Large-scale water pump unit stuffing box water leakage monitoring and automatic adjusting system

Abstract

The invention discloses a water leakage monitoring and automatic adjusting system for a stuffing box of a large water pump unit. According to the system, discrete water leakage behaviors formed by leakage liquid of a stuffing box in a preset sampling period are continuously collected through a collection unit, and original water leakage behavior data corresponding to water leakage behavior generation characteristics in unit time are obtained; and a standardized water leakage index is generated by the characterization unit. A target water leakage interval is constructed through a construction unit in combination with constraint parameters related to the operation state of the large water pump unit; the generation unit forms a water leakage deviation description quantity representing the water leakage deviation direction and deviation amplitude, and generates gland adjustment strategy parameters; the control unit converts the gland adjusting strategy parameters into control instructions capable of being recognized by a stuffing box gland executing mechanism so as to drive the executing mechanism to automatically adjust the tightness state of the gland. Quantitative monitoring and self-adaptive adjustment of the water leakage amount of the stuffing box are achieved, manual dependence is reduced, and the operation safety and reliability of a water pump unit are improved.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology, and in particular to a system for monitoring and automatically adjusting leakage of stuffing box water in large water pump units. Background Technology

[0002] During the long-term operation of large water pump units, stuffing boxes, as a crucial component for pump shaft sealing and lubrication, are widely used in various water intake pumping stations, drainage pumping stations, and industrial circulating water systems. In existing technologies, stuffing boxes typically rely on the contact pressure between the packing and the shaft sleeve to achieve sealing, while allowing a certain amount of liquid leakage to balance cooling and lubrication needs. Regarding the management of stuffing box operating status, most pumping stations still rely on manual inspections, with maintenance personnel observing leakage during equipment operation and making appropriate adjustments to the gland based on experience. While some scenarios introduce simple sensors or alarm devices, their main function remains focused on status indication or anomaly alerts, lacking a systematic automatic adjustment mechanism.

[0003] Under current technological conditions, the judgment and adjustment of stuffing box leakage status are highly dependent on manual experience and on-site conditions, making it difficult to achieve continuous, stable, and quantifiable management. On the one hand, different operating conditions, load changes, and environmental conditions can affect the leakage performance of the stuffing box, making fixed thresholds or experience-based judgments unsuitable for long-term application. On the other hand, manual adjustment is prone to response lag, inconsistent adjustment ranges, and over-tightening or under-tightening, which can easily lead to abnormal wear of the packing, damage to the bushing, or deterioration of lubrication conditions, thereby increasing the frequency of equipment maintenance and affecting the operational reliability of the pump unit.

[0004] Therefore, it is necessary to develop a technical solution that can adapt to the operating characteristics of large water pump units and achieve refined management of stuffing box leakage. Summary of the Invention

[0005] This application provides a system for monitoring and automatically adjusting leakage in the stuffing box of a large water pump unit, so as to improve the safety and reliability of the water pump unit operation.

[0006] This application provides a system for monitoring and automatically adjusting leakage in the stuffing box of a large water pump unit, including: The acquisition unit is used to continuously acquire discrete leakage behavior of the liquid leaking from the stuffing box within a preset sampling period, and obtain raw leakage behavior data corresponding to the leakage behavior characteristics per unit time. The characterization unit is used to perform time normalization and dimensional mapping on the original leakage behavior data to generate a standardized leakage index that characterizes the current leakage status of the stuffing box. The construction unit is used to construct the target leakage range corresponding to the standardized leakage index based on the constraint parameters related to the operating status of the large water pump unit, and to determine the upper and lower safety thresholds corresponding to the target leakage range. The generation unit is used to analyze the relationship between the standardized leakage index and the target leakage range, generate a leakage deviation description quantity that represents the direction and magnitude of the deviation, and generate gland adjustment strategy parameters based on the leakage deviation description quantity under the premise of satisfying the packing cooling and lubrication constraints. The control unit is used to convert the gland adjustment strategy parameters into control commands that can be recognized by the stuffing gland gland actuator, and drive the stuffing gland gland actuator to automatically adjust the tightness of the gland.

[0007] The beneficial effects of this application mainly include: (1) By continuously collecting discrete leakage behavior formed by liquid leakage from the stuffing box and performing time normalization and dimensional mapping on the original leakage behavior data, this application transforms the leakage phenomenon that originally relied on manual visual inspection and experience judgment into a calculable and comparable standardized leakage index, so that the leakage state of the stuffing box can be uniformly characterized and continuously monitored, providing a stable data basis for subsequent adjustment. (2) This application constructs the target leakage range based on the constraint parameters related to the operating status of large water pump units, and simultaneously determines the upper and lower safety thresholds, so that leakage control no longer relies on a single fixed threshold, but can be dynamically matched with changes in operating status, thereby maintaining the rationality and consistency of adjustment results under different loads, speeds or operating conditions. (3) In the process of generating gland adjustment strategy parameters, packing cooling and lubrication constraints are introduced to comprehensively analyze the direction and magnitude of leakage deviation, avoiding excessive gland tightening due to simply pursuing a reduction in leakage, thereby effectively preventing dry friction, overheating or abnormal wear of the packing, and extending the service life of the packing and bushing. (4) The pressure cover adjustment strategy parameters are directly converted into control commands that can be recognized by the actuator through the control unit, so as to realize the automatic adjustment of the pressure cover tightness, reduce the need for frequent manual inspection and manual intervention, improve the timeliness and consistency of adjustment response, and help ensure the long-term stable operation of large water pump units and reduce operation and maintenance costs. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of a large water pump unit stuffing box leakage monitoring and automatic adjustment system provided in the first embodiment of this application. Detailed Implementation

[0009] Many specific details are set forth in the following description to provide a full understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this application; therefore, this application is not limited to the specific embodiments disclosed below.

[0010] The first embodiment of this application provides a system for monitoring and automatically adjusting leakage in the stuffing box of a large water pump unit. Please refer to... Figure 1 This figure is a schematic diagram of the first embodiment of this application. The following is in conjunction with... Figure 1 The first embodiment of this application provides a detailed description of a large water pump unit stuffing box leakage monitoring and automatic adjustment system.

[0011] The large water pump unit stuffing box leakage monitoring and automatic adjustment system includes a data acquisition unit 101, a characterization unit 102, a construction unit 103, a generation unit 104, and a control unit 105.

[0012] The acquisition unit 101 is used to continuously acquire discrete leakage behavior of the liquid leaking from the stuffing box within a preset sampling period, and obtain raw leakage behavior data corresponding to the leakage behavior characteristics per unit time.

[0013] The data acquisition unit 101 is used to continuously, stably, and repeatedly collect the leakage liquid generated by the stuffing box of a large water pump unit during operation. Its core function is to transform the leakage phenomenon, which is originally difficult to quantify and relies on human experience for judgment, into a source of basic data that can be processed by subsequent units. In this invention, "leakage liquid from the stuffing box" refers to the working medium that naturally seeps from the sealing area of ​​the stuffing box to the external environment under normal operating conditions of the water pump unit, due to the necessary gap between the packing and the pump shaft to meet cooling and lubrication requirements. This leakage behavior is not equivalent to fault leakage, but a permissible operating phenomenon with engineering significance.

[0014] The term "discrete leakage behavior" refers to leakage events that occur within a preset sampling period, manifesting as dripping, intermittent flow, or discontinuous streams, and are distinguishable in time. This invention does not simply view leakage as a continuous, stable flow rate, but rather breaks it down into a series of independent behavioral units identifiable on a time axis, such as the formation and detachment of a single droplet, the formation of a small water column within a short period, or multiple droplet behaviors occurring consecutively within a certain timeframe. Those skilled in the art should understand that this discretization method does not alter the physical leakage state, but rather aims to provide a more refined and stable description of the leakage phenomenon at the data level, thereby avoiding interference from instantaneous fluctuations in the judgment results.

[0015] The acquisition unit 101 can be implemented in various ways, as long as it can identify and record the aforementioned discrete leakage behaviors without affecting the normal structure and operation of the stuffing box. For example, photoelectric sensing elements can be arranged on the leakage collection path or natural dripping location outside the stuffing box, generating an effective trigger signal when a droplet passes through the sensing area; or, through capacitive, conductive, or weight sensing methods, a detectable physical quantity change can be formed when liquid drips into a preset acquisition area. The above implementation methods are merely examples and do not constitute a limitation of the present invention. The key is that the acquisition unit 101 can reliably distinguish leakage behaviors occurring at different time points and effectively record each behavior.

[0016] The "preset sampling period" refers to the time window during which the acquisition unit 101 statistically analyzes leakage behavior. Its length can be set according to the pump unit's specifications, speed, packing type, and operating conditions; for example, it can be set to 30 seconds, 1 minute, 5 minutes, or other fixed time lengths. This invention does not limit the specific value of the sampling period, but requires consistency in the sampling period under the same operating conditions to ensure comparability between results collected at different times. Within a complete sampling period, the acquisition unit 101 cumulatively records all discrete leakage behaviors and outputs the corresponding statistical results at the end of the sampling period.

[0017] The "raw leakage behavior data" ultimately obtained by the acquisition unit 101 refers to the data set corresponding to the characteristics of leakage behavior within a unit of time. This data can be specifically represented as the number of leakage behaviors, the number of trigger signals, or the equivalent event count value recorded within a sampling period. For example, if the sampling period is set to 60 seconds, and 120 valid droplet trigger signals are detected within this time period, the corresponding raw leakage behavior data can be represented as "120 discrete leakage behaviors occurred within 60 seconds". In another implementation, if the acquisition unit combines continuous dripping into short-term events for statistical analysis, the raw leakage behavior data can also be represented as "the number of leakage events occurring within a unit of time", as long as the data can stably reflect the occurrence characteristics of leakage behavior in the time dimension.

[0018] Furthermore, the acquisition unit is specifically used for: On the natural collection or dripping path of the leaking liquid in the stuffing box, the leakage phenomena, including dripping, intermittent flow or discontinuous streams of the leaking liquid, are detected in real time, and an initial trigger signal corresponding to each leakage phenomenon is generated and the occurrence time of the initial trigger signal is recorded. Based on the time interval between adjacent initial trigger signals, the validity of the initial trigger signal is determined. When the time interval between two adjacent initial trigger signals is greater than the preset minimum behavior interval threshold, the corresponding initial trigger signal is determined to be an independent discrete leakage behavior, and a corresponding valid leakage behavior identifier is generated. Within a preset sampling period, all valid leakage behavior identifiers that are determined to be independent discrete leakage behaviors are cumulatively counted to generate a behavior statistics result that represents the total number of discrete leakage behaviors occurring within the sampling period. Based on the behavioral statistics and the corresponding sampling period, raw leakage behavior data corresponding to the discrete leakage behavior occurrence characteristics per unit time are generated.

[0019] In this embodiment, the data acquisition unit is used to transform the originally continuous, random, and difficult-to-quantify leakage phenomenon during the operation of the stuffing box into basic data with clear time characteristics and statistical significance.

[0020] During the operation of large water pump units, a certain degree of liquid leakage is permissible at the stuffing box to meet the basic cooling and lubrication requirements of the packing. This leakage typically collects naturally along the external structure of the stuffing box, the outer edge of the gland, or a dedicated flow guide structure, and is discharged in the form of droplets, intermittent small-flow outflows, or short-term discontinuous streams. The acquisition unit is positioned along these natural collection or dripping paths. Its detection position does not participate in the stress or movement of the sealing structure itself; it only senses the leaked liquid, thus avoiding interference with the normal operation of the water pump unit. "Real-time detection" means that the acquisition unit is continuously operational during the operation of the water pump unit, responding immediately to each instance of leakage, rather than periodic sampling detection.

[0021] When leaked liquid passes through the detection path, the acquisition unit generates an initial trigger signal corresponding to each leak according to a pre-set detection principle. This initial trigger signal can be understood as the most basic record of "detecting a leak," and its form can be an electrical signal, a level change, a pulse signal, or an equivalent digital event identifier, but this invention does not limit the specific signal form. Simultaneously recorded with each initial trigger signal is the precise time of its occurrence, which is marked using a unified internal time base, such as a timestamp with millisecond precision or higher. In this way, the acquisition unit not only obtains the number of leaks but also the distribution information of each leak along the time axis.

[0022] Because leaked liquid from the stuffing box can be affected by factors such as equipment vibration, liquid splashing, water mist interference, or continuous dripping within a short period during actual operation, simply treating all initial trigger signals as independent leakage events could easily lead to an overestimation of the leakage severity. Therefore, after generating an initial trigger signal, the acquisition unit does not immediately recognize it as a valid leakage event. Instead, it determines its validity based on the time interval between adjacent initial trigger signals. The time interval referred to here is the difference between the timestamps of two adjacent initial trigger signals, which is obtained by directly calculating the difference between the two recorded timestamps.

[0023] The so-called "preset minimum behavior interval threshold" refers to the minimum time interval standard used to distinguish between independent leakage behavior and continuous trigger signals within the same leakage process. This threshold can be set according to the structural characteristics of the stuffing box, the physical properties of the medium, and field operating experience, for example, set to 0.2 seconds, 0.5 seconds, or 1 second. When the time interval between two adjacent initial trigger signals is greater than this minimum behavior interval threshold, the acquisition unit considers these two leakage phenomena to be independent in time, representing two independent leakage behaviors, and determines the latter initial trigger signal as an independent discrete leakage behavior, while generating a corresponding valid leakage behavior identifier. Conversely, if the time interval between two adjacent initial trigger signals is less than or equal to the threshold, it is considered to be a continuous manifestation within the same leakage process, and is not counted separately as a new discrete leakage behavior, thereby avoiding repeated counting of continuous dripping or short-term stream flow.

[0024] After completing the above validity determination, the acquisition unit performs cumulative statistics on all valid leakage behavior identifiers determined to be independent discrete leakage behaviors within a preset sampling period. The "preset sampling period" refers to a fixed time window used to statistically analyze leakage behavior occurrences; its length can be set according to management precision requirements, such as 30 seconds, 60 seconds, or longer. Within a complete sampling period, the acquisition unit only counts valid leakage behavior identifiers, no longer focusing on the total number of initial trigger signals, thus ensuring that the statistical results reflect the occurrence of independent leakage behaviors with engineering significance. This cumulative statistical result is the behavior statistics result, used to characterize the total number of discrete leakage behaviors occurring within the sampling period.

[0025] After obtaining the behavioral statistics, the data acquisition unit further processes the statistical results by combining them with the corresponding sampling period duration, performing time feature representation processing to form raw leakage behavior data. "Raw leakage behavior data" refers to data that reflects the characteristics of discrete leakage behavior occurring within a unit of time, and can be expressed as "the number of independent leakage behaviors occurring within a unit of time." For example, if the sampling period is set to 60 seconds, and 120 valid leakage behavior identifiers are accumulated within that period, the raw leakage behavior data can be expressed as 120 discrete leakage behaviors occurring per minute; similarly, if the sampling period is 30 seconds and 60 valid leakage behaviors are counted, it can also be converted to 120 discrete leakage behaviors occurring per minute. In this way, the raw leakage behavior data eliminates the influence of differences in sampling period length, providing a direct and clear input basis for subsequent time normalization and dimensional mapping processing by the characterization unit.

[0026] Through the above continuous and clear processing flow, the acquisition unit transforms the complex physical manifestations of the leaking liquid from the stuffing box into clearly defined discrete leakage behavior data, and further forms raw leakage behavior data that can reflect the characteristics of leakage behavior occurring per unit time.

[0027] Furthermore, the acquisition unit is also used for: Based on the stuffing box structure, operating vibration level, and leakage liquid flow characteristics, a minimum behavior interval threshold was determined to distinguish between independent leakage behavior and continuous leakage process. Multiple initial trigger signals generated consecutively within the minimum behavior interval threshold are merged. Initial trigger signals that satisfy the time interval being less than or equal to the minimum behavior interval threshold are merged into the same leakage process group, and a process trigger identifier corresponding to the leakage process group is generated. Based on the process trigger identifier, the number and duration of the initial trigger signals in each leakage process group are analyzed. When the duration of the leakage process group exceeds the preset minimum duration threshold, the corresponding leakage process group is determined to be a valid leakage process, and a corresponding candidate leakage behavior identifier is generated. Based on the time interval relationship between candidate leakage behavior identifiers and adjacent leakage process groups, the candidate leakage behavior identifiers are finally confirmed. When the time interval between two adjacent candidate leakage behavior identifiers is greater than the minimum behavior interval threshold, the candidate leakage behavior identifier is confirmed and output as a valid leakage behavior identifier corresponding to an independent discrete leakage behavior.

[0028] In this embodiment, the acquisition unit, in addition to acquiring the initial trigger signal of the leaking liquid, further undertakes the function of semantic hierarchical determination of the leakage behavior. Its purpose is to accurately identify independent discrete leakage behaviors with engineering significance from a large number of continuous, instantaneous or disturbed trigger signals, thereby providing a reliable data foundation for subsequent statistics and characterization.

[0029] In the specific implementation process, the acquisition unit first determines the minimum behavioral interval threshold for distinguishing between independent leakage behaviors and continuous leakage processes based on the structure of the stuffing box, the level of operational vibration, and the flow characteristics of the leaked liquid. The stuffing box structure refers to the arrangement of the packing material, the gland structure, and the discharge path of the leaked liquid. Different structural forms result in different time characteristics required for the leaked liquid to seep out of the stuffing box and reach the detection location. The level of operational vibration refers to the intensity of mechanical vibration generated by the pump unit under current operating conditions, which affects the continuity of liquid dripping and the density of trigger signals. The flow characteristics of the leaked liquid describe whether it appears as single drops, continuous thin streams, or intermittent streams. The acquisition unit, considering all these factors, determines a time threshold that limits the time interval within which multiple trigger signals should be considered part of the same leakage process, rather than multiple independent leakage behaviors. This minimum behavioral interval threshold can be obtained through trial run calibration or empirical setting; for example, it can be determined to be 0.5 seconds or 1 second under certain stuffing box structure and vibration conditions.

[0030] After obtaining the minimum behavior interval threshold, the acquisition unit performs a merging process on multiple initial trigger signals generated consecutively within this threshold range. The merging process involves comparing the time intervals between adjacent signals according to their occurrence order. When the time interval between two adjacent initial trigger signals is less than or equal to the minimum behavior interval threshold, the acquisition unit considers these initial trigger signals to originate from the same leakage process, rather than treating them as multiple independent leakage events. Through this merging process, the acquisition unit combines several temporally consecutive and physically related initial trigger signals into a leakage process group and generates a unique corresponding process trigger identifier for this leakage process group. This process trigger identifier is used to mark a complete leakage process within the system, serving as the basic unit for subsequent analysis.

[0031] After forming a leakage process group, the acquisition unit does not immediately classify the process group as a valid leakage behavior. Instead, based on the process trigger identifier, it further analyzes the number of initial trigger signals within each leakage process group and the duration of the leakage process group. The duration referred to here is the time span from the occurrence of the first initial trigger signal to the occurrence of the last initial trigger signal within the leakage process group. The acquisition unit compares this duration with a pre-set minimum duration threshold. The minimum duration threshold is a time standard used to exclude instantaneous splashes, occasional droplets, or brief interference signals; its value can be determined according to the site environment, for example, set to 0.2 seconds or 0.3 seconds. When the duration of a leakage process group exceeds the minimum duration threshold, the acquisition unit considers the leakage process to have actual leakage significance, rather than occasional interference, thus classifying the corresponding leakage process group as a valid leakage process and generating a corresponding candidate leakage behavior identifier. Conversely, if the duration does not reach the threshold, the leakage process group is considered statistically insignificant, and no candidate leakage behavior identifier is generated.

[0032] After generating candidate leakage behavior identifiers, the acquisition unit needs to perform final confirmation to ensure that adjacent leakage processes are not misclassified as multiple independent leakage behaviors due to their close timing. To this end, the acquisition unit makes a final determination based on the time interval relationship between the candidate leakage behavior identifier and adjacent leakage process groups. Specifically, the acquisition unit calculates the time interval between the leakage process groups corresponding to two adjacent candidate leakage behavior identifiers. When this time interval is greater than the minimum behavior interval threshold, it indicates that the two leakage processes are sufficiently separated in time and meet the conditions for being counted as two independent leakage behaviors separately. In this case, the acquisition unit confirms the candidate leakage behavior identifier and outputs it as a valid leakage behavior identifier corresponding to an independent discrete leakage behavior. If the time interval is less than or equal to the minimum behavior interval threshold, it is considered that there is still continuity between the two candidate leakage behaviors, and they should be merged for processing to avoid duplicate counting of the same continuous leakage process.

[0033] The characterization unit 102 is used to perform time normalization and dimensional mapping processing on the original leakage behavior data to generate a standardized leakage volume index for characterizing the current leakage status of the stuffing box.

[0034] The characterization unit 102 is used to further process the raw leakage behavior data output by the acquisition unit 101. Its purpose is to transform the raw data, which is time-dependent, sampling period-dependent, and inconsistent in its expression, into a unified index that can stably and intuitively reflect the current leakage state of the stuffing gland, thereby providing a reliable data foundation for subsequent target interval construction and regulation strategy generation. In this invention, the characterization unit 102 does not directly participate in regulation decisions, but rather plays a core role in "data normalization" and "physical meaning mapping."

[0035] As mentioned earlier, "raw leakage behavior data" refers to the discrete leakage behavior recorded within a preset sampling period. Essentially, it typically represents the event count, trigger count, or equivalent behavior count within a certain time window. Since the sampling period length may vary in different implementation scenarios—for example, some systems use a 30-second sampling period, while others use 60 or 120 seconds—directly using raw counts for comparison or judgment can easily introduce inconsistencies in time scales. Therefore, it is necessary to first perform time normalization processing.

[0036] Time normalization refers to converting raw leakage behavior data into a unified "leakage occurrence rate per unit standard time" to eliminate the impact of sampling period differences on data comparability. Specifically, the characterization unit 102 acquires the corresponding sampling period length while obtaining the raw leakage behavior data and performs conversion based on a pre-set standard time benchmark. This standard time benchmark can be set to 1 minute, 10 seconds, or other fixed durations according to engineering needs. For example, in most pump station operation and management scenarios, 1 minute can be used as the standard time unit. For instance, if the acquisition unit 101 records 120 discrete leakage behaviors within a 60-second sampling period, the time-normalized result is 120 times per minute; if it records 90 discrete leakage behaviors within a 30-second sampling period, the time normalization result can also be converted to 180 times per minute. In this way, data obtained under different sampling periods are uniformly mapped to the same time scale, thus achieving comparability.

[0037] After time normalization, the characterization unit 102 also needs to perform dimensional mapping on the normalized data. "Dimensional mapping" here refers to converting statistical quantities, originally existing in the form of "number of behaviors" or "number of events," into continuous numerical indicators with clear engineering meaning and easy use in control logic. Since discrete leakage behavior itself is not directly equivalent to volumetric flow rate, but under the same packing structure and stable operating conditions, there is a stable correspondence between the frequency of discrete behavior and the actual leakage volume, a mapping relationship between the two can be established through pre-calibration or empirical models.

[0038] In one implementation, the dimensional mapping can be accomplished through experimental calibration. Specifically, during the shutdown or trial operation phase of the pump unit, under different known leakage conditions, the frequency of discrete leakage behaviors is recorded, thus forming a "behavior frequency – equivalent leakage" curve. During operation, the characterization unit 102 can obtain the corresponding equivalent leakage value based on the time-normalized behavior frequency through table lookup, linear interpolation, or function calculation. For example, in a specific unit, calibration reveals that when 100 discrete leakage behaviors occur per minute, the corresponding actual leakage is approximately 0.5 liters / minute; when 200 discrete leakage behaviors occur per minute, the corresponding actual leakage is approximately 1.0 liter / minute. Based on this, the characterization unit 102 can establish a proportional relationship or a piecewise mapping relationship, mapping the normalized behavior data into continuous values ​​with volumetric significance.

[0039] In another implementation, to avoid introducing specific volume units, the dimensional mapping result can be defined as a dimensionless or semi-dimensional "standardized leakage index." This index corresponds monotonically to the degree of leakage but is not necessarily equivalent to the actual flow rate. In this case, the characterization unit 102 can further map the normalized behavioral frequency to a preset numerical range, such as 0 to 1, 0 to 100, or other ranges that are easy to calculate and compare. For example, the behavioral frequency corresponding to the normal allowable leakage state can be mapped to a standardized leakage index of 50, a significantly larger leakage state can be mapped to above 80, and a significantly smaller leakage state can be mapped to below 20. Although this standardized leakage index does not directly represent the physical flow rate, it can stably and continuously reflect the changing trend of the stuffing box leakage state and facilitates interval judgment and deviation calculation by subsequent units.

[0040] It is important to emphasize that regardless of whether an equivalent volume mapping method or a dimensionless standardized index mapping method is used, as long as the generated result uniquely corresponds to the current leakage state of the stuffing box and maintains consistency in the mapping rules within the same system, it falls within the scope of the "standardized leakage index" as defined in this invention. This standardized leakage index, as the output of the characterization unit 102, is the direct input for the construction unit 103 to construct the target leakage range. Its numerical stability and continuity play a fundamental role in the reliable operation of the entire regulation system.

[0041] Through the above time normalization and dimensional mapping processing, the characterization unit 102 effectively eliminates the uncertainty of the original leakage behavior data in terms of sampling period, statistical form and expression scale, so that the leakage status of the stuffing box can be described and transmitted in a unified and standardized numerical form.

[0042] Furthermore, the characterization unit is specifically used for: The system receives raw leakage behavior data output by the acquisition unit and performs unit time conversion processing on the raw leakage behavior data based on the corresponding sampling period duration to generate a time-normalized leakage behavior frequency that characterizes the intensity of leakage behavior within the current sampling period. Based on time-normalized leakage behavior frequency, a leakage behavior frequency sequence is constructed within multiple consecutive adjacent sampling periods. Stability analysis is performed on the leakage behavior frequency sequence to identify and remove abnormal frequency values ​​that deviate from the overall trend of the frequency sequence, thereby generating an effective leakage behavior frequency after anomaly suppression processing. Based on the pre-established mapping rules between leakage behavior frequency and leakage status characterization quantity, the effective leakage behavior frequency is subjected to dimensional mapping processing and converted into a continuous value that monotonically corresponds to the leakage degree of the stuffing box, generating candidate leakage quantity characterization value. The consistency of the candidate leakage volume characterization value over multiple consecutive sampling periods is verified. When the change amplitude of the candidate leakage volume characterization value in adjacent sampling periods is less than a preset stable change threshold, the candidate leakage volume characterization value is determined as the standardized leakage volume index.

[0043] In this embodiment, the characterization unit is used to systematically process the raw leakage behavior data output by the acquisition unit. Its core objective is to transform the statistical results of leakage behavior, which are random and discrete, into a standardized leakage volume index that can stably and continuously reflect the leakage status of the stuffing box, thereby providing a reliable basis for the subsequent construction of target leakage ranges and the generation of adjustment strategies.

[0044] During system operation, the characterization unit first receives the raw leakage behavior data output by the acquisition unit. As mentioned earlier, the raw leakage behavior data refers to the number of independent, discrete leakage behaviors occurring within a preset sampling period, after validity determination. Essentially, it is a count value strongly correlated with the sampling period length. Since the sampling period length may differ under different system configurations, directly using this count value for subsequent comparisons and judgments would result in a lack of a unified time scale between the data. Therefore, while receiving the raw leakage behavior data, the characterization unit simultaneously acquires the corresponding sampling period length and performs unit-time conversion processing on the raw leakage behavior data. "Unit-time conversion processing" refers to converting the number of leakage behaviors occurring within the sampling period into the frequency of leakage behaviors occurring within a standard time unit. For example, when the sampling period is set to 60 seconds and 120 independent, discrete leakage behaviors are counted within that period, the unit-time conversion result is 120 times per minute; when the sampling period is 30 seconds and 60 independent, discrete leakage behaviors are counted, the conversion also yields 120 times per minute. Through the above conversion, the characterization unit generates a time-normalized leakage behavior frequency that characterizes the intensity of leakage behavior within the current sampling period, thereby eliminating the impact of sampling period differences on data comparability.

[0045] After obtaining the time-normalized leakage behavior frequency, the characterization unit does not directly use this frequency as the final input. Instead, it constructs a leakage behavior frequency sequence over multiple consecutive adjacent sampling periods. Here, "multiple consecutive adjacent sampling periods" refers to selecting at least two, and typically three or more, adjacent sampling periods, under the premise that the system's operating state has not changed significantly, to observe the changes in the leakage behavior frequency over time. By arranging the time-normalized leakage behavior frequencies within these adjacent sampling periods in chronological order, the characterization unit forms a leakage behavior frequency sequence to reflect the dynamic characteristics of the leakage state within a short time range.

[0046] After generating the leakage behavior frequency sequence, the characterization unit performs stability analysis on the sequence to identify and remove abnormal frequency values ​​that do not conform to the overall trend. An "abnormal frequency value" refers to an isolated data point in the frequency sequence that significantly deviates from adjacent frequency values ​​or the overall level of change. Its causes may include instantaneous vibration interference, accidental triggering by liquid splashing, or short-term operational fluctuations. Stability analysis can be implemented in various ways, such as comparing the current frequency value with the mean or median of the frequency sequence. When the deviation exceeds a preset threshold, it is determined to be an abnormal frequency value. In one example, if a frequency sequence consists of 120, 118, 121, and 160, where the first three values ​​have relatively small fluctuations, while 160 is significantly higher than the others, the characterization unit can identify 160 as an abnormal frequency value and remove it. After the above processing, the characterization unit generates an effective leakage behavior frequency after anomaly suppression processing, which more realistically reflects the stability level of the stuffing box leakage state.

[0047] After obtaining the effective leakage behavior frequency, the characterization unit further performs dimensional mapping processing on the effective leakage behavior frequency according to the pre-established mapping rules between leakage behavior frequency and leakage state characterization quantity. The "mapping rules" mentioned here refer to a set of rules or functions used to describe the correspondence between the frequency of leakage behavior and the actual leakage degree of the stuffing box. These rules can be determined through experimental calibration during equipment commissioning or long-term operating experience. The core requirement of these mapping rules is to maintain a monotonic correspondence; that is, the higher the effective leakage behavior frequency, the larger the mapped leakage state characterization quantity, and vice versa. The mapping result is a continuous numerical value used to reflect the relative magnitude of the stuffing box leakage degree; this continuous numerical value is the candidate leakage quantity characterization value. This candidate leakage quantity characterization value can have a specific physical meaning or be a dimensionless standardized value, as long as it remains consistent within the system.

[0048] After generating candidate leakage volume characterization values, the characterization unit also needs to verify the consistency of their changes over multiple consecutive sampling periods to avoid the impact of instantaneous fluctuations on the final index. The so-called "consistency verification" refers to comparing the variation amplitudes of candidate leakage volume characterization values ​​between adjacent sampling periods. When this variation amplitude is less than a preset stable variation threshold, the candidate leakage volume characterization value is considered to have sufficient stability over time. The stable variation threshold can be set based on engineering experience, for example, 5% or 10% of the candidate leakage volume characterization value. For example, if the candidate leakage volume characterization values ​​are 50, 51, and 49 in three adjacent sampling periods, with a maximum variation amplitude of 2, and this amplitude is lower than the preset stable variation threshold, then the characterization unit considers the candidate leakage volume characterization value to meet the stability requirements.

[0049] When the characterization unit determines that the variation range of a candidate leakage rate characterization value within adjacent sampling periods meets the stability requirements, it designates that candidate leakage rate characterization value as a standardized leakage rate index and uses it as input data for subsequent construction units to construct the target leakage range. If the variation range does not meet the stability requirements, the characterization unit continues to collect data from subsequent sampling periods, repeating the above analysis process until a candidate leakage rate characterization value that meets the stability conditions is obtained. Through the above processing flow, the characterization unit ensures that the output standardized leakage rate index not only reflects the actual level of the current stuffing box leakage status but also has sufficient anti-interference capability and time stability, thereby meeting the data reliability requirements of subsequent automatic adjustment logic.

[0050] Furthermore, the characterization unit is also used for: Based on the structural parameters of the stuffing box, the type of stuffing material, and the diameter of the bushing, the range of effective leakage behavior frequency is divided into segments to generate multiple non-overlapping frequency intervals, and a corresponding local mapping interval identifier is established for each frequency interval. For each frequency range, based on the lubrication and wear characteristics of the packing under the corresponding leakage intensity, a monotonic mapping relationship between the frequency range and the leakage state characterization quantity is constructed, forming a set of piecewise mapping rules composed of multiple local mapping relationships; When the effective leakage behavior frequency falls into a defined frequency range, the local mapping relationship corresponding to the frequency range is invoked to perform dimensional mapping processing on the effective leakage behavior frequency, generating an initial candidate leakage quantity characterization value corresponding to the frequency range. Based on the continuity of the change of the initial candidate leakage volume characterization value in adjacent sampling periods, the interval consistency of the initial candidate leakage volume characterization value is checked. When it is detected that the initial candidate leakage volume characterization value has a cross-interval transition in adjacent sampling periods, the boundary constraint correction of the initial candidate leakage volume characterization value is performed, and the corrected result is output as the candidate leakage volume characterization value.

[0051] In this embodiment, after completing the basic processing of the effective leakage behavior frequency, the characterization unit is further responsible for converting the frequency data into candidate leakage volume characterization values ​​that have engineering significance and can stably reflect the actual leakage state of the stuffing box.

[0052] In the specific implementation process, the characterization unit first divides the range of effective leakage behavior frequencies into segments based on the structural parameters of the stuffing box, the type of packing material, and the diameter of the bushing. The structural parameters of the stuffing box can include geometric parameters related to sealing and leakage paths, such as the inner length of the stuffing box cavity, the number of packing layers, and the effective stroke of the gland. The type of packing material refers to the specific material or combination of materials used in the packing; different materials exhibit significant differences in lubrication requirements and wear behavior under the same leakage intensity. The bushing diameter directly affects the actual leakage volume and lubrication coverage corresponding to a unit leakage behavior. The characterization unit integrates the above information to divide the overall numerical range of possible effective leakage behavior frequencies, for example, dividing the frequency range into low-frequency, mid-frequency, and high-frequency intervals, with each interval not overlapping numerically. To facilitate subsequent retrieval and identification, the characterization unit generates a unique local mapping interval identifier for each frequency interval, used to clearly identify the interval position of the current effective leakage behavior frequency.

[0053] After dividing the frequency ranges, the characterization unit constructs a monotonic mapping relationship between each defined frequency range and the leakage state characterization quantity. This monotonic mapping relationship means that within the same frequency range, the direction of change of the effective leakage behavior frequency is consistent with the direction of change of the leakage state characterization quantity; that is, as the frequency increases, the leakage state characterization quantity increases accordingly, and as the frequency decreases, the leakage state characterization quantity decreases accordingly. However, a linear mapping relationship is not required. This mapping relationship is established based on the lubrication and wear characteristics of the packing material under corresponding leakage intensities. For example, in the low-frequency range, a small amount of leakage is more sensitive to improving the lubrication state, and the mapping relationship can be designed to change steeply; while in the high-frequency range, further increases in leakage have limited effect on improving lubrication, and the mapping relationship can be designed to change more gradually. By establishing mapping relationships for each frequency range, the characterization unit ultimately forms a set of segmented mapping rules composed of multiple local mapping relationships. This set of rules covers the entire range of effective leakage behavior frequencies.

[0054] When the characterization unit receives the effective leakage behavior frequency corresponding to a certain sampling period, it first determines the specific frequency interval into which the effective leakage behavior frequency falls based on the aforementioned frequency interval division results, and obtains the local mapping interval identifier corresponding to that frequency interval. Subsequently, the characterization unit calls the local mapping relationship corresponding to the frequency interval identifier to perform dimensional mapping processing on the effective leakage behavior frequency, generating an initial candidate leakage quantity characterization value that matches the frequency interval. This initial candidate leakage quantity characterization value is a continuous numerical value used to reflect the leakage state level corresponding to the observed leakage behavior frequency under the current structural and material conditions. For example, in a specific embodiment, when the effective leakage behavior frequency is 120 times per minute and this frequency is in the mid-frequency range, the characterization unit may map it to a candidate leakage quantity characterization value of 50 through the corresponding local mapping relationship; when the frequency rises to 150 times per minute and is still within the same interval, the mapping result may become 60, thereby maintaining the monotonicity of the characterization quantity with frequency change.

[0055] To avoid frequent jumps in candidate leakage rate representation values ​​between different intervals due to sampling noise or boundary effects, the representation unit performs interval consistency checks on the continuity of changes in candidate leakage rate representation values ​​within adjacent sampling periods after generating the initial values. Continuity of change refers to the smooth trend of leakage rate representation values ​​when the system's operating state does not change abruptly, rather than exhibiting sudden changes across multiple frequency intervals within a short period. Therefore, the representation unit compares the frequency interval identifiers corresponding to the initial candidate leakage rate representation values ​​in adjacent sampling periods. When a jump across intervals is detected in the initial candidate leakage rate representation value of the current sampling period compared to the previous sampling period, the representation unit considers the change to be likely due to instantaneous fluctuations rather than a true change in leakage status.

[0056] Upon detecting the aforementioned cross-frequency range transition, the characterization unit performs boundary constraint correction on the initial candidate leakage rate characterization value. Boundary constraint correction refers to limiting the initial candidate leakage rate characterization value to the common boundary or a preset transition range between the frequency range of the previous sampling period and the current frequency range, rather than directly using the abrupt change value obtained from mapping. For example, if the candidate leakage rate characterization value of the previous sampling period corresponds to the upper boundary of the mid-frequency range, while the initial mapping result of the current sampling period falls directly into a higher position in the high-frequency range, the characterization unit can limit the corrected candidate leakage rate characterization value to near the upper boundary of the mid-frequency range or the lower boundary of the high-frequency range, thereby ensuring the continuity of the characterization quantity change. After completing the boundary constraint correction, the characterization unit outputs the corrected result as the final candidate leakage rate characterization value, which serves as the basic input for subsequent consistency verification and determination of the standardized leakage rate index.

[0057] In this application, "inter-interval transition" refers to a situation where, within adjacent sampling periods, the frequency interval identifier of the effective leakage behavior frequency, determined by the characterization unit according to the segmented mapping rule, changes, and this change crosses at least one pre-defined frequency interval boundary. Inter-interval transition is used to characterize discontinuous changes in leakage behavior frequency occurring within a short period of time, and its determination is based on the frequency interval affiliation relationship, rather than simply the magnitude of the numerical change.

[0058] Through the above processing, the characterization unit achieves a refined conversion from the effective leakage behavior frequency to the candidate leakage quantity characterization value, which not only fully reflects the influence of differences in stuffing box structure, material and size on leakage behavior, but also effectively suppresses numerical abrupt changes at the frequency boundary.

[0059] The construction unit 103 is used to construct a target leakage range corresponding to the standardized leakage index based on the constraint parameters related to the operating status of the large water pump unit, and to determine the upper and lower safety thresholds corresponding to the target leakage range.

[0060] The construction unit 103 is used to define the "reasonable leakage range" of the stuffing box under the current operating conditions based on the standardized leakage index output by the characterization unit 102 and the actual operating status of the large water pump unit. Its core function is to provide an engineering-based and directly comparable reference benchmark for subsequent deviation analysis and adjustment strategy generation.

[0061] The term "constraint parameters related to the operating status of large water pump units" refers to a set of operating condition parameters that can substantially affect the reasonable range of leakage in the stuffing box. These parameters include at least one or more of the following: pump unit speed, shaft power or load level, medium temperature, operating pressure level, and continuous operating duration. Those skilled in the art should understand that in large water pump units, the permissible leakage state of the stuffing box is not fixed but adjusts accordingly with increases in speed, shaft power, or medium temperature. Therefore, incorporating these operating condition parameters as constraints into the leakage range construction process is a necessary prerequisite for ensuring the rationality of the adjustment results.

[0062] In the specific implementation process, the construction unit 103 first obtains the relevant constraint parameters of the current operating state. This can be done by directly reading from the existing water pump monitoring system, such as reading the speed signal output by the frequency converter, reading the current or power information from the motor monitoring module, or reading the temperature value of the medium or near the bearing from the temperature sensing device. The above parameters can be obtained without adding new hardware; the construction unit 103 only calls and utilizes the existing operating state data.

[0063] After obtaining the relevant constraint parameters for the operating state, the construction unit 103 constructs the target leakage range corresponding to the standardized leakage index according to pre-set rules or models. The "target leakage range" refers to a continuous numerical range that, under the current operating state, satisfies the cooling and lubrication requirements of the packing material without causing energy loss or environmental impact due to excessive leakage. This range is defined by a lower boundary value and an upper boundary value, and its numerical unit is consistent with the standardized leakage index.

[0064] The method for constructing the target leakage range can be flexibly set according to specific engineering experience and equipment characteristics. In a typical implementation, a correspondence table of "operating state - reasonable leakage range" can be pre-established based on the operating recommendations provided by the manufacturer or historical operating statistics for different operating states. For example, under low-load operating conditions, the corresponding reasonable range for the standardized leakage index can be set to 30 to 50; under rated load operating conditions, the corresponding reasonable range can be set to 40 to 65; and under high-load or high-speed operating conditions, the corresponding reasonable range can be set to 50 to 80. When the construction unit 103 is running, it only needs to select or interpolate the corresponding range based on the current operating state parameters.

[0065] In another implementation, the construction unit 103 can also dynamically generate the target leakage range through calculation. For example, using a baseline leakage range under a certain operating condition as a reference, the upper and lower boundaries of the range are adjusted linearly or non-linearly according to the proportion of change in speed or power. For instance, if the target leakage range corresponding to the rated speed is 40 to 60 as a baseline, and the current speed increases by 20%, the construction unit 103 can shift or appropriately expand the entire range upwards according to a preset scaling factor, making the new target leakage range 48 to 72. The above-mentioned scaling relationships and adjustment methods can be determined according to the actual performance of the specific unit during the commissioning phase.

[0066] While constructing the target leakage range, the construction unit 103 also needs to further determine the upper and lower safety thresholds corresponding to the target leakage range. The upper safety threshold refers to the maximum value that the standardized leakage index is allowed to reach but should not exceed for a long period under the current operating conditions. This value is usually slightly higher than the upper boundary of the target leakage range and is used to distinguish between "acceptably large" and "abnormally large" leakage states. The lower safety threshold refers to the minimum value that the standardized leakage index is allowed to decrease but should not fall below for a long period under the current operating conditions. This value is usually slightly lower than the lower boundary of the target leakage range and is used to distinguish between "acceptably small" and "too small" leakage states that may affect lubrication.

[0067] For example, if the target leakage range obtained under a certain operating state is a standardized leakage index of 45 to 65, then the construction unit 103 can set the upper safety threshold to 75 and the lower safety threshold to 35. Thus, when the actual standardized leakage index is between 45 and 65, the system considers the leakage state to be within the ideal range; when the index is between 65 and 75 or between 35 and 45, the system considers there to be a deviation but still within the safe range; and when the index exceeds 75 or falls below 35, it indicates that the leakage state has significantly deviated from the reasonable range, providing a basis for subsequent adjustments and anomaly judgment.

[0068] It should be noted that once the target leakage range, upper safety threshold, and lower safety threshold are determined in the construction unit 103, their values ​​serve as the comparison benchmark for deviation analysis in the subsequent generation unit 104. During the same operating state, the range and threshold output by the construction unit 103 remain unchanged. Only when the constraint parameters related to the operating state change significantly will the construction unit 103 recalculate and update the corresponding target leakage range and threshold, thereby ensuring that the entire system maintains the consistency and rationality of the judgment criteria when the operating state changes.

[0069] In this way, the construction unit 103 closely links the abstract standardized leakage index with the specific operating status of large water pump units, clearly defines the numerical boundary of "reasonable leakage", and provides a clear and feasible judgment basis for the subsequent quantitative analysis of leakage deviation and the generation of pressure cover adjustment strategy.

[0070] Furthermore, the building unit is specifically used for: Based on the relevant constraint parameters of the current operating status of the large water pump unit, the operating status of the water pump unit is judged by the operating condition level, and an operating condition level identifier is generated to characterize the current operating load characteristics. The operating condition level identifier is used to limit the allowable leakage range of the stuffing box foundation under the operating status. Based on the operating condition level identifier, the basic leakage range corresponding to the current operating condition level identifier is determined from the pre-established correspondence between operating condition levels and leakage control rules, and the basic leakage range is used as the initial range of the target leakage range. Based on the initial range, and in conjunction with the minimum leakage intensity constraint required for packing cooling and lubrication and the maximum leakage intensity constraint required for sealing performance control, the upper and lower boundaries of the initial range are expanded or contracted to determine the lower and upper safety thresholds corresponding to the target leakage range, respectively. Based on the lower safety threshold, the upper safety threshold, and the distribution of the standardized leakage index over multiple consecutive sampling periods, the target leakage range is subjected to consistency verification. When the standardized leakage index is continuously and stably distributed within the target leakage range, the range width of the target leakage range is converged and corrected to form a final target leakage range suitable for the current operating state.

[0071] In this embodiment, the building unit is responsible for combining the abstract leakage volume characterization results with the specific operating conditions of the large water pump unit, and forming a target leakage range that can be directly used for subsequent adjustment and judgment, so as to ensure the rationality and consistency of the stuffing box leakage control under different operating conditions.

[0072] During system operation, the construction unit first receives and acquires constraint parameters related to the operating status of the large water pump unit. These operating status-related constraint parameters refer to a set of parameters that objectively reflect the current operating load characteristics of the water pump unit. They may include at least one or more of the following: pump shaft speed, motor output power or current value, medium flow rate, pump outlet pressure, and continuous operating duration. These parameters are not manually set but are provided in real-time by the water pump unit's existing monitoring system. The construction unit uses these parameters to comprehensively determine the current operating load level of the water pump unit. Based on this determination, the construction unit classifies the current operating status into a predefined operating condition level and generates an operating condition level identifier to characterize the current operating load characteristics. This operating condition level identifier can be in discrete form, such as low load, medium load, and high load, or it can be in numerical form, as long as it can distinguish the differences in the working environment of the stuffing box under different load conditions. The direct function of this operating condition level identifier is to limit the allowable basic leakage range of the stuffing box under the current operating status, thereby avoiding the use of the same leakage control standard under different load conditions.

[0073] After generating the operating condition level identifier, the construction unit further determines the basic leakage range matching the current operating condition level identifier based on this identifier and from the pre-established correspondence between operating condition levels and leakage control rules. The correspondence between operating condition levels and leakage control rules mentioned here is a set of rules pre-established during equipment commissioning or long-term operation based on the reasonable operating states of the stuffing box under different load conditions. This set clearly defines the reasonable numerical range of the standardized leakage rate index under each operating condition level. The construction unit selects a set of upper and lower boundary values ​​corresponding to the current operating condition level identifier from this correspondence through table lookup, indexing, or equivalent methods, and uses this range as the initial range of the target leakage range. At this point, the initial range already reflects the impact of operating load differences on leakage control, but it has not yet introduced further constraints on cooling, lubrication, and sealing performance.

[0074] After obtaining the initial range, the building unit introduces minimum leakage strength constraints required for packing cooling and lubrication, and maximum leakage strength constraints required for sealing performance control, to modify the upper and lower boundaries of the initial range. The minimum leakage strength constraint refers to the minimum leakage level that the stuffing box must maintain while ensuring a stable liquid film between the packing and the bushing, avoiding dry friction and abnormal temperature rise. The maximum leakage strength constraint refers to the maximum allowable leakage level of the stuffing box while ensuring sealing performance and preventing energy loss and environmental pollution. These two constraint values ​​can be set based on the characteristics of the packing material, the bushing material, and historical operating experience. The building unit compares the lower boundary of the initial range with the minimum leakage strength constraint. If the initial lower boundary is lower than the minimum leakage strength constraint, the lower boundary is adjusted upwards to the value corresponding to the minimum leakage strength constraint. Similarly, the upper boundary of the initial range is compared with the maximum leakage strength constraint. If the initial upper boundary is higher than the maximum leakage strength constraint, the upper boundary is adjusted downwards to the value corresponding to the maximum leakage strength constraint. Through this process, the building unit ultimately determines the lower and upper safety thresholds corresponding to the target leakage range, ensuring that the target leakage range meets both the operating load requirements and the dual engineering constraints of cooling, lubrication, and sealing performance.

[0075] After determining the lower and upper safety thresholds, the construction unit does not immediately output the target leakage range as the final result. Instead, it performs a consistency check on the target leakage range by combining the distribution of standardized leakage volume indicators over multiple consecutive sampling periods. These multiple consecutive sampling periods refer to selecting at least two adjacent sampling periods while maintaining the current operating condition level identifier, to observe the actual fluctuations of the standardized leakage volume indicators. The construction unit performs statistical analysis on the standardized leakage volume indicators within these sampling periods to determine whether they exhibit a relatively stable distribution characteristic within the target leakage range. A stable distribution means that the standardized leakage volume indicators do not exceed the target leakage range in any sampling period, and their fluctuation amplitude does not exceed a preset stability threshold.

[0076] When the construction unit determines that the standardized leakage rate index is consistently and stably distributed within the target leakage range, it indicates that the current range setting is highly compatible with the actual operating state. At this point, the construction unit performs convergence correction on the width of the target leakage range. This convergence correction can be achieved by simultaneously reducing the distance between the upper and lower boundaries of the range. For example, while keeping the range center value unchanged, the upper and lower boundaries are shrunk inward by a certain proportion, thus forming a more refined final target leakage range. In this way, the construction unit gradually converges the target leakage range during the stable operation phase of the system, improving the sensitivity and accuracy of subsequent adjustment judgments. Conversely, if the standardized leakage rate index fails to stably fall within the target leakage range within a continuous sampling period, the construction unit maintains the current range setting unchanged to avoid premature convergence and the resulting adjustment risks.

[0077] Through the aforementioned series of continuous and interconnected processing steps, the building unit can dynamically generate and optimize the final target leakage range applicable to the current operating state, based on a full consideration of differences in operating load, cooling and lubrication requirements, sealing performance limitations, and actual operational stability. This target leakage range serves as the sole reference benchmark for subsequent generation units to analyze leakage deviations, ensuring that the entire stuffing box leakage monitoring and automatic adjustment system has a clear, implementable, and engineering-significant control basis under different operating conditions.

[0078] Furthermore, the building unit is also used for: Based on the standardized leakage volume index obtained in multiple consecutive sampling periods, a corresponding leakage volume distribution sequence is constructed, and the interval occupancy ratio of the standardized leakage volume index relative to the target leakage interval in each sampling period is calculated to characterize the degree of concentration of leakage volume in the target leakage interval. Based on the changes in the occupancy ratio of the interval over multiple consecutive sampling periods, it is determined whether the standardized leakage index shows a continuous and stable distribution. When the occupancy ratio of the interval remains above the preset stable occupancy threshold within the preset stable determination period, a target leakage interval stability identifier is generated. After generating a stable identifier for the target leakage range, a convergence correction process is performed on the range width of the target leakage range based on the statistical dispersion of the standardized leakage volume index within the target leakage range. The convergence correction process reduces the range span of the target leakage range by simultaneously moving the lower safety threshold and the upper safety threshold inward. After completing the interval width convergence correction, the consistency between the corrected target leakage interval and the minimum leakage strength constraint required for packing cooling and lubrication and the maximum leakage strength constraint required for sealing performance control is checked. When the corrected target leakage interval meets the constraint conditions, it is determined as the final target leakage interval applicable to the current operating state.

[0079] In this embodiment, based on the already obtained target leakage range, lower safety threshold, and upper safety threshold, the construction unit further analyzes the historical distribution characteristics of standardized leakage indicators. This allows the target leakage range to gradually converge to a range more closely aligned with the current operating conditions as the pump unit's actual operating status changes, thus preventing the target leakage range from remaining too wide or too conservative for extended periods, which could affect adjustment accuracy. To this end, the construction unit first uses standardized leakage indicators obtained over multiple consecutive sampling periods as basic data to construct a corresponding leakage distribution sequence. Here, the "leakage distribution sequence" refers to a set of standardized leakage indicator values ​​arranged chronologically, reflecting the changes in leakage status of the stuffing box over a continuous operating period. Based on this, the construction unit performs interval assignment judgment on the standardized leakage indicators within each sampling period, determining whether the indicator falls within the current target leakage range, and calculates the interval occupancy ratio corresponding to that sampling period. The interval occupancy ratio refers to the proportion of times the standardized leakage rate falls within the target leakage interval within a preset statistical window, relative to the total number of sampling periods within that window. This ratio is used to quantitatively characterize the concentration of leakage within the target leakage interval. For example, in a statistical window containing ten consecutive sampling periods, if the standardized leakage rate falls within the target leakage interval in eight of those periods, the corresponding interval occupancy ratio is 0.8.

[0080] After obtaining the interval occupancy ratio, the construction unit further analyzes the changes of this ratio over multiple consecutive sampling periods to determine whether the standardized leakage index exhibits a continuously stable distribution. The "continuously stable distribution" refers to the standardized leakage index consistently falling within the target leakage interval for most of a continuous operating period, without frequent boundary crossings or significant deviations. To determine feasibility, the construction unit pre-sets a stability determination period and a stability occupancy threshold. The stability determination period limits the number of consecutive sampling periods participating in the stability assessment, while the stability occupancy threshold limits the minimum requirement for the interval occupancy ratio. When the interval occupancy ratio for each sampling period remains above the stability occupancy threshold throughout the entire stability determination period, the construction unit determines that the current target leakage interval has a good match with the actual leakage state and generates a target leakage interval stability identifier to characterize this state. This stability identifier serves as a crucial basis for subsequent interval convergence correction; its generation signifies that the current target leakage interval can effectively cover the main distribution range of the actual leakage state.

[0081] After generating a stable identifier for the target leakage range, the construction unit does not simply maintain the existing range unchanged. Instead, it further performs convergence correction processing on the range width of the target leakage range based on the statistical dispersion of the standardized leakage volume index within the target leakage range. The statistical dispersion referred to here is the degree of dispersion of the standardized leakage volume index within the target leakage range, used to reflect the magnitude of leakage fluctuations. Those skilled in the art will understand that this dispersion can be characterized by various statistical methods, such as calculating the difference between the maximum and minimum values ​​of the standardized leakage volume index within the target leakage range, or calculating its variance, standard deviation, etc. Based on this statistical dispersion, the construction unit determines whether the current target leakage range is excessively wide and redundant. When statistical results indicate that the standardized leakage volume index is consistently concentrated in the middle of the target leakage range and its fluctuation range is significantly smaller than the range width, the construction unit performs symmetrical or approximately symmetrical convergence correction on the target leakage range by simultaneously moving the lower safety threshold and the upper safety threshold inward, thereby reducing the range span of the target leakage range. For example, if the current target leakage range is [30, 70], and the standardized leakage index has been concentrated in the range of [40, 60] for a long time, the building unit can adjust the lower safety threshold from 30 to a position close to 40, and the upper safety threshold from 70 to a position close to 60, so as to form a new, more compact target leakage range.

[0082] After completing the convergence correction of the interval width, the construction unit also needs to perform a consistency check on the corrected target leakage interval to ensure that the interval still meets the basic physical constraints required for the operation of the stuffing box after convergence. Specifically, the construction unit compares the corrected target leakage interval with the minimum leakage intensity constraint required for packing cooling and lubrication and the maximum leakage intensity constraint required for sealing performance control, confirming that the corrected lower safety threshold is not lower than the minimum leakage intensity requirement and the corrected upper safety threshold is not higher than the maximum leakage intensity allowed by the sealing performance. When the corrected target leakage interval meets the above constraints simultaneously, the construction unit determines it as the final target leakage interval applicable to the current operating state and outputs it for the generation unit to use for subsequent leakage deviation analysis and adjustment strategy generation. Through the above process, the construction unit realizes a complete closed loop from the initial setting of the target leakage interval to the gradual convergence and optimization based on actual operating data, enabling the target leakage interval to continuously approach the actual operating requirements while ensuring safety, thereby providing a stable, reliable, and adaptive control benchmark for the entire automatic control system.

[0083] The generation unit 104 is used to analyze the relationship between the standardized leakage index and the target leakage range, generate a leakage deviation description quantity that represents the direction and magnitude of the deviation, and generate gland adjustment strategy parameters based on the leakage deviation description quantity under the premise of satisfying the packing cooling and lubrication constraints.

[0084] The generation unit 104, based on the target leakage range, upper safety threshold, and lower safety threshold already provided by the construction unit 103, quantitatively analyzes the standardized leakage volume index output by the characterization unit 102 at the current moment, and forms an adjustment basis that can be directly used for control execution. Its core function is to complete the transition from "state judgment" to "adjustment decision". In this invention, the generation unit 104 neither directly drives the actuator nor redefines the leakage range, but generates an intermediate quantity with clear physical and control implications through fine analysis of the degree of deviation of the leakage state, and further transforms the intermediate quantity into pressure cap adjustment strategy parameters.

[0085] The so-called "analysis of the relationship between the standardized leakage index and the target leakage range" refers to the generation unit 104 comparing the current standardized leakage index with the upper and lower boundaries of the target leakage range to determine the positional relationship of the current leakage state relative to the target range. This positional relationship includes at least three basic scenarios: the standardized leakage index is within the target leakage range, above the target leakage range, or below the target leakage range. When the standardized leakage index falls within the target leakage range, it indicates that the current leakage state of the stuffing box is within a reasonable range under the current operating conditions. At this time, the generation unit 104 can generate a leakage deviation description indicating "no adjustment required" or "maintain the status quo." When the standardized leakage index is higher than the upper boundary of the target leakage range, it indicates that the leakage is too large and needs to be suppressed by adjusting the tightness of the gland. When the standardized leakage index is lower than the lower boundary of the target leakage range, it indicates that the leakage is too small, which may affect the cooling and lubrication conditions of the stuffing box and needs to be restored to a reasonable leakage state through the opposite adjustment.

[0086] Based on the above judgment, the generation unit 104 further forms a "leakage deviation description quantity". In this invention, the "leakage deviation description quantity" is an information quantity that simultaneously includes the deviation direction and the deviation magnitude. The deviation direction is used to indicate the basic direction of adjustment, and the deviation magnitude is used to reflect the degree to which the current leakage state deviates from the target range. The deviation direction can be represented by symbols or logical identifiers, for example, defining excessive leakage as a positive deviation and insufficient leakage as a negative deviation, or using semantically clear methods such as "tightening direction" and "loosening direction" for differentiation. The deviation magnitude can be calculated by the numerical difference between the standardized leakage volume index and the boundary of the target leakage range, and its numerical unit is consistent with the standardized leakage volume index.

[0087] For example, under a certain operating state, the target leakage range given by the construction unit 103 is a standardized leakage index of 45 to 65. If the current standardized leakage index is 75, the generation unit 104 can calculate the deviation range as 75 minus 65, which is 10, and determine the deviation direction as excessive leakage. If the current standardized leakage index is 35, the deviation range is the difference between 35 and 45, which is 10, and the deviation direction is determined as insufficient leakage. If the current standardized leakage index is 55, the index is within the target leakage range, the deviation range can be defined as 0, and the deviation direction is neutral.

[0088] After obtaining the leakage deviation description, the generation unit 104 also needs to generate gland adjustment strategy parameters based on this leakage deviation description, while satisfying the packing cooling and lubrication constraints. The "packing cooling and lubrication constraints" refer to the requirement that, when adjusting the gland tightness, the leakage level must not remain below the minimum level necessary to ensure the formation of an effective liquid film between the packing and the bushing for an extended period. This minimum level can be limited by a safety lower limit threshold or by independently set lubrication assurance parameters. When generating the adjustment strategy, the generation unit 104 must ensure that the adjustment result does not cause the standardized leakage rate index to shift excessively below the safety lower limit threshold.

[0089] The gland adjustment strategy parameters are a set of control parameters used to guide the specific actions of the actuator, and they include at least two aspects: adjustment direction and adjustment range. The adjustment direction is directly determined by the deviation direction in the leakage deviation description. When the leakage is large, the corresponding adjustment direction is to move the gland in the direction of increasing the clamping force; when the leakage is small, the corresponding adjustment direction is to move the gland in the direction of decreasing the clamping force. The adjustment range is related to the deviation range, but it is not a simple proportional amplification. Instead, it needs to be limited in combination with lubrication constraints and the mechanical characteristics of the actuator.

[0090] In one implementation, the generation unit 104 can calculate the corresponding adjustment step size based on the deviation amplitude and a preset proportional coefficient. For example, the deviation amplitude can be multiplied by an adjustment coefficient less than 1 to obtain the target displacement or rotation angle for a single adjustment. For instance, when the deviation amplitude is 10 and the adjustment coefficient is set to 0.1, the standardized adjustment amount for a single adjustment is 1; when the deviation amplitude is 5, the corresponding adjustment amount is 0.5. This method avoids excessively large adjustments at once, which could cause system oscillations or sudden changes in lubrication conditions.

[0091] In another embodiment, the generation unit 104 can also use a tiered strategy to generate adjustment parameters. That is, when the deviation is within a small range, a fine-tuning step size is used, and when the deviation exceeds a preset threshold, a relatively large step size is used. However, in either case, the adjustment strategy parameters are limited by the packing cooling and lubrication constraints. For example, when the standardized leakage index is close to the lower safety threshold, even if the deviation direction points to further reduce leakage, the generation unit 104 will automatically limit the adjustment range to prevent the gland from being further tightened.

[0092] Through the above analysis and generation process, generation unit 104 transforms the abstract leakage deviation into gland adjustment strategy parameters with clear directionality and amplitude, conforming to engineering constraints. These adjustment strategy parameters are then used as inputs to control unit 105 to drive the stuffing box gland actuator to perform specific actions.

[0093] Furthermore, the generating unit is specifically used for: Based on the relative positional relationship between the standardized leakage volume index and the target leakage range, the current leakage state of the stuffing box is determined to be deviated, and a leakage deviation direction identifier is generated to characterize the positional relationship between the standardized leakage volume index and the target leakage range. The leakage deviation direction identifier is used to distinguish whether the leakage state is above, within, or below the target leakage range. Based on the generated leakage deviation direction identifier, the distance between the standardized leakage volume index and the boundary of the target leakage interval is calculated to generate a leakage deviation amplitude quantity to characterize the degree of leakage deviation. The leakage deviation direction identifier and the leakage deviation amplitude quantity together constitute the leakage deviation description quantity. Based on the leakage deviation description and combined with the packing cooling and lubrication constraints, the adjustment priority corresponding to different leakage deviation directions is distinguished. When the leakage deviation direction indicator indicates that the leakage amount is lower than the target leakage range, the adjustment priority is limited to avoid insufficient packing cooling. When the leakage deviation direction indicator indicates that the leakage amount is higher than the target leakage range, the adjustment priority is increased to control leakage loss. An adjustment priority indicator that matches the current leakage state is generated. Based on the adjustment priority identifier and the leakage deviation magnitude, the adjustment direction and adjustment step size of a single cap adjustment are determined according to the preset adjustment mapping rules. In the process of generating the adjustment step size, a single adjustment upper limit constraint is introduced to prevent over-adjustment. Finally, cap adjustment strategy parameters are generated to describe the adjustment behavior.

[0094] In this embodiment, the generation unit plays a key role in converting the quantitative results of the leakage status into directly executable gland adjustment decisions. Its working objective is to generate constrained decisions on the adjustment direction, adjustment range, and adjustment priority, while fully considering the safety of packing cooling and lubrication, so as to avoid the adverse effects of over-adjustment or adjustment lag on the operational reliability of the stuffing box.

[0095] During system operation, the generation unit first receives the standardized leakage rate index output by the characterization unit and the target leakage range output by the construction unit. The "standardized leakage rate index" refers to a continuous value obtained after time normalization, anomaly suppression, and dimensional mapping, which monotonically reflects the relative magnitude of the current stuffing box leakage. The "target leakage range" is defined by a pair of boundary values, representing a reasonable leakage range that satisfies both the packing cooling and lubrication requirements and maintains sealing performance under the current operating conditions. The generation unit compares the standardized leakage rate index with the upper and lower boundaries of the target leakage range to determine the deviation of the current stuffing box leakage state and generates a leakage deviation direction indicator. This leakage deviation direction indicator is a discrete state indicator used to clearly distinguish between three situations: the standardized leakage rate index is above, inside, or below the target leakage range, thus providing a directional basis for subsequent adjustment strategies.

[0096] After obtaining the leakage deviation direction indicator, the generation unit further quantitatively characterizes the degree of leakage deviation. To this end, based on the determination result of the leakage deviation direction indicator, the generation unit selects the target leakage interval boundary corresponding to the current leakage state and calculates the distance between the standardized leakage volume index and this boundary. If the leakage deviation direction indicator indicates that the standardized leakage volume index is higher than the target leakage interval, the upper boundary of the target leakage interval is selected as the reference boundary, and the difference between the standardized leakage volume index and this upper boundary is calculated. If the leakage deviation direction indicator indicates that the standardized leakage volume index is lower than the target leakage interval, the lower boundary of the target leakage interval is selected as the reference boundary, and the difference between this lower boundary and the standardized leakage volume index is calculated. If the leakage deviation direction indicator indicates that the standardized leakage volume index is within the target leakage interval, the distance value can be set to zero or a preset minimum value. This distance value is the leakage deviation amplitude, used to characterize the degree of deviation of the current leakage state from the reasonable range. The generation unit combines the leakage deviation direction indicator with the leakage deviation amplitude to form a complete leakage deviation description, so that the leakage status has both directional and amplitude descriptions.

[0097] After generating the leakage deviation description, the generation unit does not directly generate an adjustment strategy based on it. Instead, it introduces packing cooling and lubrication constraints to prioritize adjustment behaviors. Packing cooling and lubrication constraints mean that during any adjustment process, sufficient lubricating fluid must be prioritized between the packing and the bushing to prevent dry friction, abnormal heating, or packing ablation. When the leakage deviation direction indicator indicates that the current leakage is below the target leakage range, it means that the stuffing box leakage is approaching or below the safe lower limit of cooling and lubrication. In this case, the generation unit restricts the adjustment behavior, lowers the adjustment priority, and avoids further tightening of the gland, which could lead to insufficient cooling. Conversely, when the leakage deviation direction indicator indicates that the current leakage is above the target leakage range, it means that there is a risk of excessive leakage loss from the stuffing box. In this case, the generation unit increases the adjustment priority to more aggressively control the leakage. Through this method, the generation unit generates an adjustment priority indicator that matches the current leakage state based on the leakage deviation description. This adjustment priority indicator guides the determination of subsequent adjustment step sizes and adjustment frequencies.

[0098] After determining the adjustment priority identifier, the generation unit determines the adjustment direction and adjustment step size for a single cap adjustment according to the adjustment priority identifier and the leakage deviation amplitude, following a preset adjustment mapping rule. The "adjustment direction" refers to the tightening or loosening direction of the cap relative to its current state, directly corresponding to the leakage deviation direction identifier. The "adjustment step size" quantifies the magnitude of a single adjustment; its size can increase with the leakage deviation amplitude, and different mapping ratios are used under different adjustment priorities. For example, a larger mapping ratio can be used with a higher adjustment priority, making the adjustment step size more sensitive to changes in the leakage deviation amplitude; a smaller mapping ratio is used with a lower adjustment priority, making the adjustment process more gradual.

[0099] To prevent sudden changes in packing stress or sealing instability caused by excessively large adjustments in a single step, the generation unit introduces a single adjustment upper limit constraint during the adjustment step generation process. This upper limit constraint limits the maximum allowable step size for a single adjustment. Even if the theoretical step size calculated according to the adjustment mapping rules exceeds this upper limit, the generation unit will still limit the final step size within this upper limit. For example, when the adjustment step size calculated based on the leakage deviation is 3 execution units, and the single adjustment upper limit constraint is set to 1 execution unit, the final adjustment step size output by the generation unit will be 1 execution unit. Through this constraint mechanism, the generation unit effectively avoids the risks associated with over-adjustment.

[0100] After completing the above judgments and calculations, the generation unit finally outputs the capping adjustment strategy parameters to describe the current adjustment behavior. These capping adjustment strategy parameters include at least the adjustment direction, adjustment step size, and corresponding adjustment priority information, and serve as the direct input for the control unit to generate specific execution instructions.

[0101] The control unit 105 is used to convert the gland adjustment strategy parameters into control commands that can be recognized by the stuffing gland gland actuator, and drive the stuffing gland gland actuator to automatically adjust the tightness of the gland.

[0102] The control unit 105 is used to convert the gland adjustment strategy parameters output by the generation unit 104 into control commands that can be directly understood and executed by the stuffing gland gland actuator, and drive the actuator to make specific adjustments to the tightness of the gland. Its function is to realize the final implementation from "adjustment strategy" to "physical action".

[0103] The so-called "gland adjustment strategy parameters" refer to a set of adjustment indication information generated by the generation unit 104 based on the leakage deviation description and the packing cooling and lubrication constraints. This information includes at least the adjustment direction and adjustment range, and may also include the adjustment rhythm, adjustment frequency, or adjustment restrictions if necessary. After receiving the strategy parameters, the control unit 105 first analyzes the adjustment direction to determine whether the gland should move in the direction of increasing the clamping force or in the direction of decreasing the clamping force. The "gland tightness state" mentioned here refers to the axial pressure state applied by the stuffing box gland relative to the packing and bushing. Its changes directly affect the tightness of the contact between the packing and bushing and the effective cross-section of the leakage channel.

[0104] In this invention, the "stuffing gland cap actuator" refers to a mechanical or electromechanical device capable of controllably adjusting the position or force applied to the cap under the action of a control signal. This actuator can be electrically driven, such as a lead screw, gear, or worm gear mechanism driven by a motor, or it can be hydraulic or electro-hydraulic hybrid, as long as it can achieve a slight movement of the cap in a predetermined direction after receiving a control command. The control unit 105 does not limit the specific structural form of the actuator; its core requirement is that the output control command is consistent with the control interface of the selected actuator.

[0105] The process by which the control unit 105 converts the gland adjustment strategy parameters into control commands typically involves mapping abstract adjustment ranges to specific action quantities. For example, when the adjustment range given in the adjustment strategy parameters is a standardized value, the control unit 105 can convert this value into the displacement, rotation angle, or number of action steps required by the actuator based on a pre-established correspondence. For instance, when using a motor and lead screw structure as the actuator, if it is pre-calibrated that a 0.1 mm axial movement of the gland corresponds to 10 motor rotations, then when the standardized adjustment amount corresponding to the adjustment strategy parameters given by the generation unit 104 is 1, the control unit 105 can generate control commands for "10 forward rotations" or "10 reverse rotations"; when the adjustment amount is 0.5, it corresponds to generating a control command for "5 rotations". In this way, the adjustment strategy parameters are precisely mapped to actions that the actuator can directly execute.

[0106] While converting the adjustment range into specific action quantities, the control unit 105 also needs to reasonably limit the control commands based on the constraint information in the adjustment strategy parameters. For example, even if the adjustment strategy parameters given by the generation unit 104 have already taken into account the packing cooling and lubrication constraints, the control unit 105 still needs to ensure that the control commands do not exceed the mechanical stroke range of the actuator, or that the gland position does not change abruptly in a single adjustment. Therefore, the control unit 105 can verify the control commands before issuing them, and if necessary, break them down into steps so that the adjustment action is completed in the form of multiple small movements.

[0107] When the control unit 105 drives the stuffing gland actuator to make adjustments, it is also responsible for controlling the execution sequence of the adjustment action, i.e., when to start the adjustment, how long to last, and when to end the adjustment. This execution sequence can be set according to the overall operating rhythm of the system. For example, after completing an adjustment action, the control unit 105 can wait for the next sampling and characterization results to be generated, and then decide whether to continue adjusting based on the new adjustment strategy parameters. This method avoids conflicts between the adjustment action and the leakage status acquisition, thereby ensuring the stability of the entire system operation.

[0108] It should be noted that the control unit 105 itself does not make a final judgment on whether the adjustment result meets the standard; its responsibility is limited to executing the corresponding adjustment actions according to the received gland adjustment strategy parameters. After the adjustment action is completed, the tightness of the stuffing box gland changes, which in turn affects the occurrence of leakage. The new leakage state will be reflected again through the acquisition unit 101, the characterization unit 102, and other links, thus providing the input basis for the next round of adjustment. In this way, the control unit 105 forms a key connection point between strategy execution and status feedback in the system.

[0109] Furthermore, the control unit is specifically used for: The system receives the pressure adjustment strategy parameters output by the generation unit, parses the instruction elements of the pressure adjustment strategy parameters, extracts the adjustment direction, adjustment step size and adjustment priority information to characterize the current adjustment behavior, and generates a set of adjustment execution parameters for subsequent execution judgment. Based on the historical adjustment status records of the current stuffing gland, a consistency check is performed on the adjustment execution parameter set before execution. When it is detected that the adjustment direction is opposite to the most recently executed adjustment direction and the adjustment interval is less than the preset minimum execution interval threshold, the execution of this adjustment is restricted or delayed, and an execution inhibition flag is generated. When the execution suppression flag is not triggered, the adjustment step size is segmented according to the adjustment step size in the adjustment execution parameter set to generate at least two sequentially executed sub-adjustment step sizes, and the sub-adjustment step sizes are sequentially converted into segmented control commands that can be recognized by the stuffing gland cap actuator to drive the stuffing gland cap actuator to complete the adjustment of the cap tightness state segment by segment in a predetermined order. After completing the adjustment action corresponding to each sub-adjustment step, the execution feedback information of the stuffing gland actuator is obtained, and the historical adjustment status record is updated in combination with the execution feedback information. When the execution feedback information indicates that the adjustment effect is abnormal or the actuator has not reached the expected position, the execution of the subsequent sub-adjustment step is terminated, and an adjustment interruption status flag is generated as the input basis for the subsequent generation unit to adjust the adjustment strategy parameters.

[0110] In this embodiment, the control unit is located at the execution layer of the entire automatic adjustment system. Its core function is to safely, reliably and orderly convert the adjustment decision results output by the generation unit into the actual mechanical action of the stuffing gland cover, and to introduce necessary constraints and feedback mechanisms during the execution process to avoid the rationality of the front-end judgment and strategy generation due to the instability of the execution layer.

[0111] During system operation, the control unit first receives the cap adjustment strategy parameters output by the generation unit. These cap adjustment strategy parameters are a complete description of a single adjustment action, containing at least three interrelated elements: the adjustment direction indicating whether the cap should tighten or loosen, the adjustment step size quantifying the magnitude of the single adjustment, and adjustment priority information reflecting the urgency and sensitivity of the current adjustment. Upon receiving these strategy parameters, the control unit parses and processes their instruction elements, breaking down the complex strategy parameters into a structured set of adjustment execution parameters. This set of adjustment execution parameters serves as the unified data basis for subsequent execution decisions, ensuring that the control unit can clearly understand key information such as "which direction to adjust," "how much adjustment is needed," and "how fast or how cautiously the adjustment should be performed" at the execution level.

[0112] After generating the set of adjustment execution parameters, the control unit does not immediately drive the actuator. Instead, it introduces a pre-execution consistency verification mechanism to pre-judge the rationality of the current adjustment behavior. To this end, the control unit calls and reads the historical adjustment status record of the stuffing gland cap. The historical adjustment status record refers to a data set that continuously stores and updates the execution time, direction, and results of past adjustment behaviors, reflecting the trajectory of cap adjustment over time. The control unit compares the adjustment direction included in the current set of adjustment execution parameters with the most recently successfully executed adjustment direction, and simultaneously calculates the time interval between the current adjustment time and the most recent adjustment completion time. When it detects that the current adjustment direction is opposite to the most recent adjustment direction, and the time interval is less than a preset minimum execution interval threshold, the control unit determines that the current adjustment behavior carries a risk of frequent reverse adjustments. The minimum execution interval threshold mentioned here refers to the shortest time interval allowed between two adjacent reverse adjustments, which can be set based on the stiffness of the stuffing gland structure, the response speed of the cap transmission mechanism, and actual operating experience, for example, 30 seconds or 60 seconds. When the above conditions are met, the control unit restricts or delays the execution of this adjustment and generates an execution suppression flag to clearly indicate that the current adjustment behavior does not yet meet the conditions for immediate execution.

[0113] When the execution suppression flag is not triggered, or after the delay condition is lifted, the control unit enters the actual execution phase. At this time, the control unit segments the adjustment step according to the adjustment step size in the set of adjustment execution parameters. Segmentation means dividing the adjustment step size, originally planned to be completed in one go, into multiple smaller sub-adjustment steps, which are then executed sequentially. This approach aims to reduce the impact of a single mechanical action on the packing stress state and the stability of the sealing structure. For example, when the theoretical execution quantity corresponding to the adjustment step size is 3 execution units, the control unit can divide it into 3 sub-adjustment steps, each corresponding to 1 execution unit. The control unit then converts each sub-adjustment step size into segmented control commands that can be recognized by the stuffing box gland actuator, and issues them sequentially to drive the actuator to adjust the gland tightness segment by segment. Through this segmented execution method, even with large adjustment ranges, the smoothness and controllability of the adjustment process can be maintained.

[0114] After each sub-adjustment step is completed, the control unit actively acquires the execution feedback information from the stuffing gland actuator. This feedback information reflects whether the actuator has completed the expected action as instructed, and may include an execution position signal, execution stroke feedback value, or equivalent status confirmation information. The control unit compares the acquired execution feedback information with the corresponding segmented control command and updates the historical adjustment status record accordingly, ensuring that the historical record accurately reflects the execution result and status of the latest adjustment. If, after a sub-adjustment step is completed, the control unit detects that the execution feedback information indicates an abnormal adjustment effect—for example, the actuator has not reached the expected position, the execution process is stuck, or the feedback signal is abnormal—the control unit immediately terminates any subsequent unexecuted sub-adjustment steps and ceases issuing control commands. Simultaneously, the control unit generates an adjustment interruption status flag to clearly indicate that the adjustment action was not executed completely as expected.

[0115] The interruption status flag is then passed as input to the generation unit, enabling it to consider execution-level feedback when generating subsequent capping adjustment strategy parameters, thereby correcting or downgrading the adjustment strategy. Through this mechanism, the control unit forms a complete closed loop at the execution level, ensuring not only safe implementation of adjustment actions but also real-time feedback of execution results to the strategy generation stage, preventing the system from continuing to use incompatible adjustment schemes in the event of execution abnormalities.

[0116] Through the above processing, the control unit achieves precise control and safety constraints on the cap adjustment behavior.

[0117] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims of this application.

Claims

1. A large water pump unit stuffing box leakage monitoring and automatic adjusting system, characterized in that, The method comprises the following steps: The acquisition unit is configured to continuously collect discrete water leakage behaviors of the leakage liquid outside the packing gland within a preset sampling period, and obtain original water leakage behavior data corresponding to the characteristics of water leakage behaviors occurring in a unit time; The characterization unit is configured to perform time normalization and dimension mapping processing on the original water leakage behavior data, and generate a standardized water leakage quantity index for characterizing the current water leakage state of the packing gland; The construction unit is configured to construct a target water leakage interval corresponding to the standardized water leakage quantity index based on constraint parameters related to the operating state of the large water pump unit, and determine a safe upper threshold and a safe lower threshold corresponding to the target water leakage interval; The generation unit is configured to analyze the relationship between the standardized water leakage quantity index and the target water leakage interval, generate a water leakage deviation description quantity representing the deviation direction and the deviation amplitude, and generate a gland adjusting strategy parameter based on the water leakage deviation description quantity on the premise of satisfying the packing cooling and lubrication constraint; The control unit is configured to convert the gland adjusting strategy parameter into a control instruction that can be recognized by a packing gland execution mechanism, and drive the packing gland execution mechanism to automatically adjust the tightness state of the gland.

2. The large water pump unit stuffing box leakage monitoring and automatic adjusting system according to claim 1, characterized in that, The acquisition unit is specifically configured to: Real-time detect the dripping, intermittent outflow or intermittent flow of the leakage liquid outside the packing gland on the natural collection or dripping path of the leakage liquid, generate an initial trigger signal corresponding to each leakage phenomenon, and record the occurrence time of the initial trigger signal; Determine the effectiveness of the initial trigger signal based on the time interval between adjacent initial trigger signals, and determine the corresponding initial trigger signal as an independent discrete water leakage behavior when the time interval between the two adjacent initial trigger signals is greater than a preset minimum behavior interval threshold, and generate a corresponding effective water leakage behavior identifier; Within a preset sampling period, accumulate and count all effective water leakage behavior identifiers determined as independent discrete water leakage behaviors, and generate a behavior statistical result representing the total amount of discrete water leakage behaviors occurring in the sampling period; Form original water leakage behavior data corresponding to the characteristics of discrete water leakage behaviors occurring in a unit time based on the behavior statistical result and the corresponding sampling period length.

3. The large water pump unit stuffing box leakage monitoring and automatic adjusting system according to claim 1, characterized in that, The characterization unit is specifically configured to: Receive the original water leakage behavior data output by the acquisition unit, and perform unit time conversion processing on the original water leakage behavior data based on the corresponding sampling period length, to generate a time-normalized water leakage behavior frequency representing the intensity of water leakage behaviors occurring in the current sampling period; Based on the time-normalized water leakage behavior frequency, construct a water leakage behavior frequency sequence in a plurality of consecutive adjacent sampling periods, and perform stability analysis on the water leakage behavior frequency sequence, identify and remove abnormal frequency values deviating from the overall trend of the frequency sequence, and generate an effective water leakage behavior frequency after abnormality suppression processing; According to a pre-established mapping rule between water leakage behavior frequency and water leakage state characterization quantity, perform dimension mapping processing on the effective water leakage behavior frequency, convert it into a continuous numerical value monotonously corresponding to the water leakage degree of the packing gland, and generate a candidate water leakage quantity characterization value; The consistency of the change of the candidate water leakage amount characteristic value in a plurality of continuous sampling periods is verified, and when the change amplitude of the candidate water leakage amount characteristic value in adjacent sampling periods is less than a preset stable change threshold, the candidate water leakage amount characteristic value is determined as the standardized water leakage amount indicator.

4. The large water pump set stuffing box leakage monitoring and automatic adjusting system according to claim 1, characterized in that, The construction unit is specifically used for: According to the current obtained large water pump unit running state related constraint parameter, the running state of the water pump unit is judged, and the working condition grade identification for representing the current running load characteristic is generated, wherein the working condition grade identification is used to limit the basic allowable water leakage range of the stuffing box under the running state; Based on the working condition grade identification, the corresponding relationship between the working condition grade and the water leakage control rule is determined, and the basic water leakage interval corresponding to the current working condition grade identification is determined as the initial interval range of the target water leakage interval; On the basis of the initial interval range, the lowest water leakage intensity constraint required by the cooling and lubrication of the filler and the highest water leakage intensity constraint required by the sealing performance control are combined, and the upper and lower boundaries of the initial interval range are expanded or contracted, respectively, to determine the safe lower threshold and the safe upper threshold corresponding to the target water leakage interval; Based on the safe lower threshold, the safe upper threshold and the distribution of the standardized water leakage amount indicator in a plurality of continuous sampling periods, the consistency of the target water leakage interval is verified, and when the standardized water leakage amount indicator is continuously and stably distributed in the target water leakage interval, the interval width of the target water leakage interval is corrected to form the final target water leakage interval suitable for the current running state.

5. The large water pump set stuffing box leakage monitoring and automatic adjusting system according to claim 1, characterized in that, The generation unit is specifically used for: Based on the relative position relationship between the standardized water leakage amount indicator and the target water leakage interval, the deviation of the current stuffing box water leakage state is judged, and the water leakage deviation direction identification for representing the position relationship of the standardized water leakage amount indicator relative to the target water leakage interval is generated, wherein the water leakage deviation direction identification is used to distinguish whether the water leakage state is above, in or below the target water leakage interval; On the basis of generating the water leakage deviation direction identification, the distance value between the standardized water leakage amount indicator and the boundary of the target water leakage interval is calculated, the water leakage deviation amplitude quantity for representing the water leakage deviation degree is generated, and the water leakage deviation direction identification and the water leakage deviation amplitude quantity jointly constitute the water leakage deviation description quantity; Based on the water leakage deviation description quantity and combined with the filler cooling and lubrication constraint, the adjustment priority corresponding to different water leakage deviation directions is distinguished, when the water leakage deviation direction identification indicates that the water leakage amount is lower than the target water leakage interval, the adjustment priority is limited to avoid insufficient cooling of the filler, when the water leakage deviation direction identification indicates that the water leakage amount is higher than the target water leakage interval, the adjustment priority is improved to control the water leakage loss, and the adjustment priority identification matched with the current water leakage state is generated. According to the adjustment priority identifier and the water leakage deviation amplitude, an adjustment direction and an adjustment step of single cover adjustment are determined according to a preset adjustment mapping rule, and an upper limit constraint of single adjustment is introduced in the adjustment step generation process to prevent over-adjustment, and finally a cover adjustment strategy parameter used to describe the current adjustment behavior is generated.

6. The large water pump set stuffing box leakage monitoring and automatic adjusting system according to claim 1, characterized in that, The control unit is specifically used for: receiving the cover adjustment strategy parameter output by the generation unit, and performing instruction element analysis on the cover adjustment strategy parameter to extract adjustment direction, adjustment step and adjustment priority information used to represent the current adjustment behavior, and generating an adjustment execution parameter set used for subsequent execution judgment; based on the historical adjustment state record of the current stuffing box cover, performing a consistency check on the adjustment execution parameter set before execution, when the adjustment direction is opposite to the most recently executed adjustment direction and the adjustment interval time is less than the preset minimum execution interval threshold, limiting or delaying the current adjustment execution, and generating an execution inhibition identifier; when the execution inhibition identifier is not triggered, the adjustment step in the adjustment execution parameter set is segmented and split to generate at least two sequentially executed sub-adjustment steps, and the sub-adjustment steps are sequentially converted into segmented control instructions recognizable by the stuffing box cover execution mechanism to drive the stuffing box cover execution mechanism to complete the adjustment of the cover tightness state in a predetermined order; after completing the adjustment action corresponding to each sub-adjustment step, the execution feedback information of the stuffing box cover execution mechanism is obtained, and the historical adjustment state record is updated combined with the execution feedback information, when it is detected that the execution feedback information indicates that the adjustment effect is abnormal or the execution mechanism does not reach the expected position, the execution of the subsequent sub-adjustment step is terminated, and an adjustment interruption state identifier is generated as an input basis for the subsequent generation unit to adjust the adjustment strategy parameter.

7. The large water pump set stuffing box leakage monitoring and automatic adjusting system according to claim 2, characterized in that, The collection unit is also used for: based on the stuffing box structure form, the running vibration level and the flow state characteristics of the leaked liquid, determining a minimum behavior interval threshold for distinguishing between independent water leakage behaviors and continuous leakage processes; merging processing is performed on a plurality of initial trigger signals continuously generated within the limited range of the minimum behavior interval threshold, and the initial trigger signals satisfying the time interval less than or equal to the minimum behavior interval threshold are merged into the same leakage process group, and a process trigger identifier corresponding to the leakage process group is generated; based on the process trigger identifier, the number and duration of the initial trigger signals in each leakage process group are analyzed, when the duration of the leakage process group exceeds the preset minimum duration threshold, the corresponding leakage process group is determined as an effective water leakage process, and a candidate water leakage behavior identifier is generated; based on the time interval relationship between the candidate water leakage behavior identifier and the adjacent leakage process group, the candidate water leakage behavior identifier is finally confirmed, when the time interval between the adjacent two candidate water leakage behavior identifiers is greater than the minimum behavior interval threshold, the candidate water leakage behavior identifier is confirmed and output as an effective water leakage behavior identifier corresponding to an independent discrete water leakage behavior.

8. The large water pump set stuffing box leakage monitoring and automatic adjusting system according to claim 3, characterized in that, The representation unit is also used for: Based on the structure parameters of the filler function, the type of the filler material and the diameter information of the shaft sleeve, the value range of the effective water leakage behavior frequency is segmented and divided, a plurality of non-overlapping frequency intervals are generated, and a corresponding local mapping interval identifier is established for each frequency interval; For each frequency interval, a monotonic mapping relationship between the frequency interval and the water leakage state representation is constructed according to the lubrication and wear characteristics of the filler under the corresponding water leakage intensity, forming a segmented mapping rule set composed of a plurality of local mapping relationships; When the effective water leakage behavior frequency falls into the determined frequency interval, the local mapping relationship corresponding to the frequency interval is called to perform dimension mapping processing on the effective water leakage behavior frequency, and an initial candidate water leakage quantity representation value corresponding to the frequency interval is generated; Based on the continuity of the change of the initial candidate water leakage quantity representation value in the adjacent sampling period, the interval consistency of the initial candidate water leakage quantity representation value is checked, and when it is detected that the initial candidate water leakage quantity representation value has a cross-interval transition in the adjacent sampling period, the initial candidate water leakage quantity representation value is corrected by boundary limitation, and the corrected result is output as the candidate water leakage quantity representation value.

9. The large water pump set stuffing box leakage monitoring and automatic adjusting system according to claim 4, characterized in that, The construction unit is also used for: Based on the standardized water leakage quantity indicators obtained in the continuous multiple sampling periods, a corresponding water leakage quantity distribution sequence is constructed, and the interval occupancy proportion of the standardized water leakage quantity indicator in each sampling period with respect to the target water leakage interval is calculated, which is used to represent the concentration degree of the water leakage quantity in the target water leakage interval; According to the change of the interval occupancy proportion in the continuous multiple sampling periods, it is judged whether the standardized water leakage quantity indicator presents a continuous and stable distribution state, and when the interval occupancy proportion remains above the preset stable occupancy threshold in the preset stable judgment period, a target water leakage interval stability identifier is generated; After generating the target water leakage interval stability identifier, the interval width of the target water leakage interval is subjected to convergence correction processing based on the statistical dispersion degree of the standardized water leakage quantity indicator in the target water leakage interval, wherein the convergence correction processing narrows the interval span of the target water leakage interval by simultaneously moving the safety lower limit threshold and the safety upper limit threshold inward; After completing the interval width convergence correction, the modified target water leakage interval is subjected to consistency review with the lowest water leakage intensity constraint required for filler cooling and lubrication and the highest water leakage intensity constraint required for sealing performance control, and when the modified target water leakage interval meets the constraint conditions, it is determined as the final target water leakage interval suitable for the current operating state.

Images