather than just a casual result at the data level.

[0162] In some optional embodiments, the advancing mode includes: the first pointer and the second pointer are synchronously advanced; the first pointer is fast-forwarded and the second pointer is slow-forwarded; the first pointer is slow-forwarded and the second pointer is fast-forwarded; wherein the fast-forwarding and the slow-forwarding correspond to the advancing frequency of the corresponding pointer, and the advancing frequency is calculated according to the average slope of the working state change curve and the reference cleaning curve in the corresponding candidate time window.

[0163] It can be understood that the fast-forwarding or the slow-forwarding does not change the physical time axis, but refers to that different sample step sizes and traversal densities are respectively used for the working state change curve and the reference cleaning curve in the same real time sequence of the candidate time window to establish local registration.

[0164] Specifically, the candidate time window is fixed as a real time interval, and the first pointer and the second pointer are both monotonically advanced in the interval and do not back off; when the working state change curve shows a steeper average downward trend in the time window, to avoid misjudging that the state progresses faster as time misalignment, the first pointer uses a smaller sampling interval and the second pointer uses a larger sampling interval in the time window, so that the two curves are elastically corresponded in one-to-many or many-to-one in morphological evolution without changing the real time coordinate; conversely, when the average slope of the working state change curve in the time window is less than the average slope of the reference cleaning curve, the second pointer increases the traversal frequency and the first pointer reduces the traversal frequency; if the robust average slopes of the two curves in the time window are approximately consistent, the two pointers are synchronously advanced, that is, the same sample step size is used to advance in the same real time step. The above advancing frequency is adaptively calculated from the robust average slopes of the two curves in the candidate time window, and is limited by the constraints of monotonous advancing and non-back-off, to avoid pseudo-differences caused by jump point alignment.

[0165] In the embodiment, the fast-forwarding and the slow-forwarding are realized by adjusting the advancing frequency of the pointer, and the advancing frequency is not a fixed value but is dynamically calculated in combination with the average slope of the curve in the candidate time window. Specifically, the average slope of the curve reflects the rate of the water flow rate changing with time, when the average slope of the actual working state change curve in the candidate time window is greater than the average slope of the reference cleaning curve, it indicates that the actual flow rate decreases faster, and the advancing frequency of the first pointer needs to be correspondingly reduced; when the average slope of the actual working state change curve is less than the average slope of the reference cleaning curve, the advancing frequency of the second pointer needs to be reduced. Through this adaptive adjustment mode based on the slope, the two curves in the candidate time window can be more reasonably matched in the morphology, so that in the subsequent candidate division point extraction, the error accumulation caused by curve misalignment can be avoided.

[0166] Those skilled in the art can understand that the average slope of the foregoing curve can be obtained by windowed extremum statistics, piecewise linear fitting, or median trend estimation; the upper and lower limits of the push frequency and the step size can be quantified according to the sampling period, the MCU computing power, and the noise level, and only need to meet the condition of establishing a monotonic and non-backtracking local registration within a fixed real time window, and the present application does not make more limitations.

[0167] S2.3.3: When the time scaling factor is greater than or equal to a preset time scaling threshold, and the offset value of the stability index and the tolerance band is greater than a preset deviation threshold, the average time of the corresponding candidate time window is taken as a candidate division point, wherein the tolerance band is calculated according to the minimum cleaning volume of the reference cleaning curve;

[0168] Further, in the final determination process of the candidate division point, the embodiment is screened by two aspects to ensure the effectiveness of the division point:

[0169] On the one hand, the time scaling factor is required to exceed the time scaling threshold, so as to verify that the difference between the actual working state change curve and the reference cleaning curve is continuous and significant enough;

[0170] On the other hand, the offset of the stability index relative to the tolerance band is required to exceed the preset deviation threshold, so as to ensure that the offset is not a normal fluctuation in the reference range.

[0171] The tolerance band is set according to the minimum cleaning volume of the reference cleaning curve, which provides a reasonable tolerance interval for the slight offset between the curves, so as to avoid misjudgment of sensor noise or slight fluctuation of water pressure as device performance degradation. In this way, the average time of the candidate time window is selected as the candidate division point, which can smooth out the local jitter inside the window and ensure that the final division point is closer to the real turning point of the device performance state.

[0172] S2.4: For each candidate division point, the working state change curve is resampled in the neighborhood of the candidate division point, and the hysteresis review is performed according to the resampling result, to obtain a final division point;

[0173] Specifically, simply comparing with the reference cleaning curve is prone to errors, especially in the case of instantaneous water fluctuation of the user or jitter in sensor measurement, a single offset point may not represent the real change of the device performance. Therefore, this step resamples the data in the neighborhood of the candidate division point, that is, the local trend of the working state change curve is recalculated within a certain range before and after the candidate division point with higher sampling accuracy. In this way, occasional noise points can be filtered out, and a more smooth and representative local curve can be obtained.

[0174] In this embodiment, after resampling, a hysteresis back-sweeping review method is adopted, that is, not only the curve change before the candidate boundary point is investigated, but also the curve trend after the candidate boundary point is taken into account. If the candidate boundary point before and after it shows consistent trend changes, for example, the flow rate continues to increase or the attenuation slope suddenly changes, it can be confirmed that the candidate boundary point is a real performance turning point, not a false signal caused by noise. The finally screened candidate boundary point is used as the basis for dividing the first and second equivalent windows. In this way, the accuracy of the boundary point can be significantly improved, the false division caused by single-point anomalies can be reduced, and the stability and reliability of the subsequent life prediction model can be ensured.

[0175] In one example, the resampling of the working state change curve in the neighborhood of the candidate boundary point, and the hysteresis back-sweeping review according to the resampling result, comprises:

[0176] S2.4.1: Calculate the cumulative volume according to the working state change curve, resample the working state change curve corresponding to the candidate boundary point neighborhood according to the cumulative volume, and obtain a candidate pair corresponding to the uplink segment and the backhaul segment, wherein the volume span of the candidate pair is greater than or equal to the minimum cleaning volume;

[0177] Specifically, if the original time series in the neighborhood of the candidate boundary point is directly compared, it is easy to be affected by the sampling beat, short-term water impact and local noise, resulting in incomparability between different neighborhoods. In order to obtain a scale directly related to the filtration process, the cumulative volume V is first obtained by integrating the water flow with respect to time in the neighborhood, and the curve is resampled with a fixed volume step to form a Q(V) expression representing the relationship between the water flow Q and the cumulative volume V. Using volume as the resampling reference, the time stretching difference caused by different water rhythms and different sampling intervals can be converted into a unified processing scale, so that each candidate boundary point neighborhood is horizontally comparable and can directly reflect the instantaneous response within the same water volume segment.

[0178] In this embodiment, in order to reliably extract the two natural segments corresponding to the uplink segment and the backhaul segment from Q(V), first, the resampling result is locally trend-stripped in the volume neighborhood of the candidate boundary point:

[0179] The neighborhood baseline is obtained by moving the median, low-pass smoothing or piecewise linear fitting, and the fluctuation is calculated to highlight the positive and negative swings of the natural disturbance relative to the baseline.

[0180] Subsequently, the discrete first-order sign sequence of the fluctuation is scanned to identify a continuously non-negative oriented segment as an uplink segment candidate, and the continuously non-positive oriented segment immediately after it as a backhaul segment candidate; or when the orientations are opposite, they are paired according to the mirror rule to ensure that the two segments are sequentially followed in the volume axis and have opposite directions.

[0181] Further, to ensure physical meaning and numerical stability, only the candidate pair satisfying the following pair-wise constraints is reserved:

[0182] The first constraint is that the volume span of each segment is not less than the minimum cleaning volume threshold, and the volume span ratio is within the preset range, to avoid distortion caused by extremely short segments or excessive imbalance;

[0183] The second constraint is that the amplitude difference between the endpoints of the two segments has sufficient discrimination compared to the local noise level, which can be set by the neighborhood robust variance or quantile range;

[0184] The third constraint is that the monotonicity within the segment is checked by a small amount of tolerance (allowing individual sampling points to be slightly reversed), and if necessary, morphological opening / closing operation is performed on the sign sequence to remove burrs.

[0185] The uplink segment and the return segment that meet the above conditions constitute a candidate pair. If there are multiple feasible pairings in the same neighborhood, the pairing with the most sufficient volume coverage, the highest amplitude energy indicator, or the closest position to the candidate division point is preferred, and the rest are entered as backups for subsequent weighting.

[0186] The specific processing method is:

[0187] In the neighborhood interval before and after the candidate division point, the flow curve is resampled according to a fixed volume step, obtaining an uplink segment corresponding to the device performance decline, and a return segment corresponding to the recovery after cleaning.

[0188] The two curves generated in this way not only eliminate the influence of uneven time distribution in the original data, but also reflect the symmetry of device attenuation and recovery under the same volume scale. In addition, the candidate pair needs to meet the condition that its volume span is greater than or equal to the minimum cleaning volume, to ensure that the obtained uplink segment and return segment curve has sufficient coverage range, avoiding the result being not representative due to too small sample interval, thereby ensuring the robustness and accuracy of subsequent calculation.

[0189] S2.4.2: Align and interpolate the uplink segment and the return segment on the volume axis, calculate the flow difference of the uplink segment and the return segment under the consistent volume grid, and obtain the corresponding quasi-hysteresis area by integration;

[0190] Specifically, after resampling, the uplink segment and the return segment need to be projected onto a unified volume grid. Since the two curves come from different time windows, if directly compared, there will be problems of inconsistent number of sampling points and misaligned positions, so interpolation and alignment processing need to be performed on the volume axis to make the two curves comparable under the same volume coordinates.

[0191] In this embodiment, the two curves are first reconstructed by interpolation at the same volume resolution, and then the instantaneous flow difference between them is calculated point by point under a consistent volume grid. By integrating the difference curve, a numerical index can be obtained, which represents the area enclosed by the actual flow curve and the ideal return curve in the neighborhood of the candidate boundary point, i.e., the pseudo-hysteresis area. The larger the pseudo-hysteresis area, the more significant the difference between the decrease and increase in flow within the neighborhood, and the more likely it corresponds to the actual cleanliness boundary point of the device. Enhancing the robustness of the judgment by energy quantification of the entire range can avoid the problem of single-point difference being affected by noise or local anomalies. The principle is to use the hysteresis phenomenon of flow with volume as a significant feature of the device performance state transition, thereby providing a quantitative basis for the final selection of candidate boundary points.

[0192] S2.4.3: Sort the pseudo-hysteresis areas of each candidate boundary point in descending order, and output the first candidate boundary point as the final boundary point. The descending order sorting also includes weighting the pseudo-hysteresis areas according to the area of ​​the neighborhood of the candidate boundary point.

[0193] Specifically, after obtaining the pseudo-hysteresis area of ​​all candidate boundary points, this embodiment sorts these candidate boundary points in descending order and selects the candidate boundary point with the largest area as the final output boundary point.

[0194] It is understandable that the hysteresis characteristics of the neighborhoods of different candidate boundary points vary. The larger the area value, the stronger the difference between the up-travel and down-travel curves in that neighborhood, and the more it represents a significant change in the working state of the device before and after that position, thus possessing sufficiency and reliability as the final boundary point.

[0195] Furthermore, to avoid amplifying local noise due to relying solely on area values, the ranking process also uses the overall area of ​​the neighborhood of each candidate boundary point as a weighting factor to assign a weighted value to the hysteresis area. The weighted ranking result not only considers the strength of local differences but also balances the size of the neighborhood coverage, ensuring that the finally selected boundary points possess both local abrupt change characteristics and sufficient representativeness on the global curve. This strategy effectively improves the robustness of boundary point identification, avoiding misjudgments caused by local anomalies or short-term fluctuations. Through this verification mechanism, the most reasonable cleaning time can be automatically identified among multiple candidate points, improving the system's intelligence level and user experience.

[0196] S2.5: Based on the position of the final dividing point on the working state change curve, the updated equivalent window is divided to obtain a first equivalent window and a second equivalent window, and the first equivalent window is assigned a value based on the current equivalent window;

[0197] In the present embodiment, the final determined demarcation point is used as an anchor point for dividing the updating equivalent window, i.e., dividing the entire window into two parts: a first equivalent window and a second equivalent window.

[0198] It can be understood that, in the defined conditions of the present application, the device often has two phased performance in a working cycle, the filtering effect of the former stage is relatively stable, and the water flow decreases slowly, while in the latter stage, the device decay rate will significantly accelerate due to the rapid accumulation of impurities. If the window is not divided, the life prediction model can only be calculated based on the overall curve, which is easy to underestimate the early life or overestimate the late life, and finally leads to prediction deviation. Therefore, by dividing the updating equivalent window, the remaining filtering capacity of the device in different stages can be described more finely.

[0199] In specific operation, according to the position of the final demarcation point on the working state change curve, the former half is defined as the first equivalent window, and the latter half is defined as the second equivalent window, and the first equivalent window is valued according to the current equivalent window, so as to serve as a reference paragraph for subsequent life prediction. Further, the second equivalent window is used as an important input for subsequent decay curve prediction to reflect the performance characteristics of the device in the approach to failure stage. Through this division and valuation method, not only the segmented modeling of the device performance is realized, but also the life prediction process can combine the actual performance in different stages, so as to obtain a more actual prediction result.

[0200] Next, the principle of updating equivalent window correction of the method of the present application is further expanded.

[0201] Reference Figure 7 , Figure 7 The updating equivalent window correction principle diagram provided by the embodiment of the present application.

[0202] In one example, the updating equivalent window is corrected according to the first equivalent window and the second equivalent window, comprising:

[0203] S3.1: calculating the volume difference between the first equivalent window and the second equivalent window;

[0204] In the present embodiment, the cumulative volume start and end points of the two updating equivalent windows are compared, and the corresponding volume span difference is calculated, so as to obtain an index capable of representing the consistency degree of the two. Through this processing method, not only the numerical comparison of different algorithm results can be directly compared, but also an objective basis is provided for whether further correction is needed.

[0205] S3.2: if the volume difference is less than or equal to a preset tolerance threshold, the second equivalent window is used as the corrected updating equivalent window;

[0206] In the embodiment, when the calculated volume difference is within the allowed tolerance range, it indicates that the difference between the first equivalent window and the second equivalent window is small enough, and both can be considered consistent in a statistical sense. In this case, no additional complex correction operation is needed, and the second equivalent window can be directly used.

[0207] The second equivalent window is selected because it can more comprehensively reflect the characteristic information of device performance attenuation and recovery, and its physical interpretation is more consistent with the real cleaning process. In this way, the additional calculation overhead can be effectively reduced, and the distortion introduced by excessive correction of the curve can be avoided. At the same time, this determination logic can also improve the robustness of the algorithm, so that the system can maintain a relatively stable output under different working conditions. It ensures that the simplified path is used within the error range, thereby improving the calculation efficiency and practicality, and reducing the complexity of subsequent processing.

[0208] S3.3: If the volume difference is greater than the preset tolerance threshold, a correction factor is calculated according to the volume difference, and the updated equivalent window is weighted and assigned according to the correction factor to obtain a corrected updated equivalent window;

[0209] In another case, if the volume difference exceeds the tolerance threshold, it indicates that there is a large deviation between the first equivalent window and the second equivalent window. If either of them is directly used, it will cause the final updated equivalent window to deviate from the real working law.

[0210] To solve this problem, the concept of a correction factor is introduced in the embodiment. The correction factor is calculated according to the size of the volume difference, and its value can reflect the relative credibility between the two candidate windows. When the volume difference is large, the correction factor tends to increase the weight of the more stable window, and when the volume difference is relatively small but still exceeds the tolerance range, the two windows are relatively balanced. The specific operation is to combine the correction factor with the updated equivalent window for weighted assignment, thereby obtaining a corrected window that comprehensively considers the characteristics of the first equivalent window and the second equivalent window. This can avoid the deviation risk caused by relying entirely on a single result, and also makes full use of the complementary characteristics between different algorithms to form an updated equivalent window that is closer to the real working condition.

[0211] In an optional embodiment, the correction of the updated equivalent window also includes modeling and compensating for the self-cleaning process of the device after the end of each working period. Specifically, when the device stops working or enters a standby state, the internal part will have a certain degree of self-recovery due to mechanisms such as fluid pressure drop, residual impurity settlement, or reverse osmosis flushing. This process will cause a small upward trend in the working state change curve before the start of the next working period. If this feature is ignored during the correction of the updated equivalent window, it will cause a systematic deviation between the cumulative volume and the actual decay trajectory, especially in a multi-period continuous working scenario. This deviation will be continuously accumulated, ultimately affecting the accuracy of the updated equivalent window.

[0212] In this embodiment, first, the typical self-cleaning upward amplitude is extracted according to the flow difference between the end point of the device in the historical working data and the start point of the next period, and the self-cleaning recovery factor related to the cumulative volume or working time is established. Then, when updating the equivalent window, the self-cleaning recovery factor is introduced into the correction process to dynamically adjust the upper and lower boundaries of the candidate time window. For example, when it is detected that the candidate time window spans the period switching point, the self-cleaning recovery amount is introduced in the corresponding cumulative volume range of the interval, thereby offsetting the window boundary. In this way, the problem of volume calculation being too small or too large due to the self-cleaning process of the device can be avoided, and the updated equivalent window can more truly reflect the decay and recovery rules of the device in the actual working environment.

[0213] Further, the stability of the self-cleaning recovery factor can be combined in the correction process to differentiate the windows of different periods. For device models or working conditions with obvious self-cleaning recovery characteristics, the cleaning correction factor weight is appropriately increased to fully compensate for the volume difference caused by period switching; and for the case where the recovery effect is weak, a lower weight is used to avoid excessive correction. In this way, not only the adaptability of the correction method to different devices and working conditions is improved, but also the updated updated equivalent window can reflect the true working state of the device without deviating from the original data characteristics due to excessive correction. Through this improvement, the final updated equivalent window has better stability and universality in a multi-period working environment, thereby providing more reliable input data for subsequent device life prediction and maintenance strategies.

[0214] In one example, comparing the current equivalent window with the device life table obtains a device life, including:

[0215] S4.1: inputting the current equivalent window into the device life table to retrieve a corresponding life reference value;

[0216] S4.2: determining the remaining life of the device according to the life reference value, and outputting the device life.

[0217] In one example, the application provides a device life prediction system based on working state change curve, which is applied to MCU of offline fine filtration system, comprising a data acquisition module, an operation processing module and a life evaluation module, wherein:

[0218] The data acquisition module is configured to acquire state information of water sample to be filtered and working state change curve of the device in the current working period, and transmit the data to the operation processing module;

[0219] The operation processing module is arranged in the MCU, and is configured to maintain and update equivalent window and device life table, and perform update, correction, segmentation of the updated equivalent window and calculation of predicted attenuation amount;

[0220] The life evaluation module is configured to compare the corrected updated equivalent window with the device life table to obtain the remaining life of the device.

[0221] Although the embodiments of the application have been shown and described above, it should be understood by those skilled in the art that the above embodiments are exemplary and should not be construed as limiting the application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the application.

Claims

1. A device lifespan prediction method based on operating state change curves, applied to an MCU in an offline fine filtration system, characterized in that, The MCU stores a device lifetime table and a current equivalent window, wherein the current equivalent window is an updated equivalent window from the previous operating cycle of the device, and the updated equivalent window is a representation index obtained by mapping the state information of the device's working object through equivalent processing. The method is executed by the MCU after the current operating cycle ends, and specifically includes: Obtain the status information of the device's working object during the current working cycle; Based on the status information, an update equivalent window is determined. The update equivalent window is divided into a first equivalent window and a second equivalent window by monitoring the working status change curve of the device in the current working cycle. The size of the first equivalent window is the same as that of the current equivalent window, and the size of the second equivalent window is the same as that of the update equivalent window. The updated equivalent window is modified and replaced with the current equivalent window based on the first and second equivalent windows, including: Calculate the volume difference between the first equivalent window and the second equivalent window; If the volume difference is less than or equal to a preset tolerance threshold, the second equivalent window is used as the corrected updated equivalent window. If the volume difference is greater than the preset tolerance threshold, a correction factor is calculated based on the volume difference, and the updated equivalent window is weighted and assigned a value based on the correction factor to obtain the corrected updated equivalent window. The device lifespan is obtained by comparing the revised updated equivalent window with the device lifespan table.

2. The device life prediction method based on the operating state change curve according to claim 1, characterized in that, Based on the status information, determine the update equivalent window, including: Based on the status information of the working object of the device, the impurity accumulation rate of the current working cycle is obtained, wherein the status information includes at least one of turbidity value, sand content, particle size distribution and dissolved solids content; Based on the impurity accumulation rate and the current equivalent window, the predicted decay amount is calculated, and the predicted decay amount is applied to the current equivalent window to obtain an updated equivalent window.

3. The device life prediction method based on the operating state change curve according to claim 2, characterized in that, Based on the impurity accumulation rate and the current equivalent window, the predicted decay is calculated, including: The current equivalent window is divided into multiple continuous sub-windows, wherein the length of the continuous sub-windows is calculated based on the changing trend of the impurity accumulation rate; For each consecutive sub-window, the attenuation contribution value is calculated based on the corresponding impurity accumulation rate, and the attenuation contribution values ​​are summed to obtain the first attenuation curve. Align the first attenuation curve with the predicted working state change curve within the current working cycle to obtain the hysteresis compensation factor. The first attenuation curve is compensated according to the hysteresis compensation factor to obtain the second attenuation curve; The predicted attenuation amount is determined based on the descending slope of the second attenuation curve and the cumulative attenuation area.

4. The device life prediction method based on the operating state change curve according to claim 1, characterized in that, The status information also includes the influent flow rate. The update equivalent window is divided into a first equivalent window and a second equivalent window by monitoring the operating status change curve of the monitoring device within the current working cycle, including: The working status change curve of the device in the current working cycle is obtained, and the working status change curve is compared with the reference cleaning curve to obtain the candidate boundary point, wherein the reference cleaning curve is calculated based on the current equivalent window and the inlet water flow rate. For each candidate boundary point, the working state change curve is resampled in the neighborhood of the candidate boundary point, and hysteresis backscan verification is performed based on the resampling results to obtain the final boundary point. Based on the position of the final dividing point on the working state change curve, the updated equivalent window is divided to obtain a first equivalent window and a second equivalent window, and the first equivalent window is assigned a value based on the current equivalent window.

5. The device life prediction method based on the operating state change curve according to claim 4, characterized in that, The working state change curve is compared with the baseline cleaning curve to obtain candidate boundary points, including: A first pointer and a second pointer are configured in the time axis corresponding to the working state change curve and the reference cleaning curve, wherein the first pointer is used to traverse the working state change curve and the second pointer is used to traverse the reference cleaning curve; The first and second pointers are advanced within a preset candidate time window, and the time scaling factor and stability index within the corresponding candidate time window are calculated according to the advancement mode. The stability index includes the variance and range of the candidate time window. When the time scaling factor is greater than or equal to the preset time scaling threshold, and the offset value between the stability index and the tolerance band is greater than the preset deviation threshold, the average time of the corresponding candidate time window is used as the candidate boundary point, wherein the tolerance band is calculated based on the minimum cleaning volume of the benchmark cleaning curve.

6. The device life prediction method based on the operating state change curve according to claim 5, characterized in that, The propulsion mode includes: The first pointer advances synchronously with the second pointer; The first pointer advances fast while the second pointer advances slowly. The first pointer advances slowly while the second pointer advances quickly. The advance frequency of the pointers corresponding to the fast forward and the slow forward is calculated based on the average slope of the working state change curve and the reference cleaning curve in the corresponding candidate time window.

7. The device life prediction method based on the operating state change curve according to claim 4, characterized in that, The step of resampling the working state change curve within the neighborhood of the candidate boundary point and performing hysteresis scan verification based on the resampling results includes: The cumulative volume is calculated based on the working state change curve. The working state change curves corresponding to the neighborhood of the candidate boundary point are resampled based on the cumulative volume to obtain candidate pairs corresponding to the uplink and downlink segments. The volume span of the candidate pairs is greater than or equal to the minimum cleaning volume. Align and interpolate the uplink and downlink segments on the volume axis, calculate the flow difference between the uplink and downlink segments under a consistent volume grid, and obtain the corresponding pseudohysteresis area by integration. The pseudohysteresis areas of each candidate boundary point are sorted in descending order, and the first candidate boundary point is output as the final boundary point. The descending sorting also includes weighting the pseudohysteresis areas based on the area of ​​the neighborhood of the candidate boundary point.

8. The device life prediction method based on the operating state change curve according to claim 1, characterized in that, The device lifetime is obtained by comparing the revised updated equivalent window with the device lifetime table, including: The corrected updated equivalent window is input into the device life table to retrieve the corresponding life reference value; The remaining lifespan of the device is determined based on the lifespan reference value, and the device lifespan is output.

9. A device life prediction system based on operating state change curves, used to implement the device life prediction method based on operating state change curves as described in any one of claims 1-8, characterized in that, The system is applied to the MCU of an offline fine filtration system. The MCU stores a device lifespan table and a current equivalent window, where the current equivalent window is an updated equivalent window from the previous operating cycle of the device, and the updated equivalent window is a characterization index obtained by mapping the status information of the device's working object through equivalent processing. The system includes a data acquisition module, a processing module, and a lifespan assessment module, wherein: The data acquisition module is used to acquire the status information of the water sample to be filtered in the current working cycle, as well as the working status change curve of the device in the current working cycle, and transmit the data to the calculation and processing module. The computational processing module, located in the MCU, is used to maintain and update the equivalent window and the device lifetime table, and to update, correct, divide, and predict the decay amount of the equivalent window. The lifespan assessment module is used to compare the corrected updated equivalent window with the device lifespan table to obtain the device lifespan.

Images