A method and device for correcting abnormal values ​​in hydrogen concentration measurement

Abstract

The present application is applicable to the field of hydrogen measurement technology, and provides a method and device for correcting abnormal values ​​of hydrogen concentration measurements. The method for correcting abnormal values ​​of hydrogen concentration measurements includes: obtaining a set of original hydrogen concentration values ​​at a monitoring point; applying a single accumulation algorithm to obtain a set of original accumulated hydrogen concentration values; adjusting each original accumulated hydrogen concentration value in the set of original accumulated hydrogen concentration values ​​to obtain an adjusted hydrogen concentration value set; obtaining an accumulated predicted hydrogen concentration value based on the adjusted hydrogen concentration value; and obtaining a corresponding effective original hydrogen concentration predicted value based on the accumulated predicted hydrogen concentration value. The present application adjusts each original accumulated hydrogen concentration value to obtain an adjusted hydrogen concentration value set, thereby accurately screening hydrogen concentration abnormal values, and when the hydrogen concentration measurement value is an abnormal value, replacing the abnormal value with a corresponding replacement value to achieve judgment and correction of the hydrogen concentration abnormal value.

Description

Technical Field

[0001] The present application belongs to the field of hydrogen measurement technology, and in particular relates to a method and device for correcting abnormal values ​​in hydrogen concentration measurement. Background Art

[0002] For a long time, with the continuous expansion of oil and gas genesis theory and the growing global demand for clean energy, hydrogen, as an important link connecting inorganic and organic hydrocarbon generation theories and also as a promising clean energy, has gradually attracted widespread attention from the academic community.

[0003] However, when exploring hydrogen mining areas, it is first necessary to measure the hydrogen concentration in the hydrogen reservoir to determine whether the hydrogen mining area is suitable for mining. However, the existing hydrogen concentration values ​​are sometimes interfered with by the external environment during measurement. For example, the effects of temperature, humidity or magnetic field may cause the measured hydrogen concentration values ​​to be abnormal. However, the existing hydrogen concentration measurement technology is unable to screen, judge and process abnormal hydrogen concentration values, so that the measured abnormal values ​​are output together with the normal values, which seriously affects the researchers' use of the concentration values. Summary of the Invention

[0004] In view of this, the embodiments of the present application provide a method and device for correcting abnormal values ​​in hydrogen concentration measurement, which can solve the problem that existing hydrogen concentration measurement technology is unable to screen, judge and process abnormal hydrogen concentration values.

[0005] A first aspect of an embodiment of the present application provides a method for correcting an abnormal value in hydrogen concentration measurement, the method comprising:

[0006] In the hydrogen detection area, n monitoring points are selected, and a hydrogen concentration sensor is arranged at each monitoring point. The n hydrogen concentration sensors are used to detect the hydrogen concentration values ​​of the n corresponding monitoring points. The original hydrogen concentration value set at the n monitoring points is obtained and X is used. (0) Indicates that, X (0) ={X (0) (1), X (0) (2),…,X (0) (i),…,X (0) (k)}; wherein k = 1, 2, ..., i, ..., n, n is a positive integer, and i is a positive integer;

[0007] Using a one-time accumulation algorithm, the original hydrogen concentration value set X (0) Each original hydrogen concentration value in is preprocessed one by one to obtain the original cumulative hydrogen concentration value set, and X is used to calculate the total hydrogen concentration value. (1) Indicates that, X (1) ={X (1) (1), X (1)(2),…,X (1) (i),…,X (1) (k)}, where k = 1, 2, ..., i, ..., n, n is a positive integer, and i is a positive integer;

[0008] in, j=1, 2, …, n, j is a positive integer;

[0009] For the original cumulative hydrogen concentration value set X (1) Each original accumulated hydrogen concentration value in is adjusted to obtain a set of adjusted hydrogen concentration values, and X is used to calculate the hydrogen concentration value. (2) Indicates that, X (2) ={d1X (1) (1), d2X (1) (2),…,d i X (1) (i),…,d k X (1) (k)}, d i is the influence factor of the corresponding adjustment hydrogen concentration value;

[0010] The corresponding influence factor d for adjusting the hydrogen concentration value i The calculation method is:

[0011] Establish the influence factor d for adjusting the hydrogen concentration value according to the temperature at the corresponding detection point i Mathematical model:

[0012]

[0013] Wherein, T(i) is the temperature at the corresponding detection point, g1 is the first specific gravity coefficient, g2 is the temperature proportional coefficient, g3 is the second specific gravity coefficient, g4 is the temperature variation coefficient, and g5 is the temperature correction coefficient;

[0014] The hydrogen concentration value set X will be adjusted (2) Each of the adjusted hydrogen concentration values ​​is taken as the initial value and brought into the improved GM (1, 1) model to obtain the cumulative predicted hydrogen concentration value in turn.

[0015] Predict hydrogen concentration value based on accumulation Obtain the corresponding effective original hydrogen concentration prediction value for:

[0016]

[0017] Where μ(i) is the predicted value of the effective original hydrogen concentration Correction factor for

[0018] According to the original hydrogen concentration value X(0) (i) and the predicted value of effective original hydrogen concentration Calculate the residual e(i);

[0019] Compare e(i) with the preset standard value λ to determine the abnormal value of hydrogen concentration measurement. When e(i)>λ, the corresponding X (0) (i) is an abnormal value and is replaced by the corresponding value X″′ (0) (i) Replace abnormal values ​​to realize the judgment and correction of abnormal hydrogen concentration values.

[0020] In one embodiment, the corresponding replacement value X″′ (0) The calculation method for (i) is:

[0021] According to the predicted value of effective original hydrogen concentration Established X″′ (0) (i) The calculation model is:

[0022]

[0023] Where β(i) is the corresponding replacement value X″′ (0) (i) The effective coefficient.

[0024] In one embodiment, the corresponding replacement value X″′ (0) The calculation method of the effective coefficient β(i) of (i) is:

[0025] According to the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Create the corresponding replacement value X″′ (0) The mathematical model of the effective coefficient β(i) of (i) is:

[0026]

[0027] in, is the original hydrogen concentration value X (0) The average value of (i), is the predicted value of the effective original hydrogen concentration The average value of .

[0028] In one embodiment, the hydrogen concentration value set X is adjusted (2) Each of the adjusted hydrogen concentration values ​​is taken as the initial value and brought into the improved GM (1, 1) model in turn, and the cumulative predicted hydrogen concentration value is obtained as follows:

[0029] Among them, a is the development coefficient and b is the gray action.

[0030] In one embodiment, the development coefficient a and the gray action amount b are calculated by estimating them using the principle of least square method.

[0031] In one embodiment, the correction factor μ(i) of the effective original hydrogen concentration prediction value is calculated as follows:

[0032] The mathematical model of the correction factor μ(i) of the effective original hydrogen concentration prediction value is established according to the methane concentration at the corresponding detection point:

[0033]

[0034] Wherein, H(i) is the methane concentration at the corresponding detection point, f1 is the sine coefficient, f2 is the first proportional coefficient of the methane concentration, f3 is the first correction coefficient of the methane concentration, f4 is the exponential coefficient, f5 is the second proportional coefficient of the methane concentration, and f6 is the second correction coefficient of the methane concentration.

[0035] In one embodiment, the calculation method of the preset standard value λ is:

[0036] According to the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Establish a mathematical model for the preset standard value λ:

[0037] In one embodiment, the corresponding replacement value X″′ (0) The effective coefficient of (i) β(i)=1.

[0038] In one embodiment, the calculation method of the preset standard value λ is:

[0039] The mathematical model for establishing the preset standard value λ is:

[0040]

[0041] Among them, M is the learning coefficient determined by the BP neural network method.

[0042] A second aspect of an embodiment of the present application provides a hydrogen concentration measurement abnormal value correction device, which is used to execute any of the hydrogen concentration measurement abnormal value correction methods described above.

[0043] The beneficial effects of the embodiments of the present application are: by providing a method for correcting abnormal values ​​in hydrogen concentration measurements, the method includes: obtaining a set of original hydrogen concentration values ​​at a monitoring point; using a single accumulation algorithm to pre-process each original hydrogen concentration value in the set of original hydrogen concentration values ​​one by one to obtain a set of original accumulated hydrogen concentration values; adjusting each original accumulated hydrogen concentration value in the set of original accumulated hydrogen concentration values ​​to obtain a set of adjusted hydrogen concentration values; using each adjusted hydrogen concentration value in the set of adjusted hydrogen concentration values ​​as an initial value in turn to obtain an accumulated predicted hydrogen concentration value; obtaining a corresponding effective original hydrogen concentration predicted value based on the accumulated predicted hydrogen concentration value; judging the abnormal value of the hydrogen concentration measurement and replacing the abnormal value with a corresponding replacement value to achieve judgment and correction of the abnormal value of the hydrogen concentration. The present application adjusts each original accumulated hydrogen concentration value in the original accumulated hydrogen concentration value set to obtain an adjusted hydrogen concentration value set, so that the hydrogen concentration abnormal values ​​can be accurately screened, and when the hydrogen concentration measurement value is an abnormal value, the abnormal value is replaced by the corresponding replacement value to achieve the judgment and correction of the hydrogen concentration abnormal value. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0045] Figure 1 This is a schematic diagram of the steps of a method for correcting abnormal values ​​in hydrogen concentration measurement provided by one embodiment of the present application;

[0046] Figure 2 This is a schematic diagram of a learning coefficient determined by a BP neural network method according to an embodiment of the present application;

[0047] Figure 3 This is a schematic diagram of the error rate of the patent method and the existing method provided in one embodiment of the present application;

[0048] Figure 4 This is a schematic diagram of the number of experiments and error rate of a method for correcting abnormal values ​​in hydrogen concentration measurement provided in one embodiment of the present application. DETAILED DESCRIPTION

[0049] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0050] The following will be combined with the accompanying drawings in the embodiments of this application to clearly describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0051] For a long time, with the continuous expansion of oil and gas genesis theory and the growing global demand for clean energy, hydrogen, as an important link connecting inorganic and organic hydrocarbon generation theories and also as a promising clean energy, has gradually attracted widespread attention from the academic community.

[0052] However, when exploring hydrogen mining areas, it is first necessary to measure the hydrogen concentration in the hydrogen reservoir to determine whether the hydrogen mining area is suitable for mining. However, the existing hydrogen concentration values ​​are sometimes interfered with by the external environment during measurement. For example, the effects of temperature, humidity or magnetic field may cause the measured hydrogen concentration values ​​to be abnormal. However, the existing hydrogen concentration measurement technology is unable to screen, judge and process abnormal hydrogen concentration values, so that the measured abnormal values ​​are output together with the normal values, which seriously affects the researchers' use of the concentration values.

[0053] In order to solve the above technical problems, the present invention provides a method for correcting abnormal values ​​of hydrogen concentration measurement. Figure 1 As shown, the method for correcting abnormal values ​​of hydrogen concentration measurement includes: steps S100 to S600.

[0054] Specifically, step S100: in the hydrogen detection area, n monitoring points are selected, and a hydrogen concentration sensor is arranged at each monitoring point. The n hydrogen concentration sensors are used to detect the hydrogen concentration values ​​of the n corresponding monitoring points, and the original hydrogen concentration value set at the n monitoring points is obtained, and X is used to calculate the hydrogen concentration value of the n monitoring points. (0) Indicates that, X (0) ={X (0) (1), X (0) (2),…,X (0) (i),…,X (0) (k)}; wherein k = 1, 2, ..., i, ..., n, n is a positive integer, and i is a positive integer.

[0055] Step S200: Apply a one-time accumulation algorithm to the original hydrogen concentration value set X (0) Each original hydrogen concentration value in is preprocessed one by one to obtain the original cumulative hydrogen concentration value set, and X is used to calculate the total hydrogen concentration value. (1) Indicates that, X (1) ={X (1) (1), X (1) (2),…,X (1) (i),…,X (1) (k)}, where k = 1, 2, ..., i, ..., n, n is a positive integer, and i is a positive integer;

[0056] in, j=1, 2,…, n, where j is a positive integer.

[0057] Step S300: The original accumulated hydrogen concentration value set X (1) Each original accumulated hydrogen concentration value in is adjusted to obtain a set of adjusted hydrogen concentration values, and X is used to calculate the hydrogen concentration value. (2) Indicates that, X (2) ={d1X (1) (1), d2X (1) (2),…,d i X (1) (i),…,d k X (1) (k)}, where d i is the influence factor of the corresponding adjustment hydrogen concentration value;

[0058] The corresponding influence factor d for adjusting the hydrogen concentration value i The calculation method is:

[0059] Since the stability of hydrogen is directly related to temperature, a model of the effect of temperature on hydrogen concentration is established to correct the hydrogen concentration:

[0060] Establish the influence factor d for adjusting the hydrogen concentration value according to the temperature at the corresponding detection point i Mathematical model:

[0061] Wherein, T(i) is the temperature at the corresponding detection point, g1 is the first specific gravity coefficient, g2 is the temperature proportional coefficient, g3 is the second specific gravity coefficient, g4 is the temperature variation coefficient, and g5 is the temperature correction coefficient; wherein, g1 is 0.0247, g2 is 0.0793, g3 is 67.02, g4 is 182.6, and g5 is 64.73.

[0062] Step S400: Adjust the hydrogen concentration value set X (2)Each of the adjusted hydrogen concentration values ​​is taken as the initial value and brought into the improved GM (1, 1) model to obtain the cumulative predicted hydrogen concentration value in turn. for:

[0063] Among them, a is the development coefficient, b is the gray action amount;

[0064] Step S500: Predict hydrogen concentration value based on accumulation Obtain the corresponding effective original hydrogen concentration prediction value for:

[0065] Where μ(i) is the predicted value of the effective original hydrogen concentration Correction factor for

[0066] Step S600: According to the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Calculate the residual e(i):

[0067] The calculation method is:

[0068] Compare e(i) with the preset standard value λ to determine the abnormal value of hydrogen concentration measurement. When e(i)>λ, the corresponding X (0) (i) is an abnormal value and is replaced by the corresponding value X″′ (0) (i) Replace abnormal values ​​to realize the judgment and correction of abnormal hydrogen concentration values.

[0069] In this embodiment, in step S100, n monitoring points are set in the hydrogen detection area, wherein each monitoring point can be provided with a hydrogen concentration sensor by punching a hole at the detection point to measure the hydrogen concentration at the detection point. It is understood that the hydrogen concentration sensor at each detection point can detect the hydrogen concentration of the monitoring point in real time or at a fixed time and output it. For example, the original hydrogen concentration value of the first monitoring point is represented by X (0) (1) indicates that the original hydrogen concentration value at the second monitoring point is expressed as X (0) (2) indicates that, and so on, the original hydrogen concentration value of the nth monitoring point is expressed as X (0) (n) indicates that, for example, when five detection points are set in a small detection area, the original hydrogen concentration values ​​of the detection points output in sequence are expressed as: X (0) (1) X (0) (2) X (0) (3) X (0) (4) X (0) (5) indicates that. In this embodiment, X (0)(i) represents the original hydrogen concentration value at any detection point.

[0070] In step S200, a cumulative algorithm is used to calculate the original hydrogen concentration value X of each detection point. (0) (i) Perform cumulative processing, for example, the original hydrogen concentration value X at the first monitoring point (0) (1) After cumulative processing, it is X (1) (1), the original hydrogen concentration value X at the second detection point (0) (2) After cumulative processing, it is X (1) (2) By analogy, the original hydrogen concentration value X at the nth detection point (0) (n) After cumulative processing, it is X (1) (n), and when five detection points are set in a smaller detection area, the original hydrogen concentration values ​​of the detection points output in sequence are expressed as: X (0) (1) X (0) (2) X (0) (3) X (0) (4) X (0) (5) indicates that the original accumulated hydrogen concentration value output after the accumulation process is expressed as: X (1) (1) X (1) (2) X (1) (3) X (1) (4) X (1) (5) indicates that the fifth original cumulative hydrogen concentration value is the sum of the fifth and the first four original hydrogen concentration values. Similarly, the fourth original cumulative hydrogen concentration value is the sum of the fourth and the first three original hydrogen concentration values. In this embodiment, j = 1, 2, ..., n, where j is a positive integer and j is any one of the n monitoring points.

[0071] In step S300, since the stability of hydrogen is directly related to temperature, the influence factor d of the hydrogen concentration is adjusted. i Specifically, the influence factor d of the adjusted hydrogen concentration value is established. i The impact model is:

[0072] Where T(i) is the temperature at the detection point, g1 is the first specific gravity coefficient, g2 is the temperature proportional coefficient, g3 is the second specific gravity coefficient, g4 is the temperature variation coefficient, and g5 is the temperature correction coefficient. It is understandable that the temperature T(i) at each detection point needs to be measured in sequence by the temperature sensor. For example, the temperature at the first monitoring point is T(1), and the temperature at the second detection point is T(2). Where g1 is 0.0247, g2 is 0.0793, g3 is 67.02, g4 is 182.6, and g5 is 64.73.

[0073] In one embodiment, the influence factor d of adjusting the hydrogen concentration value is established. i The influence model of the model, the first specific gravity coefficient g1, the temperature proportional coefficient g2, the second specific gravity coefficient g3, the temperature change coefficient g4, the temperature correction coefficient g5, can be obtained by the temperature and the corresponding influence factor d i For example, in a simulation experiment, the temperatures at the six monitoring points are measured at 0°C, 5°C, 10°C, 15°C, 20°C, and 25°C, and the corresponding influencing factors for adjusting the hydrogen concentration value are 2%, 3.9%, 5.8%, 8%, 12%, and 18%, respectively. The first specific gravity coefficient g1, temperature proportional coefficient g2, second specific gravity coefficient g3, temperature variation coefficient g4, and temperature correction coefficient g5 in the model can be obtained at once. In addition, the influencing factor d of adjusting the hydrogen concentration value is verified. i The impact model is established to meet the needs of practical applications.

[0074] In step S400, in this embodiment, each adjusted hydrogen concentration value d i X (1) (i) Bring it into the improved GM(1,1) model and get the cumulative predicted hydrogen concentration value in turn It can be understood that each adjustment of the hydrogen concentration value d i X (1) (i) corresponds to a cumulative predicted hydrogen concentration value For example, the first adjustment of the hydrogen concentration value d1X (1) (1) Corresponding cumulative predicted hydrogen concentration value And so on, each cumulative predicted hydrogen concentration value is obtained.

[0075] In step S500, the cumulative reduction algorithm is used to obtain n effective original hydrogen concentration prediction values ​​in sequence. It can be understood that when i=5, then A correction factor μ(i) is added to each valid original hydrogen concentration prediction value to correct each valid original hydrogen concentration prediction value.

[0076] In step S600, according to the original hydrogen concentration value X(0) (i) and the predicted value of effective original hydrogen concentration Calculate the residual e(i): The calculation method is: Compare e(i) with the preset standard value λ to determine the abnormal value of hydrogen concentration measurement. When e(i)>λ, the corresponding X (0) (i) is an abnormal value and is replaced by the corresponding value X″′ (0) (i) Replace abnormal values ​​to determine and correct abnormal hydrogen concentration values. This allows abnormal hydrogen concentration values ​​to be identified and replaced, resolving the problem that existing hydrogen concentration measurement technology is unable to screen, determine, and process abnormal hydrogen concentration values.

[0077] In one embodiment, the corresponding replacement value X″′ (0) The calculation method of (i) is: based on the predicted value of the effective original hydrogen concentration Establish X″′ (0) (i) The calculation model is: Where β(i) is the corresponding replacement value X″′ (0) (i) The effective coefficient.

[0078] In one embodiment, the corresponding replacement value X″′ (0) The calculation method of the effective coefficient β(i) is as follows: Based on the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Create the corresponding replacement value X″′ (0) The mathematical model of the effective coefficient β(i) of (i) is:

[0079] in, is the original hydrogen concentration value X (0) The average value of (i), is the predicted value of the effective original hydrogen concentration In this embodiment, is the original hydrogen concentration value X (0) (i) The average value, it can be understood that The solution can be: by making each original hydrogen concentration value X (0) (i) Add and then divide by the number of first added to get, is the predicted value of the effective original hydrogen concentration The average value of The solution method can be: by making each effective original hydrogen concentration prediction value Add and then divide by the number added first.

[0080] In one embodiment, the hydrogen concentration value set X is adjusted(2) Each of the adjusted hydrogen concentration values ​​is taken as the initial value and brought into the improved GM (1, 1) model in turn, and the cumulative predicted hydrogen concentration value is obtained as follows:

[0081] Among them, a is the development coefficient and b is the gray action.

[0082] In one embodiment, the corresponding influence factor d of adjusting the hydrogen concentration value is i The calculation method is:

[0083] The stability of hydrogen is directly related to temperature. By establishing a model for the impact of temperature on hydrogen concentration, the concentration of hydrogen can be corrected: a mathematical model for the impact of temperature on the hydrogen concentration value at the monitoring point is established:

[0084] Where T(i) is the temperature at the detection point, g1 is the first specific gravity coefficient, g2 is the temperature proportional coefficient, g3 is the second specific gravity coefficient, g4 is the temperature variation coefficient, and g5 is the temperature correction coefficient. Here, g1 is 0.0247, g2 is 0.0793, g3 is 67.02, g4 is 182.6, and g5 is 64.73.

[0085] In one embodiment, the development coefficient a and the gray action amount b are calculated by estimating them using the principle of least square method:

[0086] Among them, the matrix B is: Among them, z (1) (i) = 0.4x (1) (i)+0.6x (1) (i-1), k = 1, 2, ..., i, ..., n, n is a positive integer, i is a positive integer; the matrix Y is: Then the development coefficient a and ash action amount b are calculated.

[0087] In one embodiment, the correction factor μ(i) of the effective original hydrogen concentration prediction value is calculated as follows: Since hydrogen and methane are generally produced by serpentinization reaction, the amount of hydrogen produced in the serpentinization reaction is related to the content of iron minerals (e.g., olivine). In nature, hydrogen and methane generally coexist, that is, where hydrogen exists, methane also exists, and the concentration of methane is closely related to the concentration of hydrogen. A mathematical model of the correction factor μ(i) of the effective original hydrogen concentration prediction value is established based on the methane concentration at the corresponding detection point:

[0088]

[0089] Where H(i) is the methane concentration at the corresponding detection point, f1 is the sine coefficient, f2 is the first proportional coefficient of the methane concentration, f3 is the first correction coefficient of the methane concentration, f4 is the exponential coefficient, f5 is the second proportional coefficient of the methane concentration, and f6 is the second correction coefficient of the methane concentration. Where H(i) is the methane concentration at the detection point, f1 is 0.3844, f2 is -0.0002, f3 is 1.129, f4 is 1.581, f5 is 0.0089, and f6 is -0.6444.

[0090] In one embodiment, the methane concentration H(i) at each detection point can be measured by a methane concentration sensor. In the mathematical model of the correction factor μ(i), the sinusoidal coefficient f1, the first proportional coefficient f2 of the methane concentration, the first correction coefficient f3 of the methane concentration, the exponential coefficient f4, the second proportional coefficient f5 of the methane concentration, and the second correction coefficient f6 of the methane concentration can be calculated based on the methane concentration at the detection point and the corresponding correction factor μ(i). For example, in a simulation experiment, the methane concentration at each detection point is detected to be: 7.84%, 32 .99%, 45.62%, 63.43%, 78.79%, 85.65%, and the corresponding correction factors μ(i) are 9.216%, 6.701%, 5.438%, 3.657%, 2.121%, and 1.346%, respectively. The sinusoidal coefficient f1, the first proportional coefficient f2 of methane concentration, the first correction coefficient f3 of methane concentration, the exponential coefficient f4, the second proportional coefficient f5 of methane concentration, and the second correction coefficient f6 of methane concentration are obtained at one time. Moreover, the establishment of the correction factor μ(i) meets the needs of practical applications.

[0091] In one embodiment, the calculation method of the preset standard value λ is: according to the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Establish a mathematical model for the preset standard value λ:

[0092]

[0093] In one embodiment, the corresponding replacement value X″′ (0) The effective coefficient of (i) β(i) = 1. It can be understood that That is, when e(i)>λ, the corresponding X is determined. (0) (i) is an abnormal value and is replaced with the corresponding value Instead of abnormal values, it can realize the judgment and correction of abnormal hydrogen concentration values. For example, when X (0) (1) When it is an abnormal value, use Replace and output to achieve the judgment and correction of abnormal hydrogen concentration.

[0094] In one embodiment, the calculation method of the preset standard value λ is: a mathematical model of the preset standard value λ is established as follows: Among them, M is the learning coefficient determined by the BP neural network method.

[0095] In this embodiment, reference Figure 2 As shown, the method for obtaining the learning coefficient M determined by the BP neural network method is: the original hydrogen concentration value X (0) The number of (i) number, measurement speed mv, temperature te at the detection point, humidity hu at the detection point, noise intensity noise at the detection point, methane concentration ch at the detection point, and iron ore content fe at the detection point are used as parameters of the BP neural network input layer; the middle layer of the BP neural network is a hidden layer and adopts 8 neurons; the learning coefficient M is used as the output parameter of the BP neural network output layer; the learning coefficient M is obtained through training.

[0096] In one embodiment, the training process of the BP neural network is as follows: establishing a sample set S = (s1, s2, ..., s n ), where s i =(number i , mv i ,te i ,hu i , noise i , ch i ,fe i , M i ), where i = 1, 2, 3, ..., n; the sample set S is used for training to obtain the input layer to hidden layer weight matrix w1 ij And the hidden layer to output layer weight matrix w2 ij , where w1 ij In w2ij, i=8 is the number of hidden layer neurons, j=7 is the number of input layer parameters, i=1 is the number of output layer neurons, j=8 is the number of input layer parameters, and learning and memory are performed to obtain the recognition coefficient M.

[0097] In one embodiment, Figure 3 、 Figure 4 As shown, 12 hydrogen concentration measurement experiments were conducted in different underground hydrogen storage mining areas using the patented method. The figure shows that the average error value of the error value judgment of underground hydrogen storage concentration measured using the patented method is 1.53%, the variance is 1.74, and the maximum error is 1.68%, indicating that the patented method is stable. 12 hydrogen concentration measurement experiments were conducted in different underground hydrogen storage spaces using the patented method, and the average value was recorded as the result. It can be seen that the maximum relative error is 1.42, which is more accurate than the existing measurement method.

[0098] An embodiment of the present application further provides a hydrogen concentration measurement abnormal value correction device, which is used to execute any of the hydrogen concentration measurement abnormal value correction methods described above.

[0099] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0100] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for correcting abnormal values ​​in hydrogen concentration measurement, characterized in that: The hydrogen concentration measurement abnormal value correction method includes: In the hydrogen detection area, n monitoring points are selected, and a hydrogen concentration sensor is arranged at each monitoring point. The n hydrogen concentration sensors are used to detect the hydrogen concentration values ​​of the n corresponding monitoring points. The original hydrogen concentration value set at the n monitoring points is obtained and X is used. (0) Indicates that, X (0) ={X (0 )(1), X (0) (2),…,X (0) (i),…,X (0) (k)}; where k = 1, 2, …, i, …, n, n is a positive integer, and i is a positive integer; Using a one-time accumulation algorithm, the original hydrogen concentration value set X (0) Each original hydrogen concentration value in is preprocessed one by one to obtain the original cumulative hydrogen concentration value set, and X is used to calculate the total hydrogen concentration value. (1) Indicates that, X (1) ={X (1) (1), X (1) (2),…,X (1) (i),…,X (1) (k)}, where k = 1, 2, …, i, …, n, n is a positive integer, and i is a positive integer; in, j is a positive integer; For the original cumulative hydrogen concentration value set X (1) Each original accumulated hydrogen concentration value in is adjusted to obtain a set of adjusted hydrogen concentration values, and X is used to calculate the hydrogen concentration value. (2) Indicates that, X (2) ={d1X (1) (1), d2X (1) (2),…,d i X (1) (i), ..., d k X (1) (k)}, d i is the influence factor of the corresponding adjustment hydrogen concentration value; The corresponding influence factor d for adjusting the hydrogen concentration value i The calculation method is: Establish the influence factor d for adjusting the hydrogen concentration value according to the temperature at the corresponding detection point i Mathematical model: Wherein, T(i) is the temperature at the corresponding detection point, g1 is the first specific gravity coefficient, g2 is the temperature proportional coefficient, g3 is the second specific gravity coefficient, g4 is the temperature variation coefficient, and g5 is the temperature correction coefficient; The hydrogen concentration value set X will be adjusted (2) Each of the adjusted hydrogen concentration values ​​is taken as the initial value and brought into the improved GM (1, 1) model to obtain the cumulative predicted hydrogen concentration value. Predict hydrogen concentration value based on accumulation Obtain the corresponding effective original hydrogen concentration prediction value for: Where μ(i) is the predicted value of the effective original hydrogen concentration Correction factor for According to the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Calculate the residual e(i); Compare e(i) with the preset standard value λ to determine the abnormal value of hydrogen concentration measurement. When e(i)>λ, the corresponding X (0) (i) is an abnormal value and is replaced by the corresponding value X″′ (0) (i) Replace abnormal values ​​to realize the judgment and correction of abnormal hydrogen concentration values.

2. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 1, characterized in that: The corresponding replacement value X″′ (0) The calculation method for (i) is: According to the predicted value of effective original hydrogen concentration Established X″′ (0) (i) The calculation model is: Where β(i) is the corresponding replacement value X″′ (0) (i) The effective coefficient.

3. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 2, characterized in that: The corresponding replacement value X″′ (0) The calculation method of the effective coefficient β(i) of (i) is: According to the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Create the corresponding replacement value X″′ (0) The mathematical model of the effective coefficient β(i) of (i) is: in, is the original hydrogen concentration value X (0) The average value of (i), is the predicted value of the effective original hydrogen concentration The average value of .

4. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 1, characterized in that: The hydrogen concentration value set X will be adjusted (2) Each of the adjusted hydrogen concentration values ​​is taken as the initial value and brought into the improved GM (1, 1) model in turn, and the cumulative predicted hydrogen concentration value is obtained as follows: Among them, a is the development coefficient and b is the gray action.

5. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 4, characterized in that: The calculation method of the development coefficient a and the ash action amount b is: to estimate them using the principle of the least square method.

6. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 1, characterized in that: The calculation method of the correction factor μ(i) of the effective original hydrogen concentration prediction value is: The mathematical model of the correction factor μ(i) of the effective original hydrogen concentration prediction value is established according to the methane concentration at the corresponding detection point: Wherein, H(i) is the methane concentration at the corresponding detection point, f1 is the sine coefficient, f2 is the first proportional coefficient of the methane concentration, f3 is the first correction coefficient of the methane concentration, f4 is the exponential coefficient, f5 is the second proportional coefficient of the methane concentration, and f6 is the second correction coefficient of the methane concentration.

7. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 1, characterized in that: The calculation method of the preset standard value λ is: According to the original hydrogen concentration value X (0) (i) and the predicted value of effective original hydrogen concentration Establish a mathematical model for the preset standard value λ:

8. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 2, characterized in that: The corresponding replacement value X″′ (0) The effective coefficient of (i) β(i)=1.

9. The method for correcting abnormal values ​​of hydrogen concentration measurement according to claim 1, characterized in that: The calculation method of the preset standard value λ is: The mathematical model for establishing the preset standard value λ is: Among them, M is the learning coefficient determined by the BP neural network method.

10. A device for correcting abnormal values ​​of hydrogen concentration measurement, characterized in that: The hydrogen concentration measurement abnormal value correction device is used to execute the hydrogen concentration measurement abnormal value correction method according to any one of claims 1 to 9.

Images