A fiber optic gyroscope temperature drift modeling and real-time compensation method

Abstract

A fiber optic gyroscope temperature drift modeling and real-time compensation method is provided, and the specific steps are as follows: Step 1: Design a fiber optic gyroscope temperature test, and collect the fiber optic gyroscope output value and the fiber optic gyroscope internal temperature sensor data within the full temperature range of ‑30° to 70°; Step 2: Perform wavelet threshold denoising preprocessing on the fiber optic gyroscope output obtained in step 1; Step 3: Establish a fiber optic gyroscope temperature drift segmented compensation model F with temperature change and temperature gradient as independent variables and the fiber optic gyroscope output signal after wavelet threshold denoising preprocessing as the dependent variable; Step 4: Calculate the coefficient of the temperature compensation model F by the least squares method and compensate the fiber optic gyroscope output in real time. The present invention meets the requirements of fast calculation speed and simplicity in engineering, meets the requirements of real-time compensation in engineering, and can effectively solve the problem of fiber optic gyroscope temperature drift, thereby improving the accuracy of the fiber optic gyroscope.

Description

Technical Field

[0001] The invention belongs to the field of inertial technology, and in particular relates to a fiber optic gyroscope temperature drift modeling and real-time compensation method. Background Art

[0002] Fiber optic gyroscopes have become key components of inertial navigation systems due to their all-solid-state, high sensitivity, large dynamic range, light weight, and low energy consumption. Fiber optic gyroscopes are optical sensors based on the Sagnace effect. Since the components of the device are sensitive to temperature, changes in the working environment temperature will affect the output accuracy of the fiber optic gyroscope, thereby restricting the improvement of the accuracy of the fiber optic inertial navigation system. In order to suppress this temperature drift error, many researchers have done a lot of work. On the one hand, from the hardware design, mainly starting from the internal structure of the gyroscope, the working environment temperature of the coil, light source, and optical device is made more stable through structural design improvements; on the other hand, software modeling and compensation are carried out. By analyzing the temperature characteristics of the fiber optic gyroscope, a temperature drift model is established to achieve the purpose of error compensation and improved accuracy. Among them, the software modeling method has become a hot topic of research due to its low cost, convenience and flexibility.

[0003] At present, temperature modeling mostly adopts polynomial model methods, such as the least squares method, which is widely used because of its simple principle; and with the development of computers, methods of using various neural networks to establish nonlinear models have also been studied. For example, Zhou Qi et al. studied the use of adaptive neural network fuzzy inference system to model and compensate for fiber optic gyroscope temperature drift errors; Guo Shiluo et al. proposed a method for optimizing BP neural network to model and compensate for fiber optic gyroscope temperature drift. However, these methods also have some shortcomings:

[0004] 1) In addition to drift caused by temperature, the output data of the fiber optic gyroscope is also superimposed with white noise. Therefore, whether the least squares fitting method or the BP neural network is used to approximate the temperature drift data, it will be affected by random noise and cannot obtain the highest compensation accuracy;

[0005] 2) The fitting effect of polynomial fitting is limited in more complex output situations;

[0006] 3) The neural network method has a large amount of calculation, is time-consuming and the established model cannot be obtained intuitively, which is not conducive to engineering applications.

[0007] Therefore, improvements are proposed. Summary of the invention

[0008] The technical problem solved by the present invention is to provide a method for modeling and real-time compensation of temperature drift of a fiber optic gyroscope. The present invention meets the requirements of fast calculation speed and simplicity in engineering, meets the requirements of real-time compensation in engineering, and can effectively solve the problem of temperature drift of the fiber optic gyroscope, thereby improving the accuracy of the fiber optic gyroscope.

[0009] The technical solution adopted by the present invention is a method for modeling and real-time compensation of temperature drift of an optical fiber gyroscope, and the specific steps are as follows:

[0010] Step 1: Design a fiber optic gyroscope temperature test and collect the fiber optic gyroscope output value and the fiber optic gyroscope internal temperature sensor data within the full temperature range of -30° to 70°;

[0011] Step 2: Perform wavelet threshold noise reduction preprocessing on the fiber optic gyroscope output obtained in step 1;

[0012] Step 3: Establish a fiber optic gyroscope temperature drift segmented compensation model F with temperature change and temperature gradient as independent variables and the fiber optic gyroscope output signal after wavelet threshold denoising preprocessing as the dependent variable;

[0013] Step 4: Calculate the coefficients of the temperature compensation model F by the least square method and perform real-time compensation on the FOG output.

[0014] In the above step 1, the specific experimental process of the fiber optic gyroscope temperature test is as follows:

[0015] 1) High and low temperature variable temperature test: first start at room temperature 20℃, enter the heating stage after 5 minutes, and heat up to 70℃ at a temperature change rate of 5℃ / min in the heating stage; enter the high temperature holding stage, keep it for 20 minutes, enter the cooling stage, and cool it to -30℃ at a temperature change rate of 5℃ / min; enter the low temperature holding stage, keep it for 20 minutes, enter the heating stage, and heat up to 20℃ at a temperature change rate of 5℃ / min; perform 6 cycles according to this step;

[0016] 2) Constant temperature test: First, cool down from room temperature 20°C to -30°C at a temperature change rate of 5°C / min, then increase the temperature by 1° every hour and maintain it until the temperature reaches 70°, then cool down to 20°C at a temperature change rate of 5°C / min, and the test ends;

[0017] 3) The third cycle stage from low temperature -30℃ to high temperature 70℃ is selected as the compensation data for the heating stage; the second cycle from high temperature insulation 70℃ to low temperature -30℃ is selected as the compensation data for the cooling stage.

[0018] In the above step 2, the wavelet threshold denoising preprocessing process is as follows:

[0019] (1) Select a wavelet function and determine the number of decomposition layers N, and perform N-layer wavelet decomposition on the fiber optic gyroscope output signal S;

[0020] (2) Select a threshold for each layer coefficient obtained by decomposition, and calculate the threshold of each layer;

[0021] (3) Apply thresholds to the Nth level approximate coefficients of the wavelet decomposition and the processed 1st to Nth level detail coefficients to achieve noise reduction.

[0022] In the above step 3, the fiber optic gyroscope temperature drift segmented compensation model F is as follows:

[0023]

[0024] The coefficients in the above formula satisfy the following relationship:

[0025]

[0026] Where F is the instantaneous output of the fiber optic gyroscope; F 0 is the average value of the gyro output in the first 3 minutes after the fiber optic gyro is turned on; T 1 is the temperature value measured by a temperature measuring resistor on the outer surface of the gyroscope; T 0 is the first temperature value measured by the temperature measuring resistor when the gyro is started; T 2 is the temperature value measured by the temperature sensor on the fiber optic gyroscope cavity; dT / dt is the rate of change of temperature, a 1 、a 2 、a 3 and b 1 , b 2 , b 3 are the fitting thermal coefficients of the heating stage and the cooling stage respectively; A 1j , A 2j , A 3j (j=0,1,2) and B 1j , B 2j , B 3j (j=0,1,2) is the fitting coefficient.

[0027] The advantages of the present invention compared with the prior art are:

[0028] 1. This scheme uses temperature change and temperature gradient as modeling parameters, and establishes models for the heating stage and the cooling stage respectively. The model directly uses the gyro output obtained by sampling the first 3 minutes after the gyro is turned on as the main value of the gyro output, and replaces the temperature with the temperature difference before and after the temperature measured by the same temperature sensor. Since the temperature difference is a small amount, the model can achieve real-time compensation and the overall compensation effect is very good;

[0029] 2. This solution can not only reduce the temperature drift of the fiber optic gyroscope by at least one order of magnitude, effectively solving the temperature drift problem of the fiber optic gyroscope, but also improve the stability of the temperature drift, thereby improving the accuracy of the fiber optic gyroscope;

[0030] 3. This solution has fast calculation speed and simple and easy method, which meets the real-time compensation requirements in engineering and has important engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a flow chart of the present invention;

[0032] Figure 2 This is a graph showing the change of the gyro output over time during the heating stage of the present invention;

[0033] Figure 3 This is a graph showing the change of the gyro output over time during the cooling stage of the present invention;

[0034] Figure 4 A comparison diagram of the original data of the fiber optic gyroscope in the heating stage of the present invention and the data after wavelet denoising;

[0035] Figure 5 A comparison diagram of the original data of the fiber optic gyroscope in the cooling stage of the present invention and the data after wavelet denoising;

[0036] Figure 6 It is the gyro output value before and after compensation of the model at 20°C in the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] See also Figure 1-6 , details of the embodiments of the present invention.

[0039] like Figure 1 As shown, the embodiment of the present invention discloses a method for modeling and real-time compensation of temperature drift of a fiber optic gyroscope, and the specific steps are as follows:

[0040] Step 1: Design a fiber optic gyroscope temperature test and collect the fiber optic gyroscope output value and the fiber optic gyroscope internal temperature sensor data within the full temperature range of -30° to 70°;

[0041] In order to measure the temperature change of the gyro, a temperature measuring resistor is added to the outer surface of the gyro, and a temperature sensor is installed on the shell of the gyro. The test process is carried out in a temperature box, and the gyro environment temperature is controlled by the temperature box. The data acquisition frequency in the test is 10ms, the temperature change range is -30°~70°, and the temperature change rate is 3℃ / min~5℃ / min. The experimental steps are as follows.

[0042] 1) High and low temperature variable temperature test: Start at room temperature 20℃, enter the heating stage after 5 minutes, and heat up to 70℃ at a temperature change rate of 5℃ / min. Enter the high temperature holding stage, keep it for 20 minutes, enter the cooling stage, and cool it to -30℃ at a temperature change rate of 5℃ / min; enter the low temperature holding stage, keep it for 20 minutes, enter the heating stage, and heat it to 20℃ at a temperature change rate of 5℃ / min; perform 6 cycles according to this step.

[0043] 2) Constant temperature test: First, cool down from room temperature 20℃ to -30℃ at a temperature change rate of 5℃ / min, then increase the temperature by 1° every 1h and maintain it until the temperature reaches 70°, then cool down to 20℃ at a temperature change rate of 5℃ / min, and the test ends.

[0044] 3) In this embodiment, the temperature change from low temperature -30°C to high temperature 70°C in the third cycle is selected at intervals of 10° as the compensation data for the heating stage; the temperature change from high temperature insulation of 70°C to low temperature -30°C in the second cycle is selected as the compensation data for the cooling stage; the collected north output of the fiber optic gyroscope is used as the original data. Figure 2 and Figure 3 The time variation of gyro output in the heating stage and cooling stage is given in Fig. Figure 2 and Figure 3 It can be seen that the gyro output corresponding to the same temperature during heating and cooling is different, mainly because the temperature drift of the fiber optic gyro is not only related to temperature, but also to temperature change and temperature gradient. Therefore, step 3 of this scheme establishes a segmented compensation model.

[0045] Step 2: Perform wavelet threshold noise reduction preprocessing on the FOG output obtained in step 1; the process is as follows:

[0046] (1) Select a wavelet function and determine the number of decomposition layers N, and perform N-layer wavelet decomposition on the fiber optic gyroscope output signal S;

[0047] (2) Select a threshold for each layer coefficient obtained by decomposition, and calculate the threshold of each layer;

[0048] (3) Apply thresholds to the Nth level approximate coefficients of the wavelet decomposition and the processed 1st to Nth level detail coefficients to achieve noise reduction.

[0049] This embodiment uses the north output of the fiber optic gyroscope in the heating and cooling stages obtained in step 1 as the original data for wavelet threshold denoising. The specific process is as follows:

[0050] 1) Perform three-layer wavelet decomposition on the fiber optic gyroscope output signal S. In this embodiment, the SymletsA (symN) wavelet family function is selected. The wavelet decomposition of the fiber optic gyroscope data Gy is in the form of:

[0051] [c,l]=wavedec(Gy,3,'sym2');

[0052] 2) Obtain the thresholds of each layer

[0053] This embodiment selects the Birge-Massart strategy to determine the threshold method, and uses the wdcbm function in MATLAB to obtain the thresholds of each layer. The format is:

[0054] [thr3,nkeep3]=wdcbm(c,l,3,1.3*prod(l(4,:)));

[0055] Among them, the thresholds of each layer determined by the Birge-Massart strategy return to thr3;

[0056] 3) Apply threshold and achieve noise reduction

[0057] Finally, the wavelet coefficients of the fiber optic gyroscope signal are denoised by layered soft thresholding in the following form:

[0058] [Gyd3,cxd,sxd,perf0,perfl2]=wdencmp('lvd',c,l,'sym',3,thr3,'s'); where Gyd3 is the FOG output after noise reduction.

[0059] Figure 4 and Figure 5 The output of the north-pointing fiber optic gyroscope before and after wavelet threshold denoising in the heating and cooling stages are given respectively. It can be seen from the figure that the output shape of the fiber optic gyroscope after wavelet threshold denoising is very similar to the original data, and it is obvious that the denoised data is smoother and more convergent. The standard deviations of the original output of the fiber optic gyroscope and the wavelet threshold denoising are 2.5735 and 0.0969 respectively. It can be seen that the use of wavelet transform for fiber optic gyroscope output preprocessing can effectively reduce the output noise of the fiber optic gyroscope and improve the signal-to-noise ratio.

[0060] Step 3: Establish a fiber optic gyroscope temperature segmented compensation model F with temperature change and temperature gradient as independent variables and the fiber optic gyroscope output signal after wavelet threshold denoising preprocessing as the dependent variable.

[0061] The fiber optic gyro temperature segmented compensation model is as follows:

[0062]

[0063] The coefficients in the above formula satisfy the following relationship:

[0064]

[0065] Where F is the instantaneous output of the fiber optic gyroscope; F0 is the average value of the gyro output value in the first 3 minutes after the fiber optic gyro is turned on; T 1 is the temperature value measured by the temperature measuring resistor on the outer surface of the gyroscope, T 0 is the first temperature value measured by the temperature measuring resistor when the gyro is started; T 2 is the temperature value measured by the temperature sensor on the fiber optic gyroscope cavity; dT / dt is the rate of change of temperature, a 1 、a 2 、a 3 and b 1 、b 2 、b 3 are the fitting thermal coefficients of the heating stage and the cooling stage respectively; A 1j , A 2j , A 3j (j=0,1,2) and B 1j , B 2j , B 3j (j=0,1,2) is the fitting coefficient.

[0066] Step 4: Calculate the coefficients of the temperature compensation model F by the least square method and perform real-time compensation on the output of the fiber optic gyroscope; specifically,

[0067] 1) According to model (1), the average value of the gyro output data at the corresponding time in the heating stage in step 2 and the average value of the gyro output data in the cooling stage in step 2 under an ambient temperature of -30℃ to 70℃ are respectively calculated by least squares to obtain the fitting thermal coefficient a of the heating stage and the cooling stage in model (1). 1 、a 2 、a 3 and b 1 、b 2 、b 3 , as shown in Table 1;

[0068] 2) Using thermal coefficient a 1 、a 2 、a 3 and b 1 、b 2 、b 3 According to formula (2), the fitting coefficient A can be determined by performing the least squares calculation 1j , A 2j , A 3j and B 1j , B 2j , B 3j (j=0,1,2) as shown in Table 2.

[0069] Table 1 Coefficients in the model

[0070] <![CDATA[a 1 ]]> <![CDATA[a 2 ]]> <![CDATA[a 3 ]]> <![CDATA[b 1 ]]> <![CDATA[b 2 ]]> <![CDATA[b 3 ]]> -30 0.0455 0.0105 -0.0419 0.0179 -0.0905 -0.0358 -20 -0.2806 0.2649 -0.0009 0.0206 0.1247 -0.0026 -10 -0.3698 0.0790 0.0251 0.3247 -0.0711 -0.0013 0 -0.0462 0.3898 0.0374 -0.0138 0.1259 -0.0005 10 -0.2614 0.1214 -0.1278 0.2523 0.2567 -0.0000 30 0.0407 -0.0853 -0.0126 -0.0197 0.1278 0.0246 40 0.1166 -0.1100 -0.0078 -0.0265 0.2165 0.0134 50 0.1557 -0.0923 -0.0095 -0.3478 -0.1023 0.0086 60 0.1889 -0.0835 -0.0087 -0.0246 -0.0429 0.0043 70 0.2136 -0.0679 -0.0009 -0.1032 0.1279 0.0013

[0071] Table 2 Coefficients in model (2)

[0072]

[0073]

[0074] 3) Utilization coefficient A 1j , A 2j , A 3j and B 1j , B 2j , B 3j (j=0,1,2), the gyro output can be compensated in real time according to the compensation model equations (2) and (1) in step 3.

[0075] This embodiment makes real-time compensation for the heating stage. The compensation result is as follows: Figure 6 Before compensation, the overall mean of the gyro measurement values ​​was 11.8785, and the variance was 0.0969; after compensation, the overall mean of the gyro measurement values ​​was 12.0429, and the variance was 0.0062, which was two orders of magnitude lower than before compensation, and the compensation effect was obvious.

[0076] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0077] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A method for modeling and real-time compensation of temperature drift of a fiber optic gyroscope, characterized in that: The specific steps are as follows: Step 1: Design a fiber optic gyroscope temperature test and collect the fiber optic gyroscope output value and the fiber optic gyroscope internal temperature sensor data within the full temperature range of -30° to 70°; Step 2: Perform wavelet threshold noise reduction preprocessing on the fiber optic gyroscope output obtained in step 1; Step 3: Establish a fiber optic gyroscope temperature drift segmented compensation model F with temperature change and temperature gradient as independent variables and the fiber optic gyroscope output signal after wavelet threshold denoising preprocessing as the dependent variable; Step 4: Calculate the coefficient of the temperature compensation model F by the least square method and perform real-time compensation on the output of the fiber optic gyroscope; In the above step 3, the fiber optic gyroscope temperature drift segmented compensation model F is as follows: The coefficients in the above formula satisfy the following relationship: Where F is the instantaneous output of the fiber optic gyroscope; F0 is the average value of the gyroscope output in the first 3 minutes after the fiber optic gyroscope is turned on; T1 is the temperature value measured by a temperature measuring resistor on the outer surface of the gyroscope; T0 is the first temperature value measured by the temperature measuring resistor when the gyro is started; T2 is the temperature value measured by the temperature sensor on the fiber optic gyro cavity; dT / dt is the rate of change of temperature, a1, a2, a3 and b1, b2, b3 are the fitting thermal coefficients of the heating stage and cooling stage respectively; A 1j , A 2j , A 3j (j=0,1,2) and B 1j , B 2j , B 3j (j=0,1,2) is the fitting coefficient.

2. The method for modeling and real-time compensation of temperature drift of an optical fiber gyroscope according to claim 1, characterized in that: In the above step 1, the specific process of the fiber optic gyroscope temperature test is as follows: 1) High and low temperature variable temperature test: first start at room temperature 20℃, enter the heating stage after 5 minutes, and heat up to 70℃ at a temperature change rate of 5℃ / min in the heating stage; enter the high temperature holding stage, keep it for 20 minutes, enter the cooling stage, and cool it to -30℃ at a temperature change rate of 5℃ / min; enter the low temperature holding stage, keep it for 20 minutes, enter the heating stage, and heat up to 20℃ at a temperature change rate of 5℃ / min; perform 6 cycles according to this step; 2) Constant temperature test: First, cool down from room temperature 20°C to -30°C at a temperature change rate of 5°C / min, then increase the temperature by 1° every hour and maintain it until the temperature reaches 70°, then cool down to 20°C at a temperature change rate of 5°C / min, and the test ends; 3) The third cycle stage from low temperature -30℃ to high temperature 70℃ is selected as the compensation data for the heating stage; the second cycle from high temperature insulation 70℃ to low temperature -30℃ is selected as the compensation data for the cooling stage.

3. The method for modeling and real-time compensation of temperature drift of an optical fiber gyroscope according to claim 1, characterized in that: In the above step 2, the wavelet threshold denoising preprocessing process is as follows: (1) Select a wavelet function and determine the number of decomposition layers N, and perform N-layer wavelet decomposition on the fiber optic gyroscope output signal S; (2) Select a threshold for each layer coefficient obtained by decomposition, and calculate the threshold of each layer; (3) Apply thresholds to the Nth level approximate coefficients of the wavelet decomposition and the processed 1st to Nth level detail coefficients to achieve noise reduction.

Images