A prediction method for the probability of occurrence of FOES in summer in China

Abstract

This invention discloses a method for predicting the probability of summer foEs occurrence in China, comprising the following steps: Step 1, collecting foEs observation data from various ionospheric observation stations in China and preprocessing the foEs data; Step 2, classifying the foEs data and calculating the probability of occurrence of different foEs values ​​for a single observation station in summer; Step 3, performing linear fitting between the foEs value and the foEs occurrence probability to obtain the fitting coefficient of the Philips rule for a single station; Step 4, performing polynomial fitting between the foEs value and the foEs occurrence probability, and using the prediction model to calculate the probability of foEs occurrence at any location in China. The method disclosed in this invention can predict the probability of foEs occurrence at any location in China using ionospheric vertical measurement data, providing good technical support for the design, operation, and maintenance of radio systems.

Description

Technical Field

[0001] The invention belongs to the research field of ionospheric environment monitoring, and particularly relates to a prediction method for the probability of occurrence of FOES in summer in China in the field. Background Art

[0002] Ionospheric parameters description:

[0003] Es: The sudden E layer in the ionospheric vertical measurement circuit diagram, that is, the sudden inhomogeneous structure of the E region.

[0004] foEs: The top frequency of the ordinary wave component of the Es layer continuously traced in the ionospheric vertical measurement circuit diagram.

[0005] h'Es: is the lowest false height of the foEs trace.

[0006] The ionosphere Es (Sporadic-E) layer is a short-lived, localized, relatively high-electron-density irregular patch in the E region of the ionosphere that seriously affects radio wave propagation. It is named because of its irregular appearance time. The appearance of the Es layer can cause reflection and forward scattering of radio waves in the HF or VHF band, so that communications that are usually only possible within the line-of-sight range can suddenly expand to more than 2000 km, realizing beyond-line-of-sight communications.

[0007] ITU-R P.534-5 conducted a statistical analysis of the occurrence probability of the Es layer in mid- and low-latitude regions such as Europe, Africa, America, and Asia (Far East), and gave the foEs values ​​that are equal to or exceeded for different time percentages of the foEs occurrence probability in summer, spring, autumn, and winter, as well as the time percentages of the sporadic E layer (foEs) with vertical incidence equal to or exceeding 7MHz in different seasons in mid- and low-latitude regions. It is believed that summer, daytime, and the Far East are the high-incidence seasons, times, and regions of the global Es layer. ITU-R P.534-5 plotted the curve of the occurrence probability of the Es layer with the season, time, and solar activity, and found that the critical frequency foEs of the Es layer and its probability of occurrence or exceeding the time conform to the Philips "frequency dependence" rule. This rule is completely empirical and applies to the critical frequency foEs of the mid-latitude region with a time percentage of less than 30% and a frequency higher than the conventional E layer.

[0008] Due to the lack of foEs data in China, the prediction accuracy of foEs occurrence probability is not high. In practical applications, the definition of foEs observation value and foEs value of given occurrence rate are different. foEs observation value cannot be simply used to replace foEs value of given occurrence rate. The prediction accuracy of sporadic E-layer occurrence probability is difficult to meet actual needs. Summary of the invention

[0009] The technical problem to be solved by the present invention is to use the ionospheric vertical measurement data in China to statistically analyze the variation characteristics of the probability of occurrence of summer foEs with longitude and latitude, and to provide a prediction method for the probability of occurrence of summer foEs in China.

[0010] The present invention adopts the following technical solution:

[0011] A method for predicting the probability of occurrence of FOES in summer in China, the improvement of which is that it includes the following steps:

[0012] Step 1: Collect foEs observation data from various ionospheric observation stations in China and preprocess the foEs data:

[0013] Step 11, using ionospheric vertical detection technology to obtain the ionospheric vertical frequency and height map of each ionospheric observation station, and read the h'Es and foEs values ​​of the vertical ionosphere;

[0014] Step 12, preprocessing the foEs historical measured meteorological data, parsing the foEs data and the Es layer false high data, and eliminating invalid data;

[0015] Step 2: Classify the foEs data and calculate the occurrence probability of different foEs values ​​in summer at a single observation station:

[0016] Step 21, classifying the foEs data according to season to form a foEs summer data set;

[0017] Step 22, calculate the maximum value of foEs of each observation station data set, and calculate the occurrence probability of foEs at a step frequency of 1 MHz starting from 1 MHz;

[0018] The probability of foEs occurrence is defined as:

[0019]

[0020] In the above formula, N represents the total number of valid samples; N(foEs≥f) represents the number of times the foEs value is not less than the f value, where the f value ranges from 1MHz to the maximum value of foEs;

[0021] Step 3: Perform a linear fit on the foEs value and the probability of foEs occurrence to obtain the fitting coefficient of the single-station Philips rule:

[0022] Step 31, pre-processing the foEs occurrence probability, eliminating the same foEs occurrence probability value and its foEs value;

[0023] Step 32, performing linear fitting on the foEs value and the foEs occurrence probability to calculate the fitting parameters;

[0024] The content of Philips rules is as follows:

[0025] lgp=a+bf

[0026] In the above formula, p is the probability of occurrence of foEs≥f; f is the frequency; a and b are coefficients;

[0027] Step 4: Perform polynomial fitting on the foEs value and the probability of foEs occurrence, establish a single observation station foEs occurrence probability prediction model, and use the prediction model to calculate the probability of foEs occurrence at any location in China:

[0028] Step 41, fitting the relationship between the fitting coefficient of each observation station and the change of longitude and latitude, and establishing the relationship between the fitting coefficient of each observation station and the change of longitude and latitude;

[0029] Assume that coefficients a and b are linear functions of latitude and longitude:

[0030]

[0031] Through linear fitting, the occurrence probability prediction model of foEs≥f in summer in China is obtained as follows:

[0032] lgp=a(lat,lon)+b(lat,lon)f

[0033] In the above formula, ka1 = 0.0226, ka2 = -0.0049, kb1 = -0.0013, kb2 = -9.1443 × 10 -4 , Lon represents longitude, Lat represents latitude;

[0034] Step 42, establishing a prediction model for the probability of occurrence of foEs in summer in China, and using the prediction model to calculate the probability of occurrence of foEs values ​​at any location in China.

[0035] Further, in step 21, summer is from May to August.

[0036] Furthermore, the forecast absolute error AE and root mean square error RMS are used to characterize the quality of the forecast results, which are specifically defined as:

[0037] AE=∑|AB|×100%

[0038]

[0039] In the above formula, A and B are the measured and predicted values ​​of the probability of occurrence of a certain foEs at a specific observation station, respectively.

[0040] The beneficial effects of the present invention are:

[0041] The method disclosed in the present invention can utilize the ionospheric vertical measurement data in China, classify the foEs data according to seasons, and calculate the occurrence probability of different foEs values ​​in summer at observation stations such as Beijing and the fitting coefficients of the Philips rule. By fitting the fitting coefficients of each observation station with the longitude and latitude parameters, a prediction model for the occurrence probability of foEs in summer in China is established, and the occurrence probability of foEs values ​​at any location in China is predicted, which can provide good technical support for the design and operation and maintenance of radio systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic flow diagram of the method of the present invention;

[0043] Figure 2 It is a vertical survey map of the ionosphere;

[0044] Figure 3 This is a graph showing the change in the number of Es occurrences with latitude in Manchuria, Beijing, Xi'an, Chongqing and Kunming in 2011;

[0045] Figure 4 This is a graph showing the change in the number of Es occurrences with latitude in Manchuria, Beijing, Xi'an, Chongqing and Kunming in 2016;

[0046] Figure 5 This is a graph showing the change in the number of Es occurrences with longitude in Xinjiang Uygur Autonomous Region, Lanzhou, Xi'an, Xinxiang and Qingdao in 2011;

[0047] Figure 6 This is a graph showing the change in the number of Es occurrences with longitude in Xinjiang Uygur Autonomous Region, Lanzhou, Xi'an, Xinxiang and Qingdao in 2016;

[0048] Figure 7 is the probability map of foEs occurrence in summer for Beijing, Guangzhou, Manchuria and Urumqi;

[0049] Figure 8 It is the fitting curve diagram of the occurrence probability of foEs in Beijing, Guangzhou, Manchuria and Urumqi in summer. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0051] Example 1: This example collects the foEs observation data of Beijing, Changchun, Kunming, Xinxiang, Qingdao, Lanzhou, Urumqi, Chongqing, Guangzhou, Haikou, Xi'an and Manchuria over the years, analyzes the variation characteristics of the foEs occurrence probability with longitude and latitude, calculates the Philips coefficient of the Es layer occurrence probability of a single observation station, studies the Philips coefficient of the Es layer occurrence probability in the ionosphere in China, and establishes a prediction method for the occurrence probability of foEs in summer in China. Figure 1 As shown, the following steps are included:

[0052] Step 1: Collect foEs observation data from various ionospheric observation stations in China, pre-process foEs data, and analyze foEs statistical laws:

[0053] Step 11, using ionospheric vertical detection technology to obtain the ionospheric vertical frequency and height map of each ionospheric observation station, and read the h'Es and foEs values ​​of the vertical ionosphere;

[0054] Step 12, preprocess the foEs historical measured meteorological data, parse the foEs data and the Es layer false high data, remove invalid data, and ensure that the foEs data set is complete and valid;

[0055] Analysis of the variation characteristics of foEs occurrence probability with longitude and latitude:

[0056] The observation data for China is completed by relying on the 19 ionospheric observation stations in China of the National Defense Science and Technology Industry Radio Environment Observation Station Network. Figure 2 As shown in the figure, according to the frequency-height diagram provided by the vertical measurement equipment, the foEs data in the diagram is read by manual methods, and the diagrams with almost no recognizable traces are not adopted. In order to achieve better results, the foEs historical measured meteorological data are preprocessed to eliminate invalid data and ensure that the foEs data set is complete and valid.

[0057] like Figure 3 As shown in Figure 4, the ionosphere vertical measurement foEs data of Manzhouli, Beijing, Xi'an, Chongqing and Kunming stations in 2011 and 2016 are used to analyze the latitudinal variation characteristics of foEs in high and low years of solar activity. The number of occurrences of foEs not less than 4MHz shows a decreasing trend with increasing latitude, especially the number of occurrences at Manzhouli station is significantly reduced, which may be related to the high latitude of Manzhouli station;

[0058] like Figure 5As shown in Figure 6, the foEs data of Urumqi, Lanzhou, Xi'an, Xinxiang and Qingdao stations in Xinjiang Uygur Autonomous Region in 2011 and 2016 are used to analyze the longitude variation characteristics of foEs in high and low years of solar activity. In general, the number of foEs occurrences of not less than 4MHz at these five stations is not much different, while the number of foEs occurrences of not less than 7MHz at Urumqi, Lanzhou, Xinxiang and Qingdao stations is not much different, and there is no obvious longitude effect;

[0059] Step 2: Classify the foEs data and calculate the occurrence probability of different foEs values ​​in summer at a single observation station:

[0060] Step 21, classifying the foEs data according to season to form a foEs summer data set;

[0061] Step 22, calculate the maximum value of foEs of each observation station data set, and calculate the occurrence probability of foEs at a step frequency of 1 MHz starting from 1 MHz;

[0062] The probability of foEs occurrence is defined as:

[0063]

[0064] In the above formula, N represents the total number of valid samples; N(foEs≥f) represents the number of times the foEs value is not less than the f value, where the f value ranges from 1MHz to the maximum value of foEs;

[0065] The season is defined as May to August as summer, November to February as winter, and March, April, September, and October as spring and autumn. The foEs data are classified to form the foEs summer data set.

[0066] Define p as the probability of occurrence of foEs≥f, where f is the frequency (MHz). Take Beijing, Guangzhou, Urumqi, and Manchuria as examples. Figure 7 As shown, the probability of occurrence of foEs in summer is calculated.

[0067] Step 3: Perform a linear fit on the foEs value and the probability of foEs occurrence to obtain the fitting coefficient of the single-station Philips rule:

[0068] Step 31, pre-processing the foEs occurrence probability according to the foEs statistical law, and eliminating the same foEs occurrence probability value and its foEs value;

[0069] Step 32, performing linear fitting on the foEs value and the foEs occurrence probability to calculate the fitting parameters;

[0070] ITU-R P.534-5 plotted the probability of the Es layer appearing as a function of season, time of day, and solar activity, and found that the critical frequency foEs of the Es layer and the probability of its appearing or exceeding it conform to the Philips "frequency dependence" rule (the probability of foEs appearing is no more than 50%).

[0071] The content of Philips rules is as follows:

[0072] lgp=a+bf

[0073] In the above formula, p is the probability of occurrence of foEs≥f; f is the frequency (MHz); a and b are adjustable coefficients, and b is the slope.

[0074] The foEs occurrence probability of the Urumqi station was preprocessed, and the same foEs occurrence probability values ​​and their foEs values ​​were removed to improve the fitting effect. For example, for 15MHz and 16MHz, the foEs occurrence probability lgp value was -2.2037, and 16MHz and its corresponding foEs occurrence probability value were deleted. For 1MHz and 2MHz, foEs and foEs occurrence probability obviously did not conform to the linear fitting rule and were deleted. Considering the uniformity of data processing and fitting, the foEs value was calculated from 3MHz. After processing, the foEs value and foEs occurrence probability were linearly regressed according to the Philips rule. The root mean square error between the predicted value of foEs occurrence probability and the observation station was increased from 0.2359 to 0.2015, which improved the prediction accuracy.

[0075] According to the above rules, the foEs value and the probability of foEs occurrence are fitted. The fitting coefficients according to Philips rule are shown in Table 1. Figure 8 As shown in Figure 2, taking Beijing, Guangzhou, Urumqi and Manchuria as examples, the probability of occurrence of foEs in summer is calculated.

[0076] Table 1: Coordinates and regression coefficients of each observation station

[0077]

[0078]

[0079] Step 4: Perform polynomial fitting on the foEs value and the probability of foEs occurrence, establish a single observation station foEs occurrence probability prediction model, and use the prediction model to calculate the probability of foEs occurrence at any location in China:

[0080] Step 41, fitting the relationship between the fitting coefficient of each observation station and the change of longitude and latitude, and establishing the relationship between the fitting coefficient of each observation station and the change of longitude and latitude;

[0081] Assume that coefficients a and b are linear functions of latitude and longitude:

[0082]

[0083] Through linear fitting, the occurrence probability prediction model of foEs≥f in summer in China is obtained as follows:

[0084] lgp=a(lat,lon)+b(lat,lon)f

[0085] In the above formula, ka1 = 0.0226, ka2 = -0.0049, kb1 = -0.0013, kb2 = -9.1443 × 10 -4 , Lon represents longitude, Lat represents latitude, and the unit of longitude and latitude data is degree.

[0086] Because the regression coefficients a and b of the Urumqi station are obviously different from those of other stations, in order to maintain the validity and universality of the data, the Urumqi station data are not used to calculate the regression coefficients.

[0087] Step 42, establishing a prediction model for the probability of occurrence of foEs in summer in China, and using the prediction model to calculate the probability of occurrence of foEs values ​​at any location in China.

[0088] The forecast absolute error AE and root mean square error RMS are used to characterize the quality of the forecast results, which are specifically defined as:

[0089] AE=∑|AB|×100%

[0090]

[0091] In the above formula, A and B are the measured and predicted values ​​of the probability of occurrence of a certain foEs at a specific observation station, respectively, and N is the total number of samples.

[0092] Table 2: Philips rule parameter verification in the Es layer in summer

[0093]

[0094]

[0095] Table 3: Comparison of predicted values ​​and measured values ​​(I)

[0096]

[0097] Table 4: Comparison of predicted results and measured results (II)

[0098]

[0099] Considering all sample data, the predicted absolute error AE and root mean square error RMS are 5.41% and 5.42% respectively. When foEs is 9MHz and 11MHz, the predicted AE and RMS are 2.22% and 1.88% respectively, which meets the accuracy requirements.

Claims

1. A method for predicting the probability of occurrence of FOES in summer in China, characterized in that: The steps include: Step 1: Collect foEs observation data from various ionospheric observation stations in China and preprocess the foEs data: Step 11, using ionospheric vertical detection technology to obtain the ionospheric vertical frequency and height map of each ionospheric observation station, and read the h'Es and foEs values ​​of the vertical ionosphere; Step 12, preprocessing the foEs historical measured meteorological data, parsing the foEs data and the Es layer false high data, and eliminating invalid data; Step 2: Classify the foEs data and calculate the occurrence probability of different foEs values ​​in summer at a single observation station: Step 21, classifying the foEs data according to season to form a foEs summer data set; Step 22, calculate the maximum value of foEs of each observation station data set, and calculate the occurrence probability of foEs at a step frequency of 1 MHz starting from 1 MHz; The probability of foEs occurrence is defined as: In the above formula, N represents the total number of valid samples; N(foEs≥f) represents the number of times the foEs value is not less than the f value, where the f value ranges from 1MHz to the maximum value of foEs; Step 3: Perform a linear fit on the foEs value and the probability of foEs occurrence to obtain the fitting coefficient of the single-station Philips rule: Step 31, pre-processing the foEs occurrence probability, eliminating the same foEs occurrence probability value and its foEs value; Step 32, performing linear fitting on the foEs value and the foEs occurrence probability to calculate the fitting parameters; The content of Philips rules is as follows: lgp=a+bf In the above formula, p is the probability of occurrence of foEs≥f; f is the frequency; a and b are coefficients; Step 4: Perform polynomial fitting on the foEs value and the probability of foEs occurrence, establish a single observation station foEs occurrence probability prediction model, and use the prediction model to calculate the probability of foEs occurrence at any location in China: Step 41, fitting the relationship between the fitting coefficient of each observation station and the change of longitude and latitude, and establishing the relationship between the fitting coefficient of each observation station and the change of longitude and latitude; Assume that coefficients a and b are linear functions of latitude and longitude: Through linear fitting, the occurrence probability prediction model of foEs≥f in summer in China is obtained as follows: lgp=a(lat,lon)+b(lat,lon)f In the above formula, ka1 = 0.0226, ka2 = -0.0049, kb1 = -0.0013, kb2 = -9.1443 × 10 -4 , Lon represents longitude, Lat represents latitude; Step 42, establishing a prediction model for the probability of occurrence of foEs in summer in China, and using the prediction model to calculate the probability of occurrence of foEs values ​​at any location in China.

2. The method for predicting the probability of occurrence of foEs in summer in China according to claim 1, characterized in that: In step 21, summer is from May to August.

3. The method for predicting the probability of occurrence of foEs in summer in China according to claim 1, characterized in that: The forecast absolute error AE and root mean square error RMS are used to characterize the quality of the forecast results, which are specifically defined as: AE=∑|AB|×100% In the above formula, A and B are the measured and predicted values ​​of the probability of occurrence of a certain foEs at a specific observation station, respectively.

Images