Impeller type hydraulic dynamometer state on-line monitoring and abnormality discrimination and processing method

Abstract

This invention discloses an online monitoring, anomaly detection, and handling method for an impeller-type hydraulic dynamometer, belonging to the field of diesel engine testing technology. Addressing issues such as abnormal power measurement, particulate pitting of the rotor impeller, dynamic imbalance of the rotor assembly, and abnormalities of the outlet valve during dynamometer operation, it collects, displays, and stores signals from 12 channels of 6 types of sensors. An online anomaly detection method based on thresholds, trends, and models is designed to achieve anomaly monitoring, alarm, and tiered handling. The method calculates the endothermic power to determine the rationality of the power measurement; monitors the water chamber pressure to assess the impeller morphology degradation trend; combines time-domain vibration amplitude and frequency-domain characteristic analysis to identify rotor assembly dynamic imbalance; monitors current trends to determine valve core rotation jamming; and monitors dual encoder signals to determine if the outlet valve drive shaft is disconnected. Practical application shows that the dynamometer anomaly handling time has decreased to less than 3% of the total downtime.

Description

Technical Field

[0001] This invention belongs to the field of diesel engine testing technology, and in particular relates to a method for online monitoring, anomaly identification and handling of impeller-type hydraulic dynamometers. Background Technology

[0002] The impeller-type hydraulic dynamometer is a key testing instrument used to absorb the mechanical power of rotating power equipment, measure speed and torque, and control the load working mode of power equipment. Its reliable operation is a necessary prerequisite for ensuring test safety and improving test quality and efficiency.

[0003] Currently, scholars and research institutions have conducted numerous studies on the diagnosis of hydraulic dynamometer anomalies: classification of hydraulic dynamometer anomalies based on convolutional neural networks aims for rapid, accurate, and effective feature differentiation; addressing the scarcity of bearing anomaly samples in hydraulic dynamometers, a bearing anomaly diagnosis method based on improved generative adversarial networks and integrating speed tracking processing with convolutional neural networks has been proposed, enabling the identification of anomalies in the outer ring, inner ring, and rolling elements of bearings. These neural network-based anomaly diagnosis methods have advantages such as high training efficiency and strong interpretability, and are suitable for diagnosing single-point, static, and non-propagating anomalies, but they struggle to effectively handle unseen or untrained anomaly types.

[0004] However, impeller-type hydraulic dynamometers are prone to various anomalies during actual operation, such as abnormal power measurement, particulate pitting of the rotor impeller, dynamic imbalance of the rotor assembly, and abnormal outlet valve. Currently, the handling of these problems is mostly based on passive response or planned maintenance, lacking effective online monitoring methods, which often leads to unplanned downtime and makes it difficult to achieve a shift to a proactive preventive maintenance model. Summary of the Invention

[0005] In view of this, the present invention aims to propose a method for online monitoring, anomaly identification, and handling of impeller-type hydraulic dynamometers to solve the following problems: Existing technologies lack effective online monitoring methods (including those mentioned above: such neural network-based anomaly diagnosis methods have advantages such as high training efficiency and strong interpretability, and are suitable for diagnosing single-point, static, and non-propagating anomalies, but are difficult to effectively deal with unseen or untrained anomaly types), often leading to unplanned downtime and making it difficult to achieve a shift to a proactive preventive operation and maintenance mode.

[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows: In a first aspect, the present invention provides a method for online monitoring, anomaly detection, and handling of a turbine-type hydraulic dynamometer, characterized by comprising the following steps: S1. Acquire multi-channel operating status signals: Real-time acquisition of hydraulic, thermal, mechanical and control signals through a sensor array deployed on the hydraulic dynamometer, and analog-to-digital conversion, data storage, data output and data display using a data acquisition instrument that matches differentiated sampling modules according to the steady-state or transient characteristics of the signals; S2. Perform hierarchical anomaly discrimination: Input the collected operating status signal into the discrimination module of the data acquisition instrument, and simultaneously perform threshold quantitative discrimination, trend qualitative discrimination and model logic discrimination, and output the anomaly type and alarm signal; S3. Closed-loop anomaly handling: Based on the alarm signal, perform fault location, hierarchical decision-making, targeted maintenance and static and dynamic verification, and archive the data of the whole process.

[0007] Furthermore, in S1, the sensor array includes 12 monitoring channels, specifically including: a volumetric flow meter, an inlet water T-type temperature sensor, an inlet water relative pressure sensor, a free-end rotor bearing outer ring thermal resistance sensor, a dynamometer water cavity relative pressure sensor, a stepper motor operating current sensor, a free-end rotor bearing base uniaxial vibration sensor, an outlet water T-type temperature sensor, a coupling end rotor bearing base uniaxial vibration sensor, a stepper motor encoder, a valve core connecting shaft free-end encoder, and a coupling end rotor bearing outer ring thermal resistance sensor.

[0008] Furthermore, in S1, the specific configuration of the differential sampling module is as follows: The thermocouple signal acquisition module connects the inlet water type T temperature sensor and the outlet water type T temperature sensor, with a sampling frequency ≥10Hz and a resolution ≥14 bits; The thermal resistance signal acquisition module connects the thermal resistance sensor on the outer ring of the free end rotor bearing to the thermal resistance sensor on the outer ring of the coupling end rotor bearing, with a sampling frequency ≥100Hz and a resolution ≥14 bits. The voltage and current signal acquisition module is connected to the inlet relative pressure sensor, the dynamometer water cavity relative pressure sensor, the stepper motor working current sensor, the free end rotor bearing base single-axis vibration sensor and the coupling end rotor bearing base single-axis vibration sensor, with a sampling frequency ≥10f and a resolution ≥14 bits, where f is the device reference frequency; The high-speed pulse acquisition module is connected to the volumetric flow meter, the stepper motor encoder, and the encoder at the free end of the valve core connecting shaft. The sampling frequency is ≥Max(10f,4MHz) and the resolution is ≥16 bits.

[0009] Furthermore, in S2, the specific logic for the hierarchical anomaly detection is as follows: The threshold quantitative judgment is used for single variable monitoring, and an anomaly is triggered when the measured value of the parameter exceeds the preset operating threshold. The trend qualitative discrimination is used for gradual process monitoring. By calculating the slope of the parameter time series change, a slope below the decay threshold is judged as performance degradation, and a slope above the mutation threshold is judged as sudden anomaly. The model logic judgment is used for composite parameter verification. A calculation model is established based on the physical constraints of the equipment. When the deviation between the measured value and the calculated value of the model exceeds the tolerance range, it is judged as abnormal.

[0010] Furthermore, the model logic judgment specifically includes: Power measurement model discrimination: Constructing an expression based on the law of conservation of energy In the formula, It is a dynamometer that measures rotational speed; It is a dynamometer that measures torque; It is the water flow rate; It is the temperature difference between the inlet and outlet water; It is the mechanical power of the diesel engine; The heat output power of water after heat absorption. Compare this to mechanical power. Heat absorption power of water The difference is considered as follows: a smaller difference indicates equipment performance degradation, and a larger difference indicates power measurement abnormality. Valve system synchronization model judgment: The position feedback signals of the stepper motor encoder and the encoder at the free end of the valve core connecting shaft are compared in real time. If the two are not synchronized and the diesel engine cannot follow the throttle adjustment, it is judged that the valve core movement is faulty. If the two are not synchronized but the diesel engine is operating normally and follows the throttle adjustment, it is judged that the valve core drive shaft is broken.

[0011] Furthermore, in S3, the specific steps for handling closed-loop anomalies are as follows: Anomaly localization: Infer the fault domain based on alarm signals, and alternately use component replacement and system isolation methods to pinpoint the faulty physical node; Decision-making: Based on the test progress and the level of abnormality, generate decision instructions to continue operation, stop at an opportune time, or stop immediately; Repair Implementation: Depending on the characteristics of the faulty node, the following methods may be selected for handling: overall replacement, dimensional and performance repair, or control parameter adjustment. Verification and Archiving: After handling, static calibration and dynamic load verification are performed. Once it is confirmed that the test outline indicators have been restored, a closed-loop archive including the cause of the failure, the location path, and the handling plan is generated.

[0012] Furthermore, in step S1, the data storage and data display are configured as follows: Based on the storage module, a two-dimensional array structure is used to store data. The first row of the array is the name and unit of each monitoring channel, and the first column is the timestamp. A row of real-time collected data is written at 100ms intervals. Based on the display module, it supports digital display mode and curve display mode. The digital display mode displays the 1-second accumulated average value at a frequency of 1Hz, and the curve display mode displays the 100ms accumulated average value at a frequency of 10Hz.

[0013] Secondly, based on the same concept, the present invention also provides an online monitoring and anomaly detection system for a turbine-type hydraulic dynamometer, used to implement the method, including: The sensing and acquisition unit is configured to perform step S1 as described in claim 1, for real-time acquisition of multi-channel operating status signals and differential analog-to-digital conversion and storage; An edge discrimination unit is electrically connected to the sensing and acquisition unit and is configured to execute step S2 as described in claim 1, for synchronously running threshold quantitative discrimination, trend qualitative discrimination and model logic discrimination programs and outputting alarm signals; The control and handling unit is communicatively connected to the edge discrimination unit and is configured to execute the S3 step of claim 1, which is used to respond to the alarm signal to perform fault location, hierarchical decision-making, maintenance implementation and static and dynamic verification, and generate a closed-loop handling file.

[0014] Compared with existing technologies, the online monitoring, anomaly detection, and handling method for impeller-type hydraulic dynamometers described in this invention has the following advantages: (1) In response to common abnormality types of impeller-type hydraulic dynamometers, 12 online monitoring points were designed to comprehensively and extensively monitor the abnormality types.

[0015] (2) A three-tiered progressive anomaly detection method system based on threshold, trend, and model was constructed, forming a complete anomaly detection capability from simple to systematic, effectively improving the accuracy of anomaly detection. Among them, the threshold-based detection method is suitable for single-variable anomaly scenarios; the trend-based anomaly detection method can identify degradation trends in a timely manner by slope changes when vibration or temperature is far from reaching the threshold, realizing early warning and avoiding sudden damage; the model-based anomaly detection method can distinguish normal fluctuations caused by changes in operating conditions, avoid frequent invalid alarms, and can also capture composite anomalies that are difficult to detect based on the threshold method.

[0016] (3) For gradual mechanical anomalies, based on the rate of change of the condition monitoring parameters and in combination with the test progress and the anomaly level, the anomaly can be dealt with by stopping the machine at the appropriate time or immediately, which can effectively reduce the number of unplanned shutdowns caused by dynamometer anomalies during the test and initially realize the transformation of the dynamometer operation and maintenance mode to a preventive or proactive approach.

[0017] (4) According to statistics from practical applications over the past two years, after adopting this method, the proportion of dynamometer malfunction handling time in the total downtime has decreased to below 3.0% during the test. Attached Figure Description

[0018] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the measuring point arrangement according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the overall architecture as described in an embodiment of the present invention; Figure 3 This is a schematic diagram comparing the model-based endothermic power and mechanical power according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the frequency domain characteristics of the radial vibration signal of the impeller-type dynamometer rotor assembly according to an embodiment of the present invention.

[0019] Explanation of reference numerals in the attached figures: 1. Volumetric flow meter; 2. Inlet water T-type temperature sensor; 3. Inlet water relative pressure sensor; 4. Free-end rotor bearing outer ring thermal resistance sensor; 5. Dynamometer water chamber relative pressure sensor; 6. Coupling-end rotor bearing outer ring thermal resistance sensor; 7. Stepper motor operating current sensor; 8. Free-end rotor bearing base uniaxial vibration sensor; 9. Outlet water T-type temperature sensor; 10. Coupling-end rotor bearing base uniaxial vibration sensor; 11. Stepper motor encoder; 12. Valve core connecting shaft free-end encoder. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0021] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0022] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0023] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0024] like Figures 1 to 4 As shown, this describes a method for online monitoring, anomaly detection, and handling of a turbine-type hydraulic dynamometer. It includes the following steps: S1. Acquire multi-channel operating status signals: Real-time acquisition of hydraulic, thermal, mechanical and control signals through a sensor array deployed on the hydraulic dynamometer, and analog-to-digital conversion, data storage, data output and data display using a data acquisition instrument that matches differentiated sampling modules according to the steady-state or transient characteristics of the signals; S2. Perform hierarchical anomaly discrimination: Input the collected operating status signal into the discrimination module of the data acquisition instrument, and simultaneously perform threshold quantitative discrimination, trend qualitative discrimination and model logic discrimination, and output the anomaly type and alarm signal; S3. Closed-loop anomaly handling: Based on the alarm signal, perform fault location, hierarchical decision-making, targeted maintenance and static and dynamic verification, and archive the data of the whole process.

[0025] like Figure 1 As shown, the sensor array includes 12 monitoring channels, specifically including: a volumetric flow meter 1, an inlet water T-type temperature sensor 2, an inlet water relative pressure sensor 3, a free-end rotor bearing outer ring thermal resistance sensor 4, a dynamometer water chamber relative pressure sensor 5, a stepper motor operating current sensor 7, a free-end rotor bearing base uniaxial vibration sensor 8, an outlet water T-type temperature sensor 9, a coupling end rotor bearing base uniaxial vibration sensor 10, a stepper motor encoder 11, a valve core connecting shaft free-end encoder 12, and a coupling end rotor bearing outer ring thermal resistance sensor 6. The measurement types include flow rate, temperature, pressure, angular velocity, and current.

[0026] Specifically, regarding the monitoring positions of each sensor, the sensor measuring points of the volumetric flow meter 1, the inlet water T-type temperature sensor 2, the inlet water relative pressure sensor 3, the stepper motor operating current sensor 7, the outlet water T-type temperature sensor 9, and the stepper motor encoder 11 can be arranged according to the existing instruction manual requirements; the free end rotor bearing outer ring thermal resistance sensor 4 and the coupling end rotor bearing outer ring thermal resistance sensor 6 have spring preloads added to their tail ends to ensure that the sensor is in close contact with the bearing outer ring end face when the dynamometer vibrates; the dynamometer water cavity relative pressure sensor 5 is installed at the symmetrical center of the top of the hydraulic dynamometer to prevent the measurement results from being biased towards the front or rear end; the free end rotor bearing base single-axis vibration sensor 8 and the coupling end rotor bearing base single-axis vibration sensor 10 are installed at the upper end of the axial center of the bearing, measuring only the diametrical force and not the axial force; the valve core connecting shaft free end encoder 12 is installed at the free end of the valve core drive shaft, which facilitates the two encoders to jointly determine the valve core position and whether the drive shaft is broken.

[0027] The variables measured by the volumetric flow meter 1, inlet water T-type temperature sensor 2, inlet water relative pressure sensor 3, free-end rotor bearing outer ring thermal resistance sensor 4, coupling-end rotor bearing outer ring thermal resistance sensor 6, outlet water T-type temperature sensor 9, stepper motor encoder 11, and valve core connecting shaft free-end encoder 12 are steady-state variables, i.e., conventional measurements. Their measurement accuracy and corresponding frequency can be referenced from existing standards. The variables measured by the dynamometer water chamber relative pressure sensor 5, stepper motor operating current sensor 7, free-end rotor bearing base uniaxial vibration sensor 8, and coupling-end rotor bearing base uniaxial vibration sensor 10 are transient variables. If the maximum speed of the impeller-type hydraulic dynamometer is... The number of stators is The number of stator impellers is The number of rotor impellers is Then its reference frequency is According to Nyquist's law, the theoretical frequencies of the dynamometer water cavity relative pressure sensor 5, the stepper motor operating current sensor 7, the free-end rotor bearing base uniaxial vibration sensor 8, and the coupling end rotor bearing base uniaxial vibration sensor 10 are: In engineering, it is generally taken The measurement accuracy can be referenced from existing standards.

[0028] like Figure 2As shown, the data acquisition instrument consists of a thermocouple signal acquisition module, a resistance temperature detector (RTD) signal acquisition module, a voltage and current signal acquisition module, a high-speed pulse acquisition module, an alarm module, a display module, a storage module, and an anomaly detection method implantation module. The specific configuration of the differentiated sampling module is as follows: Thermocouple signal acquisition module is connected to the inlet water T-type temperature sensor 2 and the outlet water T-type temperature sensor 9, with a sampling frequency ≥10Hz and a resolution ≥14 bits; Resistance temperature detector (RTD) signal acquisition module is connected to the free end rotor bearing outer ring RTD sensor 4 and the coupling end rotor bearing outer ring RTD sensor 6, with a sampling frequency ≥100Hz and a resolution ≥14 bits; Voltage and current signal acquisition module is connected to the inlet water relative pressure sensor 3, the dynamometer water cavity relative pressure sensor 5, the stepper motor working current sensor 7, the free end rotor bearing base single-axis vibration sensor 8, and the coupling end rotor bearing base single-axis vibration sensor 10, with a sampling frequency ≥10f and a resolution ≥14 bits, where f is the equipment reference frequency; High-speed pulse acquisition module is connected to the volumetric flow meter 1, the stepper motor encoder 11, and the valve core connecting shaft free end encoder 12, with a sampling frequency ≥Max(10f, 4MHz) and a resolution ≥16 bits.

[0029] Based on the alarm module, the data acquisition instrument uses a switch signal to control the buzzer to work or stop. When the switch signal is 1, the buzzer works; when the switch signal is 0, the buzzer stops. Based on the storage module, a two-dimensional array structure is used to store data. The first row of the array contains the name and unit of each monitoring channel, and the first column contains the timestamp. Real-time acquired data is written in rows at 100ms intervals. Based on the display module, it supports digital display mode and curve display mode. The digital display mode displays the 1-second accumulated average value at a frequency of 1Hz, and the curve display mode displays the 100ms accumulated average value at a frequency of 10Hz. Based on the anomaly detection method implantation module, the threshold quantitative detection method, trend qualitative detection method, and model detection method based on state parameters are translated into computer language and run in real time.

[0030] The specific logic of the hierarchical anomaly discrimination is as follows: the threshold quantitative discrimination, based on the state parameter threshold quantitative discrimination method, is used for single variable monitoring. When the measured value of the parameter exceeds the preset operating threshold, an anomaly is triggered; the trend qualitative discrimination, based on the state parameter trend analysis qualitative discrimination method, is used for gradual process monitoring. By calculating the slope of the parameter time series change, a slope lower than the decay threshold is judged as performance degradation, and a slope higher than the mutation threshold is judged as sudden anomaly; the model logic discrimination, based on the state parameter model discrimination method, is used for composite parameter verification. A calculation model is established based on the physical constraints of the equipment. When the deviation between the measured value and the calculated value of the model exceeds the tolerance range, an anomaly is judged.

[0031] The threshold-based quantitative discrimination method based on state parameters specifically includes: setting thresholds for the monitoring parameters of the equipment; when the equipment operates within the threshold range, it is considered to be operating well; exceeding the operating threshold indicates an abnormality. This discrimination method is applied to determine whether an abnormality exists in the dynamic imbalance problem of the impeller-type hydraulic dynamometer rotor assembly. Dynamic imbalance in the impeller-type hydraulic dynamometer rotor assembly refers to a mass eccentricity in the rotor assembly, which generates periodic centrifugal force during rotation. This centrifugal force is a sinusoidal or cosine excitation force synchronized with the rotational speed, leading to periodic vibration of the dynamometer. The vibration signal of the rotor assembly is monitored using a single-axis vibration sensor 8 on the free end rotor bearing base and a single-axis vibration sensor 10 on the coupling end rotor bearing base. When the vibration signal waveform is approximately a standard sine wave, the vibration amplitude of the time-domain signal increases rapidly with the increase of the rotor assembly speed, and the amplitude of the frequency-domain signal at 1 times the rotational frequency is more than 70% greater than that of other rotational frequencies (2nd and 3rd harmonics). Furthermore, based on the normal bearing temperature measured by the thermal resistance sensor 4 on the outer ring of the free end rotor bearing and the thermal resistance sensor 6 on the outer ring of the coupling end rotor bearing, it is determined that the rotor assembly is dynamically unbalanced. If the vibration amplitude of the time-domain signal is greater than the design specification, or the radial runout value of the spindle during static inspection is greater than the design specification, the rotor assembly needs to be repaired or replaced.

[0032] The trend-based qualitative judgment method based on state parameters specifically includes: sorting the monitored parameters (such as temperature, pressure, flow rate, and current) by time series, observing and analyzing trends, and focusing on changes in their slope. When the slope changes slowly, it indicates that the equipment is experiencing performance degradation; when the slope increases dramatically, it indicates that the equipment is malfunctioning. This judgment method is used to determine whether there is an abnormality in the particulate pitting corrosion problem of the impeller rotor of an impeller-type hydraulic dynamometer. First, it monitors whether the water pressure of the dynamometer water chamber pump, measured by the relative pressure sensor 5, decreases, and whether the required cooling water flow rate, measured by the volumetric flow meter 1, increases abnormally, to achieve the same braking effect. It also statistically analyzes whether the maximum torque measured by the dynamometer decreases under the same inlet water pressure, speed, and outlet valve opening conditions. These are comprehensive manifestations of the reduced rotor and stator working efficiency due to cavitation wear. Second, it observes whether the stability of constant speed control deteriorates and whether torque fluctuations increase. This discrimination method uses qualitative analysis of mechanical performance change trends to preliminarily determine the degradation trend of rotor-stator interaction. Quantitative analysis requires disassembling and testing the dynamometer for evaluation. This discrimination method is used to diagnose abnormal water outlet valve movements. Monitoring with the stepper motor's operating current sensor 7 reveals that under the same inlet water pressure, temperature, and operating conditions, a slow increase in current indicates an increase in valve core movement resistance torque, such as due to scaling causing jamming.

[0033] The model discrimination method based on state parameters specifically includes: establishing a physical or mathematical model (such as kinetic equations, thermodynamic models, and flow characteristic curves) between multiple parameters; comparing real-time data with model predictions; the difference indicates degradation; a small difference indicates performance degradation, while a large difference indicates equipment malfunction. This discrimination method is applied to the power measurement anomaly problem of impeller-type hydraulic dynamometers, performing power measurement model discrimination and comparing mechanical power... Heat absorption power of water The difference is considered as follows: a smaller difference indicates equipment performance degradation, while a larger difference indicates abnormal power measurement. Specifically, the rationality of speed and torque measurements is verified. Under steady-state conditions, based on the principle of energy conservation, the expression for the relationship between the mechanical energy generated by the diesel engine and the heat energy absorbed by the water is obtained (Equation 1). Considering the zero-point drift and error of the temperature sensor and flow meter measurements, when... Approximately equal to When the value is true, the power measurement is valid; otherwise, the power measurement is abnormal. The expression is as follows: (1); In the formula: The dynamometer measures the rotational speed in r / min, which is obtained by the speed sensor built into the dynamometer. The dynamometer measures torque, with the unit being N·m, and is obtained by the torque sensor built into the dynamometer. It is the water flow rate, measured in m³ / h, and obtained by volumetric flow meter 1. It is the temperature difference between the inlet and outlet water, in °C, obtained by subtracting the temperature difference between the inlet and outlet water temperature sensors from the outlet water temperature sensor 9. This refers to the mechanical power of a diesel engine, measured in kW. The heat absorption power of water is measured in kJ / s. To address the issue of abnormal water outlet valve operation, this method is used to determine the valve system synchronization model. The position feedback signals of the stepper motor encoder 11 and the encoder 12 at the free end of the valve core connecting shaft are compared in real time. If the two are not synchronized and the diesel engine cannot follow the throttle adjustment, the valve core movement is deemed to have failed. If the two are not synchronized but the diesel engine operates normally and follows the throttle adjustment, the valve core drive shaft is deemed to be broken.

[0034] The specific steps for handling closed-loop anomalies are as follows: (1) Anomaly localization: Infer the fault domain based on the alarm signal, and alternately use the component replacement method and the system isolation method to locate the physical node of the fault. Specifically, infer possible causes (electrical, mechanical, control, etc.) based on the phenomenon, and start the investigation from the simplest and most likely causes, such as whether all working conditions are met. Use the component replacement method and the system isolation method to gradually approach the cause of the anomaly. The component replacement method refers to replacing the suspected component with normal spare parts or modules, and the system isolation method refers to isolating the system in sections and zones to narrow down the scope of the anomaly.

[0035] (2) Handling decision: Based on the test progress and the level of abnormality, generate decision instructions to continue operation, stop at an opportune time, or stop immediately.

[0036] (3) Repair Implementation: Based on the characteristics of the fault node, the following methods are selected for handling: overall replacement, dimensional and performance repair, or control parameter adjustment. Specifically, overall replacement refers to disassembling a damaged component as a whole and replacing it with a new or repaired part that meets quality standards; dimensional and performance repair refers to restoring the size, shape, and performance of the damaged component through technical processes, such as machining repair or welding repair; control parameter adjustment refers to restoring performance and accuracy through adjustable links in the equipment, such as valve opening, control voltage, and control parameters, i.e., without replacing or repairing parts.

[0037] (4) Verification and archiving: After handling, static calibration and dynamic load verification are performed. After confirming that the test outline indicators have been restored, a closed-loop archive including the cause of the failure, the location path and the handling plan is generated.

[0038] An online monitoring and anomaly detection system for a turbine-type hydraulic dynamometer is provided to implement the online monitoring, anomaly detection, and handling method for the turbine-type hydraulic dynamometer, comprising: The sensing and acquisition unit is configured to execute step S1, which is used to acquire multi-channel operating status signals in real time and perform differentiated analog-to-digital conversion and storage; the edge discrimination unit is electrically connected to the sensing and acquisition unit and configured to execute step S2, which is used to synchronously run threshold quantitative discrimination, trend qualitative discrimination and model logic discrimination programs and output alarm signals; the control and handling unit is communicatively connected to the edge discrimination unit and configured to execute step S3, which is used to respond to the alarm signal to perform fault location, hierarchical decision-making, maintenance implementation and static and dynamic verification, and generate closed-loop handling files.

[0039] The beneficial effects of this invention are: (1) In response to common abnormality types of impeller-type hydraulic dynamometers, 12 online monitoring points were designed to comprehensively and extensively monitor the abnormality types.

[0040] (2) A three-tiered progressive anomaly detection method system based on threshold, trend, and model was constructed, forming a complete anomaly detection capability from simple to systematic, effectively improving the accuracy of anomaly detection. Among them, the threshold-based detection method is suitable for single-variable anomaly scenarios; the trend-based anomaly detection method can identify degradation trends in a timely manner by slope changes when vibration or temperature is far from reaching the threshold, realizing early warning and avoiding sudden damage; the model-based anomaly detection method can distinguish normal fluctuations caused by changes in operating conditions, avoid frequent invalid alarms, and can also capture composite anomalies that are difficult to detect based on the threshold method.

[0041] (3) For gradual mechanical anomalies, based on the rate of change of the condition monitoring parameters and in combination with the test progress and the anomaly level, the anomaly can be dealt with by stopping the machine at the appropriate time or immediately, which can effectively reduce the number of unplanned shutdowns caused by dynamometer anomalies during the test and initially realize the transformation of the dynamometer operation and maintenance mode to a preventive or proactive approach.

[0042] (4) According to statistics from practical applications over the past two years, after adopting this method, the proportion of dynamometer malfunction handling time in the total downtime has decreased to below 3.0% during the test.

[0043] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for online monitoring, anomaly identification, and handling of impeller-type hydraulic dynamometers, characterized in that, Includes the following steps: S1. Acquire multi-channel operating status signals: Real-time acquisition of hydraulic, thermal, mechanical and control signals through a sensor array deployed on the hydraulic dynamometer, and analog-to-digital conversion, data storage, data output and data display using a data acquisition instrument that matches differentiated sampling modules according to the steady-state or transient characteristics of the signals; S2. Perform hierarchical anomaly discrimination: Input the collected operating status signal into the discrimination module of the data acquisition instrument, and simultaneously perform threshold quantitative discrimination, trend qualitative discrimination and model logic discrimination, and output the anomaly type and alarm signal; S3. Closed-loop anomaly handling: Based on the alarm signal, perform fault location, hierarchical decision-making, targeted maintenance and static and dynamic verification, and archive the data of the whole process.

2. The method according to claim 1, characterized in that, In S1, the sensor array includes 12 monitoring channels, specifically including: a volumetric flow meter (1), an inlet water T-type temperature sensor (2), an inlet water relative pressure sensor (3), a free-end rotor bearing outer ring thermal resistance sensor (4), a dynamometer water chamber relative pressure sensor (5), a stepper motor working current sensor (7), a free-end rotor bearing base uniaxial vibration sensor (8), an outlet water T-type temperature sensor (9), a coupling end rotor bearing base uniaxial vibration sensor (10), a stepper motor encoder (11), a valve core connecting shaft free-end encoder (12), and a coupling end rotor bearing outer ring thermal resistance sensor (6).

3. The method according to claim 2, characterized in that, In S1, the specific configuration of the differential sampling module is as follows: The thermocouple signal acquisition module connects the inlet water type T temperature sensor (2) and the outlet water type T temperature sensor (9), with a sampling frequency ≥10Hz and a resolution ≥14 bits; The thermal resistance signal acquisition module connects the thermal resistance sensor (4) of the outer ring of the free end rotor bearing to the thermal resistance sensor (6) of the outer ring of the coupling end rotor bearing, with a sampling frequency ≥100Hz and a resolution ≥14 bits; The voltage and current signal acquisition module is connected to the water inlet relative pressure sensor (3), the dynamometer water chamber relative pressure sensor (5), the stepper motor working current sensor (7), the free end rotor bearing base single-axis vibration sensor (8) and the coupling end rotor bearing base single-axis vibration sensor (10), with a sampling frequency ≥10f and a resolution ≥14 bits, where f is the equipment reference frequency; The high-speed pulse acquisition module is connected to the volumetric flow meter (1), the stepper motor encoder (11) and the encoder at the free end of the valve core connecting shaft (12), with a sampling frequency ≥ Max(10f, 4MHz) and a resolution ≥ 16 bits.

4. The method according to claim 1, characterized in that, In S2, the specific logic for the hierarchical anomaly detection is as follows: The threshold quantitative judgment is used for single variable monitoring, and an anomaly is triggered when the measured value of the parameter exceeds the preset operating threshold. The trend qualitative discrimination is used for gradual process monitoring. By calculating the slope of the parameter time series change, a slope below the decay threshold is judged as performance degradation, and a slope above the mutation threshold is judged as sudden anomaly. The model logic judgment is used for composite parameter verification. A calculation model is established based on the physical constraints of the equipment. When the deviation between the measured value and the calculated value of the model exceeds the tolerance range, it is judged as abnormal.

5. The method according to claim 4, characterized in that, The model logic judgment specifically includes: Power measurement model discrimination: Constructing an expression based on the law of conservation of energy In the formula, It is a dynamometer that measures rotational speed; It is a dynamometer that measures torque; It is the water flow rate; It is the temperature difference between the inlet and outlet water; It is the mechanical power of the diesel engine; The heat absorption power of water; in contrast to mechanical power. Heat absorption power with water The difference is considered as follows: a smaller difference indicates equipment performance degradation, and a larger difference indicates power measurement abnormality. Valve system synchronization model judgment: The position feedback signals of the stepper motor encoder (11) and the free end encoder (12) of the valve core connecting shaft are compared in real time. If the two are not synchronized and the diesel engine cannot follow the throttle adjustment, it is judged that the valve core movement is failed. If the two are not synchronized but the diesel engine is in normal condition and follows the throttle adjustment, it is judged that the valve core drive shaft is broken.

6. The method according to claim 1, characterized in that, In S3, the specific steps for handling closed-loop anomalies are as follows: Anomaly localization: Infer the fault domain based on alarm signals, and alternately use component replacement and system isolation methods to pinpoint the faulty physical node; Decision-making: Based on the test progress and the level of abnormality, generate decision instructions to continue operation, stop at an opportune time, or stop immediately; Repair Implementation: Depending on the characteristics of the faulty node, the following methods may be selected for handling: overall replacement, dimensional and performance repair, or control parameter adjustment. Verification and Archiving: After handling, static calibration and dynamic load verification are performed. Once it is confirmed that the test outline indicators have been restored, a closed-loop archive including the cause of the failure, the location path, and the handling plan is generated.

7. The method according to claim 1, characterized in that, In step S1, the data storage and data display are configured as follows: Based on the storage module, a two-dimensional array structure is used to store data. The first row of the array is the name and unit of each monitoring channel, and the first column is the timestamp. A row of real-time collected data is written at 100ms intervals. Based on the display module, it supports digital display mode and curve display mode. The digital display mode displays the 1-second accumulated average value at a frequency of 1Hz, and the curve display mode displays the 100ms accumulated average value at a frequency of 10Hz.

8. An online monitoring and anomaly detection system for a turbine-type hydraulic dynamometer, used to implement the method described in any one of claims 1 to 7, characterized in that, include: The sensing and acquisition unit is configured to perform step S1 as described in claim 1, for real-time acquisition of multi-channel operating status signals and differential analog-to-digital conversion and storage; An edge discrimination unit is electrically connected to the sensing and acquisition unit and is configured to execute step S2 as described in claim 1, for synchronously running threshold quantitative discrimination, trend qualitative discrimination and model logic discrimination programs and outputting alarm signals; The control and handling unit is communicatively connected to the edge discrimination unit and is configured to execute the S3 step of claim 1, which is used to respond to the alarm signal to perform fault location, hierarchical decision-making, maintenance implementation and static and dynamic verification, and generate a closed-loop handling file.

Images