Intelligent lottery sorting method and system

Abstract

The invention relates to the technical field of sorting, in particular to an intelligent lottery sorting method and system, and the method comprises the steps: obtaining order data containing timestamps, lottery numbers, lottery numbers, storage space data and replenishment data from a lottery order analysis system, and carrying out the preprocessing; dividing the preprocessed data into a training set, training to obtain a first prediction model, and obtaining a standard data set of all defect types; and determining the optimal standard data by calculating the approximation of each order data and the standard data. And calculating an expansion coefficient value of each data point by using a time sequence similarity algorithm, and adjusting the length of the order data according to the expansion coefficient value for secondary training of the first prediction model to obtain a second prediction model. And after the real-time order data is input into the second prediction model, obtaining a sorting risk type and adjusting the sorting system to complete order sorting. According to the method, the sorting efficiency and accuracy can be improved, the system adaptability and stability are enhanced, resource allocation is optimized, and the cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of sorting technology, and in particular to a lottery intelligent sorting method and system. Background Art

[0002] In today's wave of digitalization and automation, the lottery industry is also constantly seeking technological innovation to improve operational efficiency. As an emerging technology, the lottery intelligent sorting system is gradually becoming the focus of the industry. By combining artificial intelligence, machine learning and automated equipment, the system can quickly identify and classify lottery tickets, and allocate lottery tickets to the correct channels or inventory according to real-time needs. Compared with traditional sorting methods, the intelligent sorting system not only improves sorting speed and accuracy, but also significantly reduces the human error rate, providing an efficient and reliable solution for the lottery industry. The lottery industry faces growing market demand and complex operational challenges. Traditional sorting methods often rely on manual operations or simple rule engines. This model is prone to sorting congestion, high error rates, and poor inventory management when facing a surge in order volume or peak periods. To meet these challenges, intelligent sorting systems have emerged. By introducing advanced algorithms and automation technologies, intelligent sorting systems can analyze order flow and inventory status in real time and dynamically adjust sorting strategies to avoid congestion and stagnation. At the same time, it can significantly reduce the human error rate, ensure that lottery tickets can be accurately distributed to users, and improve user experience and operational efficiency.

[0003] Although the intelligent sorting system has many advantages, it may still face some challenges in practical application. Traditional sorting systems usually adopt the FIFO (first in, first out) strategy based on the rule engine, which may cause sorting jams or even stagnation when the order volume suddenly increases. In addition, insufficient lottery tickets are also a common problem, especially in high-demand periods, which may lead to unavailability of supply during sorting, affecting user experience and sales efficiency. Summary of the invention

[0004] In order to solve the problem that the traditional sorting system adopts the FIFO strategy and lacks dynamic adjustment capability, which easily leads to sorting congestion and supply interruption when the order volume surges or the lottery ticket balance is insufficient, the present invention provides solutions in the following aspects.

[0005] In a first aspect, a lottery intelligent sorting method includes: obtaining order data from a lottery order analysis system and preprocessing, wherein the order data includes: timestamp, lottery number, number of lottery tickets, position data and replenishment data; dividing the preprocessed order data into a training set, and training a preset prediction model to obtain a trained first prediction model, and obtaining a standard data set of all defect types in the first prediction model training; calculating the approximation between each input order data and multiple defect type standard data in the training set, and obtaining the best standard data for each order data; using a time series similarity algorithm, according to the correspondence between data points in the best standard data of each input order data, and calculating the expansion coefficient value of each data point in the input order data; completing order data length adjustment according to the expansion coefficient value, and using the adjusted order data length as input data for secondary training of the first prediction model to obtain a second prediction model; inputting real-time order data into the second prediction model, obtaining the sorting risk type of the real-time order data, and adjusting the sorting system to complete order sorting.

[0006] The effect is as follows: by obtaining and preprocessing order data including timestamp, lottery number, lottery quantity, warehouse data and replenishment data from the lottery order analysis system, the first prediction model and its standard data set are trained, the approximation between each order data and the defect type standard data is calculated to obtain the best standard data, the expansion coefficient value is calculated using the time series similarity algorithm and the order data length is adjusted, and the adjusted data is used for secondary training to obtain the second prediction model. The real-time order data is input into the second prediction model, the sorting risk type is obtained and the sorting system is adjusted to complete the order sorting. It not only improves the sorting efficiency and accuracy, but also enhances the adaptability and stability of the system, while optimizing resource allocation and reducing costs.

[0007] Preferably, the preprocessing of the order data includes: The order data is labeled according to the defect type, where the defect types include: too many orders, too few orders, too many lottery tickets remaining, too few lottery tickets remaining, and normal sorting; the labeled data is encoded to complete the preprocessing of the order data.

[0008] Preferably, the first prediction model comprises: The first prediction model selects to use a CNN-LSTM network, and adopts a multi-classification cross entropy loss function as the loss function of the first prediction model; The encoded data set is divided into a training set and a validation set. The training set is organized into a matrix according to the time series, with the order data at a preset number of moments as the training set for the preset prediction model.

[0009] The effect is that by using the CNN-LSTM network as the first prediction model and using the multi-classification cross entropy loss function, the time series characteristics of the order data can be effectively processed, and the model's ability to identify different types of defects can be improved. The encoded data set is divided into a training set and a validation set, and the training set is organized into a matrix input model according to the time series, which not only improves the efficiency and accuracy of the model training, but also enhances the model's adaptability to dynamic changes in order data.

[0010] Preferably, the standard data set includes: During the training process of the first prediction model, the input order data corresponding to the lowest cross entropy loss function value is used as the standard data of the corresponding defect type, and the standard data of all defect types constitute a standard data set.

[0011] The effect is that by using standard data, order data preprocessing, risk assessment and sorting control can be more effectively performed, thereby enhancing the adaptability and stability of the system, optimizing resource allocation, reducing operating costs, and improving overall sorting efficiency and system stability.

[0012] Priority should be given to the best standard data, including: The DTW algorithm is used to calculate the approximation between each input order data and each standard data in the standard data set, and the corresponding DTW value sequence is obtained. The minimum DTW value in the DTW value sequence is obtained, the closest defect type is determined, and the corresponding standard data is used as the optimal standard data for each input order data.

[0013] The effect is that by using the DTW algorithm to calculate the approximation between each input order data and each standard data in the standard data set, the time series characteristics of the order data can be effectively processed, and even the data lengths can be accurately matched. By obtaining the minimum value in the DTW value sequence, the closest defect type between the input data and the standard data set can be accurately determined, thereby providing the most matching standard data for the order data. Not only does it improve the matching accuracy between order data and standard data, it also enhances the system's ability to identify different defect types, providing a reliable basis for subsequent sorting risk assessment and system adjustment, thereby optimizing the sorting process and improving sorting efficiency and accuracy.

[0014] Preferably, the corresponding relationship includes: A cumulative distance matrix between the input order data and the best standard data is constructed using a dynamic time warping algorithm, wherein each element in the cumulative distance matrix represents a local distance between the input order data and the best standard data at a corresponding position; Use the dynamic programming method to obtain the shortest path from the starting point to the end point in the cumulative distance matrix, and determine the corresponding relationship between the input order data and the optimal standard data based on the trajectory corresponding to the shortest path.

[0015] Preferably, the expansion coefficient value includes: The matching degree between each input order data and the standard data and actual data of each defect type is calculated respectively, and the difference is calculated for normalization processing to obtain the expansion coefficient value between each input order data and each defect type.

[0016] Preferably, the step of completing the order data length adjustment comprises the following steps: Perform peak point detection on the expansion coefficient value sequence to obtain a peak sequence, wherein the peak sequence includes peak valley points and peak points; take the two peak points closest to the two ends of the peak sequence as edge peak points, obtain the sequence numbers of the two edge peak points in the expansion coefficient value sequence respectively, calculate the distances between the edge peak points and the starting end and the end respectively, and take the minimum value to obtain the adjustment direction coefficient; According to the data on one side corresponding to the adjustment direction coefficient, the mean of the sequence of expansion coefficient values ​​corresponding to all data points is calculated, and the mean is multiplied by the expansion coefficient value of the peak value corresponding to the input order data to obtain the window adjustment coefficient. The product of the window adjustment coefficient and the adjustment direction coefficient is used as the length adjustment value; In response to the length adjustment value being less than the current data length, data is removed from the tail end so that the length of the data after removal is consistent with the length adjustment value; in response to the length adjustment value being greater than or equal to the current data length, data is added from the tail end so that the length of the data after addition is consistent with the length adjustment value.

[0017] In a second aspect, a lottery intelligent sorting system includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned lottery intelligent sorting method is implemented.

[0018] The present invention has the following effects: 1. The present invention intelligently analyzes order data and dynamically adjusts the order data length, so that the sorting system can process orders more efficiently, reduce sorting errors and blockages caused by order volume fluctuations or insufficient lottery ticket balance, and thus improve sorting efficiency and accuracy.

[0019] 2. The present invention uses a prediction model to perform risk assessment and sorting control on order data, so that the system can adapt to different order patterns and emergencies, reduce risks in the sorting process, and ensure stable operation of the system.

[0020] 3. The present invention rationally allocates sorting resources through intelligent analysis and prediction of order data, avoids waste of resources, reduces operating costs, and improves the overall performance and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a method flow chart of steps S1 to S5 in a lottery intelligent sorting method according to an embodiment of the present invention.

[0022] Figure 2 The present invention is a structural block diagram of a lottery intelligent sorting system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments.

[0024] Reference Figure 1 A lottery intelligent sorting method includes steps S1 to S5, which are as follows: S1: Obtain order data from the lottery order analysis system and perform preprocessing, wherein the order data includes: timestamp, lottery number, lottery quantity, position data and replenishment data.

[0025] Preprocess the order data, including: The order data is labeled according to the defect type, where the defect types include: too many orders, too few orders, too many lottery tickets remaining, too few lottery tickets remaining, and normal sorting; the labeled data is encoded to complete the preprocessing of the order data.

[0026] Further explanation: defect types include but are not limited to too many / too few order quantities, too many / too few lottery ticket balances, and other defect types can be marked and set by implementers according to specific implementation scenarios. Further analysis of possible problems with defect types: Too many orders: The number of orders exceeds the system's processing capacity, resulting in a jam in sorting. Too few orders: Insufficient order quantity may result in a waste of resources or low sorting efficiency. Too many lottery tickets: Too much surplus of a certain lottery type may result in an inventory backlog. Too few lottery tickets: Insufficient surplus of a certain lottery type may result in a lack of supply during sorting. Normal: There are no problems mentioned above, and the sorting process is proceeding normally.

[0027] The purpose of data annotation is to clarify which data points have which types of problems, provide the labels required for supervised learning for the prediction model, and enable the model to learn the characteristics of different defect types. By analyzing the annotated data, common problems in the system can be discovered, and sorting strategies and inventory management can be optimized.

[0028] In this embodiment, one-hot encoding is used for data encoding, and the implementer can adjust the data encoding method according to the specific implementation scenario.

[0029] S2: Divide the preprocessed order data into training sets, and train a preset prediction model to obtain a trained first prediction model, and obtain a standard data set of all defect types in the training of the first prediction model.

[0030] The first prediction model includes: The first prediction model uses the CNN-LSTM network and adopts the multi-classification cross entropy loss function as the loss function of the first prediction model; The encoded data set is divided into a training set and a validation set. The training set is organized into a matrix according to the time series, with the order data at a preset number of moments as the training set for the preset prediction model.

[0031] Specifically, in the process of building the CNN-LSTM network, the input data is firstly extracted from local features through a one-dimensional convolutional layer (CNN part). The format of the input data is (feature dimension, time step), where the feature dimension is composed of one-hot encoding and numerical features (such as the number of lottery tickets, warehouse data). A one-dimensional convolutional layer (nn.Conv1d) is used to perform local convolution on the input features to extract local time series features. The number of input channels is set to input_dim, the number of output channels is set to 16, the convolution kernel size is set to 3, and padding=1 is used to keep the number of time steps unchanged. The output of the convolutional layer introduces nonlinearity through the ReLU activation function, enabling the model to capture more complex feature relationships.

[0032] Next, in order to adapt to the input requirements of LSTM, the output tensor after convolution and activation needs to be transformed in dimension. Use permute(2, 0, 1) to rearrange the data into the format of (number of time steps, batch_size, number of feature channels), where batch_size is 1 because only one sample is input each time.

[0033] Then, the LSTM layer is used for temporal dependency modeling. The input feature dimension of the LSTM layer is 16 (the number of output channels from the CNN part), and the hidden state size is set to 64. The function of the LSTM layer is to capture the long-term dependency in the sequence and improve the ability to recognize dynamic changes in data through temporal modeling.

[0034] Finally, after the LSTM layer, the output of the last time step (i.e., the last hidden state) is taken, which carries the information of the entire sequence. Through a fully connected layer (nn.Linear), the 64-dimensional hidden state is mapped to 5 categories, corresponding to five defect types (e.g., too many orders, too few orders, too many lottery tickets, too few lottery tickets, and normal sorting).

[0035] In this embodiment, the cross entropy loss function of multiple classifications is used It indicates that, Indicates the current round in the training process of the first prediction model. Total training rounds Set to 300, implementers can adjust according to specific implementation scenarios. And adjust the loss function and network model according to specific implementation scenarios and needs. For example, you can choose a different network architecture (such as pure LSTM, GRU, etc.) or adjust the training rounds To optimize model performance.

[0036] In this embodiment, a 1:4 ratio is selected for the division of the validation set and the training set, and the division ratio can be adjusted by the implementer according to the specific implementation scenario. The preset number of moments is 200 moments of order data, and the implementer can adjust it according to the specific implementation method.

[0037] Standard data sets, including: During the training process of the first prediction model, the input order data corresponding to the lowest cross entropy loss function value is used as the standard data of the corresponding defect type, and the standard data of all defect types constitute a standard data set.

[0038] That is to say, when the cross entropy loss function value is the lowest, it means that the defect type prediction result is accurate in the inference result of the first prediction model this time, and because in the multi-classification cross entropy loss function, the better the inference result of the first prediction model, the lower the calculated loss function value.

[0039] It should be noted that when the number of orders suddenly surges, a large number of continuous order input data sets will reflect local features of the same defect type. In this case, the model may overfit these local features and ignore other features, resulting in a decrease in the generalization ability of the model and affecting the overall accuracy; on the contrary, if there are multiple defect types in the order input data, or the multi-dimensional information of the input data changes dramatically, the model may be disturbed by too much irrelevant information, making it difficult to accurately identify the main defect types, thereby affecting the prediction accuracy.

[0040] In order to solve these problems, it is necessary to adaptively adjust the length of the order input data to ensure that the model can better match the characteristics of different defect types and improve the accuracy and stability of the prediction. The specific steps are as follows: S3: Calculate the approximation between each input order data in the training set and multiple defect type standard data, and obtain the best standard data for each order data.

[0041] Best practice data, including: The DTW algorithm is used to calculate the approximation between each input order data and each standard data in the standard data set, and the corresponding DTW value sequence is obtained. The minimum DTW value in the DTW value sequence is obtained, the closest defect type is determined, and the corresponding standard data is used as the optimal standard data for each input order data.

[0042] It should be noted that the DTW value indicates the similarity between the input order data and the standard data of a certain defect type. The smaller the value, the closer the two are. However, even if the value is small, it does not mean that all features in the input data are completely useful. It only means that the input order data is most similar to the standard data of the defect type as a whole, but it does not rule out that the input data may still contain incomplete or redundant information.

[0043] It should also be noted that in the previous training, the length of the input data is fixed. However, in actual applications, the length of the input data may change. The DTW algorithm can process time series data of different lengths, while methods such as the Pearson correlation coefficient usually require the input data to be of the same length. Therefore, the DTW algorithm is used to analyze the expansion coefficient value of each data point. The specific steps are as follows: S4: Using the time series similarity algorithm, according to the correspondence between the data points in the best standard data of each input order data, the expansion coefficient value of each data point in the input order data is calculated.

[0044] It should be noted that a sequence of expansion coefficient values ​​corresponding to all data points in each input order data is obtained, wherein the sequence of expansion coefficient values ​​is one-dimensional data sorted according to the corresponding relationship between all data points in each input order data.

[0045] The corresponding relationship includes: A cumulative distance matrix between the input order data and the best standard data is constructed using a dynamic time warping algorithm, wherein each element in the cumulative distance matrix represents a local distance between the input order data and the best standard data at a corresponding position; Use the dynamic programming method to obtain the shortest path from the starting point to the end point in the cumulative distance matrix, and determine the corresponding relationship between the input order data and the optimal standard data based on the trajectory corresponding to the shortest path.

[0046] That is to say, after obtaining the corresponding relationship, if the input order data has a many-to-one situation with the best standard data, that is, multiple order input data are aligned with one best standard data, it means that the order input data is redundant and needs to be eliminated; if the order input data has a one-to-many situation with the best standard data, that is, one order input data is aligned with multiple best standard data, it means that the order input data is incomplete and needs to be supplemented; if the order input data has a one-to-one situation with the best standard data, it means that the current order input data and the best standard data have a high degree of match.

[0047] However, when expanding or deleting the input order data, it is not possible to remove or expand from the middle of the order data, because this will cause data confusion and it is impossible to determine the exact location of the inserted data. Therefore, it is necessary to obtain the expansion coefficient value corresponding to each data point in each order data.

[0048] Expansion factor values, including: The matching degree between each input order data and the standard data and actual data of each defect type is calculated respectively, and the difference is calculated for normalization processing to obtain the expansion coefficient value between each input order data and each defect type.

[0049] Specifically, the expansion coefficient value satisfies the following relationship: ; In the formula, Indicates The order data entered is the same as the The expansion factor value between defect types, Indicates The order data entered is the same as the The feature matching degree between the standard data of each defect type, Indicates The order data entered is the same as the The feature matching degree between the actual data of the defect types, Represented by natural numbers An exponential function with base .

[0050] That is to say, if , indicating that the order input data has a one-to-one relationship with the best standard data, , indicating that the input data length remains unchanged.

[0051] like , the number of optimal standard data is large, and the number of order input data is small, that is, the order input data has a one-to-many situation with the optimal standard data, which needs to be expanded. , indicating that the order input data length needs to be extended; like , the number of optimal standard data is small, and the number of order input data is large, that is, the order input data has a many-to-one situation with the optimal standard data, which needs to be expanded. , indicating that the input data length needs to be reduced.

[0052] To further explain, although the expansion coefficient value can be used to determine which data point should be expanded or deleted, it is impossible to directly insert or delete data points within the data, so it can only be inserted or deleted at the edge data points, and then the peak point detection is performed on the sequence composed of the expansion coefficient values. The specific steps are as follows: S5: Adjust the order data length according to the expansion coefficient value, and use the adjusted order data length as the input data for the secondary training of the first prediction model to obtain the second prediction model; input the real-time order data into the second prediction model, obtain the sorting risk type of the real-time order data, and adjust the sorting system to complete the order sorting.

[0053] Complete the order data length adjustment, including the following steps: Perform peak point detection on the expansion coefficient value sequence to obtain a peak sequence, wherein the peak sequence includes peak valley points and peak points; take the two peak points closest to the two ends of the peak sequence as edge peak points, obtain the sequence numbers of the two edge peak points in the expansion coefficient value sequence respectively, calculate the distances between the edge peak points to the starting end and to the end respectively, and take the minimum value to obtain the adjustment direction coefficient; Specifically, the directional coefficient is adjusted to satisfy the following relationship: ; In the formula, Indicates The order data entered corresponds to the peak sequence The peak value corresponds to the adjustment direction coefficient of the order data distance from the edge, Indicates The order data entered corresponds to the peak sequence The sequence number of the peak in the sequence of expansion factor values, Indicates The total number of data points of the order data entered, Represents the minimum function.

[0054] That is to say, Reflects the distance between the edge peak and the starting end. It reflects the distance from the edge peak to the end.

[0055] According to the data on one side corresponding to the adjustment direction coefficient, the mean of the sequence of expansion coefficient values ​​corresponding to all data points is calculated, and the mean is multiplied by the expansion coefficient value of the peak value corresponding to the input order data to obtain the window adjustment coefficient; Specifically, the window adjustment coefficient satisfies the following relationship: ; In the formula, Indicates The window adjustment factor of the input order data, It indicates the mean value of the expansion coefficient corresponding to all data on one side of the adjustment direction coefficient. Indicates The order data entered corresponds to the peak sequence The expansion factor value of the peak value.

[0056] That is to say, It reflects the overall feature matching of the current data side. The larger the mean, the longer the length needs to be. The smaller the mean, the shorter the length needs to be. The closer the mean is to 1, the longer the length needs to be. It reflects the feature matching status of the current peak point and is used to adjust the data length.

[0057] The product of the window adjustment coefficient and the adjustment direction coefficient is used as the length adjustment value. In response to the length adjustment value being less than the current data length, data is removed from the tail end so that the length of the data after removal is consistent with the length adjustment value. In response to the length adjustment value being greater than or equal to the current data length, data is added from the tail end so that the length of the data after addition is consistent with the length adjustment value.

[0058] It should be noted that after adjusting the length of the input order data, the length of the next input order data is obtained, which is then used for the secondary training of the first prediction model. The secondary training rounds are set to 1000 times, which can be adjusted by the implementer according to the specific implementation scenario to obtain the trained second prediction model, in which the loss function and other parameters remain consistent with the parameters of the first prediction model.

[0059] That is to say, the collected real-time order data is input into the prediction model to obtain the order risk assessment value of each defect type, the maximum value of the order risk assessment value is selected as the defect type of the real-time order data, and the sorting system is adjusted to complete the order sorting.

[0060] The present invention also provides a lottery intelligent sorting system. Figure 2As shown, the system includes a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a lottery intelligent sorting method according to the first aspect of the present invention is implemented. The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, whose settings and functions are known in the art, and therefore will not be described in detail here.

[0061] It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A lottery intelligent sorting method, characterized in that: include: Obtain order data from the lottery order analysis system and perform preprocessing, wherein the order data includes: timestamp, lottery number, lottery quantity, position data and replenishment data; Divide the preprocessed order data into training sets, and train the preset prediction model to obtain a trained first prediction model, and obtain a standard data set of all defect types in the training of the first prediction model; Calculate the approximation between each input order data in the training set and multiple defect type standard data, and obtain the best standard data for each order data; Use the time series similarity algorithm to calculate the expansion coefficient value of each data point in the input order data based on the corresponding relationship between the data points in the best standard data of each input order data; The order data length is adjusted according to the expansion coefficient value, and the adjusted order data length is used as input data for secondary training of the first prediction model to obtain a second prediction model; the real-time order data is input into the second prediction model to obtain the sorting risk type of the real-time order data, and the sorting system is adjusted to complete order sorting.

2. A lottery intelligent sorting method according to claim 1, characterized in that: The preprocessing of the order data includes: The order data is labeled according to the defect type, where the defect types include: too many orders, too few orders, too many lottery tickets remaining, too few lottery tickets remaining, and normal sorting; the labeled data is encoded to complete the preprocessing of the order data.

3. A lottery intelligent sorting method according to claim 1, characterized in that: The first prediction model comprises: The first prediction model selects to use a CNN-LSTM network, and adopts a multi-classification cross entropy loss function as the loss function of the first prediction model; The encoded data set is divided into a training set and a validation set. The training set is organized into a matrix according to the time series, with the order data at a preset number of moments as the training set for the preset prediction model.

4. A lottery intelligent sorting method according to claim 1, characterized in that: The standard data set includes: During the training process of the first prediction model, the input order data corresponding to the lowest cross entropy loss function value is used as the standard data of the corresponding defect type, and the standard data of all defect types constitute a standard data set.

5. A lottery intelligent sorting method according to claim 1, characterized in that: The best standard data include: The DTW algorithm is used to calculate the approximation between each input order data and each standard data in the standard data set, and the corresponding DTW value sequence is obtained. The minimum DTW value in the DTW value sequence is obtained, the closest defect type is determined, and the corresponding standard data is used as the optimal standard data for each input order data.

6. A lottery intelligent sorting method according to claim 1, characterized in that: The corresponding relationship includes: A cumulative distance matrix between the input order data and the best standard data is constructed using a dynamic time warping algorithm, wherein each element in the cumulative distance matrix represents a local distance between the input order data and the best standard data at a corresponding position; Use the dynamic programming method to obtain the shortest path from the starting point to the end point in the cumulative distance matrix, and determine the corresponding relationship between the input order data and the optimal standard data based on the trajectory corresponding to the shortest path.

7. A lottery intelligent sorting method according to claim 1, characterized in that: The expansion coefficient value includes: The matching degree between each input order data and the standard data and actual data of each defect type is calculated respectively, and the difference is calculated for normalization processing to obtain the expansion coefficient value between each input order data and each defect type.

8. The lottery intelligent sorting method according to claim 1, characterized in that: The method of completing the order data length adjustment comprises the steps of: Perform peak point detection on the expansion coefficient value sequence to obtain a peak sequence, wherein the peak sequence includes peak valley points and peak points; take the two peak points closest to the two ends of the peak sequence as edge peak points, obtain the sequence numbers of the two edge peak points in the expansion coefficient value sequence respectively, calculate the distances between the edge peak points and the starting end and the end respectively, and take the minimum value to obtain the adjustment direction coefficient; According to the data on one side corresponding to the adjustment direction coefficient, the mean of the sequence of expansion coefficient values ​​corresponding to all data points is calculated, and the mean is multiplied by the expansion coefficient value of the peak value corresponding to the input order data to obtain the window adjustment coefficient. The product of the window adjustment coefficient and the adjustment direction coefficient is used as the length adjustment value; In response to the length adjustment value being less than the current data length, data is removed from the tail end so that the length of the data after removal is consistent with the length adjustment value; in response to the length adjustment value being greater than or equal to the current data length, data is added from the tail end so that the length of the data after addition is consistent with the length adjustment value.

9. A lottery intelligent sorting system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the lottery intelligent sorting method according to claims 1-8 is implemented.

Images