Boiler operation optimization method, device, equipment, medium and program

Abstract

The invention relates to the technical field of thermal power generation, in particular to a boiler operation optimization method, device, equipment, medium and program, and the method comprises the steps: obtaining the grating fiber Bragg wavelength data of a boiler during peak regulation operation; the grating fiber Bragg wavelength data are input into a pre-trained boiler temperature prediction model, the boiler temperature prediction model outputs the predicted temperature of the boiler, the boiler temperature prediction model comprises a function relationship between wavelength and temperature, the predicted temperature of the boiler is determined by using the function relationship, and the function relationship between the wavelength and the temperature is P (x) = a0 + a1x + a2x2 +... + akxk; wherein P (x) is the predicted temperature of the boiler, x is fiber bragg grating Bragg wavelength data, a0, a1, a2,..., ak are coefficients of a polynomial, and k is the order of the polynomial; and optimizing the operation state of the boiler according to the predicted temperature. Therefore, the problems that in the related technology, when the coal-fired boiler is operated in a deep peak regulation mode, the temperature change in the boiler is difficult to accurately capture, and consequently the operation stability of the boiler is affected by local overheating are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power generation, and particularly relates to a boiler operation optimization method, device, equipment, medium and program. Background Art

[0002] Deep peak shaving is an important operation mode in the power system. It means that when the peak-valley difference of the grid load is large, the power plant reduces its output so that the operating load of the generating unit exceeds the basic peak shaving range to meet the demand of the grid load change. This method usually involves reducing the load rate to 40% to 30%, and sometimes even lower, so as to realize the peak-valley load regulation of the grid.

[0003] Optimizing and adjusting the peak shaving performance of coal-fired boilers is a conventional measure adopted by power plant operators. Specifically, through analyzing the operation data of the unit, on-site experimental tests, etc., combined with technical upgrades such as equipment transformation, the performance of the unit is optimized and improved, and the peak shaving capacity of the unit is tapped.

[0004] Commonly used optimization and adjustment measures for coal-fired boilers include: in terms of fuel, co-firing multiple coal types and co-firing natural gas; in terms of the coal pulverizing system, reducing the fineness of pulverized coal and improving the uniformity of pulverized coal, optimizing the primary air velocity, temperature and leveling of primary air, and optimizing the operation of coal mills; in terms of pulverized coal burners, optimizing the air distribution characteristics of burners and optimizing the operation quantity and position of burners; in terms of in-furnace air distribution, optimizing the in-furnace air distribution position and air volume and optimizing the air inlet angle; in other aspects, optimizing the distribution of high-temperature zones in the furnace, controlling the desuperheating water volume, optimizing the outlet flue gas temperature, optimizing the start-stop of auxiliary equipment, adjusting the coal-water ratio, etc.

[0005] In the related art, during the deep peak shaving operation of coal-fired boilers, the change of load will exacerbate the wall temperature fluctuation. Coupled with the low feed water flow rate, the water inlet uniformity of a single water-cooled wall tube of the boiler decreases and the distribution is uneven, which easily causes local overheating phenomena and increases the possibility of water-cooled wall tube bursting. At the same time, due to the small steam production, the problem of increased working medium deviation will also occur, which is likely to generate oxide scale and accelerate the material aging. At present, traditional temperature measurement methods are difficult to accurately capture the temperature changes under the complex working conditions inside the boiler, especially when the load changes sharply, and it is impossible to give early warnings of possible local overheating areas, affecting the stability of boiler operation. Summary of the Invention

[0006] The present invention provides a boiler operation optimization method, device, equipment, medium and program to solve the problems in the related art that it is difficult to accurately capture the temperature changes inside the coal-fired boiler during deep peak shaving operation, resulting in the instability of boiler operation due to local overheating, etc.

[0007] An embodiment of the first aspect of the present invention provides a method for optimizing the operation of a boiler, including the following steps: obtaining fiber Bragg grating wavelength data of the boiler during peak shaving operation; inputting the fiber Bragg grating wavelength data into a pre-trained boiler temperature prediction model, and the boiler temperature prediction model outputs the predicted temperature of the boiler. Wherein, the boiler temperature prediction model includes a functional relationship between wavelength and temperature, and the predicted temperature of the boiler is determined by using the functional relationship. The functional relationship formula between the wavelength and temperature is: P(x) = a 0 + a 1 x + a 2 x 2 +... + a k x k ; wherein, P(x) is the predicted temperature of the boiler, x is the fiber Bragg grating wavelength data, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial; optimizing the operation state of the boiler according to the predicted temperature.

[0008] Through the above technical solution, the embodiment of the present invention can input the fiber Bragg grating wavelength data of the boiler during peak shaving operation into a pre-trained boiler temperature prediction model to generate the predicted temperature of the boiler, so as to achieve accurate prediction of the boiler temperature, and thus optimize the operation state of the boiler according to the predicted temperature, avoiding situations such as wall temperature fluctuations caused by factors such as load changes during deep peak shaving operation of the boiler. By optimizing the operation state of the boiler, the overall operation performance of the boiler is improved, and adverse effects such as increased working medium deviation and material aging caused by deep peak shaving are reduced.

[0009] Optionally, the training method of the boiler temperature prediction model includes: obtaining a wavelength offset distribution image generated from the fiber Bragg grating wavelength data corresponding to the historical temperature data of the boiler; calculating the relative displacement according to any two of the wavelength offset distribution images; calculating the temperature change amount caused by strain according to the relative displacement; correcting the temperature data according to the temperature change amount; performing polynomial fitting on the modified temperature data and the corresponding fiber Bragg grating wavelength data to iteratively optimize the parameters of the polynomial until the loss function meets the preset conditions to stop the training of the boiler temperature prediction model.

[0010] Through the above technical solution, the embodiment of the present invention can preprocess the wavelength offset distribution image generated from the fiber Bragg grating wavelength data corresponding to the historical temperature data to correct the temperature data, and perform polynomial fitting based on the modified temperature data and the corresponding fiber Bragg grating wavelength data to iteratively optimize the parameters of the polynomial, improving the accuracy of the boiler temperature prediction model and achieving accurate prediction of the boiler temperature.

[0011] Optionally, calculating the relative displacement according to any two of the wavelength shift distribution images includes: performing a two-dimensional fast Fourier transform on the wavelength shift distribution image to generate a corresponding frequency-domain representation; calculating a corresponding frequency-domain phase difference according to the frequency-domain representation; performing an inverse Fourier transform according to the frequency-domain phase difference to generate a corresponding image displacement amount; and calculating the relative displacement of the wavelength shift distribution image according to the image displacement amount.

[0012] Through the above technical solution, the embodiment of the present invention can use Fourier transform and inverse Fourier transform to process the wavelength shift distribution image to determine the relative displacement of the wavelength shift distribution image, which can effectively resist the influence of noise and improve the processing efficiency and accuracy.

[0013] Optionally, calculating the temperature change amount caused by strain according to the relative displacement includes: calculating a corresponding displacement amount according to the relative displacement; calculating the wavelength change amount of the fiber Bragg grating caused by strain according to the displacement amount; and calculating the temperature change amount caused by strain according to the wavelength change amount of the fiber Bragg grating.

[0014] Through the above technical solution, the embodiment of the present invention can calculate a corresponding displacement amount according to the relative displacement, calculate the wavelength change amount of the fiber Bragg grating caused by strain according to the displacement amount, and calculate the temperature change amount caused by strain according to the wavelength change amount of the fiber Bragg grating, improving the measurement accuracy and efficiency, so as to improve the accuracy of the subsequent boiler temperature prediction model.

[0015] Optionally, the loss function is calculated as follows:

[0016]

[0017] where x i is the fiber Bragg grating wavelength data of the i-th data sample, y i is the actual temperature of the i-th data sample, P(x i ) is the predicted temperature of the i-th data sample, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial.

[0018] Optionally, optimizing the operating state of the boiler according to the predicted temperature includes: if the predicted temperature exceeds a preset safety temperature threshold, generating a corresponding boiler control parameter adjustment instruction according to the difference between the predicted temperature and the preset safety temperature threshold; and adjusting at least one of the operating parameters of each system of the boiler according to the boiler control parameter adjustment instruction to optimize the operating state of the boiler.

[0019] Through the above technical solution, the embodiments of the present invention can generate corresponding adjustment instructions for boiler control parameters according to the relationship between the predicted temperature and the preset safety temperature threshold, so as to adjust at least one of the operating parameters of each system of the boiler to optimize the operating state of the boiler, improve the overall operating performance of the boiler, avoid situations such as wall temperature fluctuations caused by factors such as load changes during deep peak shaving operation of the boiler, and reduce adverse effects such as increased working medium deviation and material aging caused by deep peak shaving.

[0020] In a second aspect of the present invention, an embodiment provides a boiler operation optimization device, including: an acquisition module for acquiring fiber Bragg grating wavelength data of the boiler during peak shaving operation; a processing module for inputting the fiber Bragg grating wavelength data into a pre-trained boiler temperature prediction model, and the boiler temperature prediction model outputs the predicted temperature of the boiler. Among them, the boiler temperature prediction model includes a functional relationship between wavelength and temperature, and uses the functional relationship to determine the predicted temperature of the boiler. The functional relationship formula between wavelength and temperature is: P(x) = a 0 + a 1 x + a 2 x 2 +... + a k x k ;

[0021] where P(x) is the predicted temperature of the boiler, x is the fiber Bragg grating wavelength data, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial; an optimization module for optimizing the operating state of the boiler according to the predicted temperature.

[0022] Optionally, the training method of the boiler temperature prediction model includes: acquiring a wavelength offset distribution image generated from fiber Bragg grating wavelength data corresponding to boiler historical temperature data; calculating the relative displacement according to any two of the wavelength offset distribution images; calculating the temperature change amount caused by strain according to the relative displacement; correcting the temperature data according to the temperature change amount; performing polynomial fitting on the modified temperature data and the corresponding fiber Bragg grating wavelength data to iteratively optimize the parameters of the polynomial until the loss function meets the preset conditions to stop the training of the boiler temperature prediction model.

[0023] Optionally, it further includes: a calculation module for performing a two-dimensional fast Fourier transform on the wavelength offset distribution image to generate a corresponding frequency domain representation; calculating the corresponding frequency domain phase difference according to the frequency domain representation; performing an inverse Fourier transform according to the frequency domain phase difference to generate a corresponding image displacement amount; calculating the relative displacement of the wavelength offset distribution image according to the image displacement amount.

[0024] Optionally, it further includes: the calculation module is further configured to calculate a corresponding displacement amount according to the relative displacement; calculate a change amount of the fiber grating wavelength caused by strain according to the displacement amount; calculate a change amount of the temperature caused by strain according to the change amount of the fiber grating wavelength.

[0025] Optionally, the loss function is calculated as follows:

[0026]

[0027] where x i is the fiber grating Bragg wavelength data of the i-th data sample, y i is the actual temperature of the i-th data sample, P(x i ) is the predicted temperature of the i-th data sample, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial.

[0028] Optionally, the optimization module is further configured to: if the predicted temperature exceeds a preset safe temperature threshold, generate a corresponding boiler control parameter adjustment instruction according to the difference between the predicted temperature and the preset safe temperature threshold; adjust at least one of the operating parameters of each system of the boiler according to the boiler control parameter adjustment instruction to optimize the operating state of the boiler.

[0029] An embodiment of the third aspect of the present invention provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the program to execute the boiler operation optimization method as described in the above embodiment.

[0030] An embodiment of the fourth aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, and the program is executed by a processor to execute the boiler operation optimization method as described in the above embodiment.

[0031] An embodiment of the fifth aspect of the present invention provides a computer program product, including a computer program or instruction, and when the computer program or instruction is executed, it realizes the boiler operation optimization method as described in the above embodiment.

[0032] Thus, the present invention has at least the following beneficial effects:

[0033] In the embodiments of the present invention, the predicted temperature of the boiler can be generated according to the fiber Bragg grating wavelength data of the boiler during peak shaving operation and input into a pre-trained boiler temperature prediction model, so as to achieve accurate prediction of the boiler temperature. Then, the operating state of the boiler can be optimized according to the predicted temperature, avoiding situations such as wall temperature fluctuations caused by factors such as load changes during deep peak shaving operation of the boiler. By optimizing the operating state of the boiler, the overall operating performance of the boiler is improved, and adverse effects such as increased working medium deviation and material aging caused by deep peak shaving are reduced.

[0034] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and / or additional aspects and advantages of the present invention will become apparent and be readily understood from the following description of the embodiments in conjunction with the accompanying drawings, where:

[0036] Figure 1 is a flowchart of a boiler operation optimization method according to an embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of the boiler operation workflow according to an embodiment of the present invention;

[0038] Figure 3 is a schematic diagram of the process of training a prediction model according to an embodiment of the present invention;

[0039] Figure 4 is a schematic diagram of the process of correcting temperature data according to an embodiment of the present invention;

[0040] Figure 5 is a block diagram example of a boiler operation optimization device according to an embodiment of the present invention;

[0041] Figure 6 is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention.

[0042] Description of reference numerals: Boiler operation optimization device 10, acquisition module 100, processing module 200, and optimization module 300. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] Embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present invention, but should not be construed as limiting the present invention.

[0044] As long as the temperature of any object itself and its surroundings is not absolute zero, the object radiates heat to its surroundings. When its temperature is equal to that of its surroundings, the radiation heat process is in a dynamic equilibrium state. Since the charged particles inside the object are excited to emit electromagnetic waves of different wavelengths as the temperature rises, the thermal energy is radiated outward in the form of electromagnetic waves. The higher the temperature of the object, the more strongly the particles are excited and the greater the radiated energy. The different frequencies of the charged particle movements result in different electromagnetic wave spectra.

[0045] Generally, it is considered that the wavelength less than 0.4μm is ultraviolet ray, the shorter wavelength is γ ray, and the wavelength from 0.4μm to 0.76μm is visible light. The wavelength greater than 0.76μm is infrared ray, among which the wavelength from 0.76 - 3μm is near-infrared ray, the wavelength from 3 - 6μm is mid-infrared ray, the wavelength from 6 - 20μm is mid-far infrared ray, and the wavelength from 20 - 1000μm is far-infrared ray. In temperature measurement, the visible light region and the infrared light region with a wavelength of 0.76 - 20μm are usually applied. Because it is a heat carrier, it is also called thermal ray. Both light and thermal ray can be regarded as electromagnetic waves, and the characteristics and laws of visible light are applicable to thermal ray.

[0046] According to the radiation law of a black body, the spectral radiant exitance of a black body (that is, the power radiated from a unit area to the hemispherical space within a unit wavelength interval near a certain wavelength) is determined by Planck's formula, that is

[0047]

[0048] In the formula, C 1 =3.741833×10 -16 Wm 2 is the first radiation constant; C 2 =1.438832×10 -2 mK is the second radiation constant.

[0049] The total radiant exitance (the exitance at all wavelengths) is given by the Stefan - Boltzmann law, that is

[0050]

[0051] In the formula, σ=5.7×10 -8 WM2K-4 is the black body radiation constant. Τ in the above two formulas is the absolute temperature. The radiant exitance of an actual object only needs to multiply the emissivity ε in the formula. The radiant exitance of an object is related to the radiation temperature Τ and the emissivity ε. As long as the radiant exitance of the object is measured and the emissivity ε of the object is known, the temperature Τ can be calculated.

[0052] Actually, the temperature measurement is obtained through the measurement of the radiation quantity. If the instrument measures the temperature only according to the radiation in a certain specific wavelength band, the obtained is the bright temperature T l; If the instrument measures the temperature based on the radiation in two or more characteristic wavelength bands, the obtained temperature is the color temperature T of the object. s ; If the temperature is measured based on the total radiation in all wavelength ranges, the obtained temperature is the total radiation temperature T of the object. r ; The vast majority of measured objects are not blackbodies, that is, the emissivity ε≠1. Therefore, the radiant energy measured by the thermometer is related not only to the temperature of the object but also to the emissivity. In addition, during the process of the radiant energy emitted by the object reaching the detector, there are losses caused by various factors. According to Planck's law, the radiation spectrum of an object is determined by the temperature of the radiator and the wavelength of the radiation line, that is, L = f(λ, T). After selecting a wavelength, the radiation intensity of the object is only related to the temperature, that is, L = f(T). As long as the radiation intensity of the object is measured, the brightness temperature of the radiator can be determined. Based on Planck's and Stefan-Boltzmann's laws, the radiation spectrum of an object is determined by the temperature of the radiator and the wavelength of the radiation line, that is, L = f(λ, T). After selecting a wavelength, the radiation intensity of the object is only related to the temperature, that is, L = f(T).

[0053] The relationship given based on the radiation law of a blackbody shows that as the temperature increases, the total energy radiated by the object increases, and the peak of the radiation spectrum will shift towards the short-wavelength direction. Therefore, the boiler temperature prediction model of this application has universality.

[0054] The following describes the boiler operation optimization method, device, electronic device, storage medium, and program of the embodiments of the present invention with reference to the accompanying drawings.

[0055] Specifically, Figure 1 is a schematic flowchart of a boiler operation optimization method provided by an embodiment of the present invention.

[0056] As Figure 1 shown, the boiler operation optimization method includes the following steps:

[0057] In step S101, obtain the fiber Bragg grating wavelength data of the boiler during peak shaving operation.

[0058] It can be understood that the embodiment of the present invention can use the fiber Bragg grating sensing module to obtain the fiber Bragg grating wavelength data of the boiler during peak shaving operation, so as to subsequently determine the predicted temperature of the boiler based on the fiber Bragg grating wavelength data.

[0059] It should be noted that the fiber Bragg grating sensing module includes a fiber Bragg grating sensor and a laser. The fiber Bragg grating sensor is arranged at the position of the boiler where temperature measurement is required, and it can collect in real time the broadband light emitted by the laser and transmitted in the temperature measurement area. The fiber Bragg grating sensor uses a fiber Bragg grating. After the broadband light is incident on the fiber Bragg grating, most of the light in the frequency band passes through this area, and only a part of the light in the frequency band will be reflected back. This part of the reflected light has a specific wavelength, which is called the Bragg wavelength. That is, the fiber Bragg wavelength data of the boiler during peak shaving operation is obtained in the present invention.

[0060] In step S102, the fiber Bragg wavelength data is input into a pre-trained boiler temperature prediction model, and the boiler temperature prediction model outputs the predicted temperature of the boiler. Among them, the boiler temperature prediction model includes the functional relationship between the wavelength and the temperature, and the predicted temperature of the boiler is determined using the functional relationship.

[0061] Among them, the functional relationship formula between the wavelength and the temperature is:

[0062] P(x) = a 0 + a 1 x + a 2 x 2 +... + a k x k ;

[0063] Among them, P(x) is the predicted temperature of the boiler, x is the fiber Bragg wavelength data, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial.

[0064] It can be understood that in the embodiments of the present invention, the fiber Bragg wavelength data can be input into a pre-trained boiler temperature prediction model, and the boiler temperature prediction model outputs the predicted temperature of the boiler to achieve accurate prediction of the boiler temperature.

[0065] It should be noted that for a set of actually collected wavelength and temperature related data points (x 1 , y 1 ), (x 2 , y 2 ),...,(x n , y n ), the polynomial function P(x) can be used to approximately describe these data points. Among them, the basic form of polynomial fitting is a k-degree polynomial as:

[0066] P(x) = a 0 + a 1 x + a 2 x2 +...+a k x k ;

[0067] wherein, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, x is the collected fiber Bragg grating wavelength data, and k is the order of the polynomial.

[0068] The purpose of fitting is to find the appropriate coefficients a 0 , a 1 , a 2 ,..., a k such that the polynomial function P(x) approximates the points in the dataset, that is, to minimize the fitting error; polynomial fitting is performed by the least squares method, specifically, the optimal polynomial coefficients are selected by minimizing the sum of the squares of the errors between the fitting function and the data points.

[0069] Specifically, the fiber Bragg grating sensor has a periodic refractive index structure in its core. When broadband light passes through the optical fiber, light of a specific wavelength is reflected by the optical fiber, and the periodic structure is modulated by the measured physical quantity, resulting in a shift of the reflection wavelength. High-sensitivity sensing is achieved by establishing a mathematical model between the measured physical quantity (boiler) and the reflection wavelength; among them, the fiber Bragg grating is a periodic refractive index modulation structure formed in the core by using the photosensitivity of the optical fiber material through technologies such as ultraviolet light exposure and femtosecond writing. When a broadband light beam is transmitted in the optical fiber and incident on the fiber Bragg grating, light of a specific wavelength that satisfies the Bragg condition will be reflected back. The central wavelength of the reflected light, the grating period, and the effective refractive index of the core satisfy the Bragg formula, where the Bragg formula is expressed as:

[0070] λ B = 2n eff Λ;

[0071] wherein, n eff is the effective refractive index of the fiber core, and Λ is the grating period.

[0072] By differentiating the above formula, we can obtain: Δλ B = 2Δn eff Λ + 2n eff ΔΛ;

[0073] wherein, Δλ B is the change in the central wavelength of the reflected light, Δn eff is the change in the effective refractive index of the fiber core, and ΔΛ is the change in the grating period.

[0074] As can be seen from the above formula, when the environmental temperature where the fiber grating is located changes, it will cause thermal expansion and thermo-optic effects of the fiber material, resulting in changes in the grating period and the effective refractive index of the core, and further causing the Bragg wavelength to drift. By measuring the change in the Bragg wavelength and according to the linear relationship between temperature and Bragg wavelength, the change in temperature can be calculated.

[0075] Therefore, in this application, the linear relationship between temperature and Bragg wavelength is described by the polynomial function P(x), and the optimal polynomial coefficients are continuously selected by minimizing the sum of the squares of the errors between the fitting function and the data points to improve the accuracy of predicted temperature.

[0076] In the embodiment of the present invention, the training method of the boiler temperature prediction model includes: obtaining a wavelength shift distribution image generated from the fiber grating Bragg wavelength data corresponding to the boiler historical temperature data; calculating the relative displacement according to any two wavelength shift distribution images; calculating the temperature change caused by strain according to the relative displacement; correcting the temperature data according to the temperature change; performing polynomial fitting according to the modified temperature data and the corresponding fiber grating Bragg wavelength data to iteratively optimize the parameters of the polynomial until the loss function meets the preset conditions to stop the training of the boiler temperature prediction model.

[0077] Among them, the loss function is calculated as follows:

[0078]

[0079] Among them, x i is the fiber grating Bragg wavelength data of the i-th data sample, y i is the actual temperature of the i-th data sample, P(x i ) is the predicted temperature of the i-th data sample, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial.

[0080] Among them, the preset condition is that the sum of squares of the loss function is lower than the target threshold, and the target threshold can be set according to actual needs without specific limitation.

[0081] It can be understood that the embodiment of the present invention can correct the temperature data by preprocessing the wavelength shift distribution image generated from the fiber grating Bragg wavelength data corresponding to the historical temperature data, and perform polynomial fitting based on the modified temperature data and the corresponding fiber grating Bragg wavelength data to iteratively optimize the parameters of the polynomial, improve the accuracy of the boiler temperature prediction model, and achieve accurate prediction of the boiler temperature.

[0082] It should be noted that, such as Figure 2 and3 As shown, after obtaining the wavelength shift distribution image generated from the historical boiler temperature data corresponding to the fiber Bragg grating wavelength data, it is necessary to preprocess the wavelength shift distribution image: image scaling or image rotation. Specifically:

[0083] (1) Image scaling:

[0084] If the image size is too large, scale the image according to actual needs. Let the scaling ratio be S, then the corresponding scaled image T 1 ′ and T 2 ′:

[0085] T 1 ′ = T 1 (Sp 1 , Sq 1 )

[0086] T 2 ′ = T 2 (Sp 2 , Sq 2 )

[0087] Among them, the value range of S is between (0, 1), and p and q are the coordinates of the image T in the two-dimensional space, used to determine the position of each pixel in the image. p represents the horizontal coordinate, and q represents the vertical coordinate. For an image, the coordinates of the upper-left pixel are (0, 0). As it moves to the right, the p value increases, and as it moves down, the q value increases. Any pixel point in the image can be located through different (p, q) combinations.

[0088] (2) Image rotation:

[0089] Let the rotation angle be θ, then perform a rotation operation on the image through the rotation matrix. For the image T 1 (p 1 , q 1 ) after rotation, the image T″ 1 (p″ 1 , q″ 1 ) is calculated by the following formula:

[0090]

[0091] Among them, (p 1 , q 1 ) are the original image coordinates, (p″ 1 , q″ 1 ) are the rotated image coordinates. Similarly, perform corresponding rotation on T 2 (p 2 , q 2 ) to obtain T″ 2 (p″ 2 , q″2 )。

[0092] In an embodiment of the present invention, calculating the relative displacement according to any two wavelength shift distribution images includes: performing a two-dimensional fast Fourier transform on the wavelength shift distribution image to generate a corresponding frequency domain representation; calculating a corresponding frequency domain phase difference according to the frequency domain representation; performing an inverse Fourier transform according to the frequency domain phase difference to generate a corresponding image displacement amount; and calculating the relative displacement of the wavelength shift distribution image according to the image displacement amount.

[0093] It can be understood that the embodiment of the present invention can use Fourier transform and inverse Fourier transform to process the wavelength shift distribution image to determine the relative displacement of the wavelength shift distribution image. The image processing is used to extract the spatial wavelength shift feature, analyze the spatial correlation between temperature and wavelength shift, and improve the accuracy of the boiler temperature prediction model.

[0094] Specifically, as Figure 4 shown, the specific steps of calculating the relative displacement according to any two wavelength shift distribution images are as follows:

[0095] 1) Perform a two-dimensional fast Fourier transform on two wavelength shift distribution images T 1 (p 1 , q 1 ) and T 2 (p 2 , q 2 ) that no longer require preprocessing; using the fast Fourier transform algorithm, obtain the frequency domain representations F 1 (u 1 , v 1 ) and F 2 (u 2 , v 2 );

[0096] where u and v are the horizontal and vertical frequency coordinates in the frequency domain, respectively, and the value ranges are restricted within the size of the image.

[0097] 2) Calculate the frequency domain phase difference: calculate the conjugate multiplication of the Fourier transform results of the two wavelength shift distribution images:

[0098]

[0099] where represents the complex conjugate of F 2 (u 2 , v 2 ), F 1 (u 1 , v 1 ) and F 2 (u 2 , v 2) They are the frequency domain representations obtained by performing two-dimensional fast Fourier transform on the wavelength offset distribution images to be matched respectively; denote the modulus of.

[0100] It should be noted that the meaning of the formula is that by multiplying F 1 (u 1 , v 1 ) by the complex conjugate of F 2 (u 2 , v 2 ) and then performing normalization processing, the denominator denotes the modulus of, and a frequency domain correlation function image R(u, v) carrying phase difference information that is only related to the phase difference between the frequency domains of the two images is obtained.

[0101] 3) Inverse Fourier transform: Perform inverse Fourier transform on the correlation function image R(u, v) to obtain a pulse-type function, where the pulse position represents the displacement amount of the image;

[0102] Use a fast algorithm to perform fast inverse Fourier transform on the phase difference image R(u, v) to generate a pulse-type function. This function will show a significant peak at a certain position, while the values at other positions are close to zero. The peak coordinates directly represent the relative displacement of the two images.

[0103] Perform inverse Fourier transform on the phase difference image R(u, v) using the fast inverse Fourier transform algorithm to obtain the result H(u, v) of the inverse Fourier transform. Among them, u and v in H(u, v) are the spatial domain coordinates of the image, and their value ranges are restricted within the size of the original image.

[0104] 4) Detect the peak: In the result of the inverse Fourier transform, find the peak position, and the coordinates of the peak are the relative displacement (Δu, Δv) of the two wavelength offset distribution images;

[0105] By traversing all the pixel values of H(u, v) and comparing to find the coordinates where the maximum value is located, the coordinates of this peak are the relative displacement of the two wavelength offset distribution images T 1 (p 1 , q 1 ) and T 2 (p 2 , q 2 );

[0106] If the displacement coordinates obtained by performing scaling operation in the preprocessing in the image acquisition and transformation unit need to be restored according to the scaling ratio S; if rotation operation is performed, the relative displacement between the original images needs to be determined in combination with the rotation angle θ.

[0107] In an embodiment of the present invention, calculating the temperature change caused by strain based on the relative displacement includes: calculating the corresponding displacement amount based on the relative displacement; calculating the wavelength change of the fiber grating caused by strain based on the displacement amount; calculating the temperature change caused by strain based on the wavelength change of the fiber grating.

[0108] It can be understood that in the embodiment of the present invention, the corresponding displacement amount can be calculated based on the relative displacement, the wavelength change of the fiber grating caused by strain can be calculated based on the displacement amount, and the temperature change caused by strain can be calculated based on the wavelength change of the fiber grating, so as to eliminate strain interference, ensure that the temperature measurement only reflects the real temperature change, and improve the accuracy of the boiler temperature prediction model.

[0109] It should be noted that according to the above mathematical principle and the grating coupled mode theory, the reflection wavelength drift amount can be effectively modulated by changing the effective refractive index and the grating period. For high-temperature sensors, through measures such as armored fiber and reasonable layout of sensors, it can be considered that the single-point stress change is not large, and the changes in fiber grating parameters caused by the thermal expansion effect and the thermo-optic effect can be expressed as the following model:

[0110]

[0111] where a is the thermal expansion coefficient of the optical fiber, ξ is the thermo-optic coefficient of the optical fiber, ΔT is a certain time period, Λ is the grating period, n eff is the effective refractive index of the fiber core, Δn eff is the change in the effective refractive index of the fiber core, and ΔΛ is the change in the grating period;

[0112] Therefore, from the above formula, it can be obtained that where Δλ B is the change in the central wavelength of the reflected light, λ B is the central wavelength of the reflected light, a is the thermal expansion coefficient of the optical fiber, ξ is the thermo-optic coefficient of the optical fiber. Therefore, in practical engineering applications, since the sensor is affected by the cross influence of temperature and stress at the same time, it is necessary to correct the relationship between wavelength and temperature for applicability.

[0113] Secondly, the temperature sensitivity coefficient in the following text covers all factors that cause changes in the system response due to temperature changes. Therefore, it includes the above-mentioned thermal expansion coefficient of the optical fiber and the thermo-optic coefficient of the optical fiber.

[0114] Specifically, as Figure 3 and Figure 4 shown, since strain interference mainly occurs during boiler mechanical vibration, pressure change or pipeline deformation, resulting in additional wavelength shift of the fiber grating, it is necessary to correct the temperature data to eliminate strain interference. Among them, the specific steps for calculating the temperature change caused by strain based on the relative displacement are as follows:

[0115] 1) Obtain the relationship between strain and displacement:

[0116] First, according to the principle of vector synthesis, calculate the relative displacement d based on the relative displacement (Δu, Δv) of two wavelength shift distribution images:

[0117]

[0118] where u and v are the horizontal and vertical frequency coordinates in the frequency domain, respectively.

[0119] Establish a mathematical model of strain and displacement by combining the physical characteristics of the fiber grating. Let the length of the fiber grating be L, the change in length due to external strain be ΔL, and the relative displacement of the image be d. Then the relationship between strain B and displacement d is expressed as:

[0120]

[0121] where B is the strain, L is the length of the fiber grating, and d is the relative displacement of the image.

[0122] 2) Calculate the change in the wavelength of the fiber grating Δx caused by strain:

[0123] Δx = x × B × W ε

[0124] where W ε is the strain sensitivity coefficient, which is determined by the specific material of the fiber grating, x is the wavelength of the fiber grating, B is the strain, and the change in the wavelength of the fiber grating Δx caused by strain is obtained. ε .

[0125] 3) Compensate for the influence of strain:

[0126] Let the total change in the wavelength of the fiber grating measured be Δx total , then the change in the wavelength of the fiber grating Δx y caused by temperature change is calculated by the following formula:

[0127] Δx y = Δx total - Δx ε

[0128] where Δx total is the total change in the wavelength of the fiber grating, and Δx ε is the change in the wavelength of the fiber grating caused by strain;

[0129] According to the relationship between the wavelength of the fiber grating and temperature:

[0130] Δx y = x × A × Δy;

[0131] Wherein, x is the wavelength of the fiber grating, A is the temperature sensitivity coefficient, which is determined by the specific material of the fiber grating, and Δx y is the change in the wavelength of the fiber grating caused by the temperature change, and thus the temperature change Δy can be calculated as follows:

[0132]

[0133] In this way, the influence of strain on temperature measurement is compensated, so that the change in the fiber grating signal caused solely by temperature change is obtained, improving the accuracy of temperature measurement to correct the temperature data.

[0134] It should also be noted that when constructing the boiler temperature prediction model in this application, the wavelength offset distribution image and temperature data in the historical data will undergo strain compensation calculation to eliminate strain interference and obtain the wavelength change data caused solely by temperature change. Therefore, it is necessary to correct the temperature data to improve the accuracy of the boiler temperature prediction model. During the actual prediction process using the boiler temperature prediction model, the currently collected fiber Bragg grating wavelength data also needs to undergo strain compensation calculation to eliminate real-time strain interference to generate accurate temperature prediction values. Therefore, the above calculation method is also applicable to the data processing of the boiler temperature prediction model.

[0135] In step S103, the operating state of the boiler is optimized according to the predicted temperature.

[0136] It can be understood that the embodiments of the present invention can optimize the operating state of the boiler according to the predicted temperature, avoiding situations such as wall temperature fluctuations caused by factors such as load changes during deep peak shaving operation of the boiler, improving the overall operating performance of the boiler by optimizing the operating state of the boiler, and reducing adverse effects such as increased working medium deviation and material aging caused by deep peak shaving.

[0137] In the embodiments of this application, optimizing the operating state of the boiler according to the predicted temperature includes: if the predicted temperature exceeds the preset safe temperature threshold, a corresponding boiler control parameter adjustment instruction is generated according to the difference between the predicted temperature and the preset safe temperature threshold; at least one of the operating parameters of each system of the boiler is adjusted according to the boiler control parameter adjustment instruction to optimize the operating state of the boiler.

[0138] Among them, the preset safe temperature threshold can be set according to the actual situation without specific limitation.

[0139] It should be noted that based on the optimization strategy of historical records and the corresponding temperature data, an optimization strategy model can be established through algorithms including but not limited to polynomial regression algorithm, Bayesian network algorithm, random forest algorithm, and principal component analysis combined with linear regression algorithm. Based on the temperature data predicted by the prediction model building module and the established optimization strategy model, an optimization strategy for boiler operation optimization can be generated; through an industrial control network including but not limited to fieldbus and industrial connection to relevant equipment for transmitting the control system of the boiler, and converting the generated optimization strategy into specific control instructions and sending them to the control system of the boiler, including but not limited to the burner control system and the feedwater control system.

[0140] For example, send instructions to the burner control system to adjust the operation quantity and position of the burner, increase the heat input in this area, and stabilize the temperature; at the same time, send instructions to the feedwater control system to finely adjust the feedwater flow rate to ensure the uniformity of the water inlet of the water wall tubes and prevent local overheating.

[0141] According to the boiler operation optimization method proposed in the embodiments of the present invention, the predicted temperature of the boiler is generated by inputting the fiber Bragg grating wavelength data of the boiler during peak shaving operation into a pre-trained boiler temperature prediction model to achieve accurate prediction of the boiler temperature, so as to optimize the operation state of the boiler according to the predicted temperature, avoid situations such as wall temperature fluctuations caused by factors such as load changes during deep peak shaving operation of the boiler, improve the overall operation performance of the boiler by optimizing the operation state of the boiler, and reduce adverse effects such as increased working medium deviation and material aging caused by deep peak shaving.

[0142] Next, the boiler operation optimization device proposed in the embodiments of the present invention will be described with reference to the accompanying drawings.

[0143] Figure 5 It is a block diagram of the boiler operation optimization device according to the embodiments of the present invention.

[0144] As Figure 5 shown, the boiler operation optimization device 10 includes: an acquisition module 100, a processing module 200, and an optimization module 300.

[0145] Among them, the acquisition module 100 is used to acquire the fiber Bragg grating wavelength data of the boiler during peak shaving operation; the processing module 200 is used to input the fiber Bragg grating wavelength data into a pre-trained boiler temperature prediction model, and the boiler temperature prediction model outputs the predicted temperature of the boiler. Among them, the boiler temperature prediction model includes the functional relationship between wavelength and temperature, and the predicted temperature of the boiler is determined using the functional relationship. The functional relationship formula between wavelength and temperature is: P(x) = a 0 + a 1 x + a 2 x 2 +... + a k xk ; where P(x) is the predicted temperature of the boiler, x is the fiber Bragg grating wavelength data, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial; the optimization module 300 is used to optimize the operating state of the boiler according to the predicted temperature.

[0146] In an embodiment of the present application, the training method of the boiler temperature prediction model includes: obtaining a wavelength shift distribution image generated from the fiber Bragg grating wavelength data corresponding to the historical temperature data of the boiler; calculating the relative displacement according to any two wavelength shift distribution images; calculating the temperature change caused by strain according to the relative displacement; correcting the temperature data according to the temperature change; performing polynomial fitting on the modified temperature data and the corresponding fiber Bragg grating wavelength data to iteratively optimize the parameters of the polynomial until the loss function meets the preset conditions to stop the training of the boiler temperature prediction model.

[0147] In an embodiment of the present application, it further includes: a calculation module is used to perform a two-dimensional fast Fourier transform on the wavelength shift distribution image to generate a corresponding frequency domain representation; calculating the corresponding frequency domain phase difference according to the frequency domain representation; performing an inverse Fourier transform on the frequency domain phase difference to generate a corresponding image displacement amount; calculating the relative displacement of the wavelength shift distribution image according to the image displacement amount.

[0148] In an embodiment of the present application, it further includes: the calculation module is further used to calculate the corresponding displacement amount according to the relative displacement; calculating the fiber Bragg grating wavelength change caused by strain according to the displacement amount; calculating the temperature change caused by strain according to the fiber Bragg grating wavelength change.

[0149] In an embodiment of the present application, the loss function is calculated as follows:

[0150]

[0151] where x i is the fiber Bragg grating wavelength data of the i-th data sample, y i is the actual temperature of the i-th data sample, P(x i ) is the predicted temperature of the i-th data sample, a 0 , a 1 , a 2 ,..., a k are the coefficients of the polynomial, and k is the order of the polynomial.

[0152] In an embodiment of the present application, the optimization module 300 is further configured to: optimize the operating state of the boiler according to the predicted temperature, including: if the predicted temperature exceeds the preset safety temperature threshold, generating a corresponding boiler control parameter adjustment instruction according to the difference between the predicted temperature and the preset safety temperature threshold; adjusting at least one of the operating parameters of each system of the boiler according to the boiler control parameter adjustment instruction to optimize the operating state of the boiler.

[0153] It should be noted that the foregoing explanation of the embodiment of the boiler operation optimization method also applies to the boiler operation optimization device of this embodiment, and will not be elaborated here.

[0154] According to the boiler operation optimization device provided by an embodiment of the present invention, the predicted temperature of the boiler is generated by inputting the fiber Bragg grating wavelength data of the boiler during peak shaving operation into a pre-trained boiler temperature prediction model, so as to achieve accurate prediction of the boiler temperature, and thus optimize the operating state of the boiler according to the predicted temperature, avoiding situations such as wall temperature fluctuations caused by factors such as load changes during deep peak shaving operation of the boiler, improving the overall operating performance of the boiler by optimizing the operating state of the boiler, and reducing adverse effects such as increased working medium deviation and material aging caused by deep peak shaving.

[0155] Figure 6 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device may include:

[0156] A memory 601, a processor 602, and a computer program stored on the memory 601 and executable on the processor 602.

[0157] When the processor 602 executes the program, it implements the boiler operation optimization method provided in the foregoing embodiment.

[0158] Further, the electronic device further includes:

[0159] A communication interface 603 for communication between the memory 601 and the processor 602.

[0160] The memory 601 is used to store a computer program executable on the processor 602.

[0161] The memory 601 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0162] If the memory 601, the processor 602, and the communication interface 603 are implemented independently, the communication interface 603, the memory 601, and the processor 602 can be interconnected through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 6 only a thick line is used to represent it in Figure 6 , but it does not mean that there is only one bus or one type of bus.

[0163] Optionally, in a specific implementation, if the memory 601, the processor 602, and the communication interface 603 are integrated on a single chip, the memory 601, the processor 602, and the communication interface 603 can communicate with each other through an internal interface.

[0164] The processor 602 may be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0165] The embodiments of the present invention also provide a computer-readable storage medium, on which a computer program or instruction is stored. When the computer program or instruction is executed by a processor, the above-mentioned boiler operation optimization method is implemented.

[0166] The embodiments of the present invention also provide a computer program product, including a computer program or instruction. When the computer program or instruction is executed, the above-mentioned boiler operation optimization method is implemented.

[0167] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0168] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, the meaning of "N" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0169] Any process or method description in the flowchart or described in other ways herein can be understood to represent a module, segment, or part of code including one or N executable instructions for implementing a customized logic function or process, and the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0170] It should be understood that each part of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following techniques known in the art: discrete logic circuits having logic gate circuits for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0171] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the method of implementing the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

Claims

1. A boiler operation optimization method, characterized in that: The following steps are involved: Obtain the fiber Bragg wavelength data of the boiler during peak load operation; The fiber Bragg grating wavelength data is input into a pre-trained boiler temperature prediction model, and the boiler temperature prediction model outputs the predicted temperature of the boiler, wherein the boiler temperature prediction model includes a functional relationship between wavelength and temperature, and the predicted temperature of the boiler is determined using the functional relationship, and the functional relationship between wavelength and temperature is: P(x)=a0+a1x+a2x 2 +...+a k x k ; Where P(x) is the predicted temperature of the boiler, x is the fiber Bragg grating wavelength data, a0, a1, a2, ..., a k are the coefficients of the polynomial, k is the order of the polynomial; The operating state of the boiler is optimized according to the predicted temperature.

2. The boiler operation optimization method according to claim 1, characterized in that: The training method of the boiler temperature prediction model includes: Obtaining a wavelength shift distribution image generated by fiber Bragg wavelength data corresponding to the boiler historical temperature data; Calculating relative displacement according to any two wavelength shift distribution images; Calculating the temperature change caused by the strain according to the relative displacement; Correcting the temperature data according to the temperature change; Polynomial fitting is performed according to the modified temperature data and the corresponding fiber Bragg wavelength data to iteratively optimize the parameters of the polynomial until the loss function meets the preset conditions to stop the training of the boiler temperature prediction model.

3. The boiler operation optimization method according to claim 2, characterized in that: The calculating the relative displacement according to any two wavelength shift distribution images comprises: Performing a two-dimensional fast Fourier transform on the wavelength shift distribution image to generate a corresponding frequency domain representation; Calculate the corresponding frequency domain phase difference according to the frequency domain representation; Performing inverse Fourier transform according to the frequency domain phase difference to generate a corresponding image displacement; The relative displacement of the wavelength shift distribution image is calculated according to the image displacement amount.

4. The boiler operation optimization method according to claim 3, characterized in that: The step of calculating the temperature change caused by the strain according to the relative displacement comprises: Calculate the corresponding displacement according to the relative displacement; Calculating the wavelength change of the fiber grating caused by the strain according to the displacement; The temperature variation caused by the strain is calculated according to the fiber grating wavelength variation.

5. The boiler operation optimization method according to claim 2, characterized in that: The loss function is calculated as follows: Among them, x i is the fiber Bragg wavelength data of the i-th data sample, y i is the actual temperature of the i-th data sample, P(x i ) is the predicted temperature of the ith data sample, a0, a1, a2, ..., a k are the coefficients of the polynomial and k is the order of the polynomial.

6. The boiler operation optimization method according to claim 1, characterized in that: Optimizing the operating state of the boiler according to the predicted temperature includes: If the predicted temperature exceeds the preset safety temperature threshold, a corresponding boiler control parameter adjustment instruction is generated according to the difference between the predicted temperature and the preset safety temperature threshold; At least one of the operating parameters of each system of the boiler is adjusted according to the boiler control parameter adjustment instruction to optimize the operating state of the boiler.

7. A boiler operation optimization device, characterized in that: include: An acquisition module is used to acquire the Bragg wavelength data of the grating fiber when the boiler is in peak operation; A processing module is used to input the grating fiber Bragg wavelength data into a pre-trained boiler temperature prediction model, and the boiler temperature prediction model outputs the predicted temperature of the boiler, wherein the boiler temperature prediction model includes a functional relationship between wavelength and temperature, and the predicted temperature of the boiler is determined using the functional relationship, and the functional relationship between wavelength and temperature is: P(x)=a0+a1x+a2x 2 +...+a k x k ; Where P(x) is the predicted temperature of the boiler, x is the fiber Bragg grating wavelength data, a0, a1, a2, ..., a k are the coefficients of the polynomial, k is the order of the polynomial; An optimization module is used to optimize the operating state of the boiler according to the predicted temperature.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the boiler operation optimization method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the boiler operation optimization method according to any one of claims 1 to 6.

10. A computer program product, characterized in that It comprises a computer program, which, when executed by a processor, is used to implement the boiler operation optimization method according to any one of claims 1 to 6.

Images