Rail transit catenary operation vehicle-ground numerical arrangement device and method

Abstract

A rail transit catenary operation vehicle ground numerical arrangement device and method, a total column chart is constructed according to the numerical queue composed of the vehicle-ground numerical value, intelligent segmentation is performed according to the corresponding frequency of each numerical value in the total column chart; the total cluster-one numerical value and the total cluster-two numerical value of each sub-section domain are obtained according to each transformation variable group of each sub-section domain; the best transformation variable group of each sub-section domain is obtained according to the total cluster-one numerical value and the cluster-two numerical value in each sub-section domain, the emphasis-univariate regression is performed on each sub-section domain according to the best transformation variable group, and the univariate regression has a large difference between the corresponding adjustment frequency of each numerical value at the beginning and end and the starting point frequency in the numerical queue, so as to disturb the frequency of each numerical value in the source numerical value, increase the decoding complexity, and make the vehicle-ground numerical value in the storage process more protective and stable.

Description

Technical Field

[0001] This invention belongs to the field of vehicle-to-ground numerical data technology for rail transit catenary maintenance vehicles, and specifically relates to a device and method for processing vehicle-to-ground numerical data of rail transit catenary maintenance vehicles. Background Technology

[0002] As a means of transporting workers, the overhead contact line maintenance vehicle for rail transit needs to have good traction and braking performance, as well as good service facilities for workers to feel comfortable and convenient. Therefore, the vehicle plays an important role in the overhead contact line maintenance vehicle system of urban rail transit.

[0003] With the development of design and manufacturing technology for rail transit overhead contact line vehicles, various types of vehicles have emerged to meet the requirements of different line conditions and environments. Based on the matching relationship between the running gear and the track, the main vehicle types include steel wheel and rail vehicles, rubber-tired vehicles, monorail vehicles, linear motor vehicles, guide rail vehicles, and maglev vehicles.

[0004] In terms of data transmission between the vehicle and the ground for rail transit overhead contact line operation vehicles, the common practice is as described in a system using wireless optical communication for data transfer between the vehicle and the ground, as mentioned in patent number "202310116068.5". The onboard FSO terminal transmits the vehicle-to-ground data to the ground control center for storage via wireless network.

[0005] During the storage of vehicle-to-ground data, to ensure the protection of the corresponding data, modulation of the vehicle-to-ground data is necessary. Previous modulation methods commonly used normal distribution modulation, which modulates the data by analyzing the probability distribution of the values. However, this type of method can also be decoded using frequency rules, so using current modulation methods often creates the risk of data leakage and malicious access. Furthermore, even if univariate regression can be used to change the frequency of the values, previous univariate regression methods used the same parameters to process all values, without focusing on the real-time collected vehicle-to-ground data. This means that using previous univariate regression methods will result in unstable fluctuations due to different distributions of the source values.

[0006] Therefore, constructing a method that can perform targeted univariate regression based on the frequency of numerical values ​​to disrupt the modulation of the frequency of numerical values ​​is a crucial task. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention proposes a device and method for processing vehicle-to-ground data from a rail transit overhead contact line maintenance vehicle. The method involves constructing a total bar chart based on a data queue of vehicle-to-ground data, and intelligently segmenting the data according to the frequency of each value within the total bar chart. It then obtains all cluster 1 and cluster 2 values ​​for each sub-segment based on each transformation parameter group. Finally, it obtains the optimal transformation parameter group for each sub-segment based on the total cluster 1 and cluster 2 values. A univariate regression with a focus is then performed on each sub-segment based on the optimal transformation parameter group. This ensures that the adjusted frequency of each value at the beginning and end of the regression differs significantly from its initial frequency in the data queue, thereby disrupting the frequency of each value within its source data. This increases the complexity of decoding and enhances the protective stability of the vehicle-to-ground data storage process.

[0008] The present invention employs the following technical solution.

[0009] A method for processing vehicle-to-ground data of a rail transit overhead contact line maintenance vehicle includes:

[0010] The onboard FSO terminal transmits the vehicle-to-ground data of the rail transit overhead contact line maintenance vehicle to the ground control center via wireless network for storage. Storage methods include:

[0011] Step 1, obtain the first numerical queue;

[0012] Step 2: Obtain each sub-domain based on the numerical queue 1;

[0013] Step 3: Obtain the total cluster 1 values ​​and total cluster 2 values ​​for each sub-domain based on each transformation parameter group for each sub-domain, thereby obtaining the estimator for each transformation parameter group; obtain the optimal transformation parameter group for each sub-domain based on the estimator for each transformation parameter group.

[0014] Step 4: Obtain the modulation value based on the optimal transformation parameter group for each sub-domain; obtain each sub-region based on the guide queue of all cluster 2 values ​​and the byte modulation coefficients within each sub-domain; obtain the jumbled queue based on each sub-region; and store the jumbled queue and the modulation value.

[0015] Preferably, the method for obtaining the first numerical queue includes:

[0016] Obtain the vehicle and ground data for each point in time within a pre-defined time period. Obtain a data queue according to the order of the points in time and treat it as data queue one. At this time, there are redundant data in data queue one.

[0017] Preferably, the preset time period is 60 minutes.

[0018] Preferably, step 2 specifically includes:

[0019] Step 2-1: Construct a total bar chart based on the frequency of each value in the numerical queue. The X coordinate of the total bar chart represents the numerical content, and the Y coordinate represents the frequency of each value. Use the frequency of each value in the total bar chart as the starting frequency of each value.

[0020] Step 2-2: Treat the queue formed by each value on the X coordinate of the total as the second value queue; at this time, there are no redundant values ​​in the second value queue.

[0021] Steps 2-3: Based on the frequency of each value and its adjacent values ​​in the numerical queue 2, obtain each segment and use this to perform intelligent segmentation of the total using a bar chart.

[0022] Preferably, the method for performing intelligent segmentation of totals using a bar chart includes:

[0023] Initially, we calculate the interval between the frequencies of the p-th value and its adjacent values, defining q. p It represents the frequency of the p-th value within the total histogram, which is also the starting frequency of the p-th value. q p+1 It is the starting frequency of the (p+1)th value, q p-1 The value q is the starting frequency of the (p-1)th value, obtained by subtracting the starting frequency of the (p+1)th value from the starting frequency of the pth value. p+1 -q p The interval value of the p-th value is obtained by subtracting the starting frequency of the p-th value from that of the (p-1)-th value. p -q p-1 If we consider the interval of the p-th value as two, then:

[0024] In q p+1 -q p and q p -q p-1 When all values ​​are zero, determine that the p-th value is not at a section break; at q p+1 -q p and q p -q p-1 When a value is non-zero, the p-th value is determined to be a segment. Process each value sequentially to obtain each segment. All values ​​between two adjacent segments form a sub-segment field. This process obtains each sub-segment field, thus achieving intelligent segmentation of the total bar chart.

[0025] Preferably, step 3 includes:

[0026] The LINEST function method is used to perform univariate regression on each sub-domain to obtain the univariate regression equation for each sub-domain. The equation for the j-th sub-domain is defined as z = L. j y+C j, put L j and C j As the starting point parameter 1 and starting point parameter 2 of the j-th sub-section domain;

[0027] Based on the total, a univariate regression with emphasis is performed on each sub-domain in the bar chart along with its starting function parameter.

[0028] Preferably, the method of performing a univariate regression with emphasis on each sub-section domain in the bar chart based on the total sum of each sub-section domain and its starting function parameter includes:

[0029] The equation for the univariate regression is defined as g(y) = by + c, where b and c are transformation parameter one and transformation parameter two, respectively. The univariate regression is performed on each subdomain by changing the values ​​of these two transformation parameters b and c. Then, based on the frequency of each value after the univariate regression, the corresponding 'value addition' or 'value removal' is performed in the value queue one according to the change in the frequency of each value in the total bar chart, thereby changing the frequency of each value.

[0030] Setting a threshold for change is Then, for the j-th subdomain:

[0031] exist When it is determined that the frequency variation of different values ​​within the subdomain is stable, the corresponding transformation parameter for the subdomain is set as the starting parameter L. j Let the transformation parameter c corresponding to this sub-section be an arbitrary quantity located between the maximum and minimum starting frequencies of the total histogram.

[0032] exist When setting the assignment range of the transformation parameter c, it is... In L j When <0, the corresponding assignment range for the transformation parameter b is set as follows: In L j When >0, the corresponding assignment range for the transformation parameter b is set as follows: Furthermore, the two transformation parameters b and c are both integers, and g j,max and g j,min These are the highest and lowest starting frequencies of the j-th sub-section domain, respectively;

[0033] Values ​​with adjustment frequencies not lower than the starting frequency are considered cluster one values; values ​​with adjustment frequencies lower than the starting frequency are considered cluster two values.

[0034] Next, based on the univariate regression performed on the j-th subdomain using different transformation parameter groups, the estimators for each transformation parameter group are obtained from the starting frequency and adjustment frequency of each value. Therefore, the estimator for the u-th transformation parameter group can be:

[0035]

[0036] Here, H u It is the estimator of the u-th transformation parameter group; n j1 This represents the number of cluster-1 values ​​within the j-th subdomain after applying the u-th transformation parameter group; n j2 This represents the number of cluster 2 values ​​in the j-th subdomain after applying the u-th transformation parameter group; Q' j,v G represents the adjustment frequency of the value of the v-th cluster within the j-th subdomain; j,v Q' represents the starting frequency of the value of the v-th cluster in the j-th subdomain; j,w G represents the adjustment frequency of the two values ​​of the w-th cluster in the j-th subdomain; j,w This represents the starting frequency of the binary value of the w-th cluster within the j-th subdomain;

[0037] The group of transformation parameters with the highest estimate is taken as the optimal group of transformation parameters.

[0038] Preferably, step 4 includes:

[0039] Let transformation parameter one and transformation parameter two in the optimal transformation parameter group be taken as the optimal transformation parameter one and optimal transformation parameter two of the j-th subdomain, and defined as b' respectively. j c' j ;

[0040] Based on the optimal transformation parameters for each subdomain, a univariate regression is performed on each subdomain. Then, the frequency value obtained after the univariate regression for each value within each subdomain is taken as the final adjusted frequency for each value. The final adjusted frequency of the k-th value within the j-th subdomain after the univariate regression with emphasis is defined as Q'. j,k The frequency of the k-th value occurring within the total histogram, which is also the starting frequency of the k-th value, is G. j,k ;

[0041] When the k-th value in the j-th subdomain is a 'cluster-one value', Q' is added before the location where this value first occurs in the value queue. j,k -G j,k The value is given; conversely, when the k-th value in the j-th sub-domain is a 'cluster two value', starting from the location where the value first occurs in the value queue, the first G values ​​are removed sequentially. j,k -Q' j,kThe value is set, and the position sequence code of the removed value in the value queue is recorded. Then, the position sequence code of each removed value is used as a guide to restore the value at the position where it was removed. When the k-th value is removed, the position sequence codes of each removed value form the guide queue of the k-th value.

[0042] Process each value in each sub-domain sequentially, and treat the processed value queue as the modulated value after univariate regression modulation.

[0043] The interval obtained by subtracting the starting parameter from the optimal transformation parameter one for each sub-segment is defined as parameter interval one; the interval obtained by subtracting the starting parameter from the optimal transformation parameter two for each sub-segment is defined as parameter interval two; parameter interval one and parameter interval two are used as the byte modulation coefficients for each sub-segment.

[0044] Given a boot queue containing the values ​​of each sub-byte field, each cluster 2 within that sub-byte field, and the corresponding parameter intervals 1 and 2 for each sub-byte field, the byte modulation coefficient of the j-th sub-byte field is Δb. j =L j -b' j , △c j =C j -c' j Take the order code group and the corresponding byte modulation coefficient of each cluster value in each sub-segment as a sub-region, that is, take each sub-segment as a sub-region, and thus obtain each sub-region.

[0045] The queue formed by the sequence code corresponding to each sub-domain is defined as the sequence queue. Then, the perturbation operation is performed through arbitrary changes to obtain the scrambled queue. The perturbation is performed at the location of each sub-region. The operable parameters of the scrambled queue are used as the perturbation coefficients ν, y0, and q of all modulation values. Here, ν represents the starting value for forming the scrambled queue, which is an arbitrary quantity in the range (0,1). y0 is any value in (3.72,3.84]. q represents the perturbation performed starting from the q-th value of the sequence queue.

[0046] Preferably, q is preset to two hundred.

[0047] Preferably, the scrambling coefficient is stored independently.

[0048] Preferably, when performing analysis on the vehicle-to-ground values, demodulation should be performed on the modulation values ​​and each sub-region. The demodulation method includes:

[0049] Initially, restoration is performed on the location of each sub-region based on the scrambling coefficient and the disordered queue; then, a total histogram is constructed based on each value and the frequency of each value within the modulation value, and the total histogram is segmented based on each sub-section field corresponding to each sub-region to obtain each sub-section field of the total histogram.

[0050] By applying the LINEST function to perform univariate regression on each subdomain and obtaining the univariate regression equation for each subdomain, the derivative of the univariate regression equation for the j-th subdomain will then correspond to b'. j The constant term of this univariate regression equation is c' j Then, the derivative and constant term of the obtained univariate regression equation are added to the corresponding byte modulation coefficients to obtain the original L of the j-th sub-byte domain. j C j That is, to obtain the starting parameter one and the starting parameter two of the j-th sub-section domain;

[0051] Next, the modulated values ​​are restored by combining the corresponding guide queues of each cluster second value, thereby obtaining the corresponding source vehicle ground values, and the demodulation process is completed.

[0052] A data processing device for track maintenance vehicles used in rail transit overhead contact lines includes:

[0053] The vehicle-mounted FSO terminal communicates with the ground control center via a wireless network.

[0054] The modules running at the ground control center include:

[0055] Get Module 1, which is used to retrieve Value Queue 1;

[0056] Module 2 is used to retrieve each sub-domain based on the numerical queue 1.

[0057] Module 3 is used to obtain the total cluster 1 values ​​and total cluster 2 values ​​of each sub-domain based on each transformation parameter group of each sub-domain, thereby obtaining the estimate of each transformation parameter group; and obtain the optimal transformation parameter group of each sub-domain based on the estimate of each transformation parameter group.

[0058] The storage module is used to obtain modulation values ​​based on the optimal transformation parameter group of each sub-domain; obtain each sub-region based on the guide queue of all cluster 2 values ​​and byte modulation coefficients in each sub-domain; obtain a jumbled queue based on each sub-region; and store the jumbled queue and modulation values.

[0059] The beneficial effects of this invention are as follows: Compared with the prior art, this invention constructs a total bar chart based on a numerical queue composed of vehicle-to-ground data, performs intelligent segmentation based on the frequency of each data in the total bar chart, obtains all cluster 1 values ​​and all cluster 2 values ​​of each sub-section based on each transformation parameter group of each sub-section, obtains the optimal transformation parameter group of each sub-section based on all cluster 1 and cluster 2 values ​​in each sub-section, and performs a univariate regression with emphasis on each sub-section based on the optimal transformation parameter group. This ensures that the adjustment frequency of each data at the beginning and end of the univariate regression is significantly different from its starting frequency in the numerical queue, thereby disrupting the frequency of each data in the source data, increasing the complexity of decoding, and making the vehicle-to-ground data more secure and stable during the storage process. Attached Figure Description

[0060] Figure 1 This is a partial flowchart of the method for processing vehicle-ground data of rail transit overhead contact line operation vehicle described in this invention;

[0061] Figure 2 This is a schematic diagram of the module of the rail transit overhead contact line operation vehicle ground data processing device described in this invention. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, any other embodiments obtained by those skilled in the art without creative effort are all within the protection scope of this invention.

[0063] like Figure 1 As shown, the present invention provides a method for processing vehicle-to-ground data for rail transit overhead contact line maintenance vehicles, comprising:

[0064] The onboard FSO terminal transmits the vehicle-to-ground data of the rail transit overhead contact line maintenance vehicle to the ground control center via wireless network for storage. Storage methods include:

[0065] Step 1, obtain the first numerical queue;

[0066] Vehicle-to-ground data refers to the values ​​of temperature, speed, humidity, wind speed, and operating power of the rail transit overhead contact line maintenance vehicle during its journey.

[0067] Previous modulation methods commonly used normal distribution modulation, which modulates data by analyzing the probability distribution of numerical values. However, these methods can also decode data based on frequency rules, often creating vulnerabilities for data leakage and malicious access. Furthermore, while univariate regression can be used to change the frequency of numerical values, previous univariate regressions applied the same parameters to all values, neglecting the specific vehicle-to-ground data collected in real-time. This resulted in unstable fluctuations due to variations in the source data distribution. Therefore, this invention uses a univariate regression method that focuses on the frequency of numerical values, thereby disrupting the modulation of the numerical frequency and ensuring the protective stability of vehicle-to-ground data.

[0068] In a preferred but non-limiting embodiment of the present invention, the method for obtaining the numerical queue one includes:

[0069] Obtain the vehicle and ground data for each point in time within a pre-defined time period. Obtain a data queue according to the order of the points in time and treat it as data queue one. At this time, there are redundant data in data queue one.

[0070] In a preferred but non-limiting embodiment of the present invention, the time period is 60 minutes.

[0071] Step 2: Obtain each sub-domain based on the numerical queue 1;

[0072] In a preferred but non-limiting embodiment of the present invention, step 2 specifically includes:

[0073] Step 2-1: Construct a total bar chart based on each value in the numerical queue and the frequency of each value. The X coordinate of the total bar chart represents the numerical content, and the Y coordinate represents the frequency of each value. In this invention, the frequency of each value in the total bar chart is taken as the starting frequency of each value.

[0074] Step 2-2: Treat the queue formed by each value on the X coordinate of the total as the second value queue; at this time, there are no redundant values ​​in the second value queue.

[0075] Steps 2-3: Current methods for varying the frequency of numerical values ​​generally use univariate regression. However, previous univariate regression methods applied the same parameters to the entire numerical distribution range to vary the frequency of different values ​​in the total histogram. This method did not focus on the real-time collected vehicle and ground values, meaning that previous univariate regression methods would result in unstable variations due to different distributions of the source values. Therefore, this invention performs intelligent segmentation on the total histogram, and then performs univariate regression with a focus based on the variation of the frequency of each value in each sub-segment. This ensures that the frequency of values ​​from the same source is significantly different after univariate regression in each sub-segment, thereby achieving better modulation performance. This invention obtains each segmentation point based on the frequency of each value and its adjacent values ​​in the numerical queue, and then performs intelligent segmentation on the total histogram.

[0076] In a preferred but non-limiting embodiment of the present invention, the method for intelligently segmenting totals using a bar chart includes:

[0077] Taking the p-th value in the numerical queue as an example, initially, the interval between the frequency of the p-th value and its adjacent values ​​is calculated, and q is defined as... p It represents the frequency of the p-th value within the total histogram, which is also the starting frequency of the p-th value. q p+1 It is the starting frequency of the (p+1)th value, q p-1 The value q is the starting frequency of the (p-1)th value, obtained by subtracting the starting frequency of the (p+1)th value from the starting frequency of the pth value. p+1 -q p The interval value of the p-th value is obtained by subtracting the starting frequency of the p-th value from that of the (p-1)-th value. p -q p-1 If we consider the interval of the p-th value as two, then:

[0078] In q p+1 -q p and q p -q p-1 When all values ​​are zero, determine that the p-th value is not at a section break; at q p+1 -q p and q p -q p-1 When a value is non-zero, the p-th value is determined to be a segment. Process each value sequentially to obtain each segment. All values ​​between two adjacent segments form a sub-segment field. This process obtains each sub-segment field, thus achieving intelligent segmentation of the total bar chart.

[0079] Step 3: Obtain the total cluster 1 values ​​and total cluster 2 values ​​for each sub-domain based on each transformation parameter group for each sub-domain, thereby obtaining the estimator for each transformation parameter group; obtain the optimal transformation parameter group for each sub-domain based on the estimator for each transformation parameter group.

[0080] In a preferred but non-limiting embodiment of the present invention, step 3 includes:

[0081] The LINEST function method is used to perform univariate regression on each sub-domain to obtain the univariate regression equation for each sub-domain. The equation for the j-th sub-domain is defined as z = L. j y+C j The present invention uses L j and C j As the starting point parameter 1 and starting point parameter 2 of the j-th subdomain.

[0082] Based on the total, perform a univariate regression with emphasis on each sub-section of the bar chart using its starting function parameter;

[0083] In a preferred but non-limiting embodiment of the present invention, a method for performing a univariate regression with emphasis on each sub-section domain in a bar chart based on the total sum of each sub-section domain and its starting function parameter includes:

[0084] The equation for the univariate regression is defined as g(y) = by + c, where b and c are transformation parameter one and transformation parameter two, respectively. This invention performs univariate regression on each subdomain by changing the values ​​of these two transformation parameters b and c. Subsequently, based on the frequency of each value after univariate regression, the corresponding 'value addition' or 'value removal' is performed in the value queue one according to the change in the frequency of each value in the total bar chart, thereby changing the frequency of each value. However, since the 'value removal' process requires registering the location of removal and the corresponding removed values, additional content will be added. Therefore, the transformation parameters of the univariate regression should be intelligently selected to ensure that the frequency change after univariate regression is not significant while reducing unnecessary value removal operations.

[0085] This invention aims to achieve a significant distinction between the frequency of each value within a subdomain and its starting frequency after performing univariate regression on each subdomain. Specifically, within a subdomain, if the frequency of a value is not high, univariate regression is used to reduce its frequency; conversely, if the frequency of a value is not high, univariate regression is used to increase its frequency. When the frequency changes of different values ​​within a subdomain are very stable, the difference between the frequencies of each value at the beginning and end of the univariate regression is often small, meaning the magnitude of the frequency change for each value is not large. In such cases, independent processing should be performed on these subdomains.

[0086] The starting function parameter corresponding to each sub-domain can characterize the frequency variation amplitude of different values ​​within each sub-domain, so setting a variation threshold is... Then, for the j-th subdomain:

[0087] exist When it is determined that the frequency variation of different values ​​within the subdomain is stable, the corresponding transformation parameter for the subdomain is set as the starting parameter L. j Let the transformation parameter c corresponding to this sub-section be an arbitrary quantity located between the maximum and minimum starting frequencies of the total histogram.

[0088] exist When setting the assignment range of the transformation parameter c, it is... In L j When <0, the corresponding assignment range for the transformation parameter b is set as follows: In L j When >0, the corresponding assignment range for the transformation parameter b is set as follows: Furthermore, the two transformation parameters b and c are both integers, and g j,max and g j,min These are the highest and lowest starting frequencies of the j-th sub-section domain, respectively;

[0089] This invention uses different groups of transformation parameters to perform univariate regression on the j-th subdomain. The frequency of each value is used as the adjustment frequency of each value. After performing univariate regression using different groups of parameters, the values ​​with adjustment frequencies not lower than the starting frequency are regarded as cluster one values; the values ​​with adjustment frequencies lower than the starting frequency are regarded as cluster two values.

[0090] Next, based on the univariate regression performed on the j-th subdomain using different transformation parameter groups, the estimators for each transformation parameter group are obtained from the starting frequency and adjustment frequency of each value. Therefore, the estimator for the u-th transformation parameter group can be:

[0091]

[0092] Here, H u It is the estimator of the u-th transformation parameter group; n j1 This represents the number of cluster-1 values ​​within the j-th subdomain after applying the u-th transformation parameter group; n j2 This represents the number of cluster 2 values ​​in the j-th subdomain after applying the u-th transformation parameter group; Q' j,v G represents the adjustment frequency of the value of the v-th cluster within the j-th subdomain; j,v Q' represents the starting frequency of the value of the v-th cluster in the j-th subdomain; j,w G represents the adjustment frequency of the two values ​​of the w-th cluster in the j-th subdomain; j,w This represents the starting frequency of the binary value of the w-th cluster within the j-th subdomain;

[0093] The group of transformation parameters with the highest estimate is taken as the optimal group of transformation parameters.

[0094] Step 4: Obtain the modulation value based on the optimal transformation parameter group for each sub-domain; obtain each sub-region based on the guide queue of all cluster 2 values ​​and the byte modulation coefficients within each sub-domain; obtain the jumbled queue based on each sub-region; and store the jumbled queue and the modulation value.

[0095] In a preferred but non-limiting embodiment of the present invention, step 4 includes:

[0096] Let transformation parameter one and transformation parameter two in the optimal transformation parameter group be taken as the optimal transformation parameter one and optimal transformation parameter two of the j-th subdomain, and defined as b' respectively. j c' j At this point, using the current conversion parameters can ensure that there is a significant difference between the adjustment frequency and the starting frequency of each numerical value, while minimizing the number of values ​​to be removed.

[0097] Based on the optimal transformation parameters for each subdomain, a univariate regression is performed on each subdomain. Then, the frequency value obtained after the univariate regression for each value within each subdomain is used as the final adjustment frequency for each value. This invention defines the final adjustment frequency of the k-th value within the j-th subdomain after the univariate regression with a focus as Q'. j,k The frequency of the k-th value occurring within the total histogram, which is also the starting frequency of the k-th value, is G. j,k ;

[0098] When the k-th value in the j-th subdomain is a 'cluster-one value', Q' is added before the location where this value first occurs in the value queue. j,k -G j,kThe value is given; conversely, when the k-th value in the j-th sub-domain is a 'cluster two value', starting from the location where the value first occurs in the value queue, the first G values ​​are removed sequentially. j,k -Q' j,k The value is set, and the position sequence code of the removed value in the value queue is recorded. Then, the position sequence code of each removed value is used as a guide to restore the value at the position where it was removed. When the k-th value is removed, the position sequence codes of each removed value form the guide queue of the k-th value.

[0099] Process each value in each sub-domain sequentially, and treat the processed value queue as the modulated value after univariate regression modulation.

[0100] The interval obtained by subtracting the starting parameter from the optimal transformation parameter one for each sub-segment is defined as parameter interval one; the interval obtained by subtracting the starting parameter from the optimal transformation parameter two for each sub-segment is defined as parameter interval two; parameter interval one and parameter interval two are used as the byte modulation coefficients for each sub-segment.

[0101] Given a boot queue containing the values ​​of each sub-byte field, each cluster 2 within that sub-byte field, and the corresponding parameter intervals 1 and 2 for each sub-byte field, the byte modulation coefficient of the j-th sub-byte field is Δb. j =L j -b' j , △c j =C j -c' j In this invention, the order code group and the corresponding byte modulation coefficient of each cluster value in each sub-segment domain are regarded as a sub-region, that is, each sub-segment domain corresponds to a sub-region, and thus each sub-region is obtained.

[0102] The queue formed by the sequence codes corresponding to each sub-domain is defined as an ordered queue. Then, a scrambling operation is performed using an arbitrary transformation method to obtain a scrambled queue. This scrambling is then applied to the location of each sub-region. The operable parameters of the scrambled queue are used as the scrambling coefficients ν, y0, and q of the entire modulation value. Here, ν represents the initial value for forming the scrambled queue, which is an arbitrary value within the range (0,1); y0 is any value within (3.72,3.84]; and q represents the value from the q-th value in the ordered queue onwards. Scrambling involves arbitrarily adding y0 to any value after the q-th value.

[0103] In the preferred but non-limiting embodiment of the present invention, q is preset to two hundred.

[0104] In a preferred but non-limiting embodiment of the present invention, the scrambling coefficient is stored independently.

[0105] In a preferred but non-limiting embodiment of the present invention, before performing analysis on the vehicle-to-ground values, demodulation is performed on the modulation values ​​and each sub-region. The demodulation method includes:

[0106] Initially, restoration is performed on the location of each sub-region based on the scrambling coefficient and the disordered queue; then, a total histogram is constructed based on each value and the frequency of each value within the modulation value, and the total histogram is segmented based on each sub-section field corresponding to each sub-region to obtain each sub-section field of the total histogram.

[0107] By applying the LINEST function to perform univariate regression on each subdomain and obtaining the univariate regression equation for each subdomain, the derivative of the univariate regression equation for the j-th subdomain will then correspond to b'. j The constant term of this univariate regression equation is c' j Then, the derivative and constant term of the obtained univariate regression equation are added to the corresponding byte modulation coefficients to obtain the original L of the j-th sub-byte domain. j C j That is, to obtain the starting parameter one and the starting parameter two of the j-th sub-section domain;

[0108] Next, the modulated values ​​are restored by combining the corresponding guide queues of each cluster second value, thereby obtaining the corresponding source vehicle ground values, and the demodulation process is completed.

[0109] This is how the vehicle-to-ground data was processed.

[0110] This invention constructs a total bar chart based on a numerical queue composed of vehicle-to-ground data, and performs intelligent segmentation based on the frequency of each data point in the total bar chart. It obtains all cluster 1 and cluster 2 values ​​for each sub-domain based on each transformation parameter group. It then obtains the optimal transformation parameter group for each sub-domain based on the total cluster 1 and cluster 2 values. Finally, it performs a univariate regression on each sub-domain based on the optimal transformation parameter group, ensuring that the adjusted frequency of each data point at the beginning and end of the regression differs significantly from its initial frequency in the numerical queue. This disrupts the frequency of each data point within its source data, increasing decoding complexity and making the vehicle-to-ground data storage process more secure and stable.

[0111] like Figure 2 As shown, the present invention provides a data processing device for track transit overhead contact line operation vehicles, comprising:

[0112] The vehicle-mounted FSO terminal communicates with the ground control center via a wireless network, so that the vehicle-mounted FSO terminal can transmit vehicle-to-ground data to the ground control center for storage via the wireless network.

[0113] The modules running at the ground control center include:

[0114] Get Module 1, which is used to retrieve Value Queue 1;

[0115] Module 2 is used to retrieve each sub-domain based on the numerical queue 1.

[0116] Module 3 is used to obtain the total cluster 1 values ​​and total cluster 2 values ​​of each sub-domain based on each transformation parameter group of each sub-domain, thereby obtaining the estimate of each transformation parameter group; and obtain the optimal transformation parameter group of each sub-domain based on the estimate of each transformation parameter group.

[0117] The storage module is used to obtain modulation values ​​based on the optimal transformation parameter group of each sub-domain; obtain each sub-region based on the guide queue of all cluster 2 values ​​and byte modulation coefficients in each sub-domain; obtain a jumbled queue based on each sub-region; and store the jumbled queue and modulation values.

[0118] The beneficial effects of this invention are as follows: Compared with the prior art, this invention constructs a total bar chart based on a numerical queue composed of vehicle-to-ground data, performs intelligent segmentation based on the frequency of each data in the total bar chart, obtains all cluster 1 values ​​and all cluster 2 values ​​of each sub-section based on each transformation parameter group of each sub-section, obtains the optimal transformation parameter group of each sub-section based on all cluster 1 and cluster 2 values ​​in each sub-section, and performs a univariate regression with emphasis on each sub-section based on the optimal transformation parameter group. This ensures that the adjustment frequency of each data at the beginning and end of the univariate regression is significantly different from its starting frequency in the numerical queue, thereby disrupting the frequency of each data in the source data, increasing the complexity of decoding, and making the vehicle-to-ground data more secure and stable during the storage process.

[0119] This disclosure may be a system, method, and / or computer program product. A computer program product may include a computer-readable appendix having computer-readable program instructions loaded thereon for causing a processor to achieve each aspect disclosed herein.

[0120] Computer-readable printed media can be tangible printed media capable of holding and displaying instructions executed by a circuit. Computer-readable printed media can be—but is not limited to—electrical printed media, magnetic printed media, optical printed media, electromagnetic printed media, semiconductor printed media, or any suitable combination thereof. Further examples of computer-readable printed media (a non-exhaustive list) include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (HD-ROM), digital multipurpose disk (DXD), memory sticks, floppy disks, mechanically encoded printed media, punch cards or recessed protrusions with instructions printed on them, or any suitable combination thereof. The computer-readable annotated medium used herein is not to be interpreted as the instantaneous message itself, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (like light pulses through power transmission cables), or electrical messages transmitted through wires.

[0121] The computer-readable program instructions expressed herein can be downloaded from computer-readable supplementary media to each computing / processing power line, or downloaded via a wireless network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external supplementary power line. The wireless network can include copper transmission cables, transmission line transmissions, wireless transmissions, routers, firewalls, switches, Wi-Fi device computers, and / or edge servers. A wireless network adapter card or wireless network port in each computing / processing power line receives the computer-readable program instructions from the wireless network and forwards the computer-readable program instructions to the computer-readable supplementary media stored in each computing / processing power line.

[0122] The computer program instructions used to execute the operations of this disclosure can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-associative instructions, microcode, firmware instructions, conditional values, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as ScalarQAL, H++, etc., and conventional procedural programming languages ​​such as H' language or similar programming languages. The computer-readable program instructions can be executed entirely on a client computer, partially on a client computer, as a single software package, or partially on a client computer and partially on a remote computing facility. Execution can be performed on-site or entirely on a remote computer or server. In the form involving a remote computer, the remote computer can connect to the client computer via any type of wireless network—including a local area network (LAb) or a wide area network (UAb)—or can connect to an external computer (such as using an Internet service provider to connect via the Internet). In some embodiments, electronic circuitry is customized using operating condition values ​​of computer-readable program instructions, such as programmable logic circuits, field-programmable gate arrays (processing platforms), or programmable logic arrays (PLAs), which can execute computer-readable program instructions to achieve every aspect of cost disclosure.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent updates can still be made to the specific embodiments of the present invention without departing from the spirit and scope of the present invention, and any modifications or equivalent updates should be covered within the scope of protection of the claims of the present invention.

Claims

1. A track traffic catenary service vehicle ground numerical arrangement method, characterized in that, The method comprises the following steps: The vehicle-ground data of the rail transit catenary operation vehicle is transmitted to the ground control center via a wireless network, and the storage method comprises the following steps: Step 1, obtaining a value queue one; Step 2, obtaining each sub-domain according to the value queue one; Step 3, obtaining the whole cluster one value and the whole cluster two value of each sub-domain according to each transformation parameter group of each sub-domain, and obtaining the estimate of each transformation parameter group; obtaining the optimal transformation parameter group of each sub-domain according to the estimate of each transformation parameter group; Step 4, obtaining the modulation value according to the optimal transformation parameter group of each sub-domain; obtaining each sub-area according to the leading queue of the whole cluster two value in each sub-area and the byte modulation coefficient; obtaining the clutter queue according to each sub-area; and performing storage on the clutter queue and the modulation value; Step 3 comprises: LINEST function method is used to perform a one-dimensional regression on each sub-section domain, and the one-dimensional regression equation of each sub-section domain is obtained. The equation of the jth sub-section domain is defined as z=L j y+C j , and L j and C j are taken as the starting point parameter one and the starting point parameter two of the jth sub-section domain. Performing a focus one-dimensional regression on each sub-domain according to the total column chart and the function parameter variable of each sub-domain, which comprises: Defining the equation of the focus one-dimensional regression as g(y)=by+c, b and c are transformation parameter one and transformation parameter two respectively, and the focus one-dimensional regression is performed on each sub-domain by changing the values of the two transformation parameters b and c; then, according to the corresponding frequency of each value after one-dimensional regression, corresponding to the change condition between the corresponding frequency of each value in the total column chart, the corresponding value addition or value removal is performed in the value queue one, so as to change the frequency of each value; Then, according to the one-dimensional regression of the jth sub-domain by using different transformation parameter groups, the starting frequency and the adjusted frequency of each value obtain the estimate of each transformation parameter group, and the estimate of the uth transformation parameter group is: Here, H u is the estimate of the u-th transformed covariate group; n j1 represents the number of cluster-one values in the j-th sub-domain after using the u-th transformed covariate group; n j2 represents the number of cluster-two values in the j-th sub-domain after using the u-th transformed covariate group; Q j,v represents the adjustment frequency of the v-th cluster-one value in the j-th sub-domain; G j,v represents the starting frequency of the v-th cluster-one value in the j-th sub-domain; Q j,w represents the adjustment frequency of the w-th cluster-two value in the j-th sub-domain; G j,w represents the starting frequency of the w-th cluster-two value in the j-th sub-domain; The transformation parameter group with the highest estimate is taken as the optimal transformation parameter group.

2. The track traffic catenary service vehicle ground numerical arrangement method according to claim 1, characterized in that, The method for obtaining the value queue one comprises: Obtaining the vehicle-ground data at each time point in a preset time period, obtaining a value queue according to the time sequence, and taking it as the value queue one, at this time, there are redundant values in the value queue one; The preset time period is 60 minutes.

3. The track traffic catenary service vehicle ground value arrangement method according to claim 1, characterized in that, Step 2 specifically comprises: Step 2-1, constructing a total column chart according to each value in the value queue one and the frequency of each value, wherein the X coordinate of the total column chart represents the value content, and the Y coordinate represents the frequency of each value, and the corresponding frequency of each value in the total column chart is taken as the starting frequency of each value; Step 2-2, taking the queue formed by each value on the X coordinate of the total column chart as the value queue two; at this time, there are no redundant values in the value queue two; Step 2-3, obtaining each sub-section according to each value in the value queue two and the corresponding frequency of the adjacent value, so as to perform intelligent segmentation on the total column chart.

4. The track traffic catenary service vehicle ground value arrangement method according to claim 3, characterized in that, The method for performing intelligent segmentation on the total column chart comprises: Initially, the interval amount between the pth value and the frequency of its adjacent value is defined as q p is the frequency of the pth value in the total column chart, that is, the start frequency of the pth value, q p+1 is the start frequency of the p+1th value, q p-1 is the start frequency of the p-1th value, and the amount obtained by subtracting the start frequency between the p+1th value and the pth value is q p+1 -q p is the amount obtained by subtracting the start frequency between the pth value and the p-1th value, and is taken as the interval amount one of the pth value, q p -q p-1 is taken as the interval amount two of the pth value, then: In q p+1 -q p and q p -q p-1 are all zero, the pth value is determined as not being a section; in q p+1 -q p and q p -q p-1 with a non-zero value, the pth value is determined as being a section; each value is processed in order, each section is obtained, the values between two adjacent sections form a sub-section domain, each sub-section domain is obtained, and intelligent sectioning of the total column chart is achieved.

5. The track traffic catenary service vehicle ground numerical arrangement method according to claim 1, characterized in that, Then, for the jth sub-domain: In the frequency of the different values in the sub-domain is stable, at which time the transformation variable two corresponding to the sub-domain is set as the start variable one L j the transformation variable two corresponding to the sub-domain is set as an arbitrary amount between the maximum start frequency and the minimum start frequency of the total column chart; exist When setting the assignment range of the transformation parameter c, it is defined as [0, g]. j，max -g j，min ]; in L j When < 0, the assignment range of the transformation parameter b is set to [0, -g]. j ]; in L j When >0, the assignment range of the transformation parameter b is set to [-g]. j [,0]; and both transformation parameters b and c are integers, g j,max and g j,min These are the highest and lowest starting frequencies of the j-th sub-section domain, respectively; The value with the adjusted frequency not lower than the starting frequency is taken as the cluster one value; and the value with the adjusted frequency lower than the starting frequency is taken as the cluster two value.

6. The track traffic catenary service vehicle ground numerical arrangement method according to claim 1, characterized in that, Step 4 comprises: b' j , c' j ; A univariate regression is performed on each subdomain according to the best transformation variable of each subdomain, and then the frequency value of each value in each subdomain after the univariate regression is taken as the final adjusted frequency of each value. The corresponding final adjusted frequency of the kth value in the jth subdomain after the emphasis univariate regression is defined as Q j,k The frequency of the kth value in the total column chart, that is, the starting frequency of the kth value, is G j,k ; So when the kth value in the jth sub-domain is a cluster-one value, the value is supplemented into Q' from the place where the value first occurs in the value queue j,k -G j,k ; otherwise, when the kth value in the jth sub-domain is a cluster-two value, G values are removed from the place where the value first occurs in the value queue j,k -Q' j,k , and the order codes of the removed values in the value queue are recorded, and then the values in the removed places are recovered using the order codes of the removed places as the guide, so that the order codes of the removed values form the guide queue of the kth value when the value is removed. Performing processing on each value of each sub-domain in sequence, and taking the value queue one obtained after processing as the modulation value after modulation by the focus one-dimensional regression; Interval quantity one is defined as the interval quantity obtained by subtracting the starting point variable from the optimal conversion variable of each sub-domain; interval quantity two is defined as the interval quantity obtained by subtracting the starting point variable two from the optimal conversion variable two of each sub-domain, and interval quantity one and interval quantity two are taken as the byte modulation coefficients of each sub-domain; A queue corresponding to each cluster two value in each subdomain and a corresponding parameter interval one and parameter interval two of each subdomain are stored. The byte modulation coefficient of the jth subdomain is Δb j = L j -b' j , Δc j = C j -c' j Each cluster two value in each subdomain is taken as a subarea, and each subarea is obtained. The queue formed by the corresponding order codes of each sub-domain is defined as an order queue, and then a scrambling operation is performed on the order queue by means of a random variation method to obtain a scrambled queue, so that scrambling is performed on the location of each sub-domain, and the operable variables of the scrambled queue are taken as the scrambling coefficients v1, y0, q of the whole modulation values, where v1 represents the starting value for forming the scrambled queue, which is an arbitrary quantity in the (0, 1) range, y0 is an arbitrary value in the (3.72, 3.84] range; q represents the scrambling performed on the qth value of the order queue.

7. The track traffic catenary service vehicle ground numerical arrangement method according to claim 6, characterized in that, Q is set to two hundred in advance.

8. The track traffic catenary service vehicle ground value arrangement method according to claim 7, characterized in that, The scrambling coefficients are independently stored; When the vehicle-to-ground values are analyzed, the modulation values and each sub-domain are demodulated, and the demodulation method comprises: First, the location of each sub-domain is restored according to the scrambling coefficients and the scrambled queue; then, a total column chart is constructed according to each value in the modulation values and the frequency of each value, the total column chart is segmented according to each sub-domain corresponding to each sub-domain of each sub-domain, and each sub-domain of the total column chart is obtained; LINEST function method is used to perform a one-dimensional regression on each sub-domain to obtain a one-dimensional regression equation for each sub-domain. Then, for the jth sub-domain, the derivative of the one-dimensional regression equation obtained for this sub-domain corresponds to b' j The constant term of the one-dimensional regression equation is c' j Then, the derivative and the constant term of the one-dimensional regression equation obtained are added to the corresponding byte modulation coefficients to obtain the original L j , C j of the jth sub-domain, that is, the starting independent variable one and the starting independent variable two of the jth sub-domain. Next, the modulation values are restored in combination with the guide queue corresponding to each cluster two value, and the corresponding source vehicle-to-ground values are obtained, and the demodulation process is completed.

9. A track transit catenary service vehicle-ground numerical value arrangement device, which executes the track transit catenary service vehicle-ground numerical value arrangement method of claim 1, characterized in that, It comprises: The vehicle-mounted FSO terminal is connected to the ground control center through wireless network communication; The modules running on the ground control center comprise: A obtaining module one is used to obtain a value queue one; An obtaining module two is used to obtain each sub-domain according to the value queue one; An obtaining module three is used to obtain the whole cluster one values and the whole cluster two values of each sub-domain according to each conversion variable group of each sub-domain, so as to obtain the estimates of each conversion variable group; and the optimal conversion variable group of each sub-domain is obtained according to the estimates of each conversion variable group; A storage module is used to obtain the modulation values according to the optimal conversion variable group of each sub-domain; to obtain each sub-domain according to the guide queue of the whole cluster two values in each sub-domain and the byte modulation coefficients; to obtain the scrambled queue according to each sub-domain; and to store the scrambled queue and the modulation values.

Images