Fan-coil abnormal state diagnosis method and system

Abstract

The application discloses a fan coil abnormal state diagnosis method and system, obtains a fan current waveform and a coil vibration signal, establishes a heat exchange efficiency benchmark based on motor load change characteristics and heat exchange pipe vibration modes, identifies an efficiency sudden drop area and positions a heat exchange dead zone expansion range, generates a heat transfer coefficient jump data to determine a heat exchange abnormal position table, obtains a supply and return water temperature difference and a water valve opening degree change, forms a water route imbalance feature through coupling analysis of water flow pulse characteristics and flow state conversion points, identifies an air flow field offset amount based on the heat exchange abnormal position table and the water route imbalance feature, determines a diagnosis focus area and generates a layered detection sequence, classifies active and passive damage, extracts damage development rates, generates damage superposition coefficients, identifies a fault coupling amplification area, generates damage synergistic effect data through chain propagation analysis, completes comprehensive diagnosis of the fan coil abnormal state, and realizes accurate identification of multiple abnormalities and accurate positioning of fault causes.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring and diagnosis of heating, ventilation and air-conditioning equipment, and in particular to a method and system for diagnosing abnormal conditions of fan coil units. Background Art

[0002] As a crucial component of air conditioning systems, fan coil units (FCUs) are directly impacted by their operational stability, impacting both indoor environmental quality and overall system performance. In practice, these units face the combined effects of multiple degradation factors during long-term operation.

[0003] Traditional equipment condition monitoring technologies primarily rely on threshold judgments based on single parameters such as temperature, pressure, and flow, lacking a deep understanding of equipment operating mechanisms. Existing diagnostic methods often fail to accurately identify fault type and severity when faced with multiple fault interactions, and even more so, struggle to establish correlations between faults. Furthermore, current technologies often rely on a post-event maintenance model, lacking the ability to predict equipment condition trends and enabling proactive maintenance management. This leads to high maintenance costs and prolonged downtime. Summary of the Invention

[0004] The present invention provides a method and system for diagnosing abnormal conditions of fan coil units. The method establishes an equipment operation benchmark through multi-source information fusion and deep correlation analysis, identifies the distribution of abnormal areas and analyzes the system imbalance pattern, determines key monitoring areas and classifies damage types, and constructs a fault propagation model to achieve accurate diagnosis and comprehensive evaluation of abnormal conditions of fan coil units.

[0005] A first aspect of the present invention provides a method for diagnosing abnormal conditions of a fan coil unit, comprising the following steps:

[0006] Obtaining a fan current waveform and a coil vibration signal, extracting a motor load variation characteristic from the current waveform, extracting a heat exchange tube vibration pattern based on the coil vibration signal, and establishing a heat exchange efficiency benchmark based on the motor load variation characteristic and the heat exchange tube vibration pattern;

[0007] identifying an efficiency sudden drop region based on the heat exchange efficiency benchmark, locating a heat exchange dead zone expansion range according to the efficiency sudden drop region, generating heat transfer coefficient jump data based on the heat exchange dead zone expansion range, and determining a heat exchange abnormality location table using the heat transfer coefficient jump data;

[0008] Obtaining the supply and return water temperature difference and the change in water valve opening, identifying the water flow pulse characteristics based on the supply and return water temperature difference, capturing the flow pattern transition point based on the change in water valve opening, and coupling the water flow pulse characteristics with the flow pattern transition point to form a waterway imbalance characteristic;

[0009] identify a flow field deviation based on the heat exchange abnormal position table and the water path imbalance feature, determine a diagnosis focus area based on the flow field deviation, and generate a layered detection sequence using the diagnosis focus area;

[0010] classify active and passive damages based on the layered detection sequence to generate active damage data and passive damage data, analyze motion trajectories of the active damage data to extract a damage development rate, and generate a damage superposition coefficient based on the damage development rate and the passive damage data;

[0011] identify a fault coupling amplification area based on the damage superposition coefficient and the heat exchange abnormal position table, generate damage synergy effect data by performing chain analysis on the fault coupling amplification area, and complete fan coil abnormal state diagnosis.

[0012] The second aspect of the application provides a fan coil abnormal state diagnosis system, comprising:

[0013] A signal acquisition module is configured to acquire fan current waveforms and coil vibration signals, extract motor load change features from the current waveforms, extract heat exchange tube vibration modes based on the coil vibration signals, and establish a heat exchange efficiency benchmark according to the motor load change features and the heat exchange tube vibration modes.

[0014] An abnormal positioning module is configured to identify an efficiency drop area based on the heat exchange efficiency benchmark, locate a heat exchange dead zone expansion range according to the efficiency drop area, generate heat transfer coefficient jump data based on the heat exchange dead zone expansion range, and determine a heat exchange abnormal position table using the heat transfer coefficient jump data.

[0015] A water path analysis module is configured to acquire supply and return water temperature differences and water valve opening degree changes, identify water flow pulse features based on the supply and return water temperature differences, capture flow state conversion points according to the water valve opening degree changes, and perform coupling analysis on the water flow pulse features and the flow state conversion points to form a water path imbalance feature.

[0016] A coupling diagnosis module is configured to identify a flow field deviation based on the heat exchange abnormal position table and the water path imbalance feature, determine a diagnosis focus area based on the flow field deviation, and generate a layered detection sequence using the diagnosis focus area.

[0017] A damage analysis module is configured to classify active and passive damages based on the layered detection sequence to generate active damage data and passive damage data, analyze motion trajectories of the active damage data to extract a damage development rate, and generate a damage superposition coefficient based on the damage development rate and the passive damage data.

[0018] A cooperative diagnosis module is configured to identify a fault coupling amplification area based on the damage superposition coefficient and the heat exchange abnormal position table, perform chain analysis on the fault coupling amplification area to generate damage cooperative effect data, and complete the fan coil abnormal state diagnosis.

[0019] The beneficial effects of the present application are embodied in the following points: first, by collecting fan current waveforms and coil vibration signals to establish a heat exchange efficiency benchmark, combined with efficiency drop area identification and heat exchange dead zone expansion range positioning, local fouling of the heat exchanger, fan blade contamination, motor performance degradation and other problems can be accurately found. Compared with the traditional method of relying only on temperature difference monitoring, this multi-signal fusion method can detect the performance decline trend of the equipment in advance and accurately locate the specific faulty components and affected areas, avoiding the continuous deterioration of equipment efficiency and the rise in energy consumption caused by missed detection. Second, by obtaining the water supply and return temperature difference and the water valve opening change to form the water imbalance characteristics, combined with the airflow field offset amount identification to determine the diagnosis focus area, it can find the uneven flow distribution, pipeline blockage, airflow short circuit and other composite problems in the water system. Through the phase analysis of the water flow pulse characteristics and the flow state transition point, the specific location and severity of the hydraulic imbalance can be identified, and the detection range can be narrowed by locating the vortex center of the airflow field, avoiding the workload of comprehensive disassembly and inspection, so that maintenance personnel can quickly lock the problem area. Finally, through the active and passive damage classification and damage development rate analysis, combined with the chain propagation analysis of the fault coupling amplification area, the active faults inside the equipment and the passive damages caused by external environmental influences can be distinguished, and the fault propagation path and development trend can be predicted. By identifying the air volume attenuation gradient, condensate retention degree, fin fouling rate and other propagation characteristics, the possibility and time node of the spread of a single fault to other components of the system can be judged in advance, effectively preventing small faults from evolving into systemic faults and reducing the risk of equipment sudden shutdown.

[0020] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0021] The drawings herein show specific examples of the technical solutions described in the present application, and constitute part of the specification together with the specific embodiments, for explaining the technical solutions, principles and effects of the present application.

[0022] Unless specifically stated or defined otherwise, the same reference signs in different drawings represent the same or similar technical features, and different reference signs may also be used to represent the same or similar technical features.

[0023] Figure 1 is a flowchart of a fan coil abnormal state diagnosis method of the present application.

[0024] Figure 2is a structural block diagram of a fan coil abnormal state diagnosis system. DETAILED DESCRIPTION

[0025] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, circuits, and

[0026] It is to be understood that the terminology "including", "comprising", "consisting" and "consisting essentially of" used in the specification and the appended claims, indicates the presence of the stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0027] Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment, but can refer to one or more but not all embodiments. Furthermore, the terms "a" or "an", as used herein, mean "one or more" unless otherwise explicitly stated.

[0028] The technical solutions of the embodiments of the present application are introduced as follows.

[0029] As shown in Figure 1 The embodiments of the present application provide a fan coil abnormal state diagnosis method, which comprises the following steps S110-S160:

[0030] In step S110, the fan current waveform and the coil vibration signal are acquired, the motor load change characteristic is extracted through the current waveform, the heat exchange pipe vibration mode is extracted based on the coil vibration signal, and the heat exchange efficiency benchmark is established according to the motor load change characteristic and the heat exchange pipe vibration mode.

[0031] Specifically, the fan current waveform and the coil vibration signal are acquired. A high-precision current sensor CSLA2CD is installed at the fan motor inlet to measure the three-phase current waveform data, with a measurement range of 0-50 A, an accuracy level of 0.5, and a sampling frequency of 5000 Hz. The data recording format is [timestamp, phase identification, instantaneous current value, effective value, power factor]. A temperature sensor PT100 is installed at the inlet and outlet pipelines to measure the water temperature, with a measurement range of 0-100℃, an accuracy of ±0.1℃, and a sampling frequency of 10 Hz. A turbine flowmeter is installed to measure the water flow, with a measurement range of 0.5-10 m³ / h, an accuracy level of 1.0, and a sampling frequency of 1 Hz. Temperature sensors are additionally arranged at the key heat transfer areas of the heat exchanger to form a multi-point temperature monitoring network. Acceleration sensors ADXL354 are installed at the inlet and outlet pipelines, the refrigerant pipeline, and the fin connection of the coil heat exchanger to acquire the coil vibration signal, with a sensitivity of 100 mV / g, a frequency response range of 0.5-1600 Hz, and a sampling frequency of 2000 Hz. The vibration signal acquisition points include the U-shaped bend, the straight pipe section, the support frame, and the fin root, and each position records the X, Y, and Z axis vibration acceleration. The sensor numbers are arranged in the order of V01-V08, and the data format is [timestamp, sensor number, X-axis acceleration, Y-axis acceleration, Z-axis acceleration, temperature compensation value].

[0032] The motor load change characteristics are extracted from the current waveform. The instantaneous power, average power, and power fluctuation amplitude of the motor are calculated using the three-phase current waveform data. The instantaneous power is obtained by summing the three-phase power, P_inst=U_A×I_A+U_B×I_B+U_C×I_C, where P_inst is the instantaneous power (W), U_A, U_B, and U_C are the three-phase voltages (V), and I_A, I_B, and I_C are the three-phase instantaneous currents (A). The average power is calculated using a sliding window with a window length of 1 second and a sliding step of 0.1 second. The load change rate is defined as the ratio of the power change to the time interval, reflecting the dynamic response characteristics of the motor load. The current harmonic analysis uses fast Fourier transform (FFT) to extract the 5th, 7th, and 11th characteristic harmonic components. The period with a harmonic content exceeding 5% is marked as an abnormal motor state. The load fluctuation coefficient is calculated by the ratio of the power standard deviation to the average value. A fluctuation coefficient less than 5% indicates a stable load, a fluctuation coefficient between 5% and 15% indicates a slight fluctuation, and a fluctuation coefficient greater than 15% indicates a severe fluctuation. The startup impact feature is identified by detecting current mutations. The startup current is usually 4-7 times the rated current, with a duration of 2-5 seconds.

[0033] The vibration mode of the heat exchange tube is extracted based on the vibration signal of the coil. The vibration frequency, amplitude and other mode characteristics of the heat exchange tube are analyzed by using the vibration signal of the coil. The vibration signal preprocessing includes low-pass filtering to remove high-frequency noise, the cutoff frequency is set to 800 Hz, and the filter order is 4th order Butterworth filter. The main frequency is identified by power spectrum analysis, and the main frequency of the heat exchange tube vibration is usually distributed in the range of 10-200 Hz, including fluid-induced vibration 10-50 Hz, structural natural frequency 80-150 Hz, and high-frequency vibration 150-200 Hz. The vibration mode analysis identifies the main bending mode and torsional mode, which is used to distinguish normal vibration and abnormal vibration mode. The vibration intensity is represented by the root mean square acceleration RMS, which reflects the overall energy level of the vibration. The temperature correlation analysis establishes the change rule of the vibration characteristics with the heat exchange temperature, and the vibration frequency usually shifts to high frequency by 2-5 Hz when the temperature rises. The vibration mode classification includes three types of steady-state vibration, intermittent vibration and impact vibration, and the classification basis is the time-varying characteristics of the vibration amplitude.

[0034] In some embodiments, the heat exchange efficiency benchmark is established based on the motor load change characteristics and the heat exchange tube vibration mode, including: generating a load fluctuation time series based on the motor load change characteristics; fusion the heat exchange tube vibration mode and the load fluctuation time series to form a vibration-load coupling curve; extract the stable working section from the vibration-load coupling curve; set the efficiency value of the stable working section as the heat exchange efficiency benchmark.

[0035] A load fluctuation time series is generated based on the motor load change characteristics. The average power, power fluctuation amplitude, load change rate and other parameters in the motor load change characteristics are used to construct a time series table. The horizontal axis of the time series is the time axis, and the time resolution is set to 0.1 second. The vertical axis is the normalized load power, and the normalization reference is the rated power of the motor. The load fluctuation curve is processed by smoothing filter, the filter window length is 5 seconds, and the short-term noise interference is eliminated. The upper and lower envelopes of the load fluctuation are extracted, and the envelope width reflects the load stability. Key event points are marked in the time series, including start time, stable running section, load mutation point, shutdown time, etc. The load gradient calculates the change slope of the load at adjacent time points, and the area with gradient greater than the threshold is marked as the transition section. The periodicity analysis identifies the periodic characteristics of the load fluctuation, and the cycle length is usually related to the refrigeration cycle period. The data of harmonic abnormal period is excluded from the time series to ensure the reliability of the benchmark establishment.

[0036] The vibration mode of the heat exchange tube and the load fluctuation time diagram form a vibration-load coupling curve. According to the time correspondence between vibration characteristics and load characteristics, the vibration intensity and load power are compared and analyzed on the same time axis. The coupling curve adopts a two-dimensional coordinate system, the horizontal axis is the normalized load power, and the vertical axis is the normalized vibration intensity. The data point density reflects the frequency of the occurrence of this working condition, and the area with high density indicates the common working condition. The coupling relationship analysis adopts correlation calculation, and the correlation coefficient r represents the correlation degree of vibration and load. Linear fitting identifies the main trend line of vibration-load, and the fitting equation is in the form of y=ax+b, where a is the coupling coefficient and b is the basic vibration level. The data points corresponding to abnormal vibration modes are marked as abnormal points and excluded from the coupling analysis. The coupling curve is divided into low load area, medium load area and high load area, and the vibration characteristics of each area are different.

[0037] The stable working section is extracted from the vibration-load coupling curve. The stable working section is defined as the running interval where the vibration and load remain relatively stable. The stability criterion includes three conditions: the load fluctuation coefficient is less than 8%, the vibration intensity changes less than 10%, and the coupling relationship deviation is less than 15%. The stable section recognition algorithm uses sliding window analysis, and the window length is set to 60 seconds. The data in the window that meets the stability criterion is marked as a candidate stable section. Continuity check ensures that the duration of the stable section is greater than 5 minutes, and short-term stability is not recognized as an effective stable section. The stability degree index quantifies the quality evaluation of the stable section, considering factors such as fluctuation amplitude, duration, consistency, etc. Multiple stable section processing identifies multiple stable sections within the running cycle, and selects the optimal stable section according to the stability degree. The stable section boundary is determined by gradient change detection, and the gradient mutation point is used as the start and end boundary of the stable section. Environmental factor correction is based on the temperature correlation analysis results to establish the influence correction of temperature on the stable section.

[0038] The efficiency value of the stable working section is set as the heat exchange efficiency benchmark. Based on the average load power and heat exchange data in the stable working section, the heat exchange efficiency η = Q / P, where η is the heat exchange efficiency (dimensionless), Q is the heat exchange (W), and P is the motor power (W). The heat exchange is calculated by temperature sensor and flow meter data, Q = m x c x ΔT, where m is the water flow (kg / s), c is the specific heat capacity of water (J / kg·K), and ΔT is the temperature difference between the inlet and outlet water (K). The benchmark efficiency is the weighted average of the efficiency in the stable working section, and the weight is distributed according to the working time. The confidence interval of the efficiency benchmark is determined by statistical analysis, and the 95% confidence interval reflects the reliability range of the benchmark. Environmental standardization is based on temperature correlation analysis, which modifies the benchmark efficiency to standard environmental conditions, with an environmental temperature of 25°C and a relative humidity of 60%. The correction formula is η_corrected = η x [1 + α x (T-25)], where α is the temperature correction coefficient. The benchmark validity is verified by repeated tests, and the benchmark efficiency deviation is less than 3% for consecutive days. The benchmark updating mechanism updates the efficiency benchmark periodically based on the accumulation of operation data, with an update period of 30 days. The multi-condition benchmark considers the efficiency difference under different load conditions, and establishes a segmented efficiency benchmark library. Finally, the heat exchange efficiency benchmark is established.

[0039] In step S120, the efficiency drop area is identified based on the heat exchange efficiency benchmark, the heat exchange dead zone expansion range is located according to the efficiency drop area, the heat transfer coefficient jump data is generated based on the heat exchange dead zone expansion range, and the heat transfer coefficient jump data is used to determine the heat exchange abnormal position table.

[0040] Specifically, the efficiency drop area is identified based on the heat exchange efficiency benchmark. The relative deviation method is used, with a deviation rate of (η_baseline-η_real) / η_baseline x 100%, where η_baseline is the benchmark efficiency (dimensionless) and η_real is the real-time efficiency (dimensionless). The area with a deviation rate greater than 15% is marked as an efficiency drop area, the area with a deviation rate of 5%-15% is an efficiency drop area, and the area with a deviation rate less than 5% is a normal efficiency area. Spatial positioning is based on the geometric coordinate system of the coil heat exchanger, and the heat exchanger is divided into grids, each grid corresponding to an efficiency monitoring point. The efficiency monitoring uses the installed temperature sensor monitoring network and turbine flow meter. The drop area identification algorithm uses connected component analysis to merge adjacent drop grids into continuous areas. Combined with the temperature compensation data of the vibration sensor ADXL354, the temperature measurement deviation of each monitoring point is corrected. The area of the region varies greatly, and the shape is mostly elliptical or irregular polygon. Time persistence analysis requires that the drop phenomenon lasts for a sufficient time to be recognized as an effective drop area.

[0041] In some embodiments, the positioning of the dead zone expansion range according to the efficiency drop region comprises: dividing the efficiency drop region into a center region and an edge region; emitting a detection path from the center region to the edge region; recording the positions of efficiency recovery points on the detection path to form a recovery point set; and connecting the points farthest apart in the recovery point set to form the dead zone expansion range.

[0042] The efficiency drop region is divided into a center region and an edge region. Based on the geometric shape of the efficiency drop region and the deviation rate distribution, the center of gravity segmentation method is used to determine the region center. The center region is defined as the region around the position of the maximum deviation rate, and the deviation rate in this region is generally greater than 25%. The edge region is the remaining part of the drop region except the center region, and the deviation rate of the edge region is in the range of 15%-25%. The segmentation algorithm uses the gradient descent method to find the maximum point of the deviation rate as the center point, and the center point coordinates (x_center, y_center) correspond to the physical position of the heat exchanger. The radius of the center region is dynamically adjusted according to the total area of the drop region. The edge region boundary is determined by the equal deviation rate line, and the 15% deviation rate contour is used as the outer boundary of the edge region. The segmentation result includes center region coordinates, radius, edge region boundary point sequence and other geometric parameters. The quality of segmentation is evaluated by the uniformity of the efficiency distribution in the region, and the standard deviation of the efficiency in the center region is considered to be effective if it meets the set threshold.

[0043] The detection path is emitted from the center region to the edge region. A radial layout is used to emit straight paths from the center of the center region to each direction of the edge region. The path angle interval is determined according to the complexity of the region to form a uniform distribution of detection rays. The length of each detection path extends to a suitable distance outside the outer boundary of the edge region to ensure that the possible expansion range of the dead zone is covered. The sampling points on the path use the temperature sensor network arranged, and the sampling point interval is consistent with the sensor arrangement interval. The detection paths are numbered and named in order of angle. The path weight is distributed according to the efficiency gradient in that direction, and the path weight is high for large gradient and low for small gradient. Multi-path parallel detection can identify the main direction and secondary direction of the dead zone expansion. The detection process uses a fixed sensor network for static monitoring.

[0044] The positions of the efficiency recovery points on the detection path form a recovery point set. A recovery point is defined as the position along the detection path where the efficiency starts to rise above the baseline efficiency of 85%. A sliding window detection is used, with a window length of several sampling points. If the efficiency continuously rises within the window and the efficiency of the last point is greater than the threshold, it is marked as a recovery point. The threshold is set to 85% of the baseline efficiency, ensuring that the recovery point represents a substantial improvement in heat exchange performance. There may be multiple recovery points on each detection path, and the recovery point farthest from the center zone is selected as the representative recovery point for that path. The recovery point coordinates are expressed in the heat exchanger coordinate system, with a precision of the sensor arrangement accuracy. The recovery point detection uses a multi-point comparison method. When the efficiency values of consecutive sampling points show an increasing trend and the last point exceeds the threshold, it is confirmed as a valid recovery point. The efficiency recovery rate and recovery amplitude at the recovery point are also recorded. The recovery point set contains the representative recovery points of all paths, forming a recovery point ring around the dead zone.

[0045] The farthest points in the recovery point set are connected to form the expansion range of the heat exchange dead zone. The farthest point pair algorithm is used to find the pair of points in the recovery point set that are farthest from each other, as the main axis endpoints of the expansion range. Multi-directional expansion is based on the main axis, finding the farthest points in the vertical and diagonal directions to form the key nodes of the expansion boundary. Based on the key point constraints, the convex hull generation method is used, with all recovery points as input, to ensure that the farthest key points are included in the convex hull boundary. The convex hull calculation uses the Graham scan algorithm, with the farthest point pair as the starting edge, connecting the peripheral recovery points in order of polar angle to generate the smallest boundary polygon that encloses all recovery points. The convex hull boundary is the boundary of the expansion range of the heat exchange dead zone, and the area inside the boundary is the dead zone influence range. The boundary smoothing process uses spline curve fitting to eliminate the sawtooth caused by sampling errors. The expansion range area is calculated by the polygon area formula, reflecting the degree of influence of the dead zone. The boundary buffer zone increases the appropriate width on the periphery of the expansion range, considering the uncertainty of the boundary. The expansion direction analysis identifies the main expansion direction of the dead zone, which is usually more obvious along the fluid flow direction. The boundary node coordinates are stored in counterclockwise order to form a closed polygon contour line.

[0046] The heat transfer coefficient jump data is generated based on the expansion range of the heat exchange dead zone. The boundary information of the expansion range of the heat exchange dead zone is used to calculate the change rule of the heat transfer coefficient near the boundary. The heat transfer coefficient is calculated by the local convective heat transfer coefficient h=q / (ΔT), where h is the convective heat transfer coefficient (W / m²·K), q is the heat flux (W / m²), and ΔT is the temperature difference between the wall and the fluid (K). The heat flux q is calculated by the temperature gradient of adjacent temperature sensors and the known thermal conductivity of the material, and ΔT is directly measured by the temperature sensor. The jump detection uses the difference analysis of the heat transfer coefficient on both sides of the boundary, and the jump amplitude = |h_inside-h_outside| / h_outside×100%, where h_inside is the heat transfer coefficient inside the dead zone, and h_outside is the heat transfer coefficient in the normal region. The position with a jump amplitude greater than 40% is marked as a significant jump point, the position with a jump amplitude between 20% and 40% is marked as a moderate jump point, and the position with a jump amplitude less than 20% is marked as a slight jump point. The spatial resolution is set as the interval of the temperature sensor arrangement, and the jump detection points are uniformly distributed along the boundary of the expansion range. Time series analysis of the jump data tracks the evolution process of the heat transfer coefficient jump and identifies the development trend of the jump.

[0047] In some embodiments, the determination of the heat transfer abnormality position table based on the heat transfer coefficient jump data includes: heat flow blockage identification of the heat transfer coefficient jump data to generate a flow line termination zone; expansion influence evaluation based on the flow line termination zone to form an expansion coefficient; spatial interpolation of the expansion coefficient to generate a continuous expansion field; and boundary extraction based on the continuous expansion field to generate a heat transfer abnormality position table.

[0048] The heat flow blockage identification of the heat transfer coefficient jump data generates a flow line termination zone. The blockage determination criterion is a region with a jump amplitude greater than 50% and consistent jump direction of adjacent detection points. The flow line tracking uses a numerical flow field analysis method to track the propagation path of the heat flow line from the inlet of the heat exchanger. The flow line termination criterion is based on the gradient mutation of the heat transfer coefficient, and when the absolute value of the gradient is greater than a set threshold, it is considered that the flow line terminates at this position. The termination zone identification clusters and merges adjacent flow line termination points, and the clustering radius is set to an appropriate distance. The geometric characteristics of the flow line termination zone include center coordinates, influence radius, termination intensity, and other parameters. The termination intensity is defined as the average value of the heat transfer coefficient jump amplitude in the region, reflecting the severity of the blockage degree. Multi-scale analysis identifies flow line termination zones of different scales, and large-scale termination zones correspond to main heat exchange obstacles, and small-scale termination zones correspond to local heat exchange defects.

[0049] The expansion coefficient is formed based on the influence evaluation of the streamline termination zone expansion. The influence degree and range of each streamline termination zone on the surrounding heat exchange performance are analyzed. The influence degree is calculated by a distance decay function, and the influence intensity I(r) = I0 x exp(-r / r0), where I(r) is the influence intensity at a distance r (dimensionless), I0 is the maximum influence intensity of the termination zone (dimensionless), and r0 is the influence decay constant. The expansion coefficient is defined as the product of the termination zone influence intensity and distance, reflecting the contribution of the region to the dead zone expansion. The weight distribution is determined according to the termination intensity and influence radius of the termination zone, and the region with high termination intensity and large influence radius has high weight. The superposition effect of multiple termination zones is calculated by the linear superposition principle, and the total expansion coefficient is equal to the weighted sum of the expansion coefficients of each termination zone. Directional analysis considers the asymmetry of the influence of fluid flow direction on expansion, and the expansion coefficient in the downstream direction is usually greater than that in the upstream direction. Time evolution analysis tracks the trend of expansion coefficient with time, and identifies the acceleration or deceleration stage of expansion.

[0050] The expansion coefficient is generated by spatial interpolation to produce a continuous expansion field. The discrete expansion coefficient is extended to a continuous spatial field distribution by using the Kriging interpolation method. The interpolation grid resolution is appropriately encrypted based on the sensor arrangement grid, covering the entire heat exchanger area. The interpolation weight is determined according to the spatial distance and data correlation, and the data points with short distance and high correlation have high weight. The boundary condition is set to zero at the boundary of the heat exchanger, ensuring the physical reasonableness of the field distribution. The interpolation accuracy is evaluated by cross-validation, and the average relative error of leave-one-out validation is less than the set threshold, which is considered to be valid. The gradient calculation of the continuous expansion field identifies the region with the most intense change in expansion intensity, and the region with large gradient usually corresponds to the boundary of heat exchange anomaly. Field smoothing processing uses Gaussian filtering to eliminate high-frequency noise in the interpolation process. Multi-level analysis generates expansion fields with different resolutions, coarse resolution for overall trend analysis, and fine resolution for local feature identification.

[0051] The boundary extraction generates the heat exchange abnormal position table according to the continuously expanding field. The boundary is extracted based on the contour analysis of the continuously expanding field, and the contour with the threshold value of the expansion coefficient is selected as the boundary of the abnormal area. The threshold value is determined by an adaptive method, and a suitable proportion of the maximum value of the expansion field is taken as the boundary threshold value. The contour tracking algorithm adopts the moving cube method to generate a smooth closed boundary curve. The boundary simplification process removes short boundary branches and retains the main abnormal area boundary. The abnormal position classification divides the extracted boundary area into main abnormal, secondary abnormal, local abnormal and other categories according to the area size. The position coding adopts the combination of the area center coordinates and the abnormal type, and the format is "type code-X coordinate-Y coordinate". The abnormal position table includes position coding, boundary coordinates, area, abnormal degree, discovery time and other fields. The table is sorted according to the priority of the abnormal degree and the area size, and the position with high abnormal degree and large area has the highest priority. The dynamic updating mechanism updates the abnormal position table regularly according to the new heat transfer coefficient data, and the update period is 24 hours.

[0052] In step S130, the supply and return water temperature difference and the water valve opening degree change are obtained, the water flow pulse characteristics are identified through the supply and return water temperature difference, the flow state conversion point is captured according to the water valve opening degree change, and the water flow pulse characteristics and the flow state conversion point are coupled and analyzed to form the water route imbalance characteristics.

[0053] Specifically, the supply and return water temperature difference and the water valve opening degree change are obtained. The supply and return water temperature difference ΔT=T_supply-T_return is calculated by using the supply and return water temperature sensor PT100 with a measurement accuracy of ±0.1℃ and a sampling frequency of 10Hz, where T_supply is the supply water temperature (℃) and T_return is the return water temperature (℃). The temperature difference data record format is [timestamp, supply water temperature, return water temperature, temperature difference value]. The water valve opening degree monitoring adopts an angular displacement sensor with a measurement range of 0-90°, a resolution of 0.1°, and a linearity of ±0.5%. The opening degree change rate is calculated by the ratio of the opening degree difference value and the time interval, reflecting the speed of valve adjustment. The opening degree data includes valve number, opening angle, change rate, adjustment direction, action time and other information. Multi-point synchronous collection covers the main water supply valve, branch regulating valve, balance valve and other key valve positions. Data storage adopts a ring buffer to maintain the last 1 hour of historical data for analysis and processing.

[0054] The water flow pulse characteristics are identified by the temperature difference between the supply and return water. The time series data of the temperature difference between the supply and return water are used to analyze the pulsation law and periodic characteristics of the water flow. The pulse detection uses a peak identification algorithm, and the fluctuation with a temperature difference change amplitude greater than 0.5°C and a duration of 3-10 seconds is identified as an effective pulse. The pulse frequency is the number of pulses occurring per unit time, and the pulse frequency is 2-5 times per minute during normal operation and can reach more than 10 times per minute during abnormal operation. The pulse amplitude is defined as the maximum temperature difference change of a single pulse, and the amplitude size reflects the strength of the water flow disturbance. The pulse waveform analysis includes the time characteristics of the rising time, peak duration, and falling time of the three stages, and the waveform characteristics are used for accuracy verification of pulse identification. The periodicity analysis uses the autocorrelation function to identify the repetition period of the pulse, and the period length is usually related to the thermal inertia and flow regulation period of the system. The pulse synchronicity analysis of the temperature difference pulse of multiple measuring points reflects the time relationship, and the synchronous pulse indicates a systematic disturbance, and the asynchronous pulse indicates a local disturbance. The synchronicity information is used for the classification processing of the pulse sequence. The filtering process removes high-frequency noise and low-frequency drift, and retains the effective pulse information.

[0055] The flow state transition points are captured according to the changes in the water valve opening. Based on the opening change data of the valve position, the time points of the flow state change of the water system are identified. The flow state transition criteria include an opening change amplitude greater than 10°, a change rate greater than 5° / s, and multiple valve coordinated action. The transition point classification includes start-up transition (system from stop to operation), regulation transition (load regulation caused by flow change), switching transition (operation mode switching), and fault transition (abnormal state caused by passive regulation). The time accuracy reaches seconds, ensuring accurate positioning of the transition point. The transition strength evaluation is calculated by the weighted sum of the opening change amount, the main valve has a large weight, and the branch valve has a small weight. The transition direction identification includes flow increase transition (opening increase), flow decrease transition (opening decrease), and mixed transition (part of the increase and part of the decrease), and the transition direction information is used for the symbol design of the transition time marker. The persistence analysis requires the transition state to remain for more than 30 seconds, and the short opening fluctuation is not identified as an effective transition. The influence range analysis identifies the pipe branches and heat exchange equipment affected by the transition point, and the influence range information is used for the weight calculation of the transition strength.

[0056] In some embodiments, the coupling analysis of the water flow pulse characteristics and the flow state transition point forms a water system imbalance feature, including: constructing a pulse time axis based on the water flow pulse characteristics; mapping the flow state transition point to the pulse time axis to form a transition time marker; dividing the pulse time axis into a pre-transition segment and a post-transition segment with the transition time marker as a boundary; comparing the phase distribution characteristics of the pre-transition segment and the post-transition segment to form a water system imbalance feature.

[0057] The pulse time axis is constructed based on the characteristics of water flow pulses. The parameters such as pulse time, pulse amplitude, duration, periodicity, etc. in the characteristics of water flow pulses are used to construct a standardized time axis coordinate system. The time axis takes the system start time as the zero point, the time resolution is set to 1 second, and the time span covers the entire running period. The pulse is marked on the time axis with the pulse time as the horizontal coordinate and the pulse amplitude as the vertical coordinate. The pulse density is calculated according to the periodicity analysis results, and the window length is set according to the identified period. The time axis standardization process unifies the pulse data of different measuring points to the same time reference, eliminating the sampling time difference. The pulse sequence is sorted according to the pulse time, forming an ordered time axis mark. The pulse intensity normalization unifies the pulse amplitude to the range of 0-1, and the normalization reference takes the maximum pulse amplitude in the observation period. The phase information calculates the phase angle φ = 2π × (t_pulse mod T) / T of each pulse relative to the reference period, where φ is the phase angle (radian), t_pulse is the pulse time (second), and T is the reference period (second).

[0058] The flow regime transition point is mapped to the pulse time axis to form a transition time mark. Based on the transition time, transition classification, transition intensity, etc. of the flow regime transition point, a transition mark symbol is added on the pulse time axis. The mapping algorithm uses a time matching method to directly correspond the time coordinates of the transition point to the corresponding position of the pulse time axis. The mark symbol distinguishes the transition classification by using different colors and shapes, the start transition uses a green triangle, the adjustment transition uses a blue circle, the switching transition uses a yellow square, and the fault transition uses a red diamond. The mark size reflects the transition intensity, the larger the transition intensity, the larger the mark symbol, and the smaller the transition intensity, the smaller the mark symbol. The time alignment accuracy reaches 1 second, ensuring the accurate correspondence of the transition mark and the pulse event. When the transition time coincides with the pulse time, a composite mark is used to display the two kinds of information. The mark sequence forms a transition event chain according to the time sequence, which is used to analyze the time law of the transition.

[0059] The pulse time axis is divided into pre-transition and post-transition segments based on the transition time mark. The continuous pulse time axis is divided into multiple time segments with each transition time mark as the dividing point. The pre-transition segment is defined as the time interval from the previous transition point to the current transition point, and the post-transition segment is defined as the time interval from the current transition point to the next transition point. The segment length statistics show the average time length of the pre-transition and post-transition segments. The first and last processing is a special processing for the start and end points of the time axis, the pre-segment of the start point is empty, and the post-segment of the end point is extended to the end of the observation period. The segment coding uses the format of "S + transition point number + front-back identifier", such as S01B representing the pre-segment of the first transition point and S01A representing the post-segment of the first transition point. The segment-in-pulse statistics calculate the number of pulses contained in each time segment, the average amplitude, etc. The segmentation results generate a time segment list, each time segment contains start and end time, contained pulses, statistical parameters, etc.

[0060] The phase distribution characteristics of the pre-conversion section and the post-conversion section are compared to form a waterway imbalance feature. The pulse sequence of the pre-conversion section is converted into a phase sequence. The pulse sequence of the post-conversion section is misaligned and superimposed on the phase sequence to form a phase difference map. The phase jump accumulation is extracted from the phase difference map. The waterway imbalance feature is formed based on the unevenness of the distribution of the phase jump accumulation.

[0061] The pulse sequence of the pre-conversion section is converted into a phase sequence. The phase angle of each pulse is calculated using the pulse time and period information contained in the pre-conversion section. The phase sequence is arranged in chronological order of pulse occurrence, and the sequence length is equal to the number of pulses in the pre-conversion section. The phase calculation uses the formula φ = 2π × (t_pulse mod T) / T, where T is the pulse period. The phase continuity processing is unfolded for the case where the phase angle crosses the 2π boundary, ensuring the continuity of the phase sequence. The sequence smoothing uses a moving average filter with a window length of 3 pulses to eliminate the phase measurement error of individual pulses. The phase trend analysis identifies the overall trend of the phase sequence, and the trend line is fitted using the least squares method. The sequence standardization adjusts the mean of the phase sequence to zero and the standard deviation to 1, facilitating comparison with the post-conversion section. The sequence interpolation generates an equally spaced phase sequence using linear interpolation for the case of uneven pulse spacing.

[0062] The pulse sequence of the post-conversion section is misaligned and superimposed on the phase sequence to form a phase difference map. The sliding matching method is used to superimpose the phase sequence of the post-conversion section on the phase sequence of the pre-conversion section at different time offsets. The offset range is set to ±50% of the sequence length, and the offset step is 1 pulse interval. The phase difference calculation is Δφ = φ_after - φ_before, where Δφ is the phase difference (radians), φ_after is the post-conversion phase angle (radians), and φ_before is the pre-conversion phase angle (radians). The phase difference map has the offset as the horizontal axis and the phase difference as the vertical axis, with color depth representing the size of the phase difference. The superposition weight is assigned according to the pulse amplitude, with high weight for large amplitude pulses and low weight for small amplitude pulses. The best matching position is determined by minimizing the mean square error of the phase difference, and the offset position with the smallest error is the best superposition position. The difference significance is evaluated by the standard deviation of the phase difference, and a large standard deviation indicates significant differences before and after conversion.

[0063] The phase jump accumulation quantity is extracted from the phase difference map. The phase jump is defined as the sudden change of phase difference between adjacent positions, and the jump threshold is set to π / 6 radian. The gradient analysis method is used to detect the jump, and the first derivative of the phase difference is calculated by sliding window, and the window length is set to 3 sampling points, and the sliding step is 1 sampling point. The central difference format is used for gradient calculation to improve numerical accuracy, and upper and lower limits of gradient are set to avoid noise interference. The jump direction identification includes positive jump (phase difference increases) and negative jump (phase difference decreases), and the direction judgment is verified by the consistency of the gradient signs of the continuous 3 points to ensure the reliability of the jump identification. The jump amplitude calculates the phase difference change of each jump point, and the amplitude size reflects the severity of the jump, and the amplitude calculation uses the difference between the local maximum and minimum before and after the jump to avoid the influence of instantaneous fluctuations. The integral method is used for accumulation quantity calculation, and the accumulation quantity C=Σ|Δφ_jump|, where C is the phase jump accumulation quantity (radian), Δφ_jump is the phase difference change of each jump point (radian), and Σ represents the summation of all jump points. The spatial distribution analysis uses the two-dimensional kernel density estimation method to identify the aggregation mode of the jump points, and the kernel function uses the Gaussian kernel, and the bandwidth is optimized and determined by the cross-validation method. The area with high aggregation degree indicates the systematic change characteristics. The time correlation analysis establishes the cross-correlation function between the jump occurrence time and the conversion time, and the peak position of the correlation coefficient is used to determine the optimal time delay, and the correlation coefficient greater than 0.7 indicates strong correlation.

[0064] The waterway imbalance feature is formed based on the unevenness of the distribution of the phase jump accumulation quantity. The Gini coefficient method is used to quantify the unevenness of the distribution of the phase jump accumulation quantity, and the calculation formula is G=(2Σi×C_i) / (n×ΣC_i)-(n+1) / n, where G is the Gini coefficient (dimensionless), C_i is the accumulation quantity value of the i-th interval (radian), n is the total number of intervals, and i is the interval number. The Gini coefficient has a value range of 0-1, and the larger the value, the more uneven the distribution, and the more serious the waterway imbalance. The interval division divides the phase difference map into 20 intervals according to the offset, and the sum of the accumulation quantity in each interval is calculated. The unevenness classification divides the Gini coefficient into three levels: mild imbalance (G<0.3), moderate imbalance (0.3≤G<0.6), and severe imbalance (G≥0.6). The imbalance mode recognition analyzes the shape characteristics of the accumulation quantity distribution to identify typical modes such as local imbalance, overall imbalance, and periodic imbalance. The stability evaluation compares the imbalance features of multiple time windows to evaluate the stability of the imbalance state. The imbalance direction analysis identifies the position and direction of the main imbalance to provide guidance for adjustment measures.

[0065] In step S140, the air flow field offset is identified based on the heat exchange abnormal position table and the waterway imbalance feature, the diagnosis focus area is determined based on the air flow field offset, and the hierarchical detection sequence is generated using the diagnosis focus area.

[0066] Specifically, the air flow field deviation is identified by coupling correlation based on the heat exchange abnormal position table and the water route imbalance feature. The coupling weight distribution is determined according to the abnormal degree in the heat exchange abnormal position table and the Gini coefficient in the water route imbalance feature, and the weight formula is W = a x A_level + b x G, wherein W is the coupling weight, A_level is the abnormal degree grade value, G is the Gini coefficient, a and b are weight coefficients and are 0.6 and 0.4 respectively. The deviation is calculated based on the difference between the air flow disturbance intensity and the normal state, and the deviation V_offset = |V_disturbed-V_normal|, wherein V_offset is the air flow deviation, V_disturbed is the disturbance state wind speed inferred from the vibration sensor data change and the temperature difference fluctuation feature, and V_normal is the normal state wind speed. The spatial correlation analysis identifies the spatial correspondence between the abnormal position in the heat exchange abnormal position table and the inferred air flow deviation, and significant air flow deviation usually occurs around the abnormal position. The time correlation analysis compares the imbalance occurrence time in the water route imbalance feature and the time consistency of the air flow deviation change, and the time difference less than 5 minutes is considered to have strong correlation. The correlation strength is determined by comprehensive evaluation of the spatial distance and the time difference, and the correlation is strong when the distance is short and the time difference is small. The influence range analysis determines the influence radius of each abnormal position on the surrounding air flow field.

[0067] In some embodiments, the determination of the diagnostic focus area based on the air flow field deviation includes: converting the air flow field deviation into a deviation vector field; finding a vortex center point in the deviation vector field; performing region growing with the vortex center point as a seed to form a preliminary focus area; and performing boundary optimization on the preliminary focus area to form a diagnostic focus area.

[0068] The air flow field deviation is converted into a deviation vector field. The vector field distribution is established in the monitoring area by using the air flow field deviation V_offset (scalar value) and the deviation direction angle θ determined by the main vibration direction of the vibration sensor and the temperature difference gradient direction. The vector field grid resolution is set to 0.5 m x 0.5 m, covering the entire monitoring area. The vector component calculation includes X direction component V_x = V_offset x cos(θ) and Y direction component V_y = V_offset x sin(θ), wherein θ is the deviation direction angle. The spatial interpolation adopts the inverse distance weighted interpolation method, sets the maximum influence radius to 3 meters, and the measurement points beyond the range do not participate in the interpolation. The discrete deviation data is expanded into a continuous vector field distribution. The vector field smoothing processing adopts a two-dimensional Gaussian filter, and the filter parameters are adaptively adjusted according to the grid density to eliminate local noise and discontinuity and ensure the reasonableness of the field distribution. The boundary condition is set to gradually change the vector value to zero at the boundary of the monitoring area. The vector field quality evaluation includes continuity check and physical reasonableness verification, divergence reflects the convergence and divergence characteristics of the air flow, and the curl Reflecting the rotational characteristics of the airflow.

[0069] The vortex center point is found in the deflection vector field. Based on the characteristics of the curl distribution of the deflection vector field, the curl calculation adopts the second-order accuracy format of the finite difference method, and the single-sided difference is used for the grid boundary. The absolute value of the curl value where ω is the curl value (1 / s), denotes the partial derivative operation. The vortex center candidate points are determined through multi-step screening. First, the curl threshold screening is performed, then the local extreme value test is performed, and finally the connectivity analysis is performed. The points with absolute value of curl greater than the threshold value 0.5 / s and local extreme value are marked as candidate points. The stream line tracking verifies the authenticity of the candidate points. The stream line integral adopts the fourth-order Runge-Kutta method, and the integral step size is adaptively adjusted, with a step size range of 0.01-0.1 meters. The stream line distribution is drawn based on the deflection vector field. The stream line tracking starts simultaneously from 8 directions around the candidate point, and the tracking length is set to 2 times the influence radius of the candidate point. The stream lines around the real vortex center are in closed ring or spiral shape. The center point precision optimization adopts the centroid iteration algorithm, taking the candidate point as the initial value, and calculating the centroid in the local high-curl area. The iteration convergence criterion is that the position change of the centroid is less than 0.05 meters for two consecutive times. The vortex intensity evaluation is calculated by the circulation, where Γ is the circulation value (m² / s), V is the vector in the deflection vector field, and the circulation integral adopts the rectangular integral rule. The integral path is selected as a circular path with the vortex center as the center. The integral path radius is automatically determined by the curl intensity distribution, and the distance where the curl decays to 50% of the maximum value is selected as the integral radius. The vortex radius is determined by multiple criteria, including curl decay criterion, flow velocity decay criterion and stream line bending criterion. The weighted average of the three criteria is used as the final radius. The vortex radius is determined by the distance where the curl decays to 50% of the peak value, and the radius size reflects the influence range of the vortex. The vortex classification adopts the double criterion of curl sign and stream line direction. The clockwise vortex (negative curl) corresponds to the sinking air flow mode, and the counterclockwise vortex (positive curl) corresponds to the ascending air flow mode. Different types of vortices correspond to different air flow anomaly modes.

[0070] The preliminary focusing area is numbered and classified according to the importance.

[0071] The preliminary focusing area is numbered and classified according to the importance. The boundary optimization is divided into three stages of preprocessing, shape regularization and post-processing based on the geometric shape characteristics of the preliminary focusing area. The irregular boundary is corrected to a regular shape to facilitate subsequent diagnostic operations. The boundary smoothing processing adopts morphological closing operation, and the structural element is adaptively selected according to the boundary complexity to eliminate jagged protrusions and small recesses. The convex hull calculation adopts Graham scan algorithm to generate the smallest enclosing polygon in polar angle order, and the smallest boundary polygon enclosing the preliminary focusing area is generated as the basis for regularization processing. The ellipse fitting adopts least squares method, considering the weight distribution of boundary points, to approximately process the shape of the focusing area and obtain the regularized elliptical boundary. The boundary buffer zone increases a 0.5-meter buffer zone outside the fitted boundary to ensure the integrity of the diagnostic coverage. The area constraint is realized through threshold checking and equal proportion scaling, and the excessively small area is appropriately expanded and the excessively large area is appropriately contracted. The shape regularity evaluation adopts geometric indicators such as circularity and compactness to comprehensively score, ensuring that the optimized area is convenient for detection operation. The overlap processing adopts region merging algorithm, and the merging strategy is determined according to the overlap area ratio. When the overlap is serious, the regions are merged. The boundary precision reaches 0.1-meter level, meeting the requirements of diagnostic positioning.

[0072] The diagnostic focus area is used to generate a hierarchical detection sequence. Based on the importance of the diagnostic focus area and the vortex intensity feature, a hierarchical detection strategy is established. The detection hierarchy includes three levels: core detection layer, key detection layer, and regular detection layer. The core detection layer targets important focus areas with a vortex intensity greater than 2.0 m² / s and an area greater than 10 m², with a detection density of 4 detection points per square meter. The key detection layer targets medium focus areas with a vortex intensity of 1.0-2.0 m² / s, with a detection density of 2 detection points per square meter. The regular detection layer covers general focus areas with a vortex intensity less than 1.0 m² / s, with a detection density of 1 detection point per square meter. The detection point arrangement uses a hexagonal close-packed method to ensure uniformity of detection coverage. The detection sequence is ordered according to the priority of the focus area, with important areas being detected first. The detection parameters include temperature, humidity, wind speed, pressure, and other environmental parameters to form comprehensive diagnostic data. The time schedule takes into account detection efficiency, with the total detection time controlled within 2 hours. Path optimization is based on the spatial distribution of focus areas, using a traveling salesman problem solving algorithm to minimize the movement distance and time of the detection equipment. The detection accuracy requirements are ±0.1℃ for the core layer, ±0.2℃ for the key layer, and ±0.5℃ for the regular layer.

[0073] In step S150, based on the hierarchical detection sequence, active and passive damage classification is performed to generate active damage data and passive damage data. Motion trajectory analysis is performed on the active damage data to extract the damage development rate. Based on the damage development rate and passive damage data, a damage superposition coefficient is generated.

[0074] In some embodiments, the active and passive damage classification based on the hierarchical detection sequence generates active damage data and passive damage data, including: identifying the source of the damage based on the hierarchical detection sequence to generate endogenous damage and exogenous damage; using the endogenous damage to assess the influence of the exogenous damage to form a damage correlation matrix; performing eigenvalue decomposition on the damage correlation matrix to generate a dominant damage vector; and according to the dominant damage vector, implementing damage classification to generate active damage data and passive damage data.

[0075] The damage source positioning and identification of the hierarchical detection sequence generates endogenous damage and exogenous damage. Based on the detection parameters such as temperature, humidity, wind speed and pressure in the hierarchical detection sequence, the abnormal propagation path analysis is adopted to track the parameter gradient change from the detection point with the strongest abnormal intensity to the surrounding. The endogenous damage identification criteria include the following characteristics: the abnormal source is located in the internal equipment of the air conditioning system, the abnormal parameters are directly related to the equipment operating state, and the abnormal intensity changes with the equipment load. The exogenous damage identification criteria include the following characteristics: the abnormal source comes from the external environment, the abnormal parameters are related to the changes of external conditions, and the abnormal intensity is not affected by the equipment adjustment. The damage intensity quantification adopts the weighted average of parameter deviation, and the intensity value I = Σ(w_i x |P_i-P_ref|) / Σ(w_i), where I is the damage intensity, w_i is the weight of the i-th parameter, P_i is the measured parameter value, and P_ref is the reference value. Spatial clustering combines adjacent abnormal detection points into damage areas, and clustering is based on the spatial distribution characteristics of the hierarchical detection sequence. Time feature analysis is based on the time series of the hierarchical detection sequence to analyze the occurrence time, duration and development trend of the damage. Endogenous damage usually shows intermittent and controllable characteristics, and exogenous damage shows persistent and uncontrollable characteristics.

[0076] The influence degree of endogenous damage on exogenous damage is evaluated to form a damage correlation matrix. Based on the position, intensity and time characteristics of endogenous damage, the influence degree of endogenous damage on exogenous damage is evaluated. The spatial influence degree is calculated by a distance attenuation function, and the influence function is I_spatial=exp(-d / d_0), where I_spatial is the spatial influence degree, d is the distance between the centers of two damages, and d_0 is the influence attenuation constant, which is 3 meters. The time influence degree is based on the time characteristic analysis results of endogenous damage and exogenous damage. Damages that occur at the same time or have a time difference of less than 10 minutes have high time influence degree. The intensity influence degree is calculated by the correlation between the intensity of endogenous damage and the intensity of exogenous damage. Damages with consistent intensity trends have high intensity influence degree. The comprehensive influence degree I_total=α×I_spatial+β×I_temporal+γ×I_intensity, where I_total is the comprehensive influence degree, I_temporal is the time influence degree, I_intensity is the intensity influence degree, and α, β, γ are weight coefficients, which are 0.4, 0.3 and 0.3 respectively. The damage correlation matrix M is an n x m matrix, n is the number of endogenous damage, m is the number of exogenous damage, and the matrix element M_ij represents the influence degree of the i-th endogenous damage on the j-th exogenous damage.

[0077] The dominant damage vector is generated by eigenvalue decomposition of the damage correlation matrix. Based on the numerical distribution characteristics of the damage correlation matrix M, the matrix is preprocessed first, including data centering and standardization, to eliminate the influence of dimension. Singular value decomposition (SVD) method is used to decompose the damage correlation matrix M into M = UΣV^T, where U is the left singular vector matrix, Σ is the singular value diagonal matrix, and V is the right singular vector matrix. The SVD decomposition uses a double diagonalization algorithm, which first converts the matrix to a double diagonal form through Householder transformation, and then uses QR iteration to solve the singular value. Numerical stability is guaranteed by condition number check, and when the matrix condition number is greater than a certain threshold, the truncated SVD method is used to improve the numerical stability. The dominant damage vector corresponds to the singular vector of the largest singular value, reflecting the main mode in the damage system. The dominant vector is determined by analyzing the energy contribution rate of the singular value, which calculates the proportion of each singular value in the total energy. The size of the singular value reflects the importance of the corresponding mode, and the cumulative contribution rate criterion is used to determine the number of singular values to be retained. Usually, the first few singular values with a cumulative contribution rate of 90% are retained. The elements of the dominant vector represent the contribution of each damage to the dominant mode, and the larger the element value, the more important the damage plays in the dominant mode. The sub-dominant vector corresponds to the second largest singular value, reflecting the secondary mode of the damage system, which is used to verify the reliability and integrity of the dominant mode. The mode contribution rate is calculated using the variance explained rate to evaluate the ability of each mode to explain the variation of the original data, and the contribution rate calculation considers the orthogonality constraint between modes.

[0078] According to the dominant damage vector, active damage data and passive damage data are generated. Based on the numerical distribution and mode contribution rate of each element in the dominant damage vector, damage classification is determined. The active damage determination standard is that the dominant vector element value is greater than 0.3 and the corresponding damage has controllability and responsiveness characteristics. The passive damage determination standard is that the absolute value of the dominant vector element value is less than 0.3 or the corresponding damage has uncontrollability and hysteresis characteristics. The classification threshold of 0.3 is dynamically adjusted in combination with the mode contribution rate, and a stricter threshold is used for dominant modes with high contribution rate. Active damage data includes damage location, damage type, control parameter, response time, adjustable range, etc. Passive damage data includes damage location, damage source, impact degree, duration, adaptation strategy, etc. Cross-validation is used to verify the classification results using the sub-dominant vector to improve the classification accuracy. Boundary damage processing is used for damages with vector element values close to the threshold, combined with time characteristics and spatial clustering results for comprehensive judgment.

[0079] Trajectory analysis is applied to active injury data to extract the injury development rate. The trajectory of injury development is constructed based on the injury location, time information and control parameter changes in active injury data. Time series analysis is used to connect the position coordinates of the same injury at different times to form a trajectory. The position accuracy is 0.1 meters, and the time accuracy is minutes, which ensures the accuracy of trajectory analysis. Kalman filter is used for trajectory smoothing to eliminate the influence of position measurement noise on the trajectory. The central difference method is used to calculate the rate, v=(s_{i+1}-s_{i-1}) / (2Δt), where v is the injury development rate, s_i is the position coordinate at the i-th time, and Δt is the time interval. The rate direction reflects the main direction of injury expansion combined with the analysis of control parameter changes in active injury data. Trajectory pattern recognition includes typical patterns such as linear expansion, radial expansion, and spiral expansion, and pattern type is used for the calculation and correction of superposition coefficient. The expansion range is calculated by the envelope area of the trajectory, and the area size is used as the spatial factor for superposition coefficient calculation. Statistical analysis of development rate calculates statistical parameters such as average rate, maximum rate, and rate variance.

[0080] Based on the injury development rate and passive injury data, the injury superposition coefficient is generated. The statistical parameters of the injury development rate and the position, intensity, duration, etc. in passive injury data are used to calculate the injury superposition effect. Spatial overlap degree analyzes the overlap degree of the expansion range of active injury and the position of passive injury, and the overlap area ratio O=A_overlap / A_total, where O is the overlap degree, A_overlap is the overlap area, and A_total is the total influence area. The rate influence factor is based on the average value of the injury development rate, and the influence factor F=1+k×v_avg, where F is the rate influence factor, k is the rate coefficient, which is 0.1 min / m, and v_avg is the average development rate. The intensity superposition considers the influence degree of active injury development and passive injury data, and the superposed intensity I_superposed=√(I_active²+I_passive²), where I_superposed is the superposed intensity, I_active is the active injury intensity, and I_passive is the passive injury intensity. Time synchronization analysis is based on the development time of active injury and the duration in passive injury data, and the injury with a time difference of less than 5 minutes is synchronized and superposed. Trajectory pattern matching analyzes the matching degree of the expansion pattern of active injury and the spatial distribution pattern of passive injury, and the superposition effect is stronger with higher matching degree. The superposition coefficient C=O×F×(I_superposed / I_max), where C is the injury superposition coefficient, and I_max is the maximum injury intensity of the system. The coefficient classification divides the superposition coefficient into three levels: mild superposition (C<0.3), moderate superposition (0.3≤C<0.7), and severe superposition (C≥0.7).

[0081] In step S160, the fault coupling amplification region is identified based on the damage superposition coefficient and the heat exchange abnormal position table, chain analysis is performed on the fault coupling amplification region to generate damage synergistic effect data, and the abnormal state diagnosis of the fan coil is completed.

[0082] Specifically, the fault coupling amplification region is identified based on the damage superposition coefficient and the heat exchange abnormal position table. The superposition level, overlap degree, and rate influence factor in the damage superposition coefficient generated in S150 are combined with the abnormal position coordinates, abnormal degree, and area in the heat exchange abnormal position table established in S120 to locate the key region in which multiple faults are coupled and amplified. The coupling determination criteria include the conditions that the damage superposition coefficient is greater than 0.5, the heat exchange abnormal region area is greater than 8 square meters, and the abnormal degree is moderate or above. Spatial coupling analysis is performed by identifying the spatial proximity of abnormal positions. Abnormal positions with a distance of less than 3 meters have a strong spatial coupling relationship. Time coupling analysis checks the consistency of abnormal occurrence times. Abnormal events with a time difference of less than 15 minutes have time coupling characteristics. The amplification effect is quantified using the multiplier effect formula M=C×(1+A_level / A_max), where M is the amplification factor, C is the damage superposition coefficient, A_level is the abnormal degree level value, and A_max is the maximum abnormal degree. The region boundary is determined by the equal amplification factor line. A continuous region with an amplification factor greater than 1.5 is defined as a fault coupling amplification region. The influence intensity classification divides the amplification region into three levels: strong coupling area (M>2.0), medium coupling area (1.5≤M≤2.0), and weak coupling area (1.2≤M<1.5). The region characteristics include center coordinates, boundary range, amplification factor, coupling type, and affected equipment parameters. The data structure of the fault coupling amplification region is [region number, center coordinates, boundary coordinates, amplification factor, coupling strength, main fault type].

[0083] In some embodiments, the chain analysis of the fault coupling amplification region to generate damage synergistic effect data includes: identifying chain propagation characteristics based on the fault coupling amplification region to determine an analysis path, the chain propagation characteristics including air volume attenuation gradient, condensate retention degree, and fin fouling rate; tracing the fault transmission process along the analysis path to form a propagation chain graph; extracting amplification coefficients of each link from the propagation chain graph; and arranging the amplification coefficients in the propagation order to generate damage synergistic effect data.

[0084] Based on the fault coupling amplification area, the chain propagation feature determination analysis path is identified. The spatial distribution and coupling strength information of the fault coupling amplification area are used to infer three chain propagation features. The wind volume attenuation gradient is inferred by the change of vibration sensor data. The vibration intensity abnormal area corresponds to the wind volume attenuation serious area. The gradient value is estimated by the ratio of vibration intensity change and distance interval. The area with gradient greater than 0.51 / s is marked as wind volume rapid attenuation area, the area with gradient 0.2-0.51 / s is marked as moderate attenuation area, and the area with gradient less than 0.21 / s is marked as slow attenuation area. The condensate retention degree is inferred by the local temperature anomaly of the temperature sensor. The temperature distribution uneven area usually accompanies the condensate retention problem. The retention index is calculated by the relative percentage of temperature deviation. The retention index greater than 30% indicates serious retention, 15%-30% indicates moderate retention, and less than 15% indicates slight retention. The fin fouling rate is inferred by the load change trend of the current sensor. The continuous increase of motor load reflects the increase of fin resistance. The fouling rate is calculated by the ratio of load increase and time interval. The fouling rate greater than 50Pa / day is fast fouling, 20-50Pa / day is moderate fouling, and less than 20Pa / day is slow fouling. The analysis path determination is based on the influence intensity grading result. The main propagation path is from the strong coupling area to the weak coupling area. The path weight is allocated according to the comprehensive influence of the three propagation features. The weight coefficients are wind volume attenuation 0.4, condensate retention 0.3, and fin fouling 0.3.

[0085] The propagation chain graph is formed by tracking the fault transmission process along the analysis path. Based on the spatial distribution and propagation direction of the analysis path, the transmission node is set at every 2 meters or at the position where the fault feature changes significantly. Each node records the fault state and propagation parameters at the current position. The propagation intensity attenuation adopts an exponential decay model, I(x)=I_0×exp(-αx), where I(x) is the propagation intensity at distance x, I_0 is the initial propagation intensity (set based on the amplification factor M of the influence intensity grading), α is the attenuation coefficient, and x is the propagation distance. The connection relationship between nodes is established based on the causal relationship of fault propagation. The fault state of the upstream node affects the fault development of the downstream node. The propagation delay time considers the time required for the fault to propagate from one node to the next. The delay time is used to determine the time window range of diagnosis. The propagation chain graph adopts a directed graph structure, with nodes representing fault states and arrows representing propagation direction and intensity. Branch processing is performed when the fault propagation appears to branch. When multiple propagation chains converge at a node, the superposition effect is used to calculate the comprehensive fault intensity of the node.

[0086] The amplification coefficient of each link is extracted from the propagation chain graph. Based on the node failure intensity distribution of the propagation chain graph, the amplification coefficient is defined as the amplification multiple of each link relative to the failure intensity of the upstream node. The direct amplification coefficient is calculated by the ratio of the current node failure intensity to the upstream node failure intensity. The cumulative amplification coefficient considers the total amplification effect from the starting node to the current node, and is used to evaluate the overall amplification degree of failure propagation. The local amplification coefficient reflects the failure amplification ability of the link itself, and does not include the influence of propagation attenuation. The environmental correction coefficient is based on temperature sensor data, considering the influence of temperature change on the amplification coefficient. The corrected amplification coefficient is obtained by multiplying the direct amplification coefficient by the environmental correction coefficient. Abnormal amplification detection identifies links with abnormally high amplification coefficients, which usually correspond to key failure nodes or amplification mechanisms. Amplification coefficient distribution analysis identifies the main contribution nodes of amplification effect by analyzing the distribution of amplification coefficients in the entire propagation chain. Link importance evaluation is based on the comprehensive evaluation of amplification coefficient and node position, and the link with high importance is used to determine the key maintenance position.

[0087] The damage synergy effect data is generated by arranging the amplification coefficients in the propagation order. Based on the amplification coefficients of each link and the space-time sequence of the propagation chain graph, the time stamp of the failure arriving at each node is used to consider the influence of propagation delay time. The spatial order is arranged according to the distance from the fault source to the impact terminal. The synergy effect is quantified by weighted calculation combined with path weight, and the synergy coefficient S_synergy=∏(w_i×A_i), where S_synergy is the synergy coefficient, A_i is the amplification coefficient of the i th link, and w_i is the corresponding path weight. Effect classification divides the synergy effect into three types: enhancement type (S_synergy>1.5), maintenance type (0.8≤S_synergy≤1.5), and attenuation type (S_synergy<0.8). Key link identification is achieved by detecting the mutation point of the amplification coefficient, and the key link information is used for key maintenance recommendations in the diagnosis report. Synergy mode analysis identifies different synergy effect modes such as linear synergy, nonlinear synergy, and threshold synergy. Effect intensity evaluation is based on the comprehensive evaluation of the size of the synergy coefficient and the influence range. Data integrity check ensures that the synergy effect data covers the complete failure propagation process.

[0088] Based on the deep analysis of the damage synergy effect data, the abnormal state and fault type of the fan coil system are comprehensively evaluated. Using the synergy coefficient, effect classification, key chain, and other information in the damage synergy effect data, the diagnostic classification divides the abnormal state into four levels: slight abnormality, medium abnormality, severe abnormality, and critical abnormality. Slight abnormality corresponds to a synergy coefficient of 1.0-1.3, which affects local performance but does not affect overall operation. Medium abnormality corresponds to a synergy coefficient of 1.3-1.8, which requires timely maintenance but does not affect normal use. Severe abnormality corresponds to a synergy coefficient of 1.8-2.5, which affects the normal operation of the system and needs to be handled immediately. Critical abnormality corresponds to a synergy coefficient greater than 2.5, which has a risk of system failure and needs to be shut down for emergency repair. Fault positioning is based on the identification results of the key chain, accurate to the device component level, including fan blades, motor bearings, heat exchange fins, condenser, control valves, and other specific positions. Root cause analysis identifies the initial cause and triggering mechanism of the fault through reverse tracking of the propagation chain diagram. Maintenance recommendations provide specific maintenance measures and time arrangements based on the chain importance evaluation results and abnormality levels. Through accurate abnormal state diagnosis, the key factors affecting system energy efficiency can be accurately identified, avoiding the blindness and over-maintenance of traditional experience-based maintenance, and achieving targeted problem solving. The diagnostic system can timely detect factors that increase hidden performance and energy consumption, such as heat exchange efficiency decline, water imbalance, and uneven airflow distribution, and develop targeted energy-saving solutions such as heat exchanger deep cleaning, water rebalancing, and accurate air volume adjustment, effectively reducing system energy consumption and prolonging equipment service life. Prognosis evaluation predicts the remaining useful life and performance degradation of the system under the current fault development trend by combining the cumulative amplification coefficient and propagation delay time. The report generation automatically generates a diagnostic report containing abnormal state, fault location, cause analysis, and maintenance recommendations. Through systematic analysis and multi-dimensional comprehensive evaluation of damage synergy effect, accurate and reliable fan coil abnormal state diagnosis is finally achieved.

[0089] To perform a fan coil abnormal state diagnosis method corresponding to the above-mentioned method embodiment, to realize the corresponding functions and technical effects. Referring to Figure 2 , Figure 2 The structure block diagram of a fan coil abnormal state diagnosis system 200 provided by an embodiment of the present application is shown. For ease of illustration, only the parts related to the present embodiment are shown. The fan coil abnormal state diagnosis system 200 provided by an embodiment of the present application comprises:

[0090] The signal acquisition module 201 is configured to acquire fan current waveform and coil vibration signals, extract motor load change characteristics from the current waveform, extract heat exchange tube vibration patterns based on the coil vibration signals, and establish a heat exchange efficiency benchmark according to the motor load change characteristics and the heat exchange tube vibration patterns.

[0091] The abnormal positioning module 202 is configured to identify an efficiency sudden drop area based on the heat exchange efficiency benchmark, locate a heat exchange dead zone expansion range according to the efficiency sudden drop area, generate heat transfer coefficient jump data based on the heat exchange dead zone expansion range, and determine a heat exchange abnormal position table by using the heat transfer coefficient jump data.

[0092] The water path analysis module 203 is configured to obtain a supply-return water temperature difference and a water valve opening degree change, identify a water flow pulse feature by using the supply-return water temperature difference, capture a flow state conversion point according to the water valve opening degree change, and perform coupling analysis on the water flow pulse feature and the flow state conversion point to form a water path imbalance feature.

[0093] The coupling diagnosis module 204 is configured to identify an air flow field deviation based on the heat exchange abnormal position table and the water path imbalance feature, determine a diagnosis focusing area based on the air flow field deviation, and generate a hierarchical detection sequence by using the diagnosis focusing area.

[0094] The damage analysis module 205 is configured to perform active and passive damage classification based on the hierarchical detection sequence to generate active damage data and passive damage data, perform motion trajectory analysis on the active damage data to extract a damage development rate, and generate a damage superposition coefficient based on the damage development rate and the passive damage data.

[0095] The cooperative diagnosis module 206 is configured to identify a fault coupling amplification area based on the damage superposition coefficient and the heat exchange abnormal position table, perform chain analysis on the fault coupling amplification area to generate damage cooperative effect data, and complete fan coil abnormal state diagnosis.

[0096] The fan coil abnormal state diagnosis system 200 described above can implement a fan coil abnormal state diagnosis method of the method embodiment described above. The optional items in the method embodiment described above are also applicable to this embodiment, and will not be described in detail here. The remaining content of the embodiment of the present application can refer to the content of the method embodiment described above, and will not be described in detail in this embodiment.

[0097] The purpose of the above embodiments is to exemplarily reproduce and deduce the technical solutions of the present application, and to completely describe the technical solutions, purposes and effects of the present application, so as to make the public understand the disclosure of the present application more thoroughly and comprehensively, and not to limit the protection scope of the present application.

[0098] The above embodiments are not based on an exhaustive enumeration of the present application, and there can be many other unlisted embodiments. Any replacement and improvement made without violating the concept of the present application is within the protection scope of the present application.

Claims

1. A method for diagnosing abnormal conditions of a fan coil unit, characterized in that: include: Obtaining a fan current waveform and a coil vibration signal, extracting a motor load variation characteristic from the current waveform, extracting a heat exchange tube vibration pattern based on the coil vibration signal, and establishing a heat exchange efficiency benchmark based on the motor load variation characteristic and the heat exchange tube vibration pattern; identifying an efficiency sudden drop region based on the heat exchange efficiency benchmark, locating a heat exchange dead zone expansion range according to the efficiency sudden drop region, generating heat transfer coefficient jump data based on the heat exchange dead zone expansion range, and determining a heat exchange abnormality location table using the heat transfer coefficient jump data; Obtaining the supply and return water temperature difference and the change in water valve opening, identifying the water flow pulse characteristics based on the supply and return water temperature difference, capturing the flow pattern transition point based on the change in water valve opening, and coupling the water flow pulse characteristics with the flow pattern transition point to form a waterway imbalance characteristic; performing coupling correlation based on the heat exchange abnormality position table and the water channel imbalance characteristics to identify the airflow field offset, determining a diagnostic focus area based on the airflow field offset, and generating a hierarchical detection sequence using the diagnostic focus area; performing active and passive damage classification based on the hierarchical detection sequence to generate active damage data and passive damage data, performing motion trajectory analysis on the active damage data to extract damage development rate, and generating a damage superposition coefficient based on the damage development rate and the passive damage data; Based on the damage superposition coefficient and the heat exchange abnormality position table, the fault coupling amplification area is identified, and a chain analysis is performed on the fault coupling amplification area to generate damage synergistic effect data, thereby completing the abnormal state diagnosis of the fan coil unit.

2. The method according to claim 1, characterized in that The establishing of a heat exchange efficiency benchmark according to the motor load variation characteristics and the heat exchange tube vibration mode includes: generating a load fluctuation timing diagram based on the motor load variation characteristics; Combining the heat exchange tube vibration mode and the load fluctuation time sequence diagram to form a vibration-load coupling curve; extracting a stable working segment from the vibration-load coupling curve; The efficiency value of the stable working section is set as the heat exchange efficiency benchmark.

3. The method according to claim 1, characterized in that The step of locating the expansion range of the heat exchange dead zone according to the efficiency sudden drop area includes: dividing the efficiency drop region into a central region and a peripheral region; transmitting a detection path from the central region to the edge region; Recording the locations of efficiency recovery points on the detection path to form a recovery point set; The farthest points in the recovery point set are connected to form a heat exchange dead zone expansion range.

4. The method according to claim 1, wherein The method of determining a heat exchange abnormality position table by using the heat transfer coefficient jump data includes: Performing heat flow blocking identification on the heat transfer coefficient jump data to generate a streamline termination area; Performing an expansion impact assessment based on the streamline termination area to form an expansion coefficient; generating a continuous expansion field by spatial interpolation of the expansion coefficients; Boundary extraction is performed based on the continuous expansion field to generate a heat exchange abnormality position table.

5. The method according to claim 1, wherein The coupling analysis of the water flow pulse characteristics and the flow state transition point to form the waterway imbalance characteristics includes: Constructing a pulse time axis based on the water flow pulse characteristics; Mapping the flow state transition point to the pulse time axis to form a transition moment mark; Dividing the pulse time axis into a pre-conversion segment and a post-conversion segment with the conversion moment mark as a boundary; The phase distribution characteristics of the front section and the back section of the conversion are compared to form a waterway imbalance characteristic.

6. The method according to claim 1, characterized in that The determining of the diagnosis focus area based on the airflow field offset includes: Converting the airflow field offset into an offset vector field; Finding a vortex center point in the offset vector field; Taking the vortex center point as a seed, regional growth is performed to form a preliminary focusing area; Boundary optimization is performed on the preliminary focusing area to form a diagnostic focusing region.

7. The method according to claim 1, characterized in that The active and passive damage classification based on the hierarchical detection sequence to generate active damage data and passive damage data includes: Performing damage source location identification on the layered detection sequence to generate endogenous damage and exogenous damage; Using the endogenous damage to evaluate the impact of the exogenous damage to form a damage correlation matrix; generating a dominant damage vector by performing eigenvalue decomposition on the damage correlation matrix; Damage classification is performed according to the dominant damage vector to generate active damage data and passive damage data.

8. The method according to claim 1, characterized in that The performing chain analysis on the fault coupling amplification region to generate damage synergy effect data includes: Determining an analysis path based on identifying chain propagation characteristics of the fault coupling amplification region, wherein the chain propagation characteristics include air volume attenuation gradient, condensate retention degree, and fin fouling rate; Tracing the fault transmission process along the analysis path to form a propagation chain diagram; Extracting the amplification factor of each chain link from the propagation chain diagram; The amplification factors are arranged in propagation order to generate damage synergy effect data.

9. The method according to claim 5, characterized in that The forming of a waterway imbalance feature by comparing the phase distribution characteristics of the pre-conversion section and the post-conversion section includes: Converting the pulse sequence of the front-end conversion section into a phase sequence; Staggering and superimposing the pulse sequence of the post-conversion segment onto the phase sequence to form a phase difference map; Extracting the phase jump cumulative amount from the phase difference map; A waterway imbalance feature is formed based on the uneven distribution of the phase jump accumulation amount.

10. A fan coil unit abnormal state diagnosis system, characterized in that: include: a signal acquisition module configured to obtain a fan current waveform and a coil vibration signal, extract a motor load variation characteristic from the current waveform, extract a heat exchange tube vibration pattern based on the coil vibration signal, and establish a heat exchange efficiency benchmark based on the motor load variation characteristic and the heat exchange tube vibration pattern; an abnormality locating module, configured to identify an efficiency sudden drop area based on the heat exchange efficiency benchmark, locate an expansion range of a heat exchange dead zone according to the efficiency sudden drop area, generate heat transfer coefficient jump data based on the expansion range of the heat exchange dead zone, and determine a heat exchange abnormality location table using the heat transfer coefficient jump data; A waterway analysis module is configured to obtain a supply / return water temperature difference and a change in water valve opening, identify water flow pulse characteristics based on the supply / return water temperature difference, capture flow pattern transition points based on the change in water valve opening, and couple the water flow pulse characteristics with the flow pattern transition points for analysis to generate waterway imbalance characteristics; a coupled diagnosis module, configured to identify an airflow field offset by performing coupled correlation based on the heat exchange abnormality position table and the water channel imbalance characteristic, determine a diagnostic focus area based on the airflow field offset, and generate a hierarchical detection sequence using the diagnostic focus area; a damage analysis module, configured to perform active and passive damage classification based on the hierarchical detection sequence to generate active damage data and passive damage data, perform motion trajectory analysis on the active damage data to extract a damage development rate, and generate a damage superposition coefficient based on the damage development rate and the passive damage data; The collaborative diagnosis module is used to identify the fault coupling amplification area based on the damage superposition coefficient and the heat exchange abnormality position table, perform chain analysis on the fault coupling amplification area to generate damage collaborative effect data, and complete the abnormal state diagnosis of the fan coil unit.

Images