Dragon boat training result model establishment system and model scoring method

Abstract

The application relates to the technical field of dragon boat training monitoring, in particular to a dragon boat training result model establishing system and a model scoring method, which comprises an environment monitoring module, a personnel monitoring module and a dragon boat monitoring module; the environment monitoring module is used for monitoring external environmental factors of a dragon boat to obtain first monitoring data; the personnel monitoring module is used for monitoring personnel states on the dragon boat to obtain second monitoring data; the dragon boat monitoring module is used for monitoring a dragon boat state to obtain third monitoring data; a data processing module is used for respectively acquiring the first monitoring data, the second monitoring data and the third monitoring data, and performing data classification and correlation to obtain a training result set; a modeling module is used for acquiring the training result set generated by the data processing module, and generating a corresponding training result model based on the training result set. The application has the effect of improving the monitoring effect and the scoring effect of training results during dragon boat water training.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of dragon boat training monitoring, in particular to a dragon boat training result model establishment system and a model scoring method. BACKGROUND

[0002] Dragon boat is a relatively traditional sports project in China, and dragon boat races have expanded from regions and schools to current international, national and provincial large-scale competitions, and more and more schools and regions have formed professional dragon boat teams.

[0003] However, as the level of dragon boat races is getting higher, the training process of dragon boat is also becoming more professional and comprehensive. Dragon boat training is divided into land training and water training. In the current related technology, there is a relatively professional and scientific monitoring means for land training, and through professional training equipment, the technical characteristics, technical advantages and disadvantages and physical conditions of athletes can be effectively detected. However, for water training, firstly, the water environment is relatively complex, and there are many influencing factors, so it is difficult to scientifically monitor the training process through manpower or related instruments, and there is a great limitation. Secondly, this also leads to that the training result of the current dragon boat training cannot be accurately and scientifically determined, and professional players cannot be scored, selected and promoted according to the training result. SUMMARY

[0004] The purpose of the application is to improve the monitoring effect and scoring effect of the training result of dragon boat water training.

[0005] In a first aspect, the application provides a dragon boat training result model establishment system, which adopts the following technical scheme:

[0006] A dragon boat training result model establishment system comprises:

[0007] An environment monitoring module is configured to monitor external environmental factors of the dragon boat to obtain first monitoring data, wherein the external environmental factors include a water environment, an underwater environment and wind power.

[0008] A personnel monitoring module is configured to monitor personnel states on the dragon boat to obtain second monitoring data, wherein the personnel states include paddle action, paddle frequency, physical indicators and drumming frequency.

[0009] A dragon boat monitoring module is configured to monitor a state of the dragon boat to obtain third monitoring data, wherein the state of the dragon boat includes a dragon boat moving speed, a dragon boat orientation angle, a dragon boat offset amount and a dragon boat acceleration.

[0010] A data processing module is electrically connected to the environment monitoring module, the personnel monitoring module and the dragon boat monitoring module to obtain corresponding first monitoring values, second monitoring values and third monitoring values, and perform data classification and association on the first monitoring values, the second monitoring values and the third monitoring values to obtain a training result set, wherein the training result set includes a plurality of training results, and specifically includes an independent training set and an associated training set.

[0011] A modeling module is configured to obtain the training result set generated by the data processing module, and generate corresponding training result models based on the training result set, wherein the training result models include a specific personnel training result model and a dragon boat overall training result model.

[0012] In some other embodiments, the data processing module further includes:

[0013] The associated items of the external environment factors, the personnel states and the dragon boat states are obtained and mapped, and the monitoring data having the same associated item are obtained and associated to obtain associated data items;

[0014] The monitoring data not having the same associated item are obtained to obtain independent data items;

[0015] The associated training set is obtained according to a plurality of the associated data items, and the independent training set is obtained according to a plurality of the independent data items.

[0016] In some other embodiments, the associated data items further include positive association and / or negative association, wherein the positive association represents that when one or more data in the plurality of the monitoring data having the same associated item changes, a positive change effect is generated on other monitoring data, and the negative association represents that when one or more data in the plurality of the monitoring data having the same associated item changes, a negative change effect is generated on other monitoring data.

[0017] In some other embodiments, a model distribution module is connected to the modeling module to obtain a plurality of the training result models, and the model distribution module is configured to obtain target information corresponding to the plurality of the training result models, wherein the target information includes personnel names, personnel numbers, dragon boat numbers, and the specific personnel training result model is sent to a corresponding personnel mobile terminal according to the target information, and all the specific personnel training result models and the dragon boat overall training result model are sent to a person in charge terminal.

[0018] In a second aspect, a model scoring method is provided, which adopts the following technical solution:

[0019] A model scoring method is used to score the training result models generated by the dragon boat training result model establishment system, and includes:

[0020] Obtaining a plurality of training results in the training result model, and classifying and setting labels for the plurality of training results according to a preset classification method to obtain a total class, a classification, and a sub-class, wherein the total class contains a plurality of classifications, the classification contains a plurality of sub-classes, the total class represents a total of all the training results in the training result model, the classification represents a set corresponding to a plurality of associated training results, and the classification represents a single training result;

[0021] Setting a preset ideal total score for the total class;

[0022] Obtaining a first weight corresponding to a plurality of classifications in the total class, and calculating a classification score corresponding to each classification based on the first weight and the preset ideal total score, wherein the classification score represents a highest score value that can be allocated to the plurality of classifications;

[0023] Obtaining a second weight corresponding to a plurality of sub-classes of each classification in the classification, and calculating a sub-class score corresponding to each sub-class based on the second weight and the classification score of the classification corresponding thereto, wherein the sub-class score represents a highest score value that can be allocated to the plurality of sub-classes;

[0024] Based on a comparison result between the training result corresponding to the sub-class and a standard result and a scoring rule corresponding to the sub-class, the sub-class score is calculated to obtain an actual score of the training result corresponding to the plurality of sub-classes.

[0025] In some other embodiments, the scoring rule includes that the smaller the value of the training result, the higher the score, and the sub-class score is calculated based on the comparison result between the training result corresponding to the sub-class and the standard result, and the method further includes the following steps:

[0026] Based on the standard result, a preset ideal minimum value and an extreme maximum value are obtained;

[0027] Judging the size between the training result and the ideal minimum value and the extreme maximum value;

[0028] If the training result is less than or equal to the ideal minimum value, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class;

[0029] If the training result is greater than or equal to the extreme maximum value, the actual score corresponding to the sub-class is zero;

[0030] If the training result is greater than the ideal minimum value and less than the limit maximum value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a first proportion item, wherein the first proportion item represents a proportion value obtained by the ideal minimum value, the limit maximum value and the training result, and is less than 1.

[0031] In some other embodiments, the scoring rule includes that the greater the value of the training result, the higher the score, and the sub-class score is calculated based on a comparison result between the training result and the standard result corresponding to the sub-class, and further includes the following steps:

[0032] Obtaining a preset ideal maximum value and limit minimum value based on the standard result;

[0033] Judging the size between the training result and the ideal maximum value and the limit minimum value;

[0034] If the training result is greater than or equal to the ideal maximum value, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class;

[0035] If the training result is less than or equal to the limit minimum value, the actual score corresponding to the sub-class is zero;

[0036] If the training result is greater than the limit minimum value and less than the ideal maximum value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a second proportion item, wherein the second proportion item represents a proportion value obtained by the ideal maximum value, the limit minimum value and the training result, and is less than 1.

[0037] In some other embodiments, the scoring rule includes that the closer the value of the training result to the median value, the higher the score, and the sub-class score is calculated based on a comparison result between the training result and the standard result corresponding to the sub-class, and further includes the following steps:

[0038] Obtaining a preset limit maximum value, limit minimum value and ideal median value based on the standard result, wherein the ideal median value is between the limit maximum value and the limit minimum value;

[0039] Judging the size between the training result and the limit maximum value, the limit minimum value and the ideal median value;

[0040] if the training result is greater than the limit minimum value and less than or equal to the ideal median value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a third proportion term, wherein the first proportion term represents a proportion value obtained by the limit minimum value, ideal median value and the training result, which is less than or equal to 1, wherein if the training result is equal to the ideal median value, the third proportion term is equal to 1;

[0041] if the training result is less than the limit maximum value and greater than or equal to the ideal median value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a fourth proportion term, wherein the fourth proportion term represents a proportion value obtained by the limit maximum value, ideal median value and the training result, which is less than or equal to 1, wherein if the training result is equal to the ideal median value, the fourth proportion term is equal to 1;

[0042] if the training result is greater than or equal to the limit maximum value or less than or equal to the limit minimum value, the actual score corresponding to the sub-class is zero.

[0043] In some other embodiments, the scoring rule includes determining the score according to the correctness of the training result, calculating the sub-class score based on the comparison result between the training result and the standard result corresponding to the sub-class, and further comprising the following steps:

[0044] determining whether the training result is correct;

[0045] if correct, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class;

[0046] if incorrect, the actual score corresponding to the sub-class is zero.

[0047] In some other embodiments, the actual scores of the training results corresponding to the sub-classes are obtained, and further comprising the following steps:

[0048] obtaining the actual scores corresponding to all the sub-classes, and calculating the actual scores corresponding to the categories and the total class;

[0049] taking the actual score corresponding to the total class as the final score corresponding to the dragon boat overall training result model in the training result model corresponding to the total class;

[0050] obtaining a personnel-specific category corresponding to personnel in the categories, and determining the final score corresponding to the specific personnel training result model according to the sum of the actual scores of the personnel-specific categories.

[0051] In summary, the present application includes at least one of the following beneficial technical effects:

[0052] 1. By means of various sensors, the external factors (environment), internal factors (personnel status) and the overall state of the dragon boat are detected in real time during the water training of the dragon boat, and the corresponding training results are obtained based on the processing and analysis of the data, and the training result model corresponding to the personnel and the overall state of the dragon boat is generated, so that each water training can be monitored and analyzed scientifically and clearly, and the result and effect of each water training can be determined.

[0053] 2. First, a number of training results are classified in combination with the corresponding association relationship and type to obtain a set relationship of three levels, and the corresponding weight is obtained in combination with the preset total score and the importance of each classification and subclass in its superior set, and the theoretical highest score corresponding to each classification and subclass is obtained in combination with the weight and the preset total score, and finally the final actual score is generated in combination with the comparison result between the training result and the standard result, so that the scoring rule of the training result model is more scientific and comprehensive. At the same time, the above scoring method makes the performance of the training result more intuitive and simple. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 is a module schematic diagram of the dragon boat training result model establishment system in the present application;

[0055] Figure 2 is a flowchart of the model scoring method in the present application. DETAILED DESCRIPTION

[0056] The following will be described in combination with the accompanying Figure 1 - the accompanying Figure 2 , the present application will be further described in detail.

[0057] As Figure 1 shown, the present application discloses a dragon boat training structure model monitoring system, which comprises:

[0058] An environment monitoring module is used for monitoring the external environmental factors of the dragon boat to obtain first monitoring data.

[0059] The external environment includes the water environment, the underwater environment, the wind force, etc. Among them, the environment monitoring module can include high-speed video shooting instruments and wind force sensors arranged on the dragon boat, water flow sensors arranged below the dragon boat, and other sensing devices, the high-speed video shooting instruments on the dragon boat can shoot the environmental conditions on the water surface during the training process, the wind force sensors can monitor the water surface wind speed in real time during the training, and the water flow sensors arranged below the dragon boat can detect the water flow speed under the water during the training in real time.

[0060] A personnel monitoring module is used for monitoring the personnel status on the dragon boat to obtain second monitoring data.

[0061] The personnel state includes a rowing action, a rowing frequency, a body index, a drumming frequency, and the like, and can further include rowing uniformity of multiple personnel, a paddle holding posture, a paddle blade rowing route, and the like.

[0062] The personnel monitoring module can include a multi-angle, multi-point high-speed video shooting instrument arranged on the dragon boat, a body index detection instrument worn on the athlete, and the like. The high-speed video shooting instrument can analyze the action made by the personnel, and the body index detection instrument can detect data such as the heartbeat and blood pressure of the personnel.

[0063] The dragon boat monitoring module is configured to monitor the state of the dragon boat to obtain third monitoring data.

[0064] The state of the dragon boat includes a dragon boat moving speed, a dragon boat orientation angle, a dragon boat offset amount, a dragon boat acceleration, and the like.

[0065] The dragon boat monitoring module can include a plurality of sensing components, such as an acceleration sensor, a speed sensor, a level, a gyroscope, and the like. The dragon boat monitoring module can detect data such as the speed, acceleration, and angle of the dragon boat.

[0066] The data processing module is electrically connected to the environment monitoring module, the personnel monitoring module, and the dragon boat monitoring module to obtain the first monitoring data, the second monitoring data, and the third monitoring data, respectively, and classifies and correlates the first monitoring data, the second monitoring data, and the third monitoring data to obtain a training result set.

[0067] The data processing module can obtain the first monitoring data, the second monitoring data, and the third monitoring data monitored by each monitoring module, and perform corresponding processing on each data to classify and correlate the data.

[0068] The training result set includes a plurality of training results, specifically, an independent training set and an associated training set.

[0069] The independent training set is characterized in that, after processing, classifying, and correlating a plurality of monitoring data, there is no data associated with or in the same classification as the independent training set. Conversely, the associated training set is characterized in that, after processing, classifying, and correlating a plurality of monitoring data, there is data associated with or in the same classification as the associated training set.

[0070] Specifically, the association indicates that there is a possibility of mutual influence between a plurality of data.

[0071] The modeling module is configured to obtain the training result set generated by the data processing module, and generate a corresponding training result model based on the training result set.

[0072] The modeling module is configured to generate a corresponding training result model according to the set of training results, which is a performance model of the training results of the personnel or the dragon boat, so that the commander or the coach can obtain the actual results in the training process according to the training result model.

[0073] The training result model includes a specific personnel training result model and a dragon boat overall training result model. The specific personnel training result model corresponds to each independent personnel on the dragon boat and is used to reflect the training results of each independent personnel. The dragon boat overall training result model corresponds to each dragon boat as a whole and is mainly used to reflect the training results of all personnel on the dragon boat, and the training results of all personnel correspond to the training performance of the dragon boat.

[0074] Through the above scheme, the external factors (environment), internal factors (personnel state) and overall state of the dragon boat during the water training of the dragon boat are detected in real time by means of various sensors, and the corresponding training results are obtained based on the processing and analysis of the data, and the training result models corresponding to the personnel and the dragon boat as a whole are generated based on the training results, so that each water training can be monitored and analyzed scientifically and clearly, and the results and effects of each water training can be determined.

[0075] In some other embodiments, the data processing module further includes:

[0076] The associated items of each external environmental factor, personnel state and dragon boat state are obtained and mapped, and the monitoring data existing in the same associated item are obtained and data associated to obtain associated data items.

[0077] The monitoring data not existing in the same associated item are obtained to obtain independent data items.

[0078] The associated training set is obtained according to a plurality of associated data items, and the independent training set is obtained according to a plurality of independent data items.

[0079] The associated item is represented by the associated content corresponding to the influence object corresponding to each data. For example, the associated item of the wind force includes the dragon boat moving speed, the dragon boat offset amount, the dragon boat acceleration, and also includes the paddle frequency. It is represented that the size and direction of the wind force may have an impact on the dragon boat moving speed, the dragon boat offset amount, the dragon boat acceleration, and the paddle frequency corresponding to the personnel will also be affected in order to maintain the speed and direction of the dragon boat. Therefore, the wind force will be associated with the data of the dragon boat moving speed, the dragon boat offset amount, the dragon boat acceleration and the paddle frequency. Therefore, after mapping, there will be an association between these data, and an associated data item will be formed.

[0080] The independent data item is represented by the fact that no other data has the same associated item as it, so the change of the data will not affect any other data. At this time, the data is represented by an independent data item.

[0081] By processing and correlation analysis of various data, the correlation training set and the independent training set can be obtained, so as to further integrate and analyze various data. Instead of regarding various data as independent data without any connection, the influence hidden danger between multiple data is fully considered, the dynamic change between multiple data is considered, and scientific result analysis data is obtained according to each training combination.

[0082] In other embodiments, the correlation data item further includes positive correlation and / or negative correlation, wherein the positive correlation is characterized by the fact that when one or more data in the multiple monitoring data of the same correlation item change, a positive change effect is generated on other monitoring data; and the negative correlation is characterized by the fact that when one or more data in the multiple monitoring data of the same correlation item change, a negative change effect is generated on other monitoring data.

[0083] For example, in the process of dragon boat race, the drummer needs to maintain a certain drumming frequency. If the drumming frequency changes, it will cause the rowing rate of each rower to change or be inconsistent, which will greatly affect the moving speed and direction of the dragon boat. Therefore, in the training process, if the drumming frequency changes, the corresponding correlation data item of the training result will be negatively correlated with the data such as the moving speed of the dragon boat, the offset of the dragon boat, the acceleration of the dragon boat, and the paddle frequency, because the change of the drumming frequency has a negative impact on these data.

[0084] If the wind force changes and the wind direction is the forward direction of the dragon boat, the increase of the wind force will have a positive impact on the moving speed and acceleration of the dragon boat, so the wind force is negatively correlated with the moving speed and acceleration of the dragon boat.

[0085] At the same time, most of the correlation data items corresponding to the data themselves have both positive correlation and negative correlation. For example, if the paddle frequency increases, it is positively correlated with the acceleration and moving speed of the dragon boat, and if the paddle frequency decreases, it is negatively correlated with the acceleration and moving speed of the dragon boat. Therefore, when judging the correlation data item, not only the specific data content corresponding to the correlation data item needs to be judged, but also the change of the correlation data item needs to be considered.

[0086] In other embodiments, a model distribution module is further included, which is connected to the modeling module to obtain a plurality of training result models. The model distribution module is used to obtain target information corresponding to the plurality of training result models, and according to the target information, the specific personnel training result model is sent to the corresponding personnel mobile terminal, and all specific personnel training models and the dragon boat overall training result model are sent to the person in charge terminal.

[0087] The model distribution module is configured to send each training result model to a corresponding mobile terminal according to target information corresponding to each training result model, wherein the specific personnel training result model is sent to a corresponding mobile terminal of each specific personnel for each specific personnel to check the training result of each specific personnel, and the all specific personnel training model and the dragon boat overall training result model are sent to a mobile terminal corresponding to a person in charge, so that the person in charge such as a coach and a team leader can check the training result of all personnel and the training result of each dragon boat team.

[0088] The target information can be obtained by checking which sensors or high-speed video shooting instruments send each monitoring information in each training result model. Since each dragon boat has independent monitoring instruments, the corresponding instrument parameters or instrument numbers of the instruments sending the monitoring data can be used to determine the corresponding target information. The target information of each specific personnel can be obtained by image analysis and processing of images shot by high-speed video shooting instruments.

[0089] As shown in Figure 2 The model scoring method is used to score the training result model generated by the dragon boat training result model supervision system, and includes the following steps:

[0090] S100, obtain a plurality of training results in a training result model, and set classification labels for the plurality of training results according to a predetermined classification method to obtain a total class, a classification, and a subclass.

[0091] The total class represents the sum of all training results in the training result model, the classification represents a set corresponding to a plurality of associated training results, and the classification represents a single training result.

[0092] Therefore, the total class includes a plurality of classifications, and each classification includes a plurality of subclasses.

[0093] For example, the total class is the overall training result of dragon boat A, the classification includes dragon boat maneuvering, and the classification of dragon boat maneuvering includes dragon boat speed, dragon boat acceleration, dragon boat orientation angle, dragon boat offset, and the like.

[0094] The plurality of training results are associated and classified in a hierarchical manner through classification, which facilitates subsequent scoring of details and totals.

[0095] S200, set a predetermined ideal total score for the total class.

[0096] The ideal total score represents the highest score in the ideal total class, for example, if the ideal total score is 100 points, it means that the sum of the scores of all training results in the training result model can reach 100 points at most.

[0097] S300, acquire the first weight corresponding to each category in the total category, and calculate the category score corresponding to each category based on the first weight and the preset ideal total score. The category score represents the highest score that can be allocated to each category.

[0098] According to the importance of the set of training results corresponding to each category in the total category, the corresponding first weight is acquired. According to the product of the first weight corresponding to each category and the ideal score, the category score corresponding to each category is acquired. The category score represents the highest score that can be allocated to each category in the ideal.

[0099] For example, there are four categories in the total category, namely A, B, C, and D. The weight of category A is 0.4, the weight of category B is 0.2, the weight of category C is 0.25, and the weight of category D is 0.15. This indicates that the ideal highest score that can be allocated to category A is 40 points, the ideal highest score that can be allocated to category B is 20 points, the ideal highest score that can be allocated to category C is 25 points, and the ideal highest score that can be allocated to category D is 15 points.

[0100] The first weight can be obtained by analytic hierarchy process or pairwise comparison method. By comparing the n training results pairwise, the corresponding matrix is obtained, and the weight of the index corresponding to each training result is obtained by standardizing the feature vector in the matrix. This weight is the first weight. In addition, the preliminary generated first weights can be submitted to experts, commanders, team leaders, etc. for evaluation. After several responsible persons analyze and adjust the corresponding first weights, the average value of the modifications of each responsible person is taken to adjust the first weights accordingly.

[0101] S400, acquire the second weight corresponding to each subcategory in its category, and calculate the subcategory score corresponding to each subcategory based on the second weight and the category score of its corresponding category. The subcategory score represents the highest score that can be allocated to each subcategory.

[0102] After acquiring the category scores corresponding to several categories, acquire the second weight corresponding to all subcategories in each category. The second weight represents the importance of several subcategories in their corresponding categories. Then, calculate the subcategory score corresponding to each subcategory based on the second weight of the subcategory and the category score of the category.

[0103] The subcategory score represents the highest score that can be allocated to the subcategory in the ideal.

[0104] The classification score of the classification A is 40 points, the classification A includes sub-classes X, Y and Z, the second weight of the sub-class X is 0.5, the second weight of the sub-class Y is 0.3, and the second weight of the sub-class Z is 0.2, so the final sub-class score of the sub-class X is 20 points, that is, the sub-class X can obtain at most 20 points, the sub-class score of the sub-class Y is 12 points, that is, the sub-class Y can obtain at most 12 points, and the sub-class score of the sub-class Z is 8 points, that is, the sub-class Z can obtain at most 8 points.

[0105] S500, based on the comparison result between the training result corresponding to the sub-class and the standard result, the sub-class score is calculated combined with the scoring rule corresponding to the sub-class, to obtain the actual score of the training result corresponding to the sub-class.

[0106] All the above scores are the highest scores that the total class, classification and sub-class can obtain in theory, so the actual score corresponding to each sub-class needs to be judged by comparing the training result corresponding to the sub-class with the preset standard result to determine how many points the sub-class can obtain in the theoretical highest score. If the training result is close to the standard result, the actual score obtained is close to the sub-class score, and vice versa, if the training result is not close to the standard result, the actual score obtained is less than the classification score.

[0107] Through the above method, first, a plurality of training results are classified combined with the corresponding association relationship and type to obtain the set relationship of the top three, and the corresponding weight is obtained combined with the preset total score and the importance of each classification and sub-class in the upper set, the theoretical highest score corresponding to each classification and sub-class is obtained combined with the weight and the preset total score, and finally the final actual score is generated combined with the comparison result between the training result and the standard result, so that the scoring rule of the training result model is more scientific and comprehensive. At the same time, the above scoring method makes the performance of the training result more intuitive and simple.

[0108] In some other embodiments, the scoring rule includes that the smaller the value of the training result, the higher the score, and the sub-class score is calculated based on the comparison result between the training result corresponding to the sub-class and the standard result, and further includes the following steps:

[0109] S510, the preset ideal minimum value and the limit maximum value are obtained based on the standard result.

[0110] S511, the size between the training result and the ideal minimum value and the limit maximum value is judged.

[0111] S512, if the training result is less than or equal to the ideal minimum value, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class.

[0112] S513, if the training result is greater than or equal to the limit maximum value, the actual score corresponding to the sub-class is zero.

[0113] S514, if the training result is greater than the ideal minimum value and less than the limit maximum value, the actual score corresponding to the sub-class is equal to the sub-class score multiplied by the first proportion item, wherein the first proportion item represents a proportion value obtained from the ideal minimum value, the limit maximum value and the training result, and is less than 1.

[0114] In the embodiments of the present application, the smaller the value corresponding to the training result is, the higher the score is. The training result corresponding to this type of scoring rule includes the dragon boat offset, wind power, etc.

[0115] When this type of training result is scored, first, the preset ideal minimum value and limit maximum value are obtained based on the standard result. The ideal minimum value represents the value corresponding to the best result, and when less than or equal to the minimum value, the training result is optimal. And the limit maximum value represents the worst training result that can be accepted, and when exceeding the maximum value, the training result is unacceptable because it is too bad.

[0116] Compare the training result with its corresponding ideal minimum value and limit maximum value. If the training result is less than or equal to the ideal minimum value, it means that the training result is very good, and at this time the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class, that is, the sub-class can get the highest score in the ideal score that can be allocated to the sub-class.

[0117] If the training result is greater than or equal to the limit maximum value, it means that the training result corresponding to the sub-class is very bad, and at this time the actual score corresponding to the sub-class is zero, that is, the training result cannot get the score.

[0118] When the training result is less than the limit maximum value and greater than the ideal minimum value, it means that the training result corresponding to the sub-class cannot get all the scores, and it needs to judge how many scores it can get in the ideal highest score according to the specific value of the training result.

[0119] Specifically, it needs to multiply the sub-class score by the corresponding first proportion item, which is not greater than 1, and represents the gap between the training result corresponding to the sub-class and the standard result. The larger the gap is, the smaller the first proportion item is, and the smaller the gap is, the larger the first proportion item is. In the embodiments of the present application, the first proportion item is:

[0120] In other embodiments, the scoring rule includes that the larger the value of the training result is, the higher the score is, and the sub-class score is calculated based on the comparison result between the training result corresponding to the sub-class and the standard result, and further includes the following steps:

[0121] S520, obtaining a preset ideal maximum value and a limit minimum value based on the standard result.

[0122] S521, judging the size between the training result and the ideal maximum value and the limit minimum value.

[0123] S522, if the training result is greater than or equal to the ideal maximum value, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class.

[0124] S523, if the training result is less than or equal to the limit minimum value, the actual score corresponding to the sub-class is zero.

[0125] S524, if the training result is greater than the limit minimum value and less than the ideal maximum value, the actual score corresponding to the sub-class is equal to the score obtained by multiplying the sub-class score by a second proportion item, wherein the second proportion item represents a proportion value obtained by the ideal maximum value, the limit minimum value and the training result, and is less than 1.

[0126] In the embodiment of the present application, the larger the value corresponding to the training result of the scoring rule is, the higher the score is. The training results corresponding to this type of scoring rule include dragon boat moving speed, dragon boat acceleration, paddle frequency, etc.

[0127] Similar to the overall steps described above, the smaller the value of the training result is, the higher the score is. The difference is that in the embodiment of the present application, the ideal maximum value and the limit minimum value are obtained based on the standard result. The ideal maximum value represents the value corresponding to the best result, and when greater than or equal to the maximum value, the training result is optimal. The limit minimum value represents the worst training result that can be accepted, and when less than the minimum value, the training result is unacceptable because it is too poor.

[0128] It should be noted that the second proportion item in the embodiment of the present application is: The closer the training result is to the ideal maximum value, the larger the second proportion item is, and the higher the actual score corresponding to the sub-class is after multiplying the classification score. On the contrary, the closer the training result is to the limit minimum value, the smaller the second proportion item is, and the lower the actual score corresponding to the sub-class is after multiplying the classification score.

[0129] In other embodiments, the scoring rule includes that the closer the value of the training result is to the median, the higher the score is, and the sub-class score is calculated based on the comparison result between the training result corresponding to the sub-class and the standard result, further comprising the following steps:

[0130] S530, obtaining a preset limit maximum value, limit minimum value and ideal median value based on the standard result, wherein the ideal median value is between the limit maximum value and the limit minimum value.

[0131] S531, judging the size between the training result and the limit maximum value, the limit minimum value and the ideal median value.

[0132] S532, if the training result is greater than the limit minimum value and less than or equal to the ideal median value, the actual score corresponding to the sub-class is equal to the score obtained by multiplying the sub-class score by the third proportion term, wherein the first proportion term represents a proportion value obtained by the limit minimum value, the ideal median value and the training result, which is less than or equal to 1, wherein if the training result is equal to the ideal median value, the third proportion term is equal to 1.

[0133] S533, if the training result is less than the limit maximum value and greater than or equal to the ideal median value, the actual score corresponding to the sub-class is equal to the score obtained by multiplying the sub-class score by the fourth proportion term, wherein the fourth proportion term represents a proportion value obtained by the limit maximum value, the ideal median value and the training result, which is less than or equal to 1, wherein if the training result is equal to the ideal median value, the fourth proportion term is equal to 1.

[0134] S534, if the training result is greater than or equal to the limit maximum value or less than or equal to the limit minimum value, the actual score corresponding to the sub-class is zero.

[0135] In the embodiments of the present application, the closer the value corresponding to the training result is to the median value, the better the corresponding training result is, and the higher the corresponding score is. The training results to which this type of scoring rules apply include drumming frequency, physical indicators, water environment, underwater environment, etc.

[0136] The reference values for this type of training result judgment include the limit maximum value, the limit minimum value and the ideal median value, wherein the ideal median value represents the value corresponding to the best training result, and the limit maximum value and the limit minimum value represent the worst training results that can be accepted, and when less than the minimum value or greater than the maximum value, the training result is unacceptable due to being too poor.

[0137] Wherein, when making specific judgments, judgments are made based on the third proportion term and the fourth proportion term.

[0138] The third proportion term is applied to the case where the training result is greater than the limit minimum value and less than or equal to the ideal median value, and at this time the third proportion term is less than or equal to 1, and specifically the third proportion term is: As can be seen from the above, the closer the value corresponding to the training result is to the ideal median value, the greater the value of the third proportion term is, and the closer the training result is to the limit minimum value, the smaller the value of the third proportion term is, and when the training result is equal to the ideal median value, the third proportion term is equal to 1. That is, the closer the value corresponding to the training result is to the ideal median value, the higher the corresponding score is, the closer the training result is to the limit minimum value, the lower the corresponding score is, and when the training result is equal to the ideal median value, the sub-class corresponding to the training result is equal to the sub-class score, that is, the ideal highest score can be obtained.

[0139] The fourth proportional term is applied to the case where the training result is less than the limit maximum value and greater than or equal to the ideal median value, and the fourth proportional term is less than or equal to 1, specifically, the fourth proportional term is: As can be seen from the above, the closer the value corresponding to the training result is to the ideal median value, the greater the value of the third proportional term, and the closer the training result is to the limit maximum value, the smaller the value of the third proportional term, and when the training result is equal to the ideal median value, the third proportional term is equal to 1. That is, the closer the value corresponding to the training result is to the ideal median value, the higher the corresponding score, the closer the training result is to the limit maximum value, the lower the corresponding score, and when the training result is equal to the ideal median value, the sub-class corresponding to the training result is equal to the sub-class score, that is, the ideal highest score can be obtained.

[0140] When the value corresponding to the training result is greater than or equal to the limit maximum value or less than or equal to the limit minimum value, the score corresponding to the training result is zero, indicating that the training result is very poor.

[0141] In other embodiments, the scoring rule includes determining the score according to the correctness of the training result, calculating the sub-class score based on the comparison result between the training result corresponding to the sub-class and the standard result, and further comprising the following steps:

[0142] S540, determining whether the training result is correct.

[0143] S541, if correct, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class.

[0144] S542, if incorrect, the actual score corresponding to the sub-class is zero.

[0145] In the embodiments of the present application, the training result does not consider the specific value, but judges whether the training result is correct based on the standard result, and scores according to the judgment result of correctness. The training result using such a scoring rule includes: paddle action, dragon boat angle, paddle water route, paddle posture, etc.

[0146] If the training result is correct, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class, that is, the sub-class can obtain the highest score in the ideal value, and if the training result is incorrect, the actual score is zero.

[0147] In other embodiments, the actual scores of the training results corresponding to the plurality of sub-classes are obtained, and further comprising the following steps: S600, obtaining the actual scores of all sub-classes, and calculating the actual scores of the plurality of categories and the total category.

[0148] After obtaining the actual scores of each sub-class, the actual scores of the plurality of sub-classes are added to obtain the actual scores of the categories and the total category.

[0149] S610, taking the actual score corresponding to the total category as the final score corresponding to the dragon boat overall training result model in the corresponding training result model thereof.

[0150] The total category corresponds to the overall score of the training result model, so the final score corresponding to the dragon boat overall training model can be calculated according to the actual score corresponding to the total category, and the final score corresponding to the dragon boat overall training model is the overall training result of the dragon boat in the training.

[0151] S620, obtaining personnel-specific categories corresponding to the personnel in the plurality of categories, and determining the final score corresponding to the specific personnel training result model according to the sum of the actual scores of the plurality of personnel-specific categories.

[0152] The training score of the specific personnel can be obtained according to the sum of the overall actual scores of the specific categories corresponding to the personnel in the plurality of categories, because part of the scores in the total category are training scores for the dragon boat itself and the plurality of specific personnel as a whole, so these scores need to be excluded and only the actual scores corresponding to the categories for the individual are obtained.

[0153] The embodiments of the specific implementation are the preferred embodiments of the present application, and are not intended to limit the protection scope of the present application, wherein the same parts are denoted by the same reference numerals. Therefore, any equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A model scoring method for scoring a training result model generated by a dragon boat training result model establishment system, characterized in that, A dragon boat training result model establishment system comprises: An environmental monitoring module for monitoring external environmental factors of a dragon boat to obtain first monitoring data, the external environmental factors including water environment, underwater environment, wind power; A personnel monitoring module for monitoring personnel states on the dragon boat to obtain second monitoring data, the personnel states including paddle action, paddle frequency, physical indicators, drumming frequency; A dragon boat monitoring module for monitoring the state of the dragon boat to obtain third monitoring data, the state of the dragon boat including the speed of the dragon boat, the angle of the dragon boat, the offset of the dragon boat, and the acceleration of the dragon boat; A data processing module electrically connected to the environmental monitoring module, the personnel monitoring module, and the dragon boat monitoring module to obtain corresponding first monitoring values, second monitoring values, and third monitoring values, and to classify and correlate the first monitoring values, the second monitoring values, and the third monitoring values to obtain a training result set, the training result set containing a plurality of training results, and specifically including an independent training set and a correlated training set; A modeling module for obtaining the training result set generated by the data processing module and generating a corresponding training result model based on the training result set, the training result model including a specific personnel training result model and a dragon boat overall training result model; The method comprises: obtaining a plurality of training results in the training result model, and setting classification labels for the plurality of training results according to a predetermined classification method to obtain a total class, a classification, and a subclass, wherein the total class contains a plurality of classifications, the classification contains a plurality of subclasses, the total class represents the sum of all training results in the training result model, the classification represents a set corresponding to a plurality of correlated training results, and the subclass represents a single training result; A preset ideal total score is set for the total class; A first weight corresponding to a plurality of classifications in the total class is obtained, and a classification score corresponding to each classification is calculated based on the first weight and the preset ideal total score, the classification score representing the highest score value that can be allocated to a plurality of classifications; A second weight corresponding to a plurality of subclasses of each classification in the classification is obtained, and a subclass score corresponding to each subclass is calculated based on the second weight and the classification score of the classification corresponding thereto, the subclass score representing the highest score value that can be allocated to a plurality of subclasses; The subclass score is calculated based on the comparison result between the training result corresponding to the subclass and the standard result and the scoring rule corresponding to the subclass to obtain the actual score of the training result corresponding to a plurality of subclasses.

2. The model scoring method of claim 1, wherein, The scoring rule includes that the smaller the value of the training result, the higher the score, and the subclass score is calculated based on the comparison result between the training result corresponding to the subclass and the standard result, and further includes the following steps: A preset ideal minimum value and an extreme maximum value are obtained based on the standard result; judging a size between the training result and the ideal minimum value and the limit maximum value; if the training result is less than or equal to the ideal minimum value, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class; if the training result is greater than or equal to the limit maximum value, the actual score corresponding to the sub-class is zero; if the training result is greater than the ideal minimum value and less than the limit maximum value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a first proportion item, wherein the first proportion item represents a proportion value obtained by the ideal minimum value, the limit maximum value and the training result, and is less than 1.

3. The model scoring method of claim 1, wherein, The scoring rule includes that the greater the value of the training result, the higher the score, and the sub-class score is calculated based on the comparison result between the training result and the standard result of the sub-class, and further includes the following steps: obtaining a preset ideal maximum value and limit minimum value based on the standard result; judging a size between the training result and the ideal maximum value and the limit minimum value; if the training result is greater than or equal to the ideal maximum value, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class; if the training result is less than or equal to the limit minimum value, the actual score corresponding to the sub-class is zero; if the training result is greater than the limit minimum value and less than the ideal maximum value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a second proportion item, wherein the second proportion item represents a proportion value obtained by the ideal maximum value, the limit minimum value and the training result, and is less than 1.

4. The model scoring method of claim 1, wherein, The scoring rule includes that the closer the value of the training result to the median value, the higher the score, and the sub-class score is calculated based on the comparison result between the training result and the standard result of the sub-class, and further includes the following steps: obtaining a preset limit maximum value, limit minimum value and ideal median value based on the standard result, wherein the ideal median value is between the limit maximum value and the limit minimum value; judging a size between the training result and the limit maximum value, the limit minimum value and the ideal median value; if the training result is greater than the limit minimum value and less than or equal to the ideal median value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a third proportion item, wherein the third proportion item represents a proportion value obtained by the limit minimum value, the ideal median value and the training result, and is less than or equal to 1, and wherein if the training result is equal to the ideal median value, the third proportion item is equal to 1; if the training result is less than the limit maximum value and greater than or equal to the ideal median value, the actual score corresponding to the sub-class is equal to a score obtained by multiplying the sub-class score by a fourth proportion item, wherein the fourth proportion item represents a proportion value obtained by the limit maximum value, the ideal median value and the training result, and is less than or equal to 1, and wherein if the training result is equal to the ideal median value, the fourth proportion item is equal to 1; If the training result is greater than or equal to the limit maximum value or less than or equal to the limit minimum value, the actual score corresponding to the sub-class is zero.

5. The model scoring method of claim 1, wherein, The scoring rule includes determining the score according to the correctness of the training result, calculating the sub-class score based on the comparison between the training result and the standard result of the sub-class, and further comprising the following steps: determining whether the training result is correct; if correct, the actual score corresponding to the sub-class is equal to the sub-class score corresponding to the sub-class; if incorrect, the actual score corresponding to the sub-class is zero.

6. The model scoring method of claim 1, wherein, Obtaining the actual score of the training result corresponding to a plurality of sub-classes further comprises the following steps: obtaining the actual score of all sub-classes and calculating the actual score of a plurality of classifications and the total class; the actual score of the total class is used as the final score corresponding to the overall training result model in the training result model corresponding to the total class; obtaining a personnel-specific classification corresponding to personnel in a plurality of classifications, and determining the final score corresponding to the specific personnel training result model according to the sum of the actual scores of a plurality of personnel-specific classifications. 7.A dragon boat training result model establishing system, characterized in that, The method of any one of claims 1 to 6 is executed, and the data processing module further comprises: obtaining the associated items of each external environmental factor, personnel state and dragon boat state and mapping, and obtaining a plurality of monitoring data existing in the same associated item and data association to obtain associated data items; obtaining monitoring data that does not exist in the same associated item to obtain independent data items; obtaining an associated training set according to a plurality of associated data items, and obtaining an independent training set according to a plurality of independent data items. 8.The dragon boat training result model establishing system according to claim 7, wherein, The associated data items further include positive association and / or negative association, wherein the positive association represents that when one or more data in a plurality of monitoring data existing in the same associated item changes, it generates a positive change effect on other monitoring data, and the negative association represents that when one or more data in a plurality of monitoring data existing in the same associated item changes, it generates a negative change effect on other monitoring data. 9.The dragon boat training result model establishing system according to claim 7, wherein, Further comprising a model distribution module connected to the modeling module to obtain a plurality of training result models, the model distribution module is used to obtain target information corresponding to a plurality of training result models, the target information includes personnel name, personnel position number, dragon boat number, and according to the target information, the specific personnel training result model is sent to the corresponding personnel mobile terminal, and all the specific personnel training result model and the overall training result model of the dragon boat are sent to the person in charge terminal.

Images