Method for testing efficiency of high-capacity high-water-head impulse turbine based on thermodynamic method and ultrasonic method

Abstract

The invention provides a high-capacity high-water-head impulse turbine efficiency testing method based on thermodynamics and an ultrasonic method. Multi-mode sensing is coordinated while water head decomposition is dynamically carried out in real time. In order to solve the problem of pressure pulsation interference under the high water head working condition, a dynamic water head decomposition compensation algorithm is designed, and ultrasonic waves are adopted to correct the jet angle in real time. A jet flow core area velocity gradient tensor is obtained through an ultrasonic array, and an impact jet flow anisotropy entropy yield model is constructed by combining local temperature rise data detected by a thermodynamic method. When the abnormal working condition is processed, the test results of the macro scale and the micro scale are dynamically fused, so that the test robustness is improved. A pulse-type alternate sampling protocol is also designed, thermodynamic and ultrasonic channels are synchronously activated at key working condition points, closed-loop testing is realized through real-time data mutual verification, and the testing precision under a transition working condition is ensured.

Description

Technical Field

[0001] The present invention belongs to the field of water turbine testing, and more particularly relates to a method for testing the efficiency of a large-capacity and high-head impulse water turbine based on a thermodynamic method and an ultrasonic method. Background Art

[0002] In the existing field of hydraulics research, particularly in the study of turbine efficiency and performance, several methods exist for testing and calculating efficiency. These traditional efficiency testing methods typically include hydraulics, thermodynamics, and water level methods. However, these existing testing methods have encountered difficulties and challenges in their application.

[0003] First, traditional hydraulic methods for measuring turbine efficiency typically perform calculations based on the Bernoulli equation and then measure the inlet and outlet flow rates using either the overflow method or a flow meter. However, these methods are highly dependent on the measurement accuracy of parameters, require complex calibration processes, and are subject to significant uncertainty in actual operation. For example, effects such as vortex flow, bubbles, solid particles, and suspended particles can affect the measurement results.

[0004] Secondly, the thermodynamic method of testing efficiency relies on accurately measuring the difference between the inlet and outlet water temperatures and converting it into power output. While this method allows for a rough efficiency estimate without precise flow measurement equipment, factors such as the nonlinear relationship between density and temperature, the concentration of air in the water, and the heat pump effect can all lead to measurement errors.

[0005] Finally, the water level method relies on accurately measuring the difference between the inlet and outlet water levels to calculate momentum changes. While this is an intuitive method, it requires the installation of an accurate water level gauge and requires extremely stable test equipment and environment to achieve an accurate efficiency assessment.

[0006] To address these issues, researchers are currently attempting to introduce new technologies and approaches into turbine efficiency testing to address the limitations of these methods, such as those based on ultrasound, MEMS sensors, and deep learning. However, while these technologies have shown some improvements, challenges remain in measurement accuracy, stability, and transient response. Furthermore, there is an urgent need to develop new integrated testing solutions to optimize and integrate these technologies, fully leveraging their respective strengths to achieve high-precision and high-stability turbine efficiency measurements. Summary of the Invention

[0007] The present invention aims to solve the problems of low accuracy, poor robustness, and high dependence on parameter measurement that may exist in the traditional impulse turbine efficiency test method in actual operation, especially the pressure pulsation interference problem under large-capacity and high-head conditions and the deficiencies in the analysis of turbulent dissipation and boundary layer stripping losses. At the same time, it also solves the temperature rise measurement deviation caused by cavitation collapse, improves the test accuracy and real-time performance, and ensures the test accuracy under transition conditions.

[0008] To achieve the above objectives, the present invention adopts the following technical solutions: a co-packaged MEMS pressure sensor and a Pt1000 temperature sensor are embedded in a high-pressure water diversion pipe section at a distance of 7 pipe diameters from the nozzle inlet. The pressure sensor has a range of 0-50 MPa and an accuracy of ±0.05% FS, and the temperature sensor has an accuracy of ±0.1K. An 8-channel anti-cavitation ultrasonic array with a center frequency of 5 MHz and a beam angle of ±15° is arranged in a stagnation point region 0.3 times the jet diameter downstream of the nozzle outlet to form a thermodynamic-ultrasonic spatial coupling measurement network. The method comprises the following steps:

[0009] (1) Based on MEMS pressure, Pt1000 temperature and electromagnetic flowmeter data, the improved Bernoulli equation is used to calculate the working water head, and the water density, kinetic energy correction coefficient and friction loss along the flow are calibrated according to the real-time water temperature;

[0010] (2) The instantaneous velocity vector distribution in the core area of ​​the nozzle jet is synchronously measured by an ultrasonic array, the jet impact angle is extracted, and the reference head calculation is corrected based on the angle feedback;

[0011] (3) A dynamic head decomposition and compensation algorithm is used to decompose the working head into static pressure head, kinetic energy head, and potential energy head components in real time. The jet flow pressure conversion coefficient is calculated in combination with the dynamic pressure measured by ultrasonic wave, thereby compensating for the pressure pulsation interference and achieving high-precision dynamic decoupling of the head.

[0012] (4) When the pressure fluctuation exceeds the set threshold, the high-frequency sampling and prediction module based on the LSTM model is activated to achieve real-time response to extreme working conditions such as water hammer, thereby improving the accuracy and robustness of the efficiency test.

[0013] In one embodiment, the dynamic head decomposition and compensation algorithm includes using the jet impact angle measured by the ultrasonic array to decouple the static head component, kinetic head component and potential head component in real time, and optimizing the kinetic head component model coefficient through a gradient descent algorithm.

[0014] In one embodiment, the center frequency of each channel of the 8-channel anti-cavitation ultrasonic array is 5 MHz, the dynamic pressure obtained by time averaging is used to calculate the static pressure head component, and the calibration range of the jet pressure conversion coefficient is 1.05 to 1.12.

[0015] In one solution, the pressure sensor and temperature sensor are co-packaged and embedded in the high-pressure water diversion pipe section at a position 7 times the pipe diameter away from the nozzle inlet to avoid the turbulent development zone and ensure measurement accuracy and stability.

[0016] In one solution, the baseline value of the working water head is dynamically corrected in real time based on the collected water temperature, total pressure and flow rate in the water pipe, combined with the kinetic energy correction coefficient (obtained by looking up the Reynolds number table) and the friction loss along the way (calculated based on the Colebrook equation).

[0017] In one embodiment, the method further includes automatically activating the high-frequency sampling mode 50ms in advance when the pressure fluctuation exceeds the design value by 15%, and using the LSTM predictor to perform high-precision early response to working conditions such as water hammer.

[0018] In one embodiment, the method uses the water pipe flow velocity data obtained by the electromagnetic flowmeter to participate in the calculation of the kinetic head component, thereby further improving the accuracy of flow velocity measurement and efficiency testing.

[0019] In one solution, the data of the thermodynamic-ultrasonic spatial coupling measurement network can effectively suppress the pressure pulsation interference under high head conditions and improve the efficiency test accuracy of large-capacity high-pressure impulse turbines through synchronous acquisition and dynamic decomposition.

[0020] Beneficial effects of the present invention:

[0021] This invention combines thermodynamic and ultrasonic methods to create a new method for testing the efficiency of large-capacity, high-head Pelton turbines. This method decomposes the operating head in real time, quantifies turbulent dissipation and boundary layer separation losses, corrects for temperature rise measurement deviations caused by cavitation collapse, and implements closed-loop testing through real-time data interverification.

[0022] The beneficial effects of the present invention are mainly manifested in the following aspects:

[0023] 1. Improved turbine efficiency test accuracy. The dynamic head decomposition and compensation algorithm enables more accurate measurement of operating head. Simultaneously, by analyzing jet velocity distribution and temperature rise data, turbulent dissipation and boundary layer separation losses can be more precisely quantified.

[0024] 2. The test process has been optimized. Through the pulsed alternating sampling protocol, the thermodynamic and ultrasonic channels can be flexibly switched under different working conditions, which can achieve rapid adjustment and response to different test conditions.

[0025] 3. The robustness of the test is improved, especially in dealing with the temperature rise measurement deviation and pressure pulsation interference caused by cavitation collapse. Through the cavitation-efficiency coupling correction model and closed-loop verification method, a new high-precision measurement tool is provided for engineering applications and scientific research experiments.

[0026] 4. Enhanced real-time measurement. By combining with deep learning, the feedback speed of measurement results is significantly improved, meeting the needs of real-time control and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0028] To facilitate understanding of the present invention, the present invention will be described more fully below with reference to the accompanying drawings. The drawings illustrate exemplary embodiments of the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as those understood by those skilled in the art to which the present invention pertains. The terms used in the present specification are for the purpose of describing specific embodiments only and are not intended to limit the present invention. To facilitate understanding of the present invention, a more comprehensive description of the present invention will be provided below with reference to the accompanying drawings. Typical embodiments of the present invention are shown in the drawings. However, the present invention may be embodied in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the present invention.

[0030] like Figure 1 As shown in FIG, a method for testing the efficiency of a large-capacity, high-head impulse turbine based on a thermodynamic method and an ultrasonic method has the following specific steps:

[0031] Step 1: Dynamic head decomposition and multimodal sensing collaboration

[0032] A micro thermodynamic parameter acquisition unit (temperature / pressure composite sensor) is embedded in the high-pressure water diversion pipe section, and an anti-cavitation ultrasonic array is deployed at 0.3 times the jet diameter downstream of the nozzle outlet to form a thermodynamic-ultrasonic spatial coupling measurement network. The innovation lies in:

[0033] A dynamic head decomposition and compensation algorithm is proposed to decompose the working head into static pressure head, kinetic energy head and potential energy head components in real time. The reference head calculation of the thermodynamic method is corrected based on the jet impact angle feedback measured by ultrasonic measurement to solve the pressure pulsation interference problem under high head conditions.

[0034] In order to achieve high-precision dynamic decomposition of water head and coordination of multi-modal sensing, a co-packaged MEMS pressure sensor (range 0-50MPa, accuracy ±0.05%FS) and a Pt1000 temperature sensor (accuracy ±0.1K) are embedded in the high-pressure water diversion pipe section at a distance of 7 times the pipe diameter from the nozzle inlet (avoiding the turbulent development zone). At the same time, an 8-channel anti-cavitation ultrasonic array (center frequency 5MHz, sound beam angle ±15°) is deployed in the stagnation point area 0.3 times the jet diameter (usually 30-80mm) downstream of the nozzle outlet to form a thermodynamic-ultrasonic spatial coupling measurement network. Its core innovation lies in the dynamic water head decomposition and compensation algorithm: first, based on the real-time acquisition of the total pipeline pressure P total , water temperature T and altitude Z, calculate the reference water head according to the improved Bernoulli equation:

[0035]

[0036] Where ρ is the density calibrated in real time according to the water temperature, g is the acceleration of gravity, α is the kinetic energy correction coefficient (obtained by looking up the Reynolds number table), V is the flow velocity of the water pipe (provided by the electromagnetic flowmeter), and h is the water flow rate. f is the friction loss along the jet (calculated based on the Colebrook equation). At this time, the ultrasonic array synchronously measures the instantaneous velocity vector distribution u in the core area of ​​the jet. i (x, y, t), the jet impact angle θ (defined as the angle between the jet axis and the runner pitch circle) is obtained through feature extraction:

[0037]

[0038] The three physical components of the working head are decoupled in real time based on this angle:

[0039] 1. Static pressure head component: H s =K p ΔP ultra / (ρg), where ΔP ultra is the dynamic pressure measured by the ultrasonic array in the stagnation zone (based on time averaging), K p is the jet flow pressure conversion coefficient (calibration range 1.05-1.12);

[0040] 2. Kinetic head weight: Where V jet The jet core velocity is directly measured by ultrasound (by Doppler frequency shift Δf = 2f0V jet cosθ / c inversion), β is the shock loss correction factor;

[0041] 3. Potential head component: H p =Z nozzle -Z ref, obtain the elevation benchmark through the high-precision RTK-GPS positioning system.

[0042] In order to suppress the high head pressure pulsation interference, a two-way feedback mechanism based on wavelet transform is introduced: the original pressure signal P of the thermodynamic method is raw (t) Perform db6 wavelet packet decomposition to the 8th layer to extract the turbulent pulsation characteristic frequency band (usually 0.1-20 Hz):

[0043]

[0044] The pulsating component is consistent with the jet turbulent energy measured by ultrasonic wave. Establish a mapping relationship (empirical relationship: c is the speed of sound), and the final corrected reference head is:

[0045]

[0046] Closed-loop control achieves dynamic tuning through angle θ: When |θ-θ0|>2° (θ0 is the design impact angle), the algorithm is based on ΔH k =k θ ·sin(θ-θ0)·H k The kinetic head component is corrected and the adjustment is fed back to the thermodynamic pressure acquisition terminal. The entire system performs real-time decoupling calculations at a frequency of 200Hz, completing head component reconstruction and cross-validation within 100ms, keeping the measurement error caused by pressure pulsation within ±0.07%.

[0047] Step 2: Entropy production rate 3D reconstruction and multi-scale loss fusion

[0048] The velocity gradient tensor in the core area of ​​the jet is obtained by ultrasonic array. Combined with the local temperature rise data detected by thermodynamic method, an anisotropic entropy production model of the impinging jet is constructed to quantify turbulent dissipation and boundary layer separation loss.

[0049] An innovative deep learning attention mechanism is adopted to dynamically weighted fuse the test results of the macro scale (overall efficiency of the thermodynamic method) and the micro scale (local loss of the ultrasonic method). When the jet splits or the braking condition is triggered, the decision weight of the ultrasonic data is automatically enhanced to improve the robustness under abnormal conditions.

[0050] On the basis of completing the dynamic head decomposition, the three-dimensional reconstruction of the entropy production rate is achieved through the spatial coupling measurement network. First, the ultrasonic array obtains the instantaneous velocity field u in the core area of ​​the jet at a sampling frequency of 10kHz. i (x, t) (where i = x, y, z), based on the velocity gradient tensor The turbulence dissipation function is constructed using the eight-channel synchronous measurement data:

[0051] At the same time, the thermodynamic unit collects the temperature field T(x,t) of the 0.2mm micro area near the runner target plate at a frequency of 1kHz, combined with the heat capacity characteristics C of the turbine material p , derive the local irreversible thermal entropy production rate:

[0052]

[0053] The anisotropic entropy production model is constructed by discretizing the fluid domain into a voxel grid of 0.5 times the jet diameter, and fusing the ultrasound and thermodynamic data at each grid center:

[0054]

[0055] Where λ is the peel strength coefficient (calibrated by PIV to be 0.18-0.35), ΔT b is the wall temperature gradient, The model is calculated from the second derivative of the near-wall velocity profile of the ultrasonic array. The model quantifies the multi-scale losses in the jet impact process: macroscopic turbulence dissipation (dominated in the core region) and microscopic boundary layer separation (dominated in the near-wall region).

[0056] For multi-scale loss fusion, a dual-path attention mechanism based on deep learning is designed. The overall efficiency of the macro-path input thermodynamic method calculation Local loss distribution of ultrasonic wave reconstruction using microscopic path input First, the hidden state is generated through the feature extractor:

[0057] h macro =LSTM([H s ,H k ,H p ,η thermo ])

[0058] h micro =CNN(L ultra )

[0059] Then the adaptive weight distribution strategy is adopted: define the working condition feature vector v = [We, Tu, I brake ](Weber number We=ρV 2 d / σ ​​characterizes jet stability and turbulence intensity Braking condition identifier I brake Dynamically generate fusion weights: α = σ(W·v+b)σ(·): The final efficiency fusion output of the Sigmoid function is: η fused =α·η thermo +(1-α)·η ultra in

[0060] Abnormal operating response is monitored by ultrasonic spectrum entropy: when the jet splits, the Shannon entropy H of the high-frequency pressure pulsation s (f)=-∑f i log2f i (Calculation frequency band 50kHz-1MHz) exceeds the threshold of 3.2 bits, at this time, α is forced to be set to <0.2, so that micro data dominates the decision. Under braking conditions (I brake =1) additionally enables Monte Carlo sampling and injects random perturbations δ~N(0,0.03η) into the ultrasonic data. ultra ) to improve robustness. This fusion mechanism can reduce the standard deviation of efficiency measurement under sudden operating conditions from ±0.8% of the traditional method to ±0.25%.

[0061] Step 3: Cavitation-efficiency coupling correction and closed-loop verification

[0062] The cavitation volume fraction is extracted using ultrasonic echo signals, and a nonlinear mapping function between cavitation loss and thermodynamic efficiency is established to correct the temperature rise measurement deviation caused by cavitation collapse.

[0063] A pulsed alternating sampling protocol is designed to synchronously activate the thermodynamic and ultrasonic channels at key operating points. Closed-loop testing is achieved through real-time data mutual verification. If the difference exceeds the threshold, the high-frequency sampling mode is triggered to ensure test accuracy under transitional conditions (such as sudden changes in guide vane opening).

[0064] After completing the multi-scale loss fusion, the cavitation effect is quantified based on the acoustic characteristics of the ultrasonic array. First, the ultrasonic echo signal with a center frequency of 5MHz is used to extract the distribution characteristics of the cavitation group through broadband dispersion spectrum analysis. The received signal s(t) is subjected to a complex analytical wavelet transform (Morlet wavelet basis):

[0065]

[0066] Calculate the sound energy attenuation coefficient in the characteristic frequency band (3.5-6.5MHz) corresponding to the scale parameter a:

[0067]

[0068] d is the sound path, and W0 is the cavitation-free reference.

[0069] Based on this, the cavitation volume fraction is derived:

[0070] is the average radius of the bubble, with a calibration value of 0.1-0.3mm

[0071] At the same time, the thermodynamic unit detects the abnormal temperature pulse ΔT in the micro area caused by the collapse of the cavitation bubble. cav(t) (time domain width 20-200μs). Establish cavitation-efficiency coupling model:

[0072]

[0073] Where V cr =0.15 is the critical cavitation volume fraction, t d is the collapse time delay (determined by cross-correlation analysis), and τ = 50 μs is the thermal diffusion time constant. The model modifies the thermodynamic efficiency value:

[0074]

[0075] In order to improve the reliability of the correction, a dual-frequency ultrasonic verification strategy is adopted: an additional 1MHz low-frequency channel penetrates the cavitation group and measures the background flow velocity u base , according to the Doppler frequency shift difference Δf H -Δf L Calculate the cavitation phase shift disturbance compensation.

[0076] A pulsed alternating sampling protocol was implemented to achieve closed-loop verification: Under steady-state conditions (guide vane opening rate of change < 1% / s), the thermodynamic channel (1Hz) and the ultrasonic channel (5Hz) were polled in a 2:1 timing sequence. When guide vane movement was detected (opening rate of change > 3% / s) or the data difference exceeded the threshold (δη > 0.5%), the synchronous pulse mode was triggered:

[0077]

[0078] At this point, the thermodynamic sampling rate is increased to 1kHz (with the transient temperature compensation algorithm enabled), and the ultrasonic sampling rate is increased to 100kHz (with full matrix capture mode enabled). The two system clocks are strictly synchronized via the PTP protocol (with an accuracy of ±10μs), and data points are acquired simultaneously at key phase points (such as 15° before the jet impacts the runner):

[0079]

[0080] The real-time verification mechanism establishes a residual detector based on the Navier-Stokes equation:

[0081]

[0082] At measuring point x s (0.1D upstream of the runner) Calculate the physical consistency index:

[0083]

[0084] When J consist >5kPa / ms or (δη takes 0.3%-0.8% dynamic threshold), activate the three-level verification response:

[0085] 1. High-frequency sampling mode: The ultrasonic sampling rate is switched to 200kHz to capture the transient flow field structure, and the thermodynamic unit enables microsecond-level temperature rise recording (ADC sampling rate 50MSPS);

[0086] 2. Cavitation field reconstruction: Invert the three-dimensional distribution of cavitation groups based on ultrasonic full matrix data and dynamically update the Vb mapping function;

[0087] 3. RAW data storage: Synchronously write the original signal stream into non-volatile memory for a duration of T store =2·τ trans , τ trans is the duration of the transition condition).

[0088] Efficiency values ​​that pass verification enter the final output queue, and the model coefficients are optimized using a gradient descent algorithm (k1 and k2 are updated every 10 abnormal triggers). When the water hammer effect is significant (pressure fluctuation >15% of the design value), the LSTM predictor output of step 2 is pre-activated, and high-frequency sampling is started 50ms in advance.

[0089] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0090] It should be understood that the detailed description of the technical solutions of the present invention using the preferred embodiments above is illustrative and not restrictive. A person skilled in the art, after reading the present specification, may modify the technical solutions described in the embodiments or replace some of the technical features therein with equivalents; such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing the efficiency of a large-capacity, high-head impulse turbine based on thermodynamics and ultrasonic methods, characterized by: A co-packaged MEMS pressure sensor and Pt1000 temperature sensor are embedded in the high-pressure water diversion pipe section at a distance of 7 pipe diameters from the nozzle inlet. The pressure sensor has a range of 0-50 MPa and an accuracy of ±0.05% FS, while the temperature sensor has an accuracy of ±0.1 K. An 8-channel anti-cavitation ultrasonic array with a center frequency of 5 MHz and a beam angle of ±15° is placed in the stagnation point area 0.3 times the jet diameter downstream of the nozzle outlet, forming a thermodynamic-ultrasonic spatial coupling measurement network. The method comprises the following steps: (1) Based on MEMS pressure, Pt1000 temperature and electromagnetic flowmeter data, the improved Bernoulli equation is used to calculate the working water head, and the water density, kinetic energy correction coefficient and friction loss along the flow are calibrated according to the real-time water temperature; (2) The instantaneous velocity vector distribution in the core area of ​​the nozzle jet is synchronously measured by an ultrasonic array, the jet impact angle is extracted, and the reference head calculation is corrected based on the angle feedback; (3) A dynamic head decomposition and compensation algorithm is used to decompose the working head into static pressure head, kinetic energy head, and potential energy head components in real time. The jet flow pressure conversion coefficient is calculated in combination with the dynamic pressure measured by ultrasonic wave, thereby compensating for the pressure pulsation interference and achieving high-precision dynamic decoupling of the head. (4) When the pressure fluctuation exceeds the set threshold, the high-frequency sampling and prediction module based on the LSTM model is activated to achieve real-time response to extreme water hammer conditions and improve the accuracy and robustness of the efficiency test.

2. The method according to claim 1, characterized in that The dynamic water head decomposition and compensation algorithm includes using the jet impact angle measured by the ultrasonic array to decouple the static pressure head component, kinetic head component and potential head component in real time, and optimizing the kinetic head component model coefficient through a gradient descent algorithm.

3. The method according to claim 1, characterized in that The center frequency of each channel of the 8-channel anti-cavitation ultrasonic array is 5 MHz. The dynamic pressure obtained by time averaging is used to calculate the static pressure head component. The calibration range of the jet pressure conversion coefficient is 1.05 to 1.

12.

4. The method according to claim 1, wherein The pressure sensor and temperature sensor are co-packaged and embedded in the high-pressure water diversion pipe section at a position 7 times the pipe diameter away from the nozzle inlet to avoid the turbulent development area and ensure the accuracy and stability of the measurement.

5. The method according to claim 1, wherein The reference value of the working water head is dynamically corrected in real time based on the collected water temperature, total pressure and flow rate in the water pipe, combined with the kinetic energy correction coefficient (obtained by looking up the Reynolds number table) and the friction loss along the way (calculated based on the Colebrook equation).

6. The method according to claim 1, characterized in that The method further includes automatically activating a high-frequency sampling mode 50 milliseconds in advance when the pressure fluctuation exceeds a design value by 15%, and utilizing an LSTM predictor to perform a high-precision advance response to a water hammer condition.

7. The method according to claim 1, characterized in that The method uses the water pipe flow velocity data obtained by the electromagnetic flowmeter to calculate the kinetic head component, thereby further improving the accuracy of flow velocity measurement and efficiency testing.

8. The method according to claim 1, characterized in that The data of the thermodynamic-ultrasonic spatial coupling measurement network can effectively suppress pressure pulsation interference in high head conditions and improve the efficiency test accuracy of large-capacity high-pressure impulse turbines through synchronous acquisition and dynamic decomposition.

Images