Cloud platform-based at-211 separation process data sharing and collaborative analysis method and system

Abstract

The application relates to the technical field of information transmission and discloses an At-211 separation process data sharing and collaborative analysis method and system based on a cloud platform. The method comprises the following steps: acquiring multi-source data of an At-211 separation process in real time, analyzing the correlation degree of the multi-source data and the At-211 separation, and obtaining separation correlation characteristic values representing the importance of various types of data for the separation; the separation correlation characteristic values are generated to accurately reflect the importance of the At-211 separation data, and a communication priority coefficient is generated through collaborative calculation combined with network state data, so that the emergency degree of data transmission can be divided, bandwidth disorder competition can be avoided, data differentiation uploading is realized through dynamic encapsulation configuration, a transmission evaluation index is generated combined with a transient transmission disturbance coefficient, a common collaborative priority coefficient is generated by fusing multiple parameters, transmission deviation is corrected, key data high-speed stable transmission is guaranteed, and the real-time performance and accuracy of cloud platform data sharing and collaborative analysis are improved.

Description

Technical Field

[0001] This invention relates to the field of information transmission technology, specifically to a cloud-based method and system for data sharing and collaborative analysis of the At-211 separation process. Background Technology

[0002] Currently, a large amount of data is generated during the At-211 separation process, such as chromatograms, activity monitoring, temperature control, and flow rate. This data is often collected and uploaded independently, and is usually uploaded in full to the cloud platform to facilitate data sharing and collaborative analysis on the platform.

[0003] However, the above analysis method still has the following drawbacks: When data is uploaded in full, the priority of the data is usually not distinguished, which leads to some key data and ordinary data competing disorderly in the limited network bandwidth. Moreover, the full upload method is extremely prone to causing critical data update delays when data is shared and collaboratively analyzed on the cloud platform, which affects the accuracy and real-time performance of collaborative analysis. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a cloud-based method and system for At-211 separation process data sharing and collaborative analysis, thus solving the aforementioned problems.

[0005] The above-mentioned technical objective of the present invention is achieved through the following technical solution:

[0006] A cloud-based method for data sharing and collaborative analysis of the At-211 separation process includes:

[0007] Step S1: Acquire multi-source data of the At-211 separation process in real time, analyze the correlation between multi-source data and At-211 separation, and obtain separation correlation feature values ​​that represent the importance of various types of data to the separation.

[0008] Step S2: Real-time acquisition of network status data of the communication link between the cloud platform and the At-211 separation device; collaborative calculation of separation correlation feature values ​​of various types of data and network status data; generation of communication priority coefficients representing the urgency of transmission of various types of data in limited network resources.

[0009] Step S3: Adjust the transmission method of various types of data according to the communication priority coefficient, and upload various types of data to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect of this transmission.

[0010] Step S4: Perform collaborative analysis on the transmission evaluation index and communication priority coefficient, correct deviations in various data transmission types, and generate a collaborative priority coefficient that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

[0011] Furthermore, the correlation between multi-source data and At-211 separation was analyzed, yielding separation correlation feature values ​​representing the importance of various data types to the separation, including:

[0012] By analyzing various types of data from multiple sources, time-series fluctuation values ​​representing the information content of each type of data are obtained;

[0013] The correlation strength between a single data source and all other data sources is calculated to obtain the cross-modal mutual information index, which represents the pivotal role of that data type in the entire sensor network.

[0014] Furthermore, the correlation between multi-source data and At-211 separation was analyzed to obtain separation correlation feature values ​​representing the importance of various types of data to the separation, including:

[0015] Perform change point analysis on the time-series fluctuation values ​​to generate a communication burst coefficient that represents the probability of sudden network bandwidth occupancy caused by this type of data;

[0016] By fusing time-series fluctuation values, cross-modal mutual information index, and communication burst coefficient, separation correlation feature values ​​representing the importance of various data to separation are obtained.

[0017] Furthermore, the separation and correlation feature values ​​of various types of data and network status data are collaboratively calculated to generate communication priority coefficients representing the urgency of transmission of various types of data within limited network resources, including:

[0018] By combining the separated correlation feature values ​​with network state data, an instantaneous communication demand index is obtained to reflect the urgency of this type of data transmission under the current network quality fluctuations.

[0019] Based on network status data, analyze network congestion trends and generate a link congestion coefficient that represents the probability of link congestion.

[0020] By fusing the separated correlation feature value with the cross-modal mutual information index and combining it with the packet loss rate in network state data for collaborative analysis, a data loss sensitivity factor representing the degree of impact of this type of data loss on communication tasks is obtained.

[0021] Furthermore, the separation and correlation feature values ​​of various types of data and network status data are collaboratively calculated to generate communication priority coefficients representing the urgency of transmission of various types of data within limited network resources. This also includes:

[0022] The transmission urgency is obtained by fusing the instantaneous communication demand index, link congestion coefficient, and data loss sensitivity factor.

[0023] Based on network status data, the transmission urgency is adjusted to generate a communication priority coefficient that represents the urgency of transmitting various types of data within limited network resources.

[0024] Furthermore, based on communication priority coefficients, the transmission methods for various types of data are adjusted, and these data types are uploaded to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the effectiveness of this transmission, including:

[0025] Based on the communication priority coefficient and the available bandwidth of the current network, various types of data are encapsulated for transmission to obtain the dynamic encapsulation configuration vector for each type of data in this transmission.

[0026] Furthermore, based on communication priority coefficients, the transmission methods for various types of data are adjusted, and these data are uploaded to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the effectiveness of this transmission. This also includes:

[0027] Various types of data are uploaded to the cloud platform in a dynamically packaged and differentiated manner. During this process, the latency jitter and throughput fluctuation of the transmission link are acquired and analyzed in real time to obtain the transient transmission disturbance coefficient, which represents the change in link quality during transmission.

[0028] The dynamic encapsulation configuration vector is fused with the transient transmission disturbance coefficient to generate a transmission evaluation index representing the transmission effect.

[0029] Furthermore, a collaborative analysis is performed on the transmission evaluation index and communication priority coefficient to correct deviations in various data transmission types, generating a collaborative priority coefficient that represents the degree of optimization in data collaborative transmission among nodes of the cloud platform, including:

[0030] The transmission delay, arrival order, and packet loss rate of various types of data during this transmission process are obtained. The transmission delay, arrival order, packet loss rate, transmission evaluation index, and communication priority coefficient are then used to calculate the transmission scheduling deviation, which represents the deviation of the transmission execution.

[0031] Furthermore, a collaborative analysis of the transmission evaluation index and communication priority coefficient is conducted to correct deviations in various data transmission methods, generating a collaborative priority coefficient that represents the degree of optimization in data collaborative transmission among nodes of the cloud platform. This also includes:

[0032] By integrating transmission scheduling deviation with link congestion coefficient and cross-modal mutual information index, a collaborative priority coefficient is generated that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

[0033] Furthermore, a cloud-based At-211 separation process data sharing and collaborative analysis system is applied to the aforementioned analysis methods, including:

[0034] The correlation analysis unit is used to acquire multi-source data of the At-211 separation process in real time, analyze the degree of correlation between multi-source data and At-211 separation, and obtain separation correlation feature values ​​that represent the importance of various types of data to the separation.

[0035] The communication analysis unit is used to acquire network status data of the communication link between the cloud platform and the At-211 separation device in real time, and to perform collaborative calculations on the separation correlation feature values ​​of various types of data and network status data to generate communication priority coefficients that represent the urgency of transmission of various types of data in limited network resources.

[0036] The transmission evaluation unit is used to adjust the transmission mode of various types of data according to the communication priority coefficient, and upload various types of data to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect of this transmission.

[0037] The collaborative optimization unit is used to perform collaborative analysis on the transmission evaluation index and communication priority coefficient, correct deviations in various data transmissions, and generate a collaborative priority coefficient that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

[0038] In summary, the present invention has the following main beneficial effects:

[0039] By analyzing time-series fluctuation values, cross-modal mutual information index, and communication burstiness coefficient, and fusing these three to obtain separation correlation feature values, the importance of various data to the separation process can be accurately reflected, enabling precise identification of key data. Furthermore, by combining network status data of the communication link between the cloud platform and the separation device, instantaneous communication demand index, link congestion coefficient, and data loss sensitivity factor are collaboratively calculated, and then fused to generate a communication priority coefficient. This allows for the reasonable allocation of data transmission urgency under limited network resources, avoiding disorderly bandwidth competition between key and ordinary data. Finally, a dynamic encapsulation configuration vector is generated based on the communication priority coefficient to manage the data... Differentiated uploading is performed, and a transmission evaluation index is generated by combining the transient transmission disturbance coefficient during the transmission process to intuitively reflect the transmission effect. Finally, the transmission scheduling deviation is calculated by co-calculating the transmission evaluation index and the communication priority coefficient. Then, the link congestion coefficient and the cross-modal mutual information index are integrated to generate a collaborative priority coefficient, which can correct the data transmission deviation in real time and optimize the data collaborative transmission efficiency of each node in the cloud platform. This solution not only ensures the high-speed and stable transmission of key At-211 separation data, but also improves the real-time performance and accuracy of data sharing and collaborative analysis in the cloud platform, facilitating the sharing and collaborative analysis of data from the At-211 separation process in the cloud platform. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating the steps of the cloud-based At-211 separation process data sharing and collaborative analysis method of the present invention.

[0041] Figure 2 This is a schematic diagram of the cloud platform-based At-211 separation process data sharing and collaborative analysis system of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] refer to Figure 1 and Figure 2 A cloud-based method for At-211 separation process data sharing and collaborative analysis, including:

[0044] Step S1: Acquire multi-source data of the At-211 separation process in real time, analyze the correlation between multi-source data and At-211 separation, and obtain separation correlation feature values ​​that represent the importance of various types of data to the separation.

[0045] Multi-source data includes: chromatographic signals, activity, flow rate, etc.

[0046] Step S2: Real-time acquisition of network status data of the communication link between the cloud platform and the At-211 separation device; collaborative calculation of separation correlation feature values ​​of various types of data and network status data; generation of communication priority coefficients representing the urgency of transmission of various types of data in limited network resources.

[0047] Network status data includes: available bandwidth, network latency, packet loss rate, etc.

[0048] Step S3: Adjust the transmission method of various types of data according to the communication priority coefficient, and upload various types of data to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect of this transmission.

[0049] Step S4: Perform collaborative analysis on the transmission evaluation index and communication priority coefficient, correct deviations in various data transmission types, and generate a collaborative priority coefficient that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

[0050] In one embodiment, the correlation between multi-source data and At-211 separation is analyzed to obtain separation correlation feature values ​​representing the importance of various types of data to the separation, including:

[0051] The analysis of various types of data from multiple sources yields time-series fluctuation values ​​representing the information content of each type of data. Specifically, this includes: processing chromatographic signals, which are acquired in real time as continuous voltage values, with each sampling period fixed at 1 second; identifying chromatographic peaks within each sampling period; extracting the peak voltage value and corresponding retention time value of the current period's chromatographic peak; simultaneously obtaining the peak retention time value of the previous complete period's chromatographic peak as the baseline retention time; and dividing the absolute value of the difference between the current period's retention time and the baseline retention time by the baseline retention time to obtain the relative peak position shift rate.

[0052] Obtain the peak voltage value of the chromatographic peak in the previous period, calculate the absolute value of the difference between the peak voltage of the current period and the peak voltage of the previous period, divide the absolute value by the peak voltage value of the previous period to obtain the relative change rate of peak height, and multiply the relative change rate of peak height by the relative shift rate of peak position to obtain the chromatographic fluctuation intensity.

[0053] The activity data is processed. When the At-211 separation process is started, the initial activity value is recorded as a reference. In each sampling period, the activity value at the current moment is collected in real time, and the sampling period is fixed at 1 second. The absolute value of the difference between the current activity value and the activity value of the previous second is calculated to obtain the activity change per second. Then, the ratio of this change to the current activity value is calculated to obtain the instantaneous activity fluctuation rate.

[0054] Then, calculate the arithmetic mean of the activity volatility over the previous ten consecutive seconds as the recent activity volatility benchmark. Calculate the absolute value of the difference between the current instantaneous activity volatility and the recent activity volatility benchmark, and divide the absolute value by the recent activity volatility benchmark to obtain the activity deviation coefficient.

[0055] The flow data is processed with a sampling period of 1 second. The collected flow is the instantaneous flow value. The arithmetic mean of the instantaneous flow values ​​of the previous ten consecutive periods is calculated as the historical mean. The absolute difference between the instantaneous flow value of the current period and the historical mean is calculated. The absolute difference is divided by the historical mean to obtain the flow fluctuation coefficient. The flow fluctuation coefficient reflects the degree of fluctuation of the flow relative to the recent average state. The larger the value, the more unstable the flow.

[0056] Based on the physicochemical mechanism of the At-211 separation process, the chromatographic signal directly determines the chemical form and purity of the separated products and is the core indicator for process control, so the chromatographic fluctuation intensity is assigned a weight of 0.55; the activity data directly reflects the yield and recovery rate of the target nuclide and is the verification indicator of the process effect, so the activity deviation coefficient is assigned a weight of 0.3; the flow rate data, as an auxiliary condition for process implementation, mainly affects the separation efficiency and stability, so the flow rate fluctuation coefficient is assigned a weight of 0.15.

[0057] After normalizing the chromatographic fluctuation intensity, activity deviation coefficient, and flow fluctuation coefficient to the 0-1 range, and then multiplying them by their corresponding weights, three time-series fluctuation values ​​representing the information content of various data types are obtained within the sampling period.

[0058] Calculate the correlation strength between a single data source and all other data sources to obtain the cross-modal mutual information index, which represents the pivotal position of this type of data in the entire sensor network. Specifically, this includes setting the length of the time analysis window to thirty seconds, with the time analysis window ending at the current sampling time and tracing back thirty sampling periods of data, where each sampling period is still 1 second.

[0059] Secondly, for each type of data source, such as chromatographic signal, activity, and flow rate, the chromatographic signal is taken as the chromatographic fluctuation intensity calculated for each sampling period, the activity is taken as the activity deviation coefficient for each sampling period, and the flow rate is taken as the flow rate fluctuation coefficient for each sampling period, forming three sequences of length thirty respectively.

[0060] Then, the correlation between a single data source and the other data sources is calculated. Taking chromatographic signals as an example, the 30 chromatographic fluctuation intensities of the chromatographic signal are used as a baseline sequence. For activity data, the 30 activity deviation coefficients are shifted forward by 1 unit, 2 units, up to 5 units, to generate 5 shifted sequences, representing the activity data lags by 1 second to 5 seconds, respectively. The Pearson correlation coefficient between the baseline sequence and each shifted sequence is calculated, and the one with the largest absolute value is taken as the delayed cross-correlation value between the chromatographic signal and the activity data. In the same way, the delayed cross-correlation value between the chromatographic signal and the flow rate is calculated, and the mean of these two delayed cross-correlation values ​​is calculated to obtain the comprehensive correlation strength of the chromatographic signal.

[0061] Then, repeat the above steps for the activity data and flow data, that is, using the activity data sequence as a benchmark, calculate the delayed cross-correlation value between it and the chromatographic and flow data and take the average value to obtain the comprehensive correlation strength of the activity data; using the flow data sequence as a benchmark, calculate the delayed cross-correlation value between it and the chromatographic and activity data and take the average value to obtain the comprehensive correlation strength of the flow data.

[0062] Normalizing the three comprehensive correlation strengths to the 0-1 interval yields the cross-modal mutual information index, which represents the pivotal role of this type of data in the entire sensor network.

[0063] In one embodiment, analyzing the correlation between multi-source data and At-211 separation to obtain separation correlation feature values ​​representing the importance of various types of data to separation further includes:

[0064] Variable point analysis is performed on the time series fluctuation values ​​to generate a communication burst coefficient that represents the probability of sudden occupancy of network bandwidth by this type of data. Specifically, this includes setting two time windows of different lengths: a short window with a length of five seconds and a long window with a length of thirty seconds. Both windows take the current sampling time as the end point and trace back the time series fluctuation values ​​for the corresponding number of seconds to form a short-time series and a long-time series, respectively.

[0065] For a short-time series, its five time-series fluctuation values ​​are used as dependent variables, and the corresponding time sequence, from the first second to the fifth second, is used as independent variables. A straight line is fitted using the least squares method to obtain the slope value of the line, which is used as the short-time slope. The short-time slope reflects the instantaneous change trend of the data information content within the most recent five seconds. A positive slope value indicates an accelerated increase in information content, while a negative value indicates a decrease. Similarly, the same operation is performed on the thirty time-series fluctuation values ​​of a long-time series to obtain the long-time slope, which reflects the overall change trend within the past thirty seconds.

[0066] The slope deviation value is obtained by calculating the difference between the short-term slope and the long-term slope. The slope deviation value eliminates the influence of the background trend and highlights the direction and magnitude of the recent abnormal changes compared with the long-term average. The maximum value of all slope deviation values ​​in the past 5 minutes is used as the benchmark value. The current slope deviation value is divided by the benchmark value, and the calculation result is normalized to the 0-1 range. This is the communication burst coefficient, which represents the probability of this type of data causing sudden occupancy of network bandwidth.

[0067] The time-series fluctuation value, cross-modal mutual information index, and communication burst coefficient are fused to obtain separation correlation feature values ​​that represent the importance of various data to separation. Specifically, the cumulative average value of the time-series fluctuation value within the first 5 minutes after the separation process starts is taken as the process baseline value. The process baseline value reflects the basic disturbance level of the current batch of separation columns. At the same time, the maximum value of the cross-modal mutual information index within this period is taken as the upper limit of network coordination, and the 95th percentile of the communication burst coefficient is taken as the burst risk threshold.

[0068] Divide the current time-series fluctuation value by the process baseline value to obtain the relative disturbance coefficient; divide the cross-modal mutual information index by the network coordination upper limit to obtain the relative coordination coefficient; and divide the communication burst coefficient by the burst risk threshold to obtain the relative risk coefficient.

[0069] After multiplying the relative synergy coefficient and the relative risk coefficient and compressing them using the hyperbolic tangent function, an intermediate variable is obtained. The relative disturbance coefficient is then multiplied with the intermediate variable as the basic variable, and the calculation result is normalized to the 0-1 interval. This result represents the separation correlation characteristic value that indicates the importance of various data to the separation.

[0070] This technical solution accurately analyzes the correlation between multi-source data and At-211 separation, calculates the separation correlation characteristic value, and clarifies the importance priority of various data such as chromatographic fluctuation intensity, activity deviation coefficient, and flow fluctuation coefficient. It solves the problem of disorderly competition for network bandwidth between key data and ordinary data in the existing full upload mode, avoids non-critical data occupying bandwidth resources, reduces the update delay of key data during cloud platform data sharing and collaborative analysis, and improves the accuracy and real-time performance of collaborative analysis.

[0071] In one embodiment, the separation and correlation feature values ​​of various types of data and network status data are jointly calculated to generate a communication priority coefficient representing the urgency of transmission of various types of data within limited network resources, including:

[0072] By combining the separated correlation feature values ​​with network state data, an instantaneous communication demand index is obtained to reflect the urgency of this type of data transmission under the current network quality fluctuations. Specifically, this includes: real-time acquisition of available bandwidth, network latency, and packet loss rate within each 1-second sampling period. Available bandwidth is recorded in megabits per second, network latency is recorded in milliseconds, and packet loss rate is represented in a 0-1 format, for example, a 2% packet loss rate is recorded as 0.02.

[0073] The maximum available bandwidth within the previous 30 seconds is used as the bandwidth benchmark; the minimum network latency within the previous 30 seconds is used as the latency benchmark. The bandwidth benchmark is divided by the current available bandwidth to obtain the bandwidth strain coefficient, which is always ≥1. The larger the value, the more strained the bandwidth resources are; the current network latency is divided by the latency benchmark to obtain the latency degradation coefficient, which is also always ≥1. The larger the value, the more severe the latency degradation; the current packet loss rate is used as the packet loss degradation coefficient.

[0074] Multiply the bandwidth strain coefficient, latency degradation coefficient, and packet loss degradation coefficient to obtain a comprehensive value;

[0075] After multiplying the separated correlation feature value by the comprehensive value, the calculation result is normalized to the 0-1 range, which is the instantaneous communication demand index of the urgency of low latency and high reliability transmission for this type of data under the current network quality fluctuation.

[0076] Based on network status data, analyze network congestion trends and generate a link congestion coefficient representing the probability of link congestion. Specifically, this involves: normalizing the available bandwidth and network latency at the current moment to the 0-1 range, dividing the former by the latter to obtain a preliminary traffic density value, using the maximum value of all preliminary traffic density values ​​in the past 5 minutes as the density benchmark, and dividing the current preliminary traffic density value by the density benchmark to obtain a traffic density coefficient. The larger the traffic density coefficient, the higher the data density carried by the network, and the higher the density, the more congested the network.

[0077] For the available bandwidth in the ten consecutive seconds preceding the current moment, treat it as a sequence of length 10. Calculate the change between two adjacent values ​​in this sequence to obtain nine changes. Add the absolute values ​​of these nine changes and divide by the average available bandwidth in these ten seconds to obtain the flow rate fluctuation intensity. Take the maximum value of all flow rate fluctuation intensities in the past 5 minutes as the fluctuation benchmark. Divide the current flow rate fluctuation intensity by the fluctuation benchmark to obtain the flow rate change coefficient. The flow rate change coefficient reflects the severity of bandwidth fluctuations. The larger the value, the more unstable the network flow rate.

[0078] Multiplying the flow density coefficient by the flow velocity variation coefficient and normalizing the result to the 0-1 range gives the link congestion coefficient, which represents the probability of link congestion.

[0079] By fusing the separation correlation feature value with the cross-modal mutual information index and combining it with the packet loss rate in network state data for collaborative analysis, a data loss sensitivity factor representing the impact of this type of data loss on communication tasks is obtained. Specifically, the separation correlation feature value at the current moment is used as the basic weight, and the cross-modal mutual information index is used as the adjustment factor of the basic weight for multiplication to obtain an intermediate value. This can amplify the influence of data that is not only important in itself but also closely related to other data sources in the network, because the absence of these data may affect the overall accuracy of multi-source data collaborative analysis.

[0080] Divide the current packet loss rate by the 90th percentile of the packet loss rate over the past 5 minutes, and normalize the result to the 0-1 range to obtain the relative packet loss risk coefficient. Multiply the median value by the relative packet loss risk coefficient, and normalize the result to the 0-1 range to obtain the data loss sensitivity factor, which represents the degree of impact of this type of data loss on the communication task.

[0081] In one embodiment, the method further includes performing collaborative calculations on the separation and correlation feature values ​​of various types of data and network status data to generate communication priority coefficients representing the urgency of transmission of various types of data within limited network resources:

[0082] The transmission urgency is obtained by fusing the instantaneous communication demand index, link congestion coefficient, and data loss sensitivity factor. Specifically, the reciprocal of the link congestion coefficient is used as a weight adjustment term. When the congestion probability is extremely high, its reciprocal approaches zero, which can effectively suppress the transmission urgency and avoid transmission pressure when the network is vulnerable.

[0083] Multiply the data loss sensitivity factor, the instantaneous communication demand index, and the weight adjustment term, and normalize the calculation result to the 0-1 interval to obtain the transmission urgency.

[0084] Based on network status data, the transmission urgency is adjusted to generate communication priority coefficients that represent the urgency of various types of data transmission within limited network resources. Specifically, this includes: within each sampling period, dividing the current available bandwidth by the maximum available bandwidth in the past 30 seconds to obtain a bandwidth adequacy index; simultaneously, using the minimum network latency in the past 30 seconds as a baseline value, dividing this baseline value by the current network latency to obtain a latency health index; the higher the latency health index, the more ideal the latency condition; subtracting 1 from the current packet loss rate to obtain a packet loss integrity index; multiplying the bandwidth adequacy index, latency health index, and packet loss integrity index together, and normalizing the result to the 0-1 range to obtain a network carrying quality coefficient; the higher the value of the network carrying quality coefficient, the better the quality of the current network link, and the more capable it is of supporting the transmission of high-urgency data.

[0085] Multiplying the transmission urgency by the network carrying capacity coefficient and normalizing the product to the 0-1 range gives the communication priority coefficient, which represents the urgency of transmitting various types of data within limited network resources. The significance of this approach is that when the network carrying capacity is high, the communication priority coefficient is close to 1, and the transmission urgency is almost completely preserved. Conversely, when the network quality deteriorates, the communication priority coefficient decreases significantly, thereby effectively suppressing the transmission urgency and preventing the injection of high-priority data streams beyond the network's carrying capacity when it is most vulnerable, which could lead to overall communication paralysis.

[0086] This technical solution separates associated feature values ​​and network status data through collaborative computing, generates communication priority coefficients, clarifies the urgency of transmission of various types of data, and dynamically matches network status to allocate transmission resources by accurately calculating and fusing instantaneous communication demand index, link congestion coefficient, and data loss sensitivity factor. This avoids non-critical data from occupying bandwidth, reduces the update delay of critical data, and improves the accuracy and real-time performance of cloud platform data sharing and collaborative analysis. At the same time, it can suppress high-urgency data transmission during network congestion and prevent overall communication paralysis.

[0087] In one embodiment, the transmission methods of various types of data are adjusted according to a communication priority coefficient, and the various types of data are uploaded to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect, including:

[0088] Based on the communication priority coefficient and the available bandwidth of the current network, various types of data are encapsulated for transmission to obtain the dynamic encapsulation configuration vector for each type of data in this transmission. Specifically, this includes: dividing the current available bandwidth by the maximum available bandwidth in the past 30 seconds in each sampling period to obtain the relative bandwidth value, and then multiplying the relative bandwidth value by the communication priority coefficient to obtain the encapsulation resource allocation base. The larger the value of the encapsulation resource allocation base, the greater the resource control rights that the data has during encapsulation.

[0089] The current data generation rate of each type of data is obtained, and the ratio of the available bandwidth to the data generation rate is calculated. If the ratio is greater than 1, it indicates that the bandwidth is sufficient. The encapsulation resource allocation base is then multiplied by the maximum protocol data unit length of that type of data to obtain the data block size. If the ratio is less than 1, it indicates that the bandwidth is tight. The encapsulation resource allocation base is then multiplied by the data generation rate and then by the sampling period to obtain the merged data block size. The data block size determines the amount of data encapsulated in each transmission packet. If the ratio is equal to 1, the data block size can be set to be comparable to the size of the original data packet to reduce encapsulation processing overhead.

[0090] Multiply the encapsulation resource allocation base by the link congestion coefficient to obtain the redundancy adjustment factor. Obtain the minimum redundancy ratio that ensures data recovery when the network quality is optimal. Multiply the redundancy adjustment factor by the minimum redundancy ratio to obtain the actual redundancy ratio. The actual redundancy ratio determines the amount of additional check data to combat packet loss.

[0091] Subtract the resource allocation base from 1 to get the compression requirement coefficient. Multiply the compression requirement coefficient by the maximum compressibility ratio of this type of data to get the actual compression ratio. The actual compression ratio determines the degree of data volume reduction.

[0092] The data block size, actual redundancy ratio, and actual compression ratio are combined into a three-dimensional vector, which is the dynamic encapsulation configuration vector for this transmission of various types of data.

[0093] In one embodiment, the transmission methods of various types of data are adjusted according to a communication priority coefficient, and the various types of data are uploaded to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect. This also includes:

[0094] Various types of data are dynamically packaged and configured to be uploaded to the cloud platform. During this process, the latency jitter and throughput fluctuation of the transmission link are acquired and analyzed in real time to obtain the transient transmission disturbance coefficient, which represents the change in link quality during transmission. Specifically, this includes: obtaining the packet loss rate, latency jitter, and throughput fluctuation of the current period. Latency jitter is obtained by calculating the absolute value of the difference between the network latency of the current period and the network latency of the previous period; throughput fluctuation is obtained by calculating the absolute value of the difference between the actual throughput of the current period and the actual throughput of the previous period, and then dividing the absolute value by the actual throughput of the previous period. The latency jitter and throughput fluctuation are normalized to the 0-1 interval and then squared to obtain their respective energy characterization values. The two are then added together and the square root is taken to obtain the comprehensive vector magnitude reflecting the magnitude of the instantaneous instability of the link. The comprehensive vector magnitude reflects the degree of deviation of latency and throughput in two orthogonal dimensions.

[0095] The arithmetic mean of all composite vector magnitudes within the 10 consecutive seconds preceding the current moment is taken as the recent disturbance baseline. The difference between the current composite vector magnitude and the recent disturbance baseline is calculated, and this difference is divided by the recent disturbance baseline to obtain the relative deviation coefficient. The ratio of the current packet loss rate to the median packet loss rate in the past minute is calculated to obtain the relative packet loss degradation index. The relative deviation coefficient and the relative packet loss degradation index are multiplied together, and the product is normalized to the 0-1 interval, which is the transient transmission disturbance coefficient representing the change in link quality during transmission.

[0096] The dynamic encapsulation configuration vector is fused with the transient transmission disturbance coefficient to generate a transmission evaluation index representing the effect of this transmission. Specifically, the data block size, actual redundancy ratio, and actual compression ratio in the dynamic encapsulation configuration vector are normalized to the 0-1 range and then multiplied to obtain the encapsulation comprehensive strength coefficient. The larger the encapsulation comprehensive strength coefficient, the more resources are invested in this transmission and the higher the requirements for network quality.

[0097] Calculate the difference between the encapsulation strength coefficient and the transient transmission disturbance coefficient. If the difference is ≥ 0, it indicates that the encapsulation strategy's ability to resist disturbances exceeds the current disturbance level, and the transmission effect is good. Then, normalize the difference to the 0-1 range, which is the transmission evaluation index representing the current transmission effect. If the difference is < 0, it indicates that the encapsulation strategy is insufficient to resist the current disturbance, and the transmission effect deteriorates. Divide the difference by the sum of the encapsulation strength coefficient and the transient transmission disturbance coefficient, and normalize the calculation result to the 0-1 range, which is the transmission evaluation index representing the current transmission effect.

[0098] By dynamically generating a dynamic encapsulation configuration vector based on communication priority coefficients and available bandwidth, differentiated uploading of multi-source data during the At-211 separation process is achieved. The dynamic encapsulation configuration vector includes data block size, actual redundancy ratio, and actual compression ratio, which can be flexibly adjusted according to the importance of the data and network status. For example, critical data can be given a larger data block size for priority transmission when bandwidth is sufficient, while data integrity and real-time performance can be ensured by merging data, increasing redundancy, or compression when bandwidth is tight, avoiding disorderly competition with ordinary data. At the same time, latency jitter and throughput fluctuations are monitored in real time, and a transmission evaluation index is finally generated to reflect the effect of each transmission, reduce the update latency of critical data, improve the accuracy and real-time performance of cloud platform collaborative analysis, and optimize network resource utilization.

[0099] In one embodiment, a collaborative analysis is performed on the transmission evaluation index and the communication priority coefficient to correct deviations in various data transmission types, generating a collaborative priority coefficient representing the degree of optimization of data collaborative transmission among nodes of the cloud platform, including:

[0100] The transmission delay, arrival order, and packet loss rate of various types of data during this transmission process are obtained. The transmission delay, arrival order, packet loss rate, transmission evaluation index, and communication priority coefficient are then used to calculate the transmission scheduling deviation, which represents the deviation of the transmission execution. Specifically, the transmission delay, arrival order, and packet loss rate of various types of data are obtained first. The transmission delay is the total time taken for a data packet to be sent and received. The arrival order is obtained by recording the sending sequence number and receiving sequence number of each data packet and counting the proportion of packets whose actual arrival order is reversed compared to the sending order to the total number of packets. The packet loss rate is still represented in the form of 0-1, for example, a 2% packet loss rate is recorded as 0.02.

[0101] Divide the transmission delay of this transmission by a reference delay benchmark to obtain the delay deviation factor. The reference delay benchmark is the minimum average transmission delay achieved by all data transmissions with the same communication priority coefficient in the past minute.

[0102] Add 1 to the order disorder degree and take the logarithm to get the order deterioration index. The larger the order deterioration index, the more serious the out-of-order phenomenon. Multiply the order deterioration index by the packet loss rate to get the comprehensive out-of-order loss factor. The comprehensive out-of-order loss factor is used to amplify the interference caused by the combined effect of out-of-order and packet loss on data stream reconstruction.

[0103] Multiplying the delay deviation factor by the comprehensive out-of-order loss factor yields the transmission execution deviation coefficient. The transmission execution deviation coefficient comprehensively reflects the joint deviation of the transmission plan from the delay degradation, out-of-order interference, and packet loss. Dividing the transmission evaluation index by the communication priority coefficient yields the plan execution matching degree. Dividing the plan execution matching degree by the transmission execution deviation coefficient and normalizing the calculation result to the 0-1 interval gives the transmission scheduling deviation degree, which represents the transmission execution deviation.

[0104] In one embodiment, the transmission evaluation index and communication priority coefficient are analyzed collaboratively to correct deviations in various data transmission types, generating a collaborative priority coefficient that represents the degree of optimization in data collaborative transmission among nodes of the cloud platform. This also includes:

[0105] The transmission scheduling deviation is integrated with the link congestion coefficient and the cross-modal mutual information index to generate a co-cooperation priority coefficient that represents the degree of data collaborative transmission optimization among the nodes of the cloud platform. Specifically, the variance of the transmission scheduling deviation, the link congestion coefficient and the cross-modal mutual information index in the past minute are calculated respectively as their dynamic volatility weights. The larger the variance, the more unstable the parameter is in the recent period and the more sensitive it is to the need for collaborative optimization.

[0106] The transmission scheduling deviation, link congestion coefficient, and cross-modal mutual information index are multiplied by their corresponding dynamic volatility weights, respectively. The geometric mean of the three product results is calculated and the result is normalized to the 0-1 range to generate a collaborative priority coefficient that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

[0107] By generating transmission scheduling deviation, the matching degree between actual execution deviations such as transmission delay, arrival order, and packet loss rate and transmission evaluation index and communication priority coefficient is comprehensively evaluated. The link congestion coefficient and cross-modal mutual information index are then integrated to finally obtain the collaborative priority coefficient. The collaborative priority coefficient accurately reflects the optimization degree of data collaborative transmission between cloud platform nodes. It prioritizes ensuring the collaborative consistency and real-time update of key data with high separation correlation characteristic values ​​between nodes, reduces the delay in key data update caused by network congestion or disordered competition, and thus improves the accuracy and real-time performance of cloud platform collaborative analysis of the At-211 separation process.

[0108] In one embodiment, a cloud-based At-211 separation process data sharing and collaborative analysis system is applied to the above-described analysis method, including:

[0109] The correlation analysis unit is used to acquire multi-source data of the At-211 separation process in real time, analyze the degree of correlation between multi-source data and At-211 separation, and obtain separation correlation feature values ​​that represent the importance of various types of data to the separation.

[0110] The communication analysis unit is used to acquire network status data of the communication link between the cloud platform and the At-211 separation device in real time, and to perform collaborative calculations on the separation correlation feature values ​​of various types of data and network status data to generate communication priority coefficients that represent the urgency of transmission of various types of data in limited network resources.

[0111] The transmission evaluation unit is used to adjust the transmission mode of various types of data according to the communication priority coefficient, and upload various types of data to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect of this transmission.

[0112] The collaborative optimization unit is used to perform collaborative analysis on the transmission evaluation index and communication priority coefficient, correct deviations in various data transmissions, and generate a collaborative priority coefficient that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

[0113] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A cloud-based method for data sharing and collaborative analysis of the At-211 separation process, characterized in that: include: Step S1: Acquire multi-source data of the At-211 separation process in real time, analyze the correlation between multi-source data and At-211 separation, and obtain separation correlation feature values ​​representing the importance of various types of data to the separation, including: By analyzing various types of data from multiple sources, time-series fluctuation values ​​representing the information content of each type of data are obtained; Calculate the correlation strength between a single data source and all other data sources to obtain the cross-modal mutual information index, which represents the pivotal position of that data type in the entire sensor network; Perform change point analysis on the time-series fluctuation values ​​to generate a communication burst coefficient that represents the probability of sudden network bandwidth occupancy caused by this type of data; By fusing time series fluctuation values, cross-modal mutual information index, and communication burst coefficient, separation correlation feature values ​​representing the importance of various data to separation are obtained; Step S2: Real-time acquisition of network status data of the communication link between the cloud platform and the At-211 separation device; collaborative calculation of the separation correlation feature values ​​of various types of data and network status data; generation of communication priority coefficients representing the urgency of transmission of various types of data in limited network resources, including: By combining the separated correlation feature values ​​with network state data, an instantaneous communication demand index is obtained to reflect the urgency of this type of data transmission under the current network quality fluctuations. Based on network status data, analyze network congestion trends and generate a link congestion coefficient that represents the probability of link congestion. By fusing the separated correlation feature value with the cross-modal mutual information index and combining it with the packet loss rate in the network state data for collaborative analysis, a data loss sensitivity factor representing the degree of impact of this type of data loss on communication tasks is obtained. Step S3: Adjust the transmission method of various types of data according to the communication priority coefficient, and upload various types of data to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect of this transmission. Step S4: Perform collaborative analysis on the transmission evaluation index and communication priority coefficient, correct deviations in various data transmission types, and generate a collaborative priority coefficient that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

2. The cloud-based At-211 separation process data sharing and collaborative analysis method according to claim 1, characterized in that, The algorithm performs collaborative calculations on the separation and correlation feature values ​​of various data types and network status data to generate communication priority coefficients that represent the urgency of transmission of various data types within limited network resources. It also includes: The transmission urgency is obtained by fusing the instantaneous communication demand index, link congestion coefficient, and data loss sensitivity factor. Based on network status data, the transmission urgency is adjusted to generate a communication priority coefficient that represents the urgency of transmitting various types of data within limited network resources.

3. The cloud-based At-211 separation process data sharing and collaborative analysis method according to claim 2, characterized in that, Based on communication priority coefficients, the transmission methods for various types of data are adjusted, and these data types are uploaded to the cloud platform in a differentiated manner. A transmission evaluation index representing the effectiveness of this transmission is generated, including: Based on the communication priority coefficient and the available bandwidth of the current network, various types of data are encapsulated for transmission to obtain the dynamic encapsulation configuration vector for each type of data in this transmission.

4. The cloud-based At-211 separation process data sharing and collaborative analysis method according to claim 3, characterized in that, Based on communication priority coefficients, the transmission methods for various types of data are adjusted, and these data are uploaded to the cloud platform in a differentiated manner. A transmission evaluation index representing the effectiveness of this transmission is generated, which also includes: Various types of data are uploaded to the cloud platform in a dynamically packaged and differentiated manner. During this process, the latency jitter and throughput fluctuation of the transmission link are acquired and analyzed in real time to obtain the transient transmission disturbance coefficient, which represents the change in link quality during transmission. The dynamic encapsulation configuration vector is fused with the transient transmission disturbance coefficient to generate a transmission evaluation index representing the transmission effect.

5. The cloud-based At-211 separation process data sharing and collaborative analysis method according to claim 4, characterized in that, A collaborative analysis of the transmission evaluation index and communication priority coefficient is performed to correct deviations in various data transmission types, generating a collaborative priority coefficient that represents the degree of optimization in data collaborative transmission among nodes of the cloud platform, including: The transmission delay, arrival order, and packet loss rate of various types of data during this transmission process are obtained. The transmission delay, arrival order, packet loss rate, transmission evaluation index, and communication priority coefficient are then used to calculate the transmission scheduling deviation, which represents the deviation of the transmission execution.

6. The cloud-based At-211 separation process data sharing and collaborative analysis method according to claim 5, characterized in that, The transmission evaluation index and communication priority coefficient are analyzed collaboratively to correct deviations in various data transmission types, generating a collaborative priority coefficient that represents the degree of optimization in data collaborative transmission among nodes of the cloud platform. This also includes: By integrating transmission scheduling deviation with link congestion coefficient and cross-modal mutual information index, a collaborative priority coefficient is generated that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

7. A cloud-based At-211 separation process data sharing and collaborative analysis system, applied in the analysis method described in any one of claims 1-6, characterized in that, include: The correlation analysis unit is used to acquire multi-source data of the At-211 separation process in real time, analyze the degree of correlation between multi-source data and At-211 separation, and obtain separation correlation feature values ​​that represent the importance of various types of data to the separation. The communication analysis unit is used to acquire network status data of the communication link between the cloud platform and the At-211 separation device in real time, and to perform collaborative calculations on the separation correlation feature values ​​of various types of data and network status data to generate communication priority coefficients that represent the urgency of transmission of various types of data in limited network resources. The transmission evaluation unit is used to adjust the transmission mode of various types of data according to the communication priority coefficient, and upload various types of data to the cloud platform in a differentiated manner to generate a transmission evaluation index representing the transmission effect of this transmission. The collaborative optimization unit is used to perform collaborative analysis on the transmission evaluation index and communication priority coefficient, correct deviations in various data transmissions, and generate a collaborative priority coefficient that represents the degree of optimization of data collaborative transmission among nodes of the cloud platform.

Images