A hot water system water replenishment valve and water supply pump state abnormality alarm system and method

Abstract

The present application relates to hot water system fault diagnosis technical field, and disclose a kind of hot water system water replenishing valve and water supply pump state abnormality alarm system and method, including valve action monitoring module, pump operation monitoring module, data acquisition and synchronization module, multi-time scale modeling analysis module and alarm diagnosis module, valve action monitoring module extracts opening response time, closing response time and stroke curve form by displacement sensor and instruction acquisition interface, compare with factory reference parameter deviation, identify valve mechanical failure early signs and realize early warning before pipeline pressure and flow parameter appear detectable abnormality, multi-time scale modeling analysis module establishes seasonal statistical value according to heating season and non-heating season, separate seasonal periodic fluctuation and long-term monotone degradation trend from long-term operation data, obtain equipment real degradation residual trend after deducting seasonal normal deviation, alarm diagnosis module receives evaluation result and executes hierarchical alarm.

Description

Technical Field

[0001] This invention relates to the field of hot water system fault diagnosis technology, and in particular to a hot water system water supply valve and water supply pump status abnormal alarm system and method. Background Technology

[0002] In hot water systems, the water supply valve and the water supply pump jointly undertake the functions of constant pressure water supply and hydraulic circulation. The water supply valve receives commands from the control system to execute opening and closing actions, and its physical action status, such as valve stem displacement and action response time, can directly characterize the valve's health level. The water supply pump provides power for circulation, and the two form a hydraulic coupling through pipelines. Moreover, the system operation is deeply modulated by time-varying factors such as seasonal changes and diurnal load fluctuations. The frequency of action of the water supply valve and the operating conditions of the water supply pump show significant seasonal differences and non-steady-state characteristics across the entire time scale. Existing monitoring methods mostly rely on indirect parameters such as pipeline pressure and flow rate, making it difficult to directly perceive the time-domain dynamic response of valve actions, and also lacking the ability to model and analyze the evolution of equipment operating characteristics across seasons.

[0003] In the prior art, such as the invention patent with publication number CN112880000A, a fault diagnosis and alarm method for an automatic water replenishment and pressure stabilization system of a heat exchanger unit's secondary network is disclosed. This method obtains the operating status information of the circulating pump by setting a continuous operating time threshold for the water replenishment pump, an upper limit threshold for the secondary return water pressure, and a lower limit threshold. Based on the comparison between the operating status of the water replenishment pump, the cumulative flow of water replenishment, and the secondary return water pressure with the corresponding thresholds, the network fault is diagnosed. This solution achieves automatic diagnosis of water replenishment system faults to a certain extent.

[0004] The above-mentioned and existing related technologies often have the following drawbacks: 1. Existing technologies only collect indirect parameters such as pipeline pressure and flow rate, and cannot obtain direct evidence of the valve's physical operating state. This makes it impossible to distinguish between the valve's normal response to control commands and its own mechanical faults. There is a logical blind spot in the identification of valve faults. Early signs of mechanical faults in water supply valves are changes in time-domain characteristics such as prolonged action response time and distortion of stroke curves. These changes exist long before the pipeline parameters show detectable abnormalities. However, existing technologies are completely unable to perceive such dynamic response characteristics, so alarms are only triggered after the fault has fully developed.

[0005] 2. Existing technology treats the annual operating data as a uniform steady-state process, lacking the ability to model the evolution of the operating characteristics of the water supply valve-water pump combination at multiple time scales. It cannot distinguish between normal parameter deviations caused by seasonal changes in system operating conditions and abnormal trends caused by long-term slow degradation of equipment performance, resulting in a large number of false alarms when switching between heating and non-heating seasons, while completely missing out on cross-seasonal gradual faults. Summary of the Invention

[0006] The technical problem to be solved by the present invention is that the existing technology has the disadvantage of being unable to distinguish between the normal response control command of the water supply valve and its own mechanical failure, and unable to distinguish between the normal deviation of the seasonal operating conditions of the hot water system and the slow degradation of equipment performance. To this end, we propose an alarm system and method for abnormal status of the water supply valve and the water supply pump of the hot water system.

[0007] To achieve the above objectives, this application adopts the following technical solution: a hot water system water supply valve and water supply pump status abnormality alarm system, comprising: The valve action monitoring module includes a displacement sensing unit, a command acquisition unit, a zero-position calibration unit, and a signal conditioning unit, which are used to measure the valve stem displacement stroke in real time and record the command time synchronously. The time-domain feature analysis module is connected to the valve action monitoring module. It has a built-in benchmark parameter library and morphological comparison algorithm. It is used to extract response time parameters and curve morphological parameters from the valve stem displacement stroke, compare the deviation with the benchmark parameters, and output the valve mechanical condition assessment results before the pipeline pressure and flow are abnormal. The pump operation monitoring module includes a pressure sensing unit, a current sensing unit, and a flow sensing unit, which are used to acquire the operating status parameters of the water supply pump. A multi-timescale modeling and analysis module is connected to the pump operation monitoring module. It has a built-in seasonal pattern recognition submodule and a degradation trend separation submodule. The seasonal pattern recognition submodule stores the parameter statistics under different seasonal operating conditions. The degradation trend separation submodule is used to retrieve the corresponding statistical values ​​as a comparison benchmark according to the current seasonal operating conditions, calculate the difference between the current operating parameters and the comparison benchmark to deduct seasonal factors, and obtain the residual trend that reflects the true degradation of equipment performance. The data acquisition and synchronization module uses a synchronous acquisition unit to synchronously acquire the above data and generate a multi-dimensional synchronous dataset, which is then provided to the time-domain feature analysis module and the multi-time-scale modeling and analysis module, respectively. The alarm diagnosis module receives the valve mechanical condition assessment results and the residual trend, respectively, and has a built-in fault causal transmission rule library. When only abnormal valve mechanical condition assessment results are received, it outputs a water supply valve mechanical fault alarm; when only abnormal residual trend is received, it outputs a water supply pump performance degradation alarm; when both abnormalities are received simultaneously, it compares the start time order of the degradation trends of the water supply valve and the water supply pump, and based on the fault transmission path based on hydraulic coupling relationship in the rule library, it identifies the device with the earlier start time as the degradation source device and outputs a graded alarm signal containing fault root cause location.

[0008] Preferably, the displacement sensing unit is an LVDT linear displacement sensor installed on the valve stem of the water supply valve; the command acquisition unit is a command acquisition interface installed on the control loop of the water supply valve actuator, connected in parallel with the control system signal line; the zero-position calibration unit is a reference position calibration block fixed at the end of the valve stem corresponding to the fully closed position; the signal conditioning unit is a differential signal conditioning circuit with a built-in common-mode noise suppression unit, used to convert the displacement analog signal into a differential signal and transmit it to the data acquisition and synchronization module.

[0009] Preferably, the time-domain feature analysis module stores the baseline parameter library containing the baseline opening response time, baseline closing response time, and baseline stroke curve shape calibrated at the factory of the water supply valve; the response time parameters include the opening response time and the closing response time, wherein the opening response time is the time from the issuance of the opening command recorded by the command acquisition unit to the valve stem displacement reaching the preset proportion of the full stroke, and the closing response time is the time from the issuance of the closing command to the valve stem displacement returning to zero; the shape comparison algorithm is a dynamic time warping comparison algorithm, used to nonlinearly align the real-time extracted curve shape with the baseline curve shape and calculate the shape correlation coefficient.

[0010] Preferably, the time-domain feature analysis module determines that the valve mechanical condition assessment result is abnormal under the following conditions: the opening response time is greater than the product of the benchmark opening response time and the first preset ratio, or the closing response time is greater than the product of the benchmark closing response time and the second preset ratio, or the morphological correlation coefficient is lower than the preset threshold; wherein the typical value of the first preset ratio and the second preset ratio is 1.3.

[0011] Preferably, the pressure sensing unit is a pressure sensor installed on the outlet pipeline of the water supply pump; the current sensing unit is a current transformer installed on the power supply circuit of the water supply pump; the flow sensing unit is a flow sensor installed on the return water pipeline of the system; and the operating status parameters of the water supply pump include outlet pressure, operating current, and system makeup water flow.

[0012] Preferably, the synchronization acquisition unit of the data acquisition and synchronization module is a multi-channel synchronous data acquisition card, and the analog-to-digital converters of each channel share the same sampling clock; the data acquisition and synchronization module also has a differential receiving circuit, which is used to restore the differential signal output by the signal conditioning unit to a single-ended displacement signal, and to perform anti-aliasing filtering and timing alignment processing on the acquired data to generate a multi-dimensional synchronous dataset with consistent timestamps.

[0013] Preferably, the parameter statistics stored in the seasonal pattern recognition submodule include: the statistical values ​​of the frequency of action and the response time of the water supply valve under heating and non-heating season conditions, and the statistical values ​​of the outlet pressure and operating current of the water supply pump under heating and non-heating season conditions; wherein the statistical value of the daily number of actions of the water supply valve is greater in the heating season than in the non-heating season; the seasonal pattern recognition submodule is also used to incorporate the newly generated operating data of the current season into the calculation at the end of each season and to perform sliding updates on the stored statistical values.

[0014] Preferably, the degradation trend separation submodule is used to obtain the time series of daily cumulative action counts of the water supply valve and the time series of daily cumulative operating time of the water supply pump within the past twelve months. A local weighted regression smoothing algorithm is used to decompose the time series into trend components, seasonal cycle components, and residual components. A long-term monotonic degradation trend line is extracted from the trend component, and the slope of this trend line over the past thirty days is calculated. When the slope exceeds a preset degradation rate threshold, a performance degradation trend is determined to exist. The method for deducting seasonal factors is as follows: the daily cumulative action counts of the water supply valve or the daily cumulative operating time, daily average operating current, or daily average outlet pressure of the water supply pump are subtracted from the corresponding seasonal operating conditions' action frequency statistics, operating time statistics, operating current statistics, or outlet pressure statistics, respectively, to obtain a daily residual value series after deducting seasonal factors. Then, a residual trend reflecting the true degradation of equipment performance is extracted from this series.

[0015] Preferably, the fault causal transmission rule base built into the alarm diagnosis module includes: Rule 1, when the impeller wear of the water supply pump causes a decrease in head, the frequency of operation of the water supply valve increases abnormally; Rule 2, when internal leakage occurs on the sealing surface of the water supply valve, the running time or running frequency of the water supply pump increases abnormally; when the alarm diagnosis module compares the order of start times, if the start time of the water supply pump degradation trend is earlier than that of the water supply valve, then the source device is determined to be the water supply pump according to Rule 1, otherwise the source device is determined to be the water supply valve according to Rule 2; the graded alarm signal includes the fault type, fault location and fault severity, and is sent to the upper monitoring system through relay output or communication bus.

[0016] Preferably, the above-mentioned system's abnormal status alarm method includes the following steps: S1: The valve stem displacement stroke and command time are obtained through the valve action monitoring module, the water supply pump operation status parameters are obtained through the pump operation monitoring module, and a multi-dimensional synchronous dataset is generated through the data acquisition and synchronization module. S2: Through the time-domain feature analysis module, the response time parameters and curve shape parameters are extracted from the multidimensional synchronous dataset, and the deviation is compared with the benchmark parameters to output the valve mechanical condition evaluation result. S3: Using the multi-timescale modeling and analysis module, the corresponding statistical values ​​are retrieved based on the current seasonal operating conditions as a comparison benchmark. The difference between the current operating parameters and the benchmark is calculated to deduct seasonal factors and obtain the residual trend. S4: Through the alarm diagnosis module, when both abnormal valve mechanical condition assessment results and abnormal residual trends are received simultaneously, the starting time of the degradation trends of the water supply valve and the water supply pump is compared. Based on the built-in fault causal transmission rule library, the device with the earlier starting time is identified as the degradation source device, and a graded alarm signal containing fault root cause location is output. Technical effects and advantages of the present invention: 1. This invention directly acquires the physical action status of the water supply valve stem through a valve action monitoring module, extracts time-domain characteristic parameters such as opening response time, closing response time, and stroke curve shape, and compares the deviation with the factory-calibrated benchmark parameters. It can identify early signs of valve mechanical failure before detectable anomalies appear in indirect parameters such as pipeline pressure and flow rate, achieving a fundamental leap from delayed alarms based on indirect parameters to early warnings based on physical actions. This solves the shortcomings of existing technologies, such as logical blind spots in valve fault identification and alarms that are only triggered after the fault has fully developed.

[0017] 2. This invention establishes seasonal statistical values ​​for both heating and non-heating seasons through a multi-timescale modeling and analysis module. It utilizes a degradation trend separation submodule to separate seasonal periodic fluctuation components and long-term monotonic degradation trend components from long-term operational data. After deducting seasonal normal offsets, it obtains the residual trend of the true degradation of equipment performance. This solves the shortcomings of existing technologies, such as generating numerous false alarms during the transition between heating and non-heating seasons and completely missing cross-seasonal progressive faults. Simultaneously, when degradation trends occur simultaneously in the water supply valve and water pump, the alarm diagnosis module distinguishes between degradation source equipment and passively responding equipment through a built-in fault causal propagation rule library, outputting an alarm signal containing fault root cause location, effectively avoiding ineffective maintenance and maintenance delays. Attached Figure Description

[0018] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts: Figure 1 This is a schematic diagram of the overall system module structure of the present invention; Figure 2 This is a flowchart of the abnormal status alarm method of the present invention; Figure 3 This is a flowchart of the degradation trend analysis algorithm for the multi-timescale modeling and analysis module of the present invention; Figure 4 This is a comparison chart of the early warning time of valve failure according to the present invention and the early warning time of traditional monitoring methods; Figure 5 This is a comparison chart of the number of false alarms between the seasonal switching false alarm prevention system of the present invention and the traditional fixed threshold alarm system; Figure 6 This is a comparison chart showing the progressive degradation identification of water supply pumps during the non-heating season and the degradation trend identification of traditional monitoring methods, based on the present invention. Detailed Implementation

[0019] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0020] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings, such as... Figures 1 to 6 As shown, the hot water system water supply valve and water supply pump status abnormal alarm system and method of the present invention can be applied to industrial scenarios that require constant pressure water supply and hydraulic circulation, such as centralized heating heat exchange stations and building hot water supply systems.

[0021] Before proceeding with a detailed description, the technical terms used in this specific embodiment will be defined to facilitate understanding.

[0022] A water supply valve is an automatic valve installed on the return water pipe of a hot water system or on the suction side of a water supply pump. It receives instructions from the control system to perform opening and closing actions and is used to supplement water supply to the system when the system pressure is lower than the set value. Its valve stem displacement stroke directly reflects the actual action state of the valve. A water supply pump is a water pump equipment that provides power for the circulation of hot water system. Its outlet pressure, operating current and other parameters can characterize the operating conditions and performance status of the pump.

[0023] Time-domain characteristic parameters refer to the set of parameters extracted from the valve stem displacement time history data of the water supply valve, reflecting the valve's action time characteristics and motion pattern. These include opening response time, closing response time, full stroke displacement time, and valve stem displacement-time curve morphology. These time-domain characteristic parameters can characterize early signs of valve mechanical failure before detectable anomalies appear in indirect parameters such as pipeline pressure and flow. For specific parameter definitions and extraction methods, please refer to the detailed description of the time-domain characteristic analysis module later. The benchmark time-domain characteristic parameter library refers to the set of time-domain characteristic parameters calibrated by the water supply valve under factory test conditions, providing a reference benchmark for comparing deviations of time-domain characteristic parameters. For specific parameters and applications, please refer to the detailed description of the time-domain characteristic analysis module later.

[0024] Multi-timescale modeling refers to a modeling method that simultaneously performs feature analysis and trend extraction on long-term operating data of water supply valves and water pumps from multiple time dimensions. By separating the seasonal fluctuation component with an annual cycle and the long-term monotonic degradation trend component spanning multiple seasonal cycles, it can distinguish between normal parameter deviations caused by seasonal changes in system operating conditions and abnormal trends caused by long-term slow degradation of equipment performance. The structure and algorithm steps of the seasonal pattern recognition submodule and the degradation trend separation submodule are described in detail in the following section on the multi-timescale modeling and analysis module.

[0025] The fault causal propagation rule base refers to the set of rules that store the known fault causal propagation paths between the water supply valve and the water supply pump. For specific rule content, please refer to the detailed description of the alarm diagnosis module later.

[0026] Early signs of valve mechanical failure refer to the abnormal time-domain characteristics exhibited in the valve stem displacement time history data before mechanical failures such as valve core jamming, sealing surface wear, or actuator performance degradation fully develop. These abnormal changes include prolonged opening response time, delayed closing response time, and distortion of the stroke curve. These abnormal changes exist long before detectable anomalies appear in indirect parameters such as pipeline pressure and flow.

[0027] Hydraulic coupling refers to the mutual influence between the water supply valve and the water supply pump formed by the connection of pipelines. The opening and closing state of the water supply valve changes the suction pressure condition of the water supply pump, and the operating condition of the water supply pump affects the pressure difference and operating frequency of the water supply valve. Due to the existence of hydraulic coupling, the performance change of any component will be reflected in the operating parameters of the other component through hydraulic transmission, making it particularly difficult to accurately distinguish the source of the fault. The specific manifestation of the hydraulic coupling transmission path is described in detail in the alarm diagnosis module later.

[0028] Seasonal normal parameter deviation refers to the normal changes in the operating parameters of a hot water system caused by seasonal changes rather than equipment performance degradation. Slow equipment degradation refers to the trend of equipment performance gradually declining over long-term operation. One of the core improvements of this invention is to accurately distinguish between the two types of changes by establishing a seasonal baseline model and separating the seasonal periodic fluctuation components. The specific implementation method is described in detail in the multi-timescale modeling and analysis module below.

[0029] Before going into a detailed description of each module, let me first explain the method for setting the key indicators in this system. The key indicators involved in this system can be divided into two categories according to their nature: one is the benchmark parameters derived from the factory test of the water supply valve, and the other is the seasonal statistical values ​​established through statistics during the long-term operation of the system.

[0030] The method for setting factory reference parameters: The reference opening response time, reference closing response time, and reference stroke curve morphology of the water supply valve are obtained from the standard test procedure before the valve leaves the factory. The specific steps are as follows: Step 1: Set up a standard test environment. Install the water supply valve and its matching actuator on a dedicated test bench and connect it to a standard hydraulic power source. Use room temperature clean water as the test medium. Maintain the system pressure at the rated working pressure of the water supply valve. Ensure that the displacement sensor is the same model as the actual system configuration, fix it to the end of the valve stem and connect it to the high-speed data acquisition system. Connect the command acquisition interface in parallel with the command signal line of the test controller.

[0031] Step Two: Execute the standard test sequence, which includes two types: single-action test and continuous-action test. In the single-action test, the test controller issues an opening command, and after the valve stem displacement reaches its full stroke, it issues a closing command until the valve stem displacement returns to zero. This process is repeated multiple times, and the valve stem displacement time history data and command signal time of each action are completely recorded. The continuous-action test simulates actual working conditions. The test controller issues a continuous sequence of opening and closing commands with randomly varying command intervals. The valve stem displacement time history data and command signal time of the entire process are completely recorded.

[0032] Step 3: Calculation and extraction of baseline parameters. The baseline opening response time is the arithmetic mean of the opening response times in multiple single-action tests; the baseline closing response time is the arithmetic mean of the closing response times in multiple single-action tests; the baseline stroke curve morphology is formed by selecting multiple full-stroke actions with the most stable and uninterrupted motion speed from the continuous motion test data, extracting their displacement-time curves, aligning and averaging them using a dynamic time normalization algorithm, and generating a standardized displacement-time average curve.

[0033] Once calibrated, the aforementioned benchmark parameters are written into the benchmark time-domain feature parameter library built into the time-domain feature analysis module, serving as an absolute reference standard for determining whether the mechanical condition of the valve has degraded during in-service operation.

[0034] The method for setting seasonal statistical values: In the multi-timescale modeling and analysis module, the seasonal pattern recognition submodule stores the statistical values ​​of action frequency, action response time, outlet pressure, and operating current. These values ​​are obtained through seasonal statistical analysis of long-term system operating data. The specific steps are as follows: Step 1: Define the rules for dividing the seasons, and clearly define the start and end dates of the heating season and the non-heating season based on the climate conditions and heating regulations of the system's operating location.

[0035] Step 2: Collect long-term operating data. After the system is put into operation, daily operating data is continuously recorded and stored. The recorded content includes the daily cumulative number of actions of the water supply valve, the opening and closing response time of each action, the daily average outlet pressure of the water supply pump, and the daily average operating current of the water supply pump. The data collection period covers at least one full year.

[0036] Step 3: Group and statistically analyze the data by season. The collected long-term data is divided into heating season dataset and non-heating season dataset according to the above rules. Within each seasonal dataset, the following calculations are performed: Action frequency statistics, which is the median of the cumulative number of daily actions of the water supply valve on all valid days of the season; Action response time statistics, which is the high quantile of the opening and closing response times among all valid action records of the season; Outlet pressure statistics and operating current statistics, which are the arithmetic averages of the daily average outlet pressure and daily average operating current of the water supply pump on all valid days of the season, respectively.

[0037] Step 4: Writing and updating initial statistical values ​​online. The calculated initial statistical values ​​for the heating season and non-heating season are written into the seasonal pattern recognition submodule to serve as the comparison benchmark for the first operating year. Thereafter, at the end of each season, the newly generated operating data of that season are included in the calculation, and the stored statistical values ​​are updated by sliding. The statistical values ​​can smoothly track the slow changes in equipment status, thus ensuring the accuracy of the degradation trend analysis results.

[0038] Reference Figure 1 As shown, this invention provides a technical solution: a status anomaly alarm system for a hot water system's water supply valve and pump, comprising a valve action monitoring module, a pump operation monitoring module, a data acquisition and synchronization module, a time-domain feature analysis module, a multi-time-scale modeling and analysis module, and an alarm diagnosis module. The valve action monitoring module and the pump operation monitoring module acquire the operating status signals of the water supply valve and the pump, respectively. The data acquisition and synchronization module is used for synchronous acquisition and time-series alignment of the signals. The time-domain feature analysis module is used to extract time-domain features from the valve stem displacement data and evaluate the valve's mechanical state. The multi-time-scale modeling and analysis module is used to separate seasonal fluctuations and extract equipment degradation trends. The alarm diagnosis module is used to comprehensively evaluate the results and output graded alarm signals. The detailed composition and functions of each module are described below: The valve action monitoring module includes a displacement sensor mounted on the valve stem of the water supply valve, an instruction acquisition interface mounted on the control circuit of the water supply valve actuator, and a reference position calibration block mounted on the valve body of the water supply valve. The displacement sensor is an LVDT linear displacement sensor, the range of which is selected according to the actual stroke of the water supply valve. It is installed at the end of the valve stem to measure the valve stem displacement stroke in real time and output the valve stem displacement time history data. The instruction acquisition interface is connected in parallel with the control signal line of the water supply valve to synchronously record the time of the opening and closing instruction signals sent by the control system to the water supply valve. The reference position calibration block is fixed on the valve body of the water supply valve and located at the end of the valve stem corresponding to the fully closed position of the valve stem. It provides an absolute zero calibration reference for the displacement sensor and eliminates zero drift caused by installation errors or thermal expansion and contraction of the valve body.

[0039] Furthermore, the valve action monitoring module also includes a differential signal conditioning circuit. The displacement analog signal output by the displacement sensor is converted into a differential signal by the differential signal conditioning circuit and then transmitted to the data acquisition and synchronization module. The differential signal conditioning circuit has a built-in common mode noise suppression unit to suppress electromagnetic interference generated by the frequency converter and water pump motor, ensuring that the valve stem displacement measurement accuracy is better than the preset displacement resolution.

[0040] This module directly acquires the valve stem displacement stroke through a displacement sensor and synchronously acquires the control command timing through a command acquisition interface. It ensures displacement measurement accuracy through differential signal conditioning circuitry and common-mode noise suppression unit, and eliminates the influence of actuation speed fluctuations on curve shape comparison through a dynamic time warping comparison algorithm. This enables the system to identify early signs of valve mechanical failure before detectable anomalies appear in indirect parameters such as pipeline pressure and flow. It also provides a basis for distinguishing between normal valve response to control commands and its own mechanical failures. Specifically, the reference position calibration block provides an absolute zero-point reference for the displacement sensor, eliminating zero-point drift during long-term operation; the differential signal conditioning circuitry and common-mode noise suppression unit solve the problem of reliable acquisition of weak signals in strong electromagnetic interference environments in industrial settings; and the dynamic time warping comparison algorithm solves the problem of misjudgment of curve shape caused by fluctuations in valve actuation speed. The specific deviation comparison and judgment logic is described in detail in the time-domain feature analysis module below.

[0041] The pump operation monitoring module includes a pressure sensor installed on the outlet pipeline of the water supply pump, a current transformer installed on the power supply circuit of the water supply pump, and a flow sensor installed on the return water pipeline of the system, which respectively measure the outlet pressure of the water supply pump, the operating current and the system makeup water flow in real time.

[0042] The data acquisition and synchronization module is electrically connected to both the valve action monitoring module and the pump operation monitoring module. It employs an industrial-grade multi-channel synchronous data acquisition card, with each channel's analog-to-digital converter sharing the same sampling clock to ensure data synchronization superior to the preset time accuracy. This module receives the valve stem displacement differential signal output from the differential signal conditioning circuit in the valve action monitoring module, as well as the command signal timing data output from the command acquisition interface and the water pump operation status parameters output from the pump operation monitoring module. It performs synchronous acquisition using a unified time reference and performs anti-aliasing filtering and time-series alignment processing on the acquired data to generate a multi-dimensional synchronous dataset with consistent timestamps. The differential signal acquisition channel has a built-in differential receiving circuit that restores the differential signal to a single-ended displacement signal before sending it to the analog-to-digital converter, further improving the anti-interference capability of signal transmission.

[0043] The time-domain feature analysis module is connected to the data acquisition and synchronization module and is implemented by an embedded industrial control computer. This module extracts the time-domain feature parameters of the water supply valve from the valve stem displacement time history data in the multi-dimensional synchronous dataset and generates the valve mechanical condition assessment results. It has a built-in reference time-domain feature parameter library and dynamic time warping comparison algorithm calibrated at the factory of the water supply valve.

[0044] The time-domain characteristic parameters are defined as follows: Enable response time This refers to the time required from the issuance of the opening command to the valve stem displacement reaching the preset proportion of its full stroke; the closing response time. This refers to the time required from the issuance of the closing command to the valve stem displacement returning to zero; the total stroke displacement time. ; and the morphological characteristics of the valve stem displacement-time curve. .

[0045] The reference time-domain characteristic parameter library includes: reference start-up response time. This refers to the time required for the water supply valve to reach 90% of its full stroke after receiving an opening command under factory testing conditions; the reference closing response time. This refers to the time required for the water supply valve to return to zero displacement from receiving a closing command under factory testing conditions; and the morphological characteristics of the reference stroke curve. This refers to the standard displacement-time curve of the valve stem displacement from zero position to full stroke under factory test conditions.

[0046] Dynamic time warping alignment algorithm is used to extract the morphological features of the stroke curve in real time. Shape characteristics of the reference stroke curve After performing nonlinear time axis alignment to eliminate the difference in curve time axis scaling caused by valve action speed fluctuations, the similarity between the two is calculated, and the morphological correlation coefficient is output.

[0047] The deviation comparison and judgment logic is as follows: Let The first preset ratio, This is the second preset ratio. A threshold is preset for the morphological correlation coefficient. The extracted time-domain feature parameters are compared with the baseline parameters. If... If so, it is determined that the valve core has a tendency to jam; if If the valve core return is abnormal, it is determined that there is an anomaly; if the morphological correlation coefficient calculated by the dynamic time warping comparison algorithm is lower than 1, it is considered that there is an anomaly in the valve core return. If this is the case, it is determined that the smoothness of valve operation has decreased. and The typical value is 1.3.

[0048] This module extracts time-domain characteristic parameters such as the response time and stroke curve shape of the valve stem action and compares them with the factory benchmark. It uses a dynamic time warping comparison algorithm to eliminate the influence of action speed fluctuations on the curve shape comparison. It can identify early signs of valve mechanical failure before detectable anomalies appear in indirect parameters such as pipeline pressure and flow, thus achieving early warning of valve mechanical failure. The method for determining whether the water supply valve is in a normal response to control commands or in a mechanical failure state is as follows: when the opening response time, closing response time and stroke curve shape are all within the allowable deviation range, the water supply valve is considered to be in a normal response to control commands. When the opening response time or closing response time exceeds the allowable deviation range or the correlation coefficient between the stroke curve shape and the benchmark shape is lower than a preset threshold, the water supply valve is considered to have a mechanical failure.

[0049] The multi-timescale modeling and analysis module is connected to the data acquisition and synchronization module and is implemented by an embedded industrial control computer. It has a built-in seasonal pattern recognition submodule and degradation trend separation submodule.

[0050] The seasonal pattern recognition submodule stores baseline models of normal operating parameters for the water supply valve and water pump under different seasonal conditions, established based on historical operating data. The seasonal conditions include at least heating season conditions and non-heating season conditions. The baseline models of normal operating parameters include statistical values ​​of the frequency of action and the response time of the water supply valve under different seasonal conditions, as well as statistical values ​​of the outlet pressure and operating current of the water pump under different seasonal conditions. Taking the daily number of actions of the water supply valve as an example, the baseline value of the daily number of actions in the heating season baseline model is significantly greater than the baseline value of the daily number of actions in the non-heating season baseline model.

[0051] Reference Figure 3 As shown, the degradation trend separation submodule uses an adaptive trend extraction algorithm based on a sliding window to perform degradation trend analysis in the following steps: Step S11: Obtain long-term operating time series. Obtain the daily cumulative number of actuations of the water supply valve and the daily cumulative operating time series of the water supply pump within the most recent first preset duration. The first preset duration is a continuous twelve months. Taking the daily cumulative number of actuations of the water supply valve as an example, the time series is expressed as follows: in, The total number of days within the first preset duration. For the first The cumulative number of times the water supply valve operates per day.

[0052] Step S12: Seasonal trend decomposition of the time series. The time series is decomposed using a locally weighted regression smoothing algorithm, which decomposes the original time series into trend components, seasonal cycle components, and residual components. in, As a trend component, it reflects the long-term monotonic change trend of equipment performance; It is a seasonal cycle component, reflecting seasonal fluctuations with an annual cycle; The residual component reflects random fluctuations.

[0053] Step S13: Extract the long-term monotonic degradation trend line from the trend components. Extract long-term monotonic degradation trend lines spanning multiple seasonal cycles. A linear regression model was used to fit the trend components: Where a is the slope of the degradation trend line, b is the intercept, and t is the time index.

[0054] Step S14: Calculate the slope of the degradation trend. Calculate the slope of the degradation trend line within the most recent second preset time period. The second preset time period is the most recent thirty days. Let the time index interval corresponding to the most recent thirty days be... The slope of the degradation trend within this interval is: When the slope When the preset degradation rate threshold is exceeded, the device is determined to have a performance degradation trend.

[0055] Step S15: Subtract seasonal factors and analyze residual degradation trends. The degradation trend separation submodule first determines whether the system is in the heating season or non-heating season based on the current date, and retrieves the corresponding statistical value from the seasonal pattern recognition submodule as a comparison benchmark. Let the seasonal benchmark value corresponding to the current date be... The residual value after deducting seasonal factors is: Then examine the residual value sequence Repeat steps S11 to S14 to obtain the actual degradation residual trend of the equipment after deducting the seasonal normal offset.

[0056] This module establishes seasonal statistical values ​​and uses the steps described above to separate seasonal periodic fluctuation components from the time series of operating parameters. After deducting normal parameter offsets caused by seasonal factors, it extracts the true degradation residual trend of the equipment, thus obtaining the true degradation residual trend of equipment performance. The method for determining whether the current parameter offset belongs to seasonal normal parameter offset or slow equipment degradation is as follows: when the residual trend is within the normal fluctuation range, it is determined to be seasonal normal parameter offset; when the residual trend shows a monotonic change exceeding the preset degradation rate threshold, it is determined to be slow equipment degradation. Thus, on the one hand, it accurately distinguishes between normal parameter offsets caused by seasonal changes in system operating conditions and abnormal trends caused by long-term slow degradation of equipment performance. On the other hand, when the number of daily operations of the water supply valve slowly increases during the heating season due to gradual wear of the sealing surface, even if the absolute value of this increase is smaller than the statistical value difference between the heating season and the non-heating season, this module can accurately identify the degradation trend. It is neither masked by the high statistical value during the heating season nor caused by false alarms due to large changes in statistical values ​​during seasonal transitions.

[0057] The alarm diagnosis module is connected to the time-domain feature analysis module and the multi-time-scale modeling analysis module respectively. It is connected to the upper-level monitoring system through relay output or communication bus. It is used to receive valve mechanical condition assessment results and equipment performance degradation trend analysis results, execute hierarchical alarm logic, and output hierarchical alarm signals containing fault type, fault location and fault severity.

[0058] The alarm diagnosis module has a built-in fault causal transmission rule base, which stores known fault causal transmission paths between the water supply valve and the water supply pump. Rule 1: When the impeller of the water supply pump wears, its head-flow characteristic curve shifts downward, causing a drop in pump outlet pressure, which in turn causes a drop in the inlet pressure of the water supply valve. The water supply valve opens more frequently to supply water under the command of the control system. In this causal transmission path, the performance degradation of the water supply pump is the cause, and the abnormal operating frequency of the water supply valve is the effect. Rule 2: When the sealing surface of the water supply valve leaks internally, the control system will command the water supply pump to increase its running time or frequency in order to maintain system pressure. In this causal transmission path, the sealing failure of the water supply valve is the cause, and the abnormal operating parameters of the water supply pump are the effect.

[0059] After receiving the valve mechanical condition assessment results from the time-domain feature analysis module and the degradation trend analysis results from the multi-time-scale modeling analysis module, the alarm diagnosis module performs fault tracing according to the following steps: Step S21: Receive early warning signals, specifically the valve mechanical fault early warning signal output by the time-domain feature analysis module and the equipment performance degradation trend early warning signal output by the multi-time-scale modeling and analysis module.

[0060] Step S22: Determine the number of warning signals and whether the two modules issue warning signals simultaneously. If only one module issues a warning signal, proceed to step S23; if both modules issue warning signals simultaneously, proceed to step S24.

[0061] Step S23: Direct location of single early warning. If only the time-domain feature analysis module issues a valve mechanical fault early warning signal, the fault source is directly located as the water supply valve; if only the multi-time-scale modeling and analysis module issues an equipment performance degradation trend early warning signal, the fault source is directly located as the water supply pump, and then step S26 is executed.

[0062] Step S24: Call the fault causal propagation rule library, call rule one and rule two in the built-in fault causal propagation rule library, and obtain the start time of the abnormal trend of the water supply valve action frequency and the start time of the abnormal trend of the water supply pump operating parameters.

[0063] Step S25: Compare the start times and determine the root cause of the fault. Compare the order of the start times of the degradation trends of the two devices. If the start time of the abnormal trend of the water supply pump operating parameters is earlier than the start time of the abnormal trend of the water supply valve actuation frequency, then it is determined that it meets Rule 1, and the root cause of the fault is the water supply pump, and the water supply valve responds passively due to hydraulic coupling. If the start time of the abnormal trend of the water supply valve actuation frequency is earlier than the start time of the abnormal trend of the water supply pump operating parameters, then it is determined that it meets Rule 2, and the root cause of the fault is the water supply valve, and the water supply pump responds passively due to control compensation.

[0064] Step S26: Output a graded alarm signal. Based on the fault tracing results, output a graded alarm signal that includes the fault type, fault location, and fault severity.

[0065] This module, through its built-in fault causal transmission rule library, distinguishes between the source equipment and the passive response equipment when the degradation trends of the water supply valve and the water supply pump occur simultaneously. It determines the order of degradation trends and the hydraulic coupling transmission path, and outputs a graded alarm signal containing the location of the fault root cause. This fundamentally solves the problem of systematic attribution error caused by the lack of fault causal transmission path analysis capability in existing technologies.

[0066] Reference Figure 2 As shown, based on the above system, the status anomaly alarm method of the present invention includes the following steps: Step S1: The valve action monitoring module measures the displacement stroke of the water supply valve stem in real time through the displacement sensor, and synchronously records the time of the opening and closing commands issued by the control system to the water supply valve through the command acquisition interface; the pump operation monitoring module measures the outlet pressure of the water supply pump, the operating current and the system water supply flow rate in real time through the pressure sensor, the current transformer and the flow sensor respectively; the data acquisition and synchronization module performs synchronous acquisition and timing alignment of the above signals with a unified time reference.

[0067] Step S2: The time-domain feature analysis module extracts time-domain feature parameters such as the opening response time, closing response time, full stroke displacement time, and stroke curve shape of the water supply valve from the aligned valve stem displacement time history data. The extracted time-domain feature parameters are compared with the benchmark opening response time, benchmark closing response time, and benchmark stroke curve shape features calibrated at the factory of the water supply valve. After nonlinear time axis alignment of the stroke curve shape with the benchmark stroke curve shape features by a dynamic time warping comparison algorithm, the shape correlation coefficient is calculated. When the opening response time exceeds the first preset proportion of the benchmark value, or the closing response time exceeds the second preset proportion of the benchmark value, or the shape correlation coefficient is lower than the preset threshold, it is determined that there is a mechanical abnormality in the water supply valve, and a valve mechanical fault warning signal is generated. When the opening response time, closing response time, and stroke curve shape are all within the allowable deviation range, it is determined that the water supply valve responds normally to the control command. The above determination is achieved before the detection of abnormalities in indirect parameters such as pipeline pressure and flow, realizing the early warning of valve mechanical faults.

[0068] Step S3: The seasonal pattern recognition submodule in the multi-timescale modeling and analysis module determines the current seasonal and day / night operating conditions of the system based on the current date and time, and retrieves the corresponding action frequency statistics, action response time statistics, outlet pressure statistics, and operating current statistics as comparison benchmarks; the degradation trend separation submodule compares the current operating parameters with the retrieved comparison benchmarks, separates the seasonal periodic fluctuation components from the time series of operating parameters, deducts the normal parameter offsets caused by seasonal factors, extracts the residual trend reflecting the true degradation of equipment performance, and calculates the slope of the degradation trend. When the residual trend shows a single value exceeding the preset degradation rate threshold, When the adjustment changes, it is judged as slow equipment degradation. When the residual trend is within the normal fluctuation range, it is judged as seasonal normal parameter deviation. In this way, the normal parameter deviation caused by seasonal changes in system operating conditions and the abnormal trend caused by long-term slow degradation of equipment performance are accurately distinguished. The degradation trend separation submodule adopts an adaptive trend extraction algorithm based on sliding window. When the number of daily actions of the water supply valve increases slowly during the heating season due to the gradual wear of the sealing surface, even if the absolute value of the increase is smaller than the statistical value difference between the heating season and the non-heating season, the degradation trend can be accurately identified. It is not masked by the high statistical value during the heating season, nor does it produce false alarms due to the large change in statistical value during seasonal transition.

[0069] Step S4: The alarm diagnosis module integrates the valve mechanical condition assessment results and the equipment performance degradation trend analysis results, and executes hierarchical alarm logic: when only a valve mechanical fault warning signal is detected, an alarm for the mechanical fault of the water supply valve is output; when only the performance degradation trend of the water supply pump exceeds the preset degradation rate threshold, an alarm for the performance degradation of the water supply pump is output; when the time-domain characteristic degradation trend of the water supply valve and the performance degradation trend of the water supply pump occur simultaneously, by judging the order of degradation trends and the hydraulic coupling transmission path, the degradation source equipment and the passive response equipment are distinguished, and a hierarchical alarm signal containing the fault root cause location is output. The alarm diagnosis module calls the rules in the built-in fault causal transmission rule library to compare the order of the start time of the degradation trends of the two devices, fundamentally solving the systematic attribution error problem caused by the existing technology of diagnosing the water supply valve and the water supply pump as independent equipment separately.

[0070] Example 1 This embodiment is an example of an early warning application for wear of the sealing surface of the water supply valve during the heating season. The secondary network system of a heat exchange station in a centralized heating community is equipped with one water supply pump and one electric water supply valve. The reference opening response time and reference closing response time of the water supply valve are known values ​​as specified by the manufacturer. The installation and debugging of each module of this invention are completed before the start of the heating season. The displacement sensor is an LVDT linear displacement sensor, which is installed at the end of the valve stem of the water supply valve. The data acquisition and synchronization module adopts a multi-channel synchronous data acquisition card, and each channel shares the same sampling clock. The time domain feature analysis module and the multi-time scale modeling and analysis module are run by an embedded industrial control computer. The alarm diagnosis module is connected to the upper-level monitoring system of the heat exchange station through a communication bus.

[0071] The system has been in continuous operation since the start of the heating season. In the initial stage of operation, the actual response time of the water supply valve monitored by the time-domain feature analysis module was within the allowable deviation range of the baseline value multiplied by the corresponding preset ratio. The correlation coefficient of the stroke curve shape remained at a normal level. The system did not issue a valve mechanical failure warning signal. The seasonal pattern recognition submodule in the multi-time-scale modeling and analysis module determined that the system was in the heating season operating condition and selected the heating season baseline model. The degradation trend separation submodule analyzed various operating parameters. The residual fluctuation amplitude after deducting the heating season baseline value was normal, and the degradation trend slope did not exceed the threshold.

[0072] As the heating season progressed into its later stages, the time-domain feature analysis module detected a continuous increase in the water supply valve's closing response time, exceeding the threshold determined by multiplying the baseline closing response time by the corresponding preset ratio. Simultaneously, the correlation coefficient of the stroke curve shape decreased significantly. Based on this, a valve mechanical fault warning signal was generated, indicating that the water supply valve exhibited abnormal valve core return or a trend of sealing surface wear. At this time, the system water supply pressure and flow rate reflected by the pipeline pressure sensor and flow sensor had not yet shown significant abnormalities, and the system water supply remained within the normal range. If only traditional indirect parameter monitoring methods were relied upon, this fault symptom would be completely ignored. Meanwhile, after deducting the seasonal baseline value for the heating season, the multi-timescale modeling and analysis module found that the slope of the residual trend of the daily actuation frequency of the water supply valve exceeded the threshold, while the residual trend of the water supply pump operating parameters showed no abnormalities. The alarm diagnosis module, combining the analysis results of the two modules, determined that only the time-domain feature analysis module and the water supply valve degradation trend issued warnings simultaneously, directly locating the fault source as the water supply valve and outputting a graded alarm signal with the fault type being water supply valve sealing surface wear and the fault location being the water supply valve core sealing surface.

[0073] Guided by the alarm signal, the maintenance personnel disassembled and inspected the water supply valve on the scheduled downtime. They found local pitting and slight wear on the valve core sealing surface, which was consistent with the judgment of the alarm diagnosis module. The fault was detected and dealt with in an early stage, which on the one hand prevented the large internal leakage caused by further wear of the sealing surface, and on the other hand avoided the additional energy loss caused by the abnormally long running time of the water supply pump due to internal leakage. The maintenance time was much shorter than the maintenance time required to replace the entire valve after the fault worsened.

[0074] Example 2 This embodiment is an example of a false alarm prevention application when switching between heating and non-heating seasons. Another heat exchange station serves a mixed user area including office buildings and residences. There are two typical operating modes throughout the year: heating season and non-heating season. During the heating season, the water supply valve operates at high frequency. During the non-heating season, the system only maintains the minimum constant pressure water supply requirement, the frequency of water supply valve operation is significantly reduced, and the water supply pump operates intermittently in low-frequency mode.

[0075] After the heating season ends and the system transitions to the non-heating season, the seasonal pattern recognition submodule in the multi-timescale modeling and analysis module automatically switches to the baseline model of normal operating parameters during the non-heating season based on the date. The daily actuation frequency of the water supply valve and the benchmark value of the water supply pump outlet pressure are simultaneously and significantly reduced. The degradation trend separation submodule pre-deducts the benchmark value of the non-heating season during comparison, so that this significant change is identified as a normal seasonal parameter shift, and the system does not generate any false alarms. In contrast, traditional fixed threshold alarm systems often trigger abnormal alarms in the water supply system during seasonal transitions because the water supply volume is significantly lower than the lower limit of the heating season threshold. Maintenance personnel need to manually disable the alarm signal before normal operation can be restored.

[0076] Throughout the entire non-heating season, neither the time-domain feature analysis module nor the multi-time-scale modeling analysis module detected any anomalies, and the system continued to operate stably, verifying the ability of this invention to prevent false alarms under seasonal switching conditions.

[0077] Example 3 This embodiment is an application example of identifying and tracing the source of failure in the progressive degradation of water supply pumps during the non-heating season. In the heat exchange station described in the second application embodiment, as the system enters the later stage of the non-heating season, the degradation trend separation submodule detects a slow and monotonically increasing residual trend of the water supply pump operating current after deducting the seasonal baseline value. The slope of the degradation trend exceeds the preset degradation rate threshold. At this time, the water supply pump outlet pressure and the system makeup water flow rate are both within the normal operating range during the non-heating season in absolute value. Traditional monitoring methods cannot detect this anomaly at all. The time domain feature analysis module does not detect any time domain feature anomalies in the makeup water valve during the same period, and the valve mechanical condition assessment result is normal. The alarm diagnosis module integrates the above results and determines that there is only a warning of the performance degradation trend of the water supply pump. It directly locates the fault source as the water supply pump and outputs a graded alarm signal with the fault type being progressive wear of the water supply pump bearing or slight scaling of the impeller.

[0078] Based on the alarm information, maintenance personnel conducted a preventative inspection of the water supply pump and found early signs of grease deterioration in the pump bearing. After cleaning and replacing the grease, the operating current of the water supply pump returned to the normal baseline level. This preventative maintenance was completed during the low-load period of the non-heating season, which prevented the fault from further developing into bearing seizure and causing a heating interruption accident at the beginning of the heating season. It also avoided the social impact and high maintenance costs caused by emergency repairs during the heating season.

[0079] As can be seen from the above three application examples, the system and method of the present invention can issue accurate early warnings at the very early stage of failure of the water supply valve and water supply pump, and effectively avoid batch false alarms caused by seasonal transitions when switching between heating and non-heating seasons. At the same time, the system can accurately distinguish and locate the root cause of the failure through the fault causal transmission rule base, and the accuracy of the status monitoring of the water supply valve and water supply pump of the hot water system and the operation and maintenance efficiency are significantly improved.

[0080] Working principle: During system operation, the valve action monitoring module acquires the time history data of the valve stem displacement of the water supply valve through the displacement sensor and synchronously records the timing of control commands through the command acquisition interface. The pump operation monitoring module acquires the operating parameters of the water supply pump. The data acquisition and synchronization module performs synchronous acquisition and time sequence alignment of the above signals. The time domain feature analysis module extracts time domain feature parameters such as opening response time and closing response time from the valve stem displacement time history data, compares them with the benchmark time domain feature parameter library, and generates a valve mechanical fault early warning signal before detectable anomalies appear in pipeline indirect parameters. At the same time, the seasonal pattern recognition submodule in the multi-time scale modeling and analysis module determines the seasonal operating conditions and selects the corresponding baseline model. The degradation trend separation submodule separates the seasonal periodic fluctuation components, deducts the seasonal normal offset, and extracts the true degradation residual trend of the equipment, accurately distinguishing between seasonal normal offset and slow equipment degradation.

[0081] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A status malfunction alarm system for a hot water system's water supply valve and water pump, characterized in that, include: The valve action monitoring module includes a displacement sensing unit, a command acquisition unit, a zero-position calibration unit, and a signal conditioning unit, which are used to measure the valve stem displacement stroke in real time and record the command time synchronously. The time-domain feature analysis module is connected to the valve action monitoring module. It has a built-in benchmark parameter library and morphological comparison algorithm. It is used to extract response time parameters and curve morphological parameters from the valve stem displacement stroke, compare the deviation with the benchmark parameters, and output the valve mechanical condition assessment results before the pipeline pressure and flow are abnormal. The pump operation monitoring module includes a pressure sensing unit, a current sensing unit, and a flow sensing unit, which are used to acquire the operating status parameters of the water supply pump. A multi-timescale modeling and analysis module is connected to the pump operation monitoring module. It has a built-in seasonal pattern recognition submodule and a degradation trend separation submodule. The seasonal pattern recognition submodule stores the parameter statistics under different seasonal operating conditions. The degradation trend separation submodule is used to retrieve the corresponding statistical values ​​as a comparison benchmark according to the current seasonal operating conditions, calculate the difference between the current operating parameters and the comparison benchmark to deduct seasonal factors, and obtain the residual trend that reflects the true degradation of equipment performance. The data acquisition and synchronization module uses a synchronous acquisition unit to synchronously acquire the above data and generate a multi-dimensional synchronous dataset, which is then provided to the time-domain feature analysis module and the multi-time-scale modeling and analysis module, respectively. The alarm diagnosis module receives the valve mechanical condition assessment results and the residual trend, respectively, and has a built-in fault causal transmission rule library. When only abnormal valve mechanical condition assessment results are received, it outputs a water supply valve mechanical fault alarm; when only abnormal residual trend is received, it outputs a water supply pump performance degradation alarm; when both abnormalities are received simultaneously, it compares the start time order of the degradation trends of the water supply valve and the water supply pump, and based on the fault transmission path based on hydraulic coupling relationship in the rule library, it identifies the device with the earlier start time as the degradation source device and outputs a graded alarm signal containing fault root cause location.

2. The abnormal status alarm system for the water supply valve and water supply pump of the hot water system according to claim 1, characterized in that: The displacement sensing unit is an LVDT linear displacement sensor installed on the valve stem of the water supply valve; the command acquisition unit is a command acquisition interface installed on the control loop of the water supply valve actuator, connected in parallel with the control system signal line; the zero-position calibration unit is a reference position calibration block fixed at the end of the valve stem corresponding to the fully closed position, used to eliminate zero-point drift; the signal conditioning unit is a differential signal conditioning circuit with a built-in common-mode noise suppression unit, used to convert the displacement analog signal into a differential signal and transmit it to the data acquisition and synchronization module.

3. The abnormal status alarm system for the water supply valve and water supply pump of the hot water system according to claim 1, characterized in that: The time-domain feature analysis module stores the baseline parameter library containing the baseline opening response time, baseline closing response time, and baseline stroke curve shape of the water supply valve as calibrated at the factory. The response time parameters include the opening response time and the closing response time. The opening response time is the time from the issuance of the opening command recorded by the command acquisition unit to the valve stem displacement reaching the preset proportion of the full stroke. The closing response time is the time from the issuance of the closing command to the valve stem displacement returning to zero. The shape comparison algorithm is a dynamic time warping comparison algorithm, which is used to nonlinearly align the real-time extracted curve shape with the baseline curve shape and calculate the shape correlation coefficient.

4. The abnormal status alarm system for the water supply valve and water supply pump of the hot water system according to claim 3, characterized in that: The time-domain feature analysis module determines that the valve mechanical condition assessment result is abnormal under the following conditions: the opening response time is greater than the product of the benchmark opening response time and the first preset ratio, or the closing response time is greater than the product of the benchmark closing response time and the second preset ratio, or the morphological correlation coefficient is lower than the preset threshold; wherein the typical value of the first preset ratio and the second preset ratio is 1.

3.

5. The abnormal status alarm system for the water supply valve and water supply pump of the hot water system according to claim 1, characterized in that: The pressure sensing unit is a pressure sensor installed on the outlet pipeline of the water supply pump; the current sensing unit is a current transformer installed on the power supply circuit of the water supply pump; the flow sensing unit is a flow sensor installed on the return water pipeline of the system; the operating status parameters of the water supply pump include outlet pressure, operating current and system makeup water flow.

6. The abnormal status alarm system for the water supply valve and water supply pump of the hot water system according to claim 1, characterized in that: The data acquisition and synchronization module's synchronization acquisition unit is a multi-channel synchronous data acquisition card, with each channel's analog-to-digital converter sharing the same sampling clock. The data acquisition and synchronization module also has a differential receiving circuit, used to restore the differential signal output by the signal conditioning unit to a single-ended displacement signal, and to perform anti-aliasing filtering and timing alignment processing on the acquired data to generate a multi-dimensional synchronous dataset with consistent timestamps.

7. The abnormal status alarm system for the water supply valve and water supply pump of the hot water system according to claim 1, characterized in that: The seasonal pattern recognition submodule stores the following parameter statistics: the frequency of operation and response time of the water supply valve under heating and non-heating season conditions, and the outlet pressure and operating current of the water supply pump under heating and non-heating season conditions; wherein the daily number of operation of the water supply valve is greater in the heating season than in the non-heating season; the seasonal pattern recognition submodule is also used to incorporate the newly generated operating data of the current season into the calculation at the end of each season and to perform sliding updates on the stored statistical values.

8. The alarm system for abnormal status of the water supply valve and water supply pump in a hot water system according to claim 1, characterized in that: The degradation trend separation submodule is used to obtain the daily cumulative number of actuations of the water supply valve and the daily cumulative operating time of the water supply pump in the past twelve months. The time series is decomposed into trend components, seasonal cycle components and residual components using a local weighted regression smoothing algorithm. The long-term monotonic degradation trend line is extracted from the trend component, and the slope of the trend line in the past thirty days is calculated. When the slope exceeds the preset degradation rate threshold, it is determined that there is a performance degradation trend. The method for deducting seasonal factors is as follows: the daily cumulative number of actuations of the water supply valve or the daily cumulative operating time, daily average operating current or daily average outlet pressure of the water supply pump are respectively subtracted from the statistical values ​​of the actuation frequency, operating time, operating current or outlet pressure under the corresponding seasonal operating conditions to obtain the daily residual value sequence after deducting seasonal factors. Then, the residual trend reflecting the true degradation of equipment performance is extracted from the sequence.

9. The alarm system for abnormal status of the water supply valve and water supply pump in a hot water system according to claim 1, characterized in that: The fault causal transmission rule base built into the alarm diagnosis module includes: Rule 1, when the impeller wear of the water supply pump causes a decrease in head, the frequency of operation of the water supply valve increases abnormally; Rule 2, when internal leakage occurs on the sealing surface of the water supply valve, the running time or running frequency of the water supply pump increases abnormally; when the alarm diagnosis module compares the order of start times, if the start time of the water supply pump degradation trend is earlier than that of the water supply valve, then the source device is determined to be the water supply pump according to Rule 1, otherwise the source device is determined to be the water supply valve according to Rule 2; the graded alarm signal includes the fault type, fault location and fault severity, and is sent to the upper monitoring system through relay output or communication bus.

10. A method for alarming abnormal status of a water supply valve and a water supply pump in a hot water system, applied to the alarm system for abnormal status of a water supply valve and a water supply pump in a hot water system as described in any one of claims 1 to 9, characterized in that, Includes the following steps: S1: The valve stem displacement stroke and command time are obtained through the valve action monitoring module, the water supply pump operation status parameters are obtained through the pump operation monitoring module, and a multi-dimensional synchronous dataset is generated through the data acquisition and synchronization module. S2: Through the time-domain feature analysis module, the response time parameters and curve shape parameters are extracted from the multidimensional synchronous dataset, and the deviation is compared with the benchmark parameters to output the valve mechanical condition evaluation result. S3: Using the multi-timescale modeling and analysis module, the corresponding statistical values ​​are retrieved based on the current seasonal operating conditions as a comparison benchmark. The difference between the current operating parameters and the benchmark is calculated to deduct seasonal factors and obtain the residual trend. S4: When the alarm diagnosis module receives both the abnormal valve mechanical condition assessment result and the abnormal residual trend, it compares the start time of the degradation trend of the water supply valve and the water supply pump. Based on the built-in fault causal transmission rule library, it identifies the device with the earlier start time as the degradation source device and outputs a graded alarm signal containing the fault root cause location.

Images