High-speed sampling measurement data processing method based on modified MT filtering algorithm

Abstract

The invention provides a high-speed sampling measurement data processing method based on a modified MT filtering algorithm, and the method comprises the steps: firstly carrying out the prediction and judgment of outliers through employing a forward differential polynomial, and carrying out the processing of received test data; effective data processed by a seven-point second-order forward differential polynomial is further processed by using an improved Kalman filtering algorithm, noise in original data is identified and replaced, and rapid filtering and processing of the noise are realized. The technical problem that in the prior art, due to the fact that a traditional kalman filtering algorithm lacks anti-interference performance on measured data noise, a data analysis result seriously deviates from the real state of a system is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a high-speed sampling measurement data processing method based on a modified MT filtering algorithm. Background Art

[0002] The telemetry test data obtained from flight tests is an important part of flight tests. The quality of real-time processing and post-processing of test data directly affects the judgment of the success or failure of the test. However, when acquiring data, even if high-precision measurement equipment is used, the measurement data often contains 1% to 2% abnormal data due to environmental influences, telemetry equipment failures, process interference, and human factors. In severe weather conditions, 10% to 20% of the data deviates seriously from the true value. The existence of these abnormal data, i.e., noise, brings large errors to the estimation of the actual flight state of the aircraft, navigation, and flight performance analysis, reduces the credibility of the test data, and brings great difficulties to the subsequent data analysis and processing. Therefore, when processing data, it is first necessary to quickly identify the noise in the test data and replace it with reasonable and reliable data to ensure the correctness of the test data analysis results. The traditional Kalman filter algorithm lacks the anti-interference ability of the measurement data noise. Even if it contains a small amount of burrs, it will have a serious adverse effect on the Kalman filter effect, and even cause the data analysis results to deviate seriously from the true state of the system. Summary of the invention

[0003] The present invention provides a high-speed sampling measurement data processing method based on a modified MT filtering algorithm, which can solve the technical problem in the prior art that the traditional Kalman filtering algorithm lacks anti-interference ability to measurement data noise, resulting in data analysis results seriously deviating from the actual state of the system.

[0004] According to one aspect of the present invention, a method for processing high-speed sampling measurement data based on a modified MT filtering algorithm is provided, the method comprising:

[0005] S1, establish an improved Kalman filter algorithm and initialize the corresponding filter;

[0006] S2, first collect the first six non-wild point data y in the original collected data 1 ~y 6 And store it in the outlier elimination data buffer;

[0007] S3, based on the first six data stored in the outlier elimination data buffer, use the 7-point second-order forward differential polynomial sliding fitting to determine the original collected data y at the current moment iIs it an outlier? If not, it is stored in the outlier elimination data buffer using the first-in-first-out principle and goes to S4. If yes, it is eliminated and goes to S5, where i is a positive integer greater than 7;

[0008] S4, the original collected data y at the current moment i Normal output;

[0009] S5, judging whether the data stored in the current outlier elimination data buffer reaches 7, and when the data stored in the current outlier elimination data buffer reaches 7, using the improved Kalman filter algorithm to filter the 7 data in the current outlier elimination data buffer to obtain corresponding filtered data, and using the filtered data to replace the original collected data y at the current moment i The filtered data is then output and used to replace the seventh data in the current outlier elimination data buffer for the next Kalman filtering.

[0010] Furthermore, based on the first six data stored in the outlier elimination data buffer, the original collected data y at the current moment is determined by sliding fitting using a 7-point second-order forward differential polynomial. i Whether it is a wild point, including:

[0011] The interpolated data of the first six data stored in the outlier elimination data buffer and the original collected data y at the current moment are calculated in sequence using the 7-point second-order forward differential polynomial. i The interpolated data of

[0012] According to the original collected data y at the current moment i And the corresponding interpolated data calculates the corresponding new information;

[0013] Determine whether the absolute value of the calculated new information is greater than the wild point threshold. If so, determine whether the original collected data y at the current moment i is an outlier. If not, then determine the original collected data y at the current moment i Not for wild spots.

[0014] Furthermore, the interpolated data of the first six data stored in the outlier elimination data buffer and the original collected data y at the current moment are calculated in sequence using the 7-point second-order forward differential polynomial through the following formula: i The interpolated data:

[0015]

[0016] In the above formula, The interpolated data of the first six data stored in the wild point elimination data buffer, for y i The interpolated data.

[0017] Furthermore, according to the original collected data y at the current moment i And the corresponding interpolated data calculates the corresponding new information:

[0018]

[0019] In the above formula, v i For new information.

[0020] Furthermore, the outlier threshold is determined by the following formula:

[0021]

[0022] In the above formula, E represents the outlier threshold.

[0023] Furthermore, the method further comprises:

[0024] When S3 determines the original collected data y at the current moment i When it is an outlier, determine the original collected data y for the next m consecutive moments i+m Respectively with the original collected data y at the previous moment i+m-1 Is the absolute value of the difference less than the wild point threshold? If so, determine whether y i+m is a wild point, if not, then judge y i+m Not for wild spots.

[0025] Furthermore, the method further comprises: when the original collected data y at consecutive m moments i+m When all are wild points, it is considered that the original collected data y at the m+1th moment i+m+1 Not for wild spots.

[0026] Furthermore, m=3.

[0027] By applying the technical solution of the present invention, a high-speed sampling measurement data processing method based on a modified MT filtering algorithm is provided. The method combines a 7-point second-order forward differential polynomial sliding fitting with an improved Kalman filtering algorithm. First, the forward differential polynomial is used to predict and judge wild points, and the received test data is processed. The effective data processed by the 7-point second-order forward differential polynomial is further processed using the improved Kalman filtering algorithm. The noise in the original data is identified and replaced, and rapid filtering and processing of the noise is achieved, that is, abnormal information in the test data is identified and replaced with reasonable and reliable data, the noise in the data can be effectively and quickly eliminated, and the continuity and validity of the data after the noise is eliminated can be reconstructed, the data error is reduced, the burrs in the data are smoothed, the convergence of the Kalman filter can be guaranteed when a large amount of noise occurs, part of the noise is effectively replaced and predicted, the reliability of the data processing results is improved, and real and effective test results are obtained. When the process data model is unknown, this method can be used to quickly filter out nonlinear and complex distributed noise in the data during flight test data processing, and has a very good processing effect. The processed data can better reflect the actual operating status of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The included drawings are used to provide a further understanding of the embodiments of the present invention, which constitute a part of the specification, are used to illustrate the embodiments of the present invention, and together with the text description, explain the principles of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 The flowchart of the high-speed sampling measurement data processing method based on the modified MT filtering algorithm provided according to the specific embodiment of the present invention is shown. DETAILED DESCRIPTION

[0030] It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0032] Unless otherwise specifically stated, the relative arrangement of the parts and steps described in these embodiments, numerical expressions and numerical values ​​do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be regarded as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0033] According to a specific embodiment of the present invention, a high-speed sampling measurement data processing method based on a modified MT filtering algorithm is provided, the method comprising:

[0034] S1, establish an improved Kalman filter algorithm and initialize the corresponding filter;

[0035] S2, first collect the first six non-wild point data y in the original collected data 1 ~y 6 And store it in the wild point elimination data buffer, that is, the buffer size is 7, the first six non-wild point data y 1 ~y 6 It is a tested normal point;

[0036] S3, based on the first six data stored in the outlier elimination data buffer, use the 7-point second-order forward differential polynomial sliding fitting to determine the original collected data y at the current moment i Is it an outlier? If not, it is stored in the outlier elimination data buffer using the first-in-first-out principle and goes to S4. If yes, it is eliminated and goes to S5, where i is a positive integer greater than 7, that is, the value range of i is [7, N];

[0037] S4, the original collected data y at the current moment i Normal output;

[0038] S5, judging whether the data stored in the current outlier elimination data buffer reaches 7, and when the data stored in the current outlier elimination data buffer reaches 7, using the improved Kalman filter algorithm to filter the 7 data in the current outlier elimination data buffer to obtain corresponding filtered data, and using the filtered data to replace the original collected data y at the current moment i The filtered data is then output and used to replace the seventh data in the current outlier elimination data buffer for the next Kalman filtering.

[0039] By applying this configuration, a high-speed sampling measurement data processing method based on the modified MT filtering algorithm is provided. The method combines the sliding fitting of the 7-point second-order forward differential polynomial and the improved Kalman filtering algorithm. First, the forward differential polynomial is used to predict and judge the wild points, and the received test data is processed. The effective data processed by the 7-point second-order forward differential polynomial is further processed by the improved Kalman filtering algorithm, and the noise in the original data is identified and replaced to achieve rapid filtering and processing of the noise, that is, the abnormal information in the test data is identified and replaced with reasonable and reliable data, the noise in the data can be effectively and quickly eliminated, and the continuity and validity of the data after the noise is eliminated can be reconstructed, the data error can be reduced, the burrs in the data can be smoothed, the convergence of the Kalman filter can be guaranteed when a large amount of noise appears, part of the noise can be effectively replaced and predicted, the reliability of the data processing results can be improved, and real and effective test results can be obtained. When the process data model is unknown, the method is used to quickly filter out nonlinear and complex distributed noise in the data during flight test data processing, which has a very good processing effect. The processed data can better reflect the actual operating status of the aircraft. Compared with the prior art, the technical solution of the present invention can solve the technical problem that the traditional Kalman filtering algorithm in the prior art lacks the anti-interference ability to the measurement data noise, resulting in the data analysis results seriously deviating from the real state of the system.

[0040] Furthermore, in the embodiment of the present invention, based on the first six data stored in the outlier elimination data buffer, a 7-point second-order forward differential polynomial sliding fitting is used to determine the original collected data y at the current moment. i Whether it is a wild point, including:

[0041] The interpolated data of the first six data stored in the outlier elimination data buffer and the original collected data y at the current moment are calculated in sequence using the 7-point second-order forward differential polynomial. i The interpolated data of

[0042] According to the original collected data y at the current moment i And the corresponding interpolated data calculates the corresponding new information;

[0043] Determine whether the absolute value of the calculated new information is greater than the wild point threshold. If so, determine whether the original collected data y at the current moment i is an outlier. If not, then determine the original collected data y at the current moment i Not for wild spots.

[0044] As a specific embodiment of the present invention, the interpolated data of the first six data stored in the outlier elimination data buffer and the original collected data y at the current moment are calculated in sequence by using the 7-point second-order forward differential polynomial through the following formula: i The interpolated data:

[0045]

[0046] In the above formula, is the interpolated data of the first six data stored in the outlier elimination data buffer. Formula (1) is calculated point by point in chronological order. for y i The interpolated data of the current moment y i That is, the data currently collected using the cycle.

[0047] Further, in the embodiment of the present invention, according to the original collected data y at the current moment i And the corresponding interpolated data calculates the corresponding new information:

[0048]

[0049] In the above formula, v i For new information.

[0050] In addition, in the embodiment of the present invention, the outlier threshold is determined by the following formula:

[0051]

[0052] In the above formula, E represents the outlier threshold.

[0053] In actual flight tests, the values ​​of consecutive wild points are relatively close. Therefore, in the embodiment of the present invention, when the original collected data y at the current moment is determined in S3, i When it is an outlier, determine the original collected data y for the next m consecutive moments i+m Respectively with the original collected data y at the previous moment i+m-1 Is the absolute value of the difference less than the wild point threshold? If so, determine whether y i+m is a wild point, if not, then judge y i+mThat is, when the data of the i-th cycle is an outlier, the value of the i+1-th cycle that satisfies formula (4) is also an outlier. Specifically, formula (4) is used to eliminate continuous outliers:

[0054] |y i+m -y i+m-1 |<E (4)

[0055] Here, m is the number of continuous wild points. In general, the number of continuous wild points in flight tests rarely exceeds 4. Therefore, in the embodiment of the present invention, m=3, that is, the maximum number of continuous wild points is 3. In this way, normal step signals can be avoided from being eliminated as wild points.

[0056] Furthermore, in the embodiment of the present invention, when the original collected data y at consecutive m moments i+m When all are wild points, it is considered that the original collected data y at the m+1th moment i+m+1 That is, when the number of consecutive wild points in the flight data exceeds 3, it is considered that the original y of the i+4th period i+4 It is not a wild point, and the number of consecutive wild points returns to zero.

[0057] The first-in-first-out principle is used to continuously update the data buffer for wild point elimination, and the rationality of the data is judged. According to step 2, it is determined whether the current cycle data is a wild point. If the current cycle data is noise, it is eliminated and the subsequent calculation is performed according to the output of the previous set of data in the current cycle. Otherwise, the data is considered to be normal data for subsequent calculation.

[0058] The above steps are processed in a loop. According to the first-in-first-out stack principle, the content of the data buffer is updated continuously. The data in the data buffer is the valid data currently received. The noise of up to 3 consecutive cycles in the original data can be effectively removed and replaced. However, long continuous noise cannot be processed. Some data still have glitches and need further correction.

[0059] Therefore, the present invention adopts the improved Kalman filtering algorithm combined with 7-point second-order forward differential polynomial sliding fitting to further correct the data.

[0060] The Kalman filter algorithm can effectively smooth and estimate the data. It is the optimal estimation method for linear systems under the minimum mean square error criterion, and can make linear and unbiased estimates at the same time. It uses the concept of state space, takes the data processing process as a linear output, and uses the state equation to describe the input-output relationship. With the help of the state transfer equation of the data, it predicts the estimated value at the current moment based on the estimated output value at the previous moment, and filters and corrects the predicted estimate through the error with the current input, outputs the estimated output at the current moment, and obtains the filtering result of each point through the continuous prediction-correction linear recursive method.

[0061] State estimation using Kalman filtering can be divided into three steps: initialization, state prediction, and state update.

[0062] A) Initialize the Kalman filter

[0063] First, the state equation and observation equation of the linear dynamic measurement system are established as follows:

[0064] Equation of state:

[0065] X t =φ t X t-1 +ω t (5)

[0066] Observation equation:

[0067] Y t =H t X t +υ t (6)

[0068] Where: t is the discrete time; X t , X t-1 Respectively represent the state vector at time t and time t-1; Y t is the state observation vector at time t; φ t is the state transfer matrix; H t is the observation matrix; ω t is the state noise vector, υ t is the observation noise vector, ω t and t Independent of each other.

[0069] B) Using 7-point second-order forward differential polynomial sliding fitting to remove single-point isolated noise and perform state prediction on the input raw data;

[0070] Based on the state equation and observation equation of the above system, the following Kalman filter equation can be established:

[0071] From the known state vector at time t-1 To predict the state at time t,

[0072] The single-step prediction state equation is:

[0073]

[0074] The single-step forecast error covariance matrix is:

[0075]

[0076] The filter gain matrix is ​​calculated as:

[0077]

[0078] Using the observed value at the tth moment, the predicted result is corrected to obtain the state filtering equation:

[0079]

[0080] in, is the sample information, Q t-1 represents the process noise covariance matrix, K t is the filter gain, see formula 9, R t represents the measurement noise covariance matrix.

[0081] At the same time, the state filter error covariance matrix at time t can be obtained:

[0082]

[0083] The above is the standard Kalman filter process. From formula 10, we can see that when the new input raw data (that is, the newly added sample point Y t ) The Kalman filter is estimated in the form of linear combination. When the new sample point Y t When it is not an outlier, the sample new information can correctly correct the filter estimate at time t, and the correction factor is the filter gain K t ; When the new sample point Y t When it is isolated noise, the outliers can be eliminated by sliding fitting the 7-point second-order forward differential polynomial based on the outlier elimination data buffer to make a correct correction to the filter estimate at time t; when the new sample point Y t It is continuous noise, bringing abnormal new information K t The error correction of the system state forecast is performed multiple times, causing the filter estimation value to be abnormally offset. To address this problem, in the embodiment of the present invention, the filter gain matrix K of the Kalman filter is t Further improvement is made by directly using the one-step prediction value as the state vector at time t At the same time, the error covariance matrix P t|t The corresponding improvement is also made, using the observed value Y at the tth moment t The prediction results are corrected and the improved state prediction equations are derived as shown in Formulas 12 and 13.

[0084]

[0085] in, Indicates P t|t-1 The first column of the matrix.

[0086] and Defined as a function, the function form is defined as:

[0087]

[0088] C) state update to obtain the best estimate of the newly measured test data;

[0089] From this, we can get the innovation equation and innovation variance of the Kalman filter:

[0090] Innovation equation:

[0091] Innovation Variance:

[0092] When the improved Kalman filter algorithm is used to filter the test data, the new value of the filter is detected in real time according to the new sample value (ie, new data). and the innovation variance C t , to achieve smooth filtering and elimination of measurement data. From formula 10, it can be seen that the new value of the new added sample point (that is, the original data at the current moment) is It is the Kalman filter input, which is the Kalman filter estimation in the form of linear combination As a function of the prediction error, its statistical properties have a decisive influence on the statistics of the Kalman filter output. t Under normal circumstances, the new information correctly corrects the filter estimate at time t, and the correction factor is the filter gain K t ; When the new sample Y t When there is an abnormality (i.e. noise), the abnormal information will also be K t The correction multiple of the Kalman filter state prediction error is corrected, causing serious distortion of the filter estimate.

[0093] In summary, the main steps of the high-speed sampling measurement data processing method based on the modified MT filtering algorithm provided by the present invention can be referred to Figure 1 The specific implementation of each step has been described in detail in the above embodiments, and will not be described in detail here. It should be known to those skilled in the art that this example is only an application mode to help understand the high-speed sampling measurement data processing method based on the modified MT filtering algorithm provided by the present invention, and does not limit it in any way.

[0094] like Figure 1As shown in the figure, the improved Kalman filtering algorithm is combined with the 7-point second-order forward differential polynomial sliding fitting, and the advantages of both are used to quickly remove the noise in the test data. The process is as follows:

[0095] 1) Initialize the Kalman filter;

[0096] 2) The input raw data is fitted by a 7-point second-order forward differential polynomial sliding fit to eliminate single-point isolated noise;

[0097] 3) Optimal estimation of newly measured test data. The 7-point second-order forward differential polynomial sliding fitting process and Kalman filtering algorithm are combined to process the input data, obtain the estimated value of the current cycle in real time, and predict the estimated value of the next cycle. When the data is normal, it is used as a normal result for the user; when the data is noisy, the memory function of the algorithm, that is, the final data in the data buffer, is used to estimate the output data with the state parameters at the time when the noise appears, and the estimated result is provided to the user.

[0098] In summary, the present invention provides a high-speed sampling measurement data processing method based on a modified MT filtering algorithm. The method combines a 7-point second-order forward differential polynomial sliding fitting with an improved Kalman filtering algorithm. First, a forward differential polynomial is used to predict and judge wild points, and the received test data is processed. The effective data processed by the 7-point second-order forward differential polynomial is further processed using the improved Kalman filtering algorithm. The noise in the original data is identified and replaced, and rapid filtering and processing of the noise is achieved, that is, the abnormal information in the test data is identified and replaced with reasonable and reliable data, the noise in the data can be effectively and quickly eliminated, and the continuity and validity of the data after the noise is eliminated can be reconstructed, the data error is reduced, the burrs in the data are smoothed, and the convergence of the Kalman filter can be guaranteed when a large amount of noise occurs. Part of the noise is effectively replaced and predicted, the reliability of the data processing results is improved, and real and effective test results are obtained. When the process data model is unknown, the method is used to quickly filter out nonlinear and complex distributed noise in the data during flight test data processing, which has a very good processing effect. The processed data can better reflect the actual operating status of the aircraft. Compared with the prior art, the technical solution of the present invention can solve the technical problem that the traditional Kalman filtering algorithm in the prior art lacks the anti-interference ability to the measurement data noise, resulting in the data analysis results seriously deviating from the real state of the system.

[0099] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0100] In addition, it should be noted that the use of terms such as "first" and "second" to limit components is only for the convenience of distinguishing the corresponding components. If not otherwise stated, the above terms have no special meaning and therefore cannot be understood as limiting the scope of protection of the present invention.

[0101] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A high-speed sampling measurement data processing method based on a modified MT filtering algorithm, characterized in that: The method comprises: S1, establish an improved Kalman filter algorithm and initialize the corresponding filter; S2, first collect the first six non-wild point data y1-y6 in the original collected data and store them in the wild point elimination data buffer; S3, based on the first six data stored in the outlier elimination data buffer, use the 7-point second-order forward differential polynomial sliding fitting to determine the original collected data y at the current moment i Is it an outlier? If not, it is stored in the outlier elimination data buffer using the first-in-first-out principle and goes to S4. If yes, it is eliminated and goes to S5, where i is a positive integer greater than 7; S4, the original collected data y at the current moment i Normal output; S5, judging whether the data stored in the current outlier elimination data buffer reaches 7, and when the data stored in the current outlier elimination data buffer reaches 7, using the improved Kalman filter algorithm to filter the 7 data in the current outlier elimination data buffer to obtain corresponding filtered data, and using the filtered data to replace the original collected data y at the current moment i The filtered data is then output and used to replace the seventh data in the current outlier elimination data buffer for the next Kalman filtering.

2. The method according to claim 1, characterized in that Based on the first six data stored in the outlier elimination data buffer, the original collected data y at the current moment is determined by sliding fitting using a 7-point second-order forward differential polynomial. i Whether it is a wild point, including: The interpolated data of the first six data stored in the outlier elimination data buffer and the original collected data y at the current moment are calculated in sequence using the 7-point second-order forward differential polynomial. i The interpolated data of According to the original collected data y at the current moment i And the corresponding interpolated data calculates the corresponding new information; Determine whether the absolute value of the calculated new information is greater than the wild point threshold. If so, determine whether the original collected data y at the current moment i is an outlier. If not, then determine the original collected data y at the current moment i Not for wild spots.

3. The method according to claim 2, characterized in that The interpolated data of the first six data stored in the outlier elimination data buffer and the original collected data y at the current moment are calculated in sequence using the 7-point second-order forward differential polynomial through the following formula: i The interpolated data: In the above formula, The interpolated data of the first six data stored in the wild point elimination data buffer, for y i The interpolated data.

4. The method according to claim 3, characterized in that According to the original collected data y at the current moment i And the corresponding interpolated data calculates the corresponding new information: In the above formula, v i For new information.

5. The method according to claim 2, characterized in that: The outlier threshold is determined by the following formula: In the above formula, E represents the outlier threshold.

6. The method according to claim 1, characterized in that The method further comprises: When S3 determines the original collected data y at the current moment i When it is an outlier, determine the original collected data y for the next m consecutive moments i+m The original data y of the previous moment are i+m-1 Is the absolute value of the difference less than the wild point threshold? If so, determine whether y i+m is a wild point, if not, then judge y i+m Not for wild spots.

7. The method according to claim 6, characterized in that The method further comprises: when the original collected data y at consecutive m moments i+m When all are wild points, it is considered that the original collected data y at the m+1th moment i+m+1 Not for wild spots.

8. The method according to claim 7, characterized in that m＝3。

Images