A method for synchronous matching of sensor response time in UCTD under complex turbulent environment

Abstract

The present invention relates to the field of sensor data processing, and discloses a method for synchronous matching of sensor response time in UCTD under complex turbulent environment, comprising the following steps: using a temperature-controlled water flow tank to generate different flow velocities and turbulence intensities, placing a UCTD probe in the temperature-controlled water flow tank to measure the temperature and conductivity under different flow velocities and turbulence intensities, and recording the response time of the temperature and conductivity sensors in UCTD under different flow velocities and turbulence intensities; establishing a temperature and conductivity sensor response time model by performing polynomial regression analysis on the recorded experimental data; using a gradient descent method to minimize the error between the model prediction value and the actual value, and optimizing the parameters in the model; using the optimized model to establish a synchronous response time model, and synchronously matching the response time difference of the temperature sensor and the conductivity sensor. The method disclosed by the present invention greatly improves the accuracy and reliability of UCTD data in marine environment monitoring.

Description

Technical Field

[0001] The present invention relates to the field of sensor data processing, and in particular to a sensor response time synchronization matching method in UCTD under complex turbulent environment. Background Art

[0002] In marine environmental monitoring, the Underway Conductivity-Temperature-Depth (UCTD) is widely used to obtain temperature-salinity-depth profile data in real time, which is particularly important in the fields of marine science research and climate change. UCTD relies on the rapid descent of the probe to divert seawater in the conductivity sensor, avoiding the complexity of traditional centrifugal pumping devices and providing high-precision vertical profile data of temperature-salinity-depth without wasting time. However, due to their working characteristics, temperature sensors and conductivity sensors are affected by flow velocity fluctuations and turbulence effects during the descent process, resulting in increased response time differences between sensors, making it impossible to synchronously measure the same water mass, and thus resulting in errors in the calculated salinity, etc. This is particularly significant in complex turbulent environments, which seriously restricts the application effect of UCTD in complex marine environments.

[0003] Traditional technologies mainly rely on filtering algorithms and smoothing to reduce the errors caused by the asynchronous response time of temperature and conductivity sensors. The existing MCT (Match conductivity and temperature response times) method and position correction method for synchronous matching of the response time of temperature and conductivity sensors are both synchronous matching methods of sensor response time under the assumption of steady flow. They fail to fully consider the dynamic impact of complex turbulent environment on sensor response time, resulting in poor adaptability under fluctuating flow velocity and turbulence intensity, and difficulty in handling flow conditions with high Reynolds number or high turbulence intensity, resulting in measurement errors and data inconsistency, and unable to reflect the real marine environment. Therefore, it is urgent to achieve synchronous matching of temperature and conductivity sensor response time in complex turbulent environment by more accurately analyzing the impact of turbulence on sensor response time. Summary of the invention

[0004] To solve the above technical problems, the present invention provides a sensor response time synchronization matching method in UCTD under complex turbulent environment, so as to ensure that the response time of temperature sensor and conductivity sensor is as synchronized as possible under complex turbulent environment, thereby improving the accuracy of marine environment monitoring.

[0005] To achieve the above object, the technical solution of the present invention is as follows:

[0006] The sensor response time synchronization matching method in UCTD under complex turbulent environment includes the following steps:

[0007] Step 1, using a temperature-controlled water flow tank to generate different flow velocities and turbulence intensities, recording the Reynolds number, Strouhal number and shear stress under different turbulence intensities, placing a UCTD probe in the temperature-controlled water flow tank to measure the temperature and conductivity under different flow velocities and turbulence intensities, and recording the response time of the temperature sensor and the conductivity sensor in the UCTD under different flow velocities and turbulence intensities;

[0008] Step 2, establishing a temperature sensor response time model and a conductivity sensor response time model by performing polynomial regression analysis on the recorded experimental data;

[0009] Step 3, using the gradient descent method to minimize the error between the model predicted response time and the actual response time, and optimize the parameters in the model;

[0010] Step 4: Use the optimized model to establish a synchronous response time model to synchronously match the response time difference between the temperature sensor and the conductivity sensor.

[0011] In the above scheme, the temperature-controlled water flow tank is equipped with an electric heater, a centrifugal pump and a laser Doppler velocimeter, and fluorescent particles are added to the temperature-controlled water flow tank to mix evenly with the water body, and different flow rates are simulated by changing the speed of the centrifugal pump. , use electric heater to adjust water temperature, and use water of different temperatures to change the viscosity of water In order to control the turbulence intensity, a laser Doppler velocimeter is used to obtain the flow velocity of the water body. and vibration frequency The Reynolds number is calculated by To characterize the turbulent state, <2000 is laminar flow, when 2000 < <4000 is transition flow. >4000 is turbulent.

[0012] In the above scheme, the Reynolds number The calculation formula is as follows:

[0013] ;

[0014] in, is the density of water, is the flow velocity of the water body, is the width of the temperature-controlled water flow channel, is the viscosity of water;

[0015] Strouhal number The calculation formula is:

[0016] ;

[0017] in, is the vibration frequency of the water body;

[0018] Shear stress The calculation formula is:

[0019] ;

[0020] in, is the gradient of flow velocity in the vertical direction, is the vertical coordinate.

[0021] In the above scheme, in step 1, a mobile platform equipped with a winch drum is set up, a rope of a specified length is wound on the UCTD probe by a fast winding machine, the UCTD probe is vertically dropped into the temperature-controlled water flow tank, and then moved along the predetermined path of the temperature-controlled water flow tank, the rope is quickly pulled out from the winch drum and the UCTD probe at the same time, and the UCTD probe performs a vertical rapid descending motion near the dropping point, and the time from the start of the temperature step change to the reaching of 63% of the final value of the temperature sensor in the UCTD under different flow velocities and turbulence intensities, that is, the response time of the temperature sensor, and the time from the start of the conductivity step change to the reaching of 63% of the final value, that is, the response time of the conductivity sensor, are recorded, and the response time curves of the temperature sensor and the conductivity sensor under different flow velocities and turbulence intensities are plotted.

[0022] In the above scheme, in step 2, the established temperature sensor response time model is as follows:

[0023] ;

[0024] in, is the predicted response time of the temperature sensor response time model, is the flow rate, is the Reynolds number, is the Strouhal number, is the shear stress, , , , is the constant to be optimized, , , , is the index to be optimized;

[0025] The established conductivity sensor response time model is as follows:

[0026] ;

[0027] in, is the predicted response time of the conductivity sensor response time model, , , , is the constant to be optimized, , , , is the index to be optimized.

[0028] In the above scheme, the specific method of step 3 is as follows:

[0029] Step 301, define the error function:

[0030] The error function formula corresponding to the temperature sensor response time model is:

[0031] ;

[0032] in, It is The actual response time of the temperature sensor for each data point, It is The predicted response time of the temperature sensor response time model with data points, variables , , , It is flow velocity, Reynolds number, Strouhal number and shear stress for each data point, , , , is the constant to be optimized, , , , is the index to be optimized, is the total number of observed data points, is the index of the data point;

[0033] The error function formula corresponding to the conductivity sensor response time model is:

[0034] ;

[0035] in, It is The actual response time of the conductivity sensor for each data point, It is The predicted response time of the conductivity sensor response time model for each data point, , , , is the constant to be optimized, , , , is the index to be optimized;

[0036] Step 302, using the chain rule to calculate the partial derivative of each parameter, with the parameter For example:

[0037] ;

[0038] Step 303, use the gradient descent method to update the parameters until the error is minimized, and the update rule is as follows:

[0039] ;

[0040] in, It is Parameters in iterations The value of is the learning rate, which is used to control the step size of each update, is the error function corresponding to the temperature sensor response time model with respect to the parameter The partial derivative of

[0041] Step 304, update the parameters by repeated iteration until the error function Convergence, that is, the update amplitude of the parameter is less than the preset threshold ,satisfy:

[0042] ;

[0043] in, Set to 0.0001;

[0044] Step 305: Other parameters , , , , , , , , , , , , , , The calculation method of minimizing the error function by gradient descent is the same as the parameter ; Thus, an optimized temperature sensor response time model and an optimized conductivity sensor response time model are obtained.

[0045] In the above scheme, step 4 is as follows:

[0046] The optimized model is used to establish a synchronous response time model through the differential method of response time difference, and the response time difference of the temperature sensor and the conductivity sensor is synchronously matched. The synchronous response time model is:

[0047] ;

[0048] in, is the synchronous response time of the temperature sensor and the conductivity sensor, is the predicted response time of the temperature sensor response time model, is the predicted response time of the conductivity sensor response time model.

[0049] Through the above technical solution, the sensor response time synchronization matching method in UCTD under complex turbulent environment provided by the present invention has the following beneficial effects:

[0050] 1. The present invention sets different flow rates and different turbulence intensity conditions (Reynolds number , Strouhal number , shear stress ), using UCTD probe to measure the changes in water parameters temperature and conductivity, and compare the obtained response time experimental data with the corresponding flow rate , Reynolds number , Strouhal number , shear stress Through polynomial regression analysis, the response time model of the temperature sensor and conductivity sensor is fitted, and the gradient descent method is used to optimize the parameters to minimize the error. This data optimization method can accurately fit the response time of the sensor in a complex turbulent environment and dynamically adjust the model parameters according to the actual fluid conditions, ensuring the adaptability of UCTD in various turbulent environments;

[0051] 2. The present invention can effectively reduce the response time difference caused by velocity fluctuation and turbulence effect by synchronously matching the response time of the temperature sensor and the conductivity sensor of UCTD under different turbulence intensities, thereby greatly reducing the measurement error. This synchronous matching method can ensure that the temperature and conductivity sensors are measured synchronously under the same water mass conditions, reducing the data deviation of parameters such as salinity caused by the asynchronous response time caused by velocity fluctuation and turbulence effect, greatly improving the accuracy and reliability of UCTD data in marine environment monitoring;

[0052] 3. The synchronous matching method adopted by the present invention is based on model optimization and has good scalability. With the emergence of new marine environment monitoring requirements and the advancement of sensor technology, the method can further improve its accuracy and adaptability by properly adjusting the model and optimization algorithm. Therefore, the present invention is not only applicable to UCTD systems under current technical conditions, but can also continue to play a role in future marine monitoring systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0054] Figure 1 A schematic diagram of a flow chart of a method for synchronous matching of sensor response time in UCTD under a complex turbulent environment disclosed in an embodiment of the present invention;

[0055] Figure 2 Schematic diagram of the structure of the temperature-controlled water flow channel.

[0056] In the figure, 1. temperature-controlled water flow trough; 2. electric heater; 3. centrifugal pump; 4. laser Doppler velocimeter; 5. winch drum; 6. mobile platform; 7. UCTD probe; 8. rope. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0058] The present invention provides a method for synchronously matching sensor response time in UCTD under complex turbulent environment. Figure 1 As shown, the following steps are included:

[0059] Step 1: Use a temperature-controlled water flow tank to generate different flow rates and turbulence intensities, record the Reynolds number, Strouhal number and shear stress under different turbulence intensities, place the UCTD probe in the temperature-controlled water flow tank to measure the temperature and conductivity under different flow rates and turbulence intensities, and record the response time of the temperature sensor and conductivity sensor in the UCTD under different flow rates and turbulence intensities.

[0060] Step 101: An electric heater, a centrifugal pump and a laser Doppler velocimeter (LDV) are installed in a temperature-controlled water flow tank, and fluorescent particles are added to the temperature-controlled water flow tank and mixed with the water. Figure 2 As shown, the electric heater 2 is installed on the side wall of the temperature-controlled water flow tank 1 to heat the water flow to the required temperature, thereby changing the viscosity of the water body and further affecting the turbulence intensity. The laser Doppler velocimeter 4 is installed in the middle of the temperature-controlled water flow tank 1 to obtain the flow velocity by measuring the frequency change of the laser beam reflected on the fluorescent particles in the water body. and vibration frequency The centrifugal pump 3 is installed on the side wall of the temperature-controlled water flow tank 1, and the speed of the centrifugal pump 3 is changed to simulate different flow rates. , by calculating the Reynolds number To characterize the turbulent state, <2000 is laminar flow, when 2000 < <4000 is transition flow. When the flow rate is >4000, it is turbulent, ensuring that the flow state of the water in the temperature-controlled water flow tank covers the range from laminar flow to strong turbulence, ensuring the comprehensiveness and diversity of the experimental data.

[0061] There is usually a positive correlation between turbulence intensity and Reynolds number. When the Reynolds number increases, the turbulence intensity increases, and the vortex and flow disturbance intensify. According to the Reynolds number, the flow state of the water in the temperature-controlled water flow channel can be roughly determined, and then the impact on the response of the sensor can be analyzed.

[0062] The flow rate range should be selected as , Reynolds number The calculation formula is as follows:

[0063] ;

[0064] in, is the density of water (kg / m³), is the flow velocity of the water body (m / s), is the characteristic length of the guide tube (the width of the temperature-controlled water flow channel, m), is the dynamic viscosity of water (Pa·s).

[0065] According to the set flow rate and water characteristics (vibration frequency ), calculate the Strouhal number To describe the vibration frequency caused by turbulence Water velocity The coupling effect of the sensor can be further studied by simulating the periodic changes caused by turbulence. When the turbulence frequency is close to the sampling frequency of the sensor, resonance will occur, resulting in fluctuations in the sensor response time. In a strong turbulent environment, it is necessary to adjust the sampling frequency of the sensor to avoid resonance and achieve better response time matching.

[0066] Strouhal number The calculation formula is:

[0067] ;

[0068] in, is the vibration frequency of the water body (Hz);

[0069] According to the set flow rate and water characteristics (water viscosity ), calculate the shear stress :

[0070] ;

[0071] in, is the gradient of flow velocity in the vertical direction ( ), is the vertical coordinate.

[0072] Shear stress The size and viscosity of the water and flow rate The gradient is closely related, especially under turbulent conditions, the flow velocity gradient fluctuates greatly, resulting in drastic changes in shear stress, which can be used to analyze the effect of turbulence on the attachment on the sensor surface and further analyze the effect of adhesion on the sensor response time.

[0073] Step 102, a mobile platform 6 equipped with a winch drum 5 is set up, and a wire rope 8 of a specified length is wound on the UCTD probe 7 by a fast winding machine, such as Figure 2 As shown, the UCTD probe 7 is vertically dropped into the temperature-controlled water flow tank 1, and then moved along the predetermined path of the temperature-controlled water flow tank 1 to simulate the navigation process of the ship. Due to the resistance of the water body to the rope 8, the rope 8 is quickly pulled out from the winch drum 5 and the UCTD probe 7 at the same time, and the UCTD probe 7 performs a vertical rapid descending movement near the dropping point to simulate the working process of UCTD. The time from the start of the temperature step change to the reaching of 63% of the final value of the temperature sensor in UCTD under different flow velocities and turbulence intensities, that is, the response time of the temperature sensor, and the time from the start of the conductivity step change to the reaching of 63% of the final value, that is, the response time of the conductivity sensor, are recorded, and the response time curves of the temperature sensor and the conductivity sensor under different flow velocities and turbulence intensities are plotted.

[0074] In complex turbulent environments, the flow rate Fluctuation and vibration frequency The turbulence will lead to periodic changes in the fluid properties on the sensor surface, making the response time more complicated. In order to establish an accurate response time model, the mixing effect of turbulence on the fluid needs to be considered, especially for UCTD sensors, where turbulence will accelerate the interaction between the fluid and the sensor surface, affecting signal acquisition and processing. Therefore, the response time is not only a simple function of flow rate, but also includes flow rate changes, turbulence intensity, and the fluid boundary layer effect near the sensor surface.

[0075] Step 2: Establish a temperature sensor response time model and a conductivity sensor response time model by performing polynomial regression analysis on the recorded experimental data.

[0076] By comparing the experimental data of response time under different turbulence intensities with the corresponding flow velocity , Reynolds number , Strouhal number , shear stress Polynomial regression analysis was performed to establish the response time model of the temperature sensor and the response time model of the conductivity sensor.

[0077] The established temperature sensor response time model is as follows:

[0078] ;

[0079] in, is the predicted response time of the temperature sensor response time model, , , , is the constant to be optimized, , , , is the index to be optimized;

[0080] The established conductivity sensor response time model is as follows:

[0081] ;

[0082] in, is the predicted response time of the conductivity sensor response time model, , , , is the constant to be optimized, , , , is the index to be optimized.

[0083] The experimental data of response time under different turbulence intensities were recorded and compared with the corresponding flow velocity , Reynolds number , Strouhal number , shear stress Through polynomial regression analysis, the response time model of the temperature sensor and the response time model of the conductivity sensor are established, which can effectively reduce the response time difference caused by flow velocity fluctuations and turbulence effects, thereby greatly reducing the measurement error and improving the measurement accuracy and reliability of UCTD in complex turbulent environments.

[0084] Step 3: Use the gradient descent method to minimize the error between the model predicted response time and the actual response time and optimize the parameters in the model.

[0085] The specific method is as follows:

[0086] Step 301, define the error function as the sum of squares of the residuals between the actual value and the model predicted value:

[0087] The error function formula corresponding to the temperature sensor response time model is:

[0088] ;

[0089] in, It is The actual response time of the temperature sensor for each data point, It is The predicted response time of the temperature sensor response time model with data points, variables , , , It is flow velocity, Reynolds number, Strouhal number and shear stress for each data point, , , , is the constant to be optimized, , , , is the index to be optimized, is the total number of observed data points, is the index of the data point;

[0090] The error function formula corresponding to the conductivity sensor response time model is:

[0091] ;

[0092] in, It is The actual response time of the conductivity sensor for each data point, It is The predicted response time of the conductivity sensor response time model for each data point, , , , is the constant to be optimized, , , , is the index to be optimized;

[0093] Step 302, using the chain rule to calculate the partial derivative of each parameter, with the parameter For example:

[0094] ;

[0095] Step 303, use the gradient descent method to update the parameters until the error is minimized, and the update rule is as follows:

[0096] ;

[0097] in, It is Parameters in iterations The value of is the learning rate, which is used to control the step size of each update, is the error function corresponding to the temperature sensor response time model with respect to the parameter The partial derivative of

[0098] Step 304, update the parameters by repeated iteration until the error function Convergence, that is, the update amplitude of the parameter is less than the preset threshold ,satisfy:

[0099] ;

[0100] in, Set to 0.0001;

[0101] Step 305: Other parameters , , , , , , , , , , , , , , The calculation method of minimizing the error function by gradient descent is the same as the parameter ; Thus, an optimized temperature sensor response time model and an optimized conductivity sensor response time model are obtained.

[0102] Step 4: Use the optimized model to establish a synchronous response time model to synchronously match the response time difference between the temperature sensor and the conductivity sensor.

[0103] The details are as follows:

[0104] Step 401: Use the optimized model to establish a synchronous response time model by using the differential method of response time difference, and synchronously match the response time difference of the temperature sensor and the conductivity sensor. The synchronous response time model is:

[0105] ;

[0106] in, is the synchronous response time of the temperature sensor and the conductivity sensor; is the predicted response time of the temperature sensor response time model, is the predicted response time of the conductivity sensor response time model.

[0107] The specific correction formula for data processing is as follows:

[0108] Temperature correction formula:

[0109] ;

[0110] Conductivity correction formula:

[0111] ;

[0112] in, is the system sampling interval, The original temperature sampling sequence is The value of the sampling point, The original temperature sampling sequence is The value of the sampling point, is the synchronous response time model, is the sampling sequence after temperature correction, The original conductivity sampling sequence is The value of the sampling point, The original conductivity sampling sequence is The value of the sampling point, This is the sampling sequence after conductivity correction.

[0113] Use the optimized model to build a synchronous response time model , the response time difference of the temperature sensor and the conductivity sensor is synchronized and matched. This synchronization matching method can ensure that the temperature and conductivity sensors are measured synchronously under the same water mass conditions, reducing the data deviation of derived parameters such as salinity caused by the asynchronous sensor response time caused by flow velocity fluctuations and turbulence effects, greatly improving the accuracy and reliability of UCTD data in marine environment monitoring, and enabling the sensor to better adapt to the complex and changeable marine environment.

[0114] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for synchronous matching of sensor response time in UCTD under complex turbulent environment, characterized in that: The steps include: Step 1, using a temperature-controlled water flow tank to generate different flow velocities and turbulence intensities, recording the Reynolds number, Strouhal number and shear stress under different turbulence intensities, placing a UCTD probe in the temperature-controlled water flow tank to measure the temperature and conductivity under different flow velocities and turbulence intensities, and recording the response time of the temperature sensor and the conductivity sensor in the UCTD under different flow velocities and turbulence intensities; Step 2, establishing a temperature sensor response time model and a conductivity sensor response time model by performing polynomial regression analysis on the recorded experimental data; Step 3, using the gradient descent method to minimize the error between the model predicted response time and the actual response time, and optimize the parameters in the model; Step 4: Use the optimized model to establish a synchronous response time model to synchronously match the response time difference between the temperature sensor and the conductivity sensor.

2. The sensor response time synchronization matching method in UCTD under complex turbulent environment according to claim 1 is characterized in that: The temperature-controlled water flow tank is equipped with an electric heater, a centrifugal pump and a laser Doppler velocimeter. Fluorescent particles are added to the temperature-controlled water flow tank to mix evenly with the water. Different flow rates are simulated by changing the speed of the centrifugal pump. , use electric heater to adjust water temperature, and use water of different temperatures to change the viscosity of water In order to control the turbulence intensity, a laser Doppler velocimeter is used to obtain the flow velocity of the water body. and vibration frequency The Reynolds number is calculated by To characterize the turbulent state, <2000 is laminar flow, when 2000 < <4000 is transition flow. >4000 is turbulent.

3. The sensor response time synchronization matching method in UCTD under complex turbulent environment according to claim 2 is characterized in that: Reynolds number The calculation formula is as follows: ； in, is the density of water, is the flow velocity of the water body, is the width of the temperature-controlled water flow channel, is the viscosity of water; Strouhal number The calculation formula is: ； in, is the vibration frequency of the water body; Shear stress The calculation formula is: ； in, is the gradient of flow velocity in the vertical direction, is the vertical coordinate.

4. The sensor response time synchronization matching method in UCTD under complex turbulent environment according to claim 1 is characterized in that: In step 1, a mobile platform equipped with a winch drum is set up, a rope of a specified length is wound on the UCTD probe by a fast winding machine, the UCTD probe is vertically placed in the temperature-controlled water flow tank, and then moved along the predetermined path of the temperature-controlled water flow tank, the rope is quickly pulled out from the winch drum and the UCTD probe at the same time, and the UCTD probe performs a vertical rapid descending motion near the placement point, and the time from the start of the temperature step change to the reaching of 63% of the final value of the temperature sensor in the UCTD under different flow velocities and turbulence intensities, that is, the response time of the temperature sensor, and the time from the start of the conductivity step change to the reaching of 63% of the final value, that is, the response time of the conductivity sensor, are recorded, and the response time curves of the temperature sensor and the conductivity sensor under different flow velocities and turbulence intensities are plotted.

5. The sensor response time synchronization matching method in UCTD under complex turbulent environment according to claim 1 is characterized in that: In step 2, the temperature sensor response time model is established as follows: ； in, is the predicted response time of the temperature sensor response time model, is the flow rate, is the Reynolds number, is the Strouhal number, is the shear stress, , , , is the constant to be optimized, , , , is the index to be optimized; The established conductivity sensor response time model is as follows: ； in, is the predicted response time of the conductivity sensor response time model, , , , is the constant to be optimized, , , , is the index to be optimized.

6. The sensor response time synchronization matching method in UCTD under complex turbulent environment according to claim 5 is characterized in that: The specific method of step 3 is as follows: Step 301, define the error function: The error function formula corresponding to the temperature sensor response time model is: ； in, It is The actual response time of the temperature sensor for each data point, It is The predicted response time of the temperature sensor response time model with data points, variables , , , It is flow velocity, Reynolds number, Strouhal number and shear stress for each data point, , , , is the constant to be optimized, , , , is the index to be optimized, is the total number of observed data points, is the index of the data point; The error function formula corresponding to the conductivity sensor response time model is: ； in, It is The actual response time of the conductivity sensor for each data point, It is The predicted response time of the conductivity sensor response time model for each data point, , , , is the constant to be optimized, , , , is the index to be optimized; Step 302, using the chain rule to calculate the partial derivative of each parameter, with the parameter For example: ； Step 303, use the gradient descent method to update the parameters until the error is minimized, and the update rule is as follows: ； in, It is Parameters in iterations The value of is the learning rate, which is used to control the step size of each update, is the error function corresponding to the temperature sensor response time model with respect to the parameter The partial derivative of Step 304, update the parameters by repeated iteration until the error function Convergence, that is, the update amplitude of the parameter is less than the preset threshold ,satisfy: ； in, Set to 0.0001; Step 305: Other parameters , , , , , , , , , , , , , , The calculation method of minimizing the error function by gradient descent is the same as the parameter ; Thus, an optimized temperature sensor response time model and an optimized conductivity sensor response time model are obtained.

7. The sensor response time synchronization matching method in UCTD under complex turbulent environment according to claim 5 is characterized in that: Step 4 is as follows: The optimized model is used to establish a synchronous response time model through the differential method of response time difference, and the response time difference of the temperature sensor and the conductivity sensor is synchronously matched. The synchronous response time model is: ； in, is the synchronous response time of the temperature sensor and the conductivity sensor, is the predicted response time of the temperature sensor response time model, is the predicted response time of the conductivity sensor response time model.

Images