A method and system for remote collaborative optimization of process parameters

Abstract

This invention relates to the field of industrial automation control technology and discloses a method and system for remote collaborative optimization of process parameters. The method first acquires the parameter deviation values ​​and transmission delay duration during the remote collaborative process of process parameters. Fluctuation periods are extracted through time-domain analysis to construct the degree of collaboration and the degree of delay impact. Then, combining the correlation between the degree of collaboration and the degree of delay impact in each period and the average state, the current degree of remote collaborative difference is obtained. Next, the decreasing trend of parameter deviation and the increasing trend of transmission delay at adjacent sampling times are analyzed to construct an adjustment trend degree. Combined with the degree of collaborative difference, the parameter optimization priority is determined, and the feedback adjustment amount is obtained. Finally, using the feedback adjustment amount and the actual adjustment amount, the remote collaborative process of process parameters is optimized through a collaborative controller. This method can improve the accuracy and adaptability of remote collaboration.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, specifically to a method and system for remote collaborative optimization of process parameters. Background Technology

[0002] In modern industrial production and complex manufacturing systems, the precise control of process parameters directly affects product quality, production efficiency, and system operational stability. With the rapid development of industrial internet technology, distributed production models are becoming increasingly prevalent, and remote collaboration of process parameters between multiple geographically dispersed production units or control nodes has become the norm. This remote collaboration model breaks down geographical limitations, enabling optimized resource allocation and efficient linkage of production processes, but it also faces numerous technical challenges.

[0003] Traditional remote collaborative control methods for process parameters often rely on preset fixed control strategies or simple feedback adjustment mechanisms to achieve parameter adjustments. However, in practical applications, the operating environment of remote collaborative systems is highly dynamic and uncertain: due to fluctuations in the network transmission environment, parameter transmission inevitably experiences delays during remote transmission, and the duration of these delays varies randomly depending on factors such as network load and signal interference. This transmission delay may lead to deviations in parameter synchronization between geographically dispersed nodes, thereby affecting the accuracy of collaborative control; differences in equipment status and operating conditions among different nodes cause parameter deviations during execution, and the patterns of these deviations are complex and difficult to predict accurately using fixed models.

[0004] In existing technologies, some remote collaborative control methods attempt to reduce the impact of latency by increasing network transmission rates or enhancing hardware performance. However, this approach is costly and cannot fundamentally solve the parameter coordination problem in dynamic environments. Other methods introduce adaptive control algorithms to adjust parameters, but most of these algorithms focus only on a single factor (such as transmission delay or parameter deviation), failing to comprehensively consider the correlation and dynamic changes between the two. For example, some algorithms adjust only based on the magnitude of parameter deviation, ignoring the impact of transmission delay on deviation correction; while others, although considering latency, fail to dynamically optimize based on the changing trends of parameter deviation, resulting in adjustment strategies lagging behind changes in the actual system state.

[0005] Furthermore, in multi-parameter collaborative control scenarios, different process parameters have varying degrees of impact on the overall system performance. Applying a uniform optimization strategy to all parameters can lead to unreasonable resource allocation and untimely optimization of key parameters, thereby affecting the overall operational efficiency of the remote collaborative system. Simultaneously, existing methods lack in-depth analysis of the dynamic characteristics during parameter collaboration when evaluating the operational status of collaborative systems, making it difficult to accurately quantify the degree of collaboration differences at different times, resulting in insufficient scientific rigor and specificity in optimization decisions.

[0006] As production systems become more complex and the scale of remote collaboration expands, the aforementioned problems become increasingly prominent. The cumulative effect of parameter deviations and transmission delays can trigger a chain reaction, causing serious consequences such as fluctuations in product quality, production interruptions, and even equipment damage. Therefore, how to comprehensively consider the dynamic characteristics of parameter deviations and transmission delays in a dynamically changing network environment and under complex production conditions to achieve precise optimization of remote collaborative process parameters has become an urgent problem to be solved in the field of industrial control. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for remote collaborative optimization of process parameters to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides a method for remote collaborative optimization of process parameters, the method comprising:

[0009] Obtain the deviation values ​​of parameters and the transmission delay during the remote collaboration of process parameters;

[0010] Time-domain analysis was performed on the deviation value and transmission delay duration to extract the fluctuation period of the deviation value and transmission delay duration.

[0011] By analyzing the magnitude of frequent changes in the deviation values ​​of parameters in the remote collaborative system and the magnitude of fluctuations in the transmission delay duration over time within each fluctuation cycle, the degree of collaboration and the degree of delay impact are constructed respectively.

[0012] Based on the correlation between the degree of coordination and the degree of delay impact of each fluctuation cycle before the current moment, as well as the average state of the degree of coordination and the average state of the degree of delay impact, the degree of remote coordination difference of parameters in the remote coordination system at the current moment is obtained.

[0013] Analyze the decreasing trend of parameter deviation values ​​and the increasing trend of transmission delay time at multiple adjacent sampling times at the current time to construct the adjustment trend degree of parameters at the current time. Combined with the degree of remote collaboration difference of parameters in the remote collaboration system, obtain the parameter optimization priority of the remote collaboration system at the current time. Utilize the change difference of parameter optimization priority and the baseline adjustment amount and preset adjustment amount threshold in the collaboration controller at the current time to obtain the feedback adjustment amount in the collaboration controller at the current time.

[0014] Based on the feedback adjustment and actual adjustment within the collaborative controller at the current moment, the remote collaborative process of process parameters is controlled and optimized using the collaborative controller.

[0015] Preferably, the extraction of the fluctuation period of the parameter deviation value and the transmission delay duration further includes:

[0016] Obtain the sequence of deviation values ​​and transmission delay durations of all historical sampling times within a preset time period before the current time, arranged in chronological order, and use it as the deviation value sequence and transmission delay sequence for the current time.

[0017] A time-domain analysis is performed on the deviation value sequence at the current moment to obtain the fluctuation characteristics of all time points in the time domain. The interval of the time point corresponding to the maximum fluctuation characteristic is taken as the deviation value fluctuation period in the short period before the current moment. Correspondingly, for the transmission delay sequence, a sliding window analysis is used to obtain the transmission delay fluctuation period in the short period before the current moment.

[0018] Preferably, the construction of the degree of synergy further includes:

[0019] The deviation values ​​within each deviation value fluctuation period before the current time are arranged in ascending order of time to form the deviation value fluctuation period sequence of each deviation value fluctuation period. Correspondingly, for the transmission delay duration within each delay fluctuation period, the delay period sequence of each transmission delay fluctuation period is obtained.

[0020] For each deviation value periodic sequence, obtain the first-order change sequence of the deviation value periodic sequence, count the number of all non-zero elements in the first-order change sequence, and record it as the fluctuation frequency. Calculate the average value of the absolute values ​​of all non-zero elements in the first-order change sequence, and record it as the average fluctuation amplitude. The product of the fluctuation frequency and the average fluctuation amplitude is used as the degree of coordination within each deviation value fluctuation period.

[0021] Preferably, the construction of the delay impact degree further includes:

[0022] For each delay period sequence, the average of the absolute values ​​of all elements in the first-order change sequence of the delay period sequence is calculated as the delay influence degree of each transmission delay fluctuation period.

[0023] Preferably, the process of determining the degree of remote collaboration difference of parameters in the remote collaboration system at the current moment is as follows: combining the correlation level between the collaboration degree sequence and the delay impact degree sequence at the current moment, as well as the average state of the elements in the collaboration degree sequence and the average state of the elements in the delay impact degree sequence, wherein the collaboration degree of all deviation value fluctuation cycles is arranged in ascending order of time to form the collaboration degree sequence at the current moment, and the delay impact degree of all transmission delay fluctuation cycles is arranged in ascending order of time to form the delay impact degree sequence at the current moment.

[0024] Preferably, the construction of the adjustment trend degree on the current time parameter includes:

[0025] The multiple times with the closest time interval to the current time are all taken as the adjacent sampling times of the current time. Based on the changes in parameter deviation values ​​and transmission delay duration of the multiple adjacent sampling times, the decreasing trend of the deviation value and the increasing trend of the delay at the current time are obtained.

[0026] The product of the current deviation value's downward trend and the delayed upward trend is determined as the adjustment trend degree of the parameter at the current moment.

[0027] Preferably, the current deviation value decreasing trend and the delay increasing trend further include:

[0028] The parameter deviation values ​​and transmission delay durations at multiple adjacent sampling times are arranged in ascending order of time to form the deviation value trend sequence and delay trend sequence at the current time. The first-order change sequences of the deviation value trend sequence and delay trend sequence at the current time are calculated respectively.

[0029] The sum of the absolute values ​​of all negative values ​​in the first-order change sequence of the deviation value trend sequence is taken as the current deviation value downward trend amount, and the sum of all positive values ​​in the first-order change sequence of the delayed trend sequence is taken as the current delayed upward trend amount.

[0030] Preferably, the calculation process of the parameter optimization priority of the remote collaboration system at the current moment is as follows: the parameter optimization priority at the current moment is obtained by mapping the product of the degree of remote collaboration difference of the parameters in the remote collaboration system at the current moment and the degree of adjustment trend of the parameters at the current moment through an exponential function.

[0031] Preferably, the calculation process of the feedback adjustment amount in the current moment's collaborative controller is as follows: it is determined by combining the baseline adjustment amount in the current moment's collaborative controller, the difference between the parameter optimization priority at the current moment and the parameter optimization priority at the previous sampling moment, and a preset adjustment amount threshold.

[0032] Preferably, the present invention also includes a remote collaborative optimization system for process parameters, the system including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the method described above.

[0033] Compared with the prior art, the beneficial effects of the present invention are:

[0034] Regarding adaptability to dynamic environments, this method no longer relies on fixed control strategies. Instead, it extracts fluctuation periods by performing time-domain analysis on parameter deviations and transmission delay durations, thereby constructing the degree of coordination and the degree of delay impact. This periodic-based analysis captures the inherent patterns of parameter changes and delay fluctuations, making the assessment of the remote collaborative system's status more closely aligned with actual operating conditions. By analyzing the correlation between the degree of coordination and the degree of delay impact within each fluctuation period, as well as their average state, the degree of remote coordination differences at the current moment can be accurately obtained. This provides a more precise basis for subsequent parameter adjustments and effectively addresses the dynamic changes in the network environment and production conditions.

[0035] Regarding the targeted nature of parameter optimization, this method innovatively constructs a parameter optimization priority system. By analyzing the decreasing trend of parameter deviation values ​​and the increasing trend of transmission delay duration across multiple adjacent sampling times at the current moment, an adjustment trend degree is formed. Combined with the degree of remote collaboration differences, the optimization priority of each parameter is determined. This process enables the system to identify key parameters that have a significant impact on the overall collaborative effect, avoiding resource waste caused by applying a uniform optimization strategy to all parameters. Simultaneously, determining the feedback adjustment amount based on the difference in optimization priority, the baseline adjustment amount, and the preset adjustment threshold ensures that the adjustment measures match the dynamic changes of the parameters, making the optimization of key parameters more timely and accurate, and improving the overall operational efficiency of the collaborative system.

[0036] Regarding the real-time performance of control optimization, this method combines feedback adjustment and actual adjustment through a collaborative controller, forming a closed-loop dynamic optimization mechanism. The determination of the feedback adjustment comprehensively considers changes in parameter optimization priorities and the system's baseline state, enabling rapid response to immediate changes in parameter deviations and transmission delays, thus reducing the lag in the adjustment strategy. This real-time adjustment mechanism can quickly take targeted measures when parameter deviations show an increasing trend or transmission delays suddenly increase, avoiding the cumulative effect of deviations and delays, thereby reducing production risks caused by parameter mismatches and ensuring product quality stability.

[0037] Furthermore, this method demonstrates significant advantages in multi-parameter collaborative scenarios. By dynamically prioritizing the optimization of different parameters, resources can be flexibly allocated and adjusted based on the real-time status of the system. This ensures that the adjustments of each parameter can cooperate and work synergistically in complex collaborative environments, avoiding conflicts and interference between parameters. This holistic optimization approach not only improves the control accuracy of individual parameters but also enhances the coordination of the entire remote collaborative system. It enables closer coordination of process parameters between multiple remote production units or control nodes, further improving the continuity and consistency of the production process.

[0038] This remote collaborative optimization method for process parameters comprehensively improves the performance of remote collaborative control of process parameters through in-depth analysis of dynamic characteristics, precise priority ranking, and real-time closed-loop adjustment, and can better meet the complex needs of distributed collaboration in modern industrial production. Attached Figure Description

[0039] Figure 1 This is a schematic diagram illustrating the working principle of the remote collaborative optimization method for process parameters described in this invention.

[0040] Figure 2 A flowchart for constructing collaboration;

[0041] Figure 3 A flowchart for determining the degree of difference in remote collaboration;

[0042] Figure 4 A flowchart was constructed to adjust the trend degree;

[0043] Figure 5 A flowchart for calculating the downward trend of the deviation value and the upward trend of the delay. Detailed Implementation

[0044] 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.

[0045] Please see Figures 1-5 This invention provides a method for remote collaborative optimization of process parameters, the method comprising the following steps:

[0046] The system acquires the deviation values ​​of process parameters and the transmission delay during remote collaborative processing. The parameter deviation value is the difference between the actual process parameters and the preset standard process parameters, and the transmission delay is the time interval between the sending end and the receiving end during the remote transmission of the process parameters.

[0047] Time-domain analysis is performed on the deviation value and transmission delay duration to extract their fluctuation periods. Time-domain analysis can identify the periodic fluctuation patterns by analyzing the curves of deviation value and transmission delay duration over time, thereby determining their respective fluctuation periods.

[0048] By analyzing the magnitude of frequent changes in parameter deviations and the fluctuation magnitude of transmission delay over time in the remote collaborative system during each fluctuation period, a collaborative degree and a delay impact degree are constructed. The collaborative degree reflects the activity level of parameter deviations within the fluctuation period, while the delay impact degree reflects the degree of influence of transmission delay fluctuations on the system within the fluctuation period.

[0049] Based on the correlation between the degree of coordination and the degree of delay impact in each fluctuation cycle prior to the current moment, as well as the average state of the degree of coordination and the average state of the degree of delay impact, the degree of remote coordination difference of parameters in the remote coordination system at the current moment is obtained. The degree of remote coordination difference comprehensively considers the historical performance and interrelationship of the degree of coordination and the degree of delay impact, and is used to measure the difference in parameter coordination in the current system.

[0050] By analyzing the decreasing trend of parameter deviation values ​​and the increasing trend of transmission delay duration across multiple adjacent sampling times, an adjustment trend degree for the parameters at the current moment is constructed. Combined with the degree of remote coordination differences in parameters within the remote coordination system, the parameter optimization priority of the remote coordination system at the current moment is obtained. Using the difference in parameter optimization priority, along with the baseline adjustment amount and preset adjustment threshold within the coordination controller at the current moment, the feedback adjustment amount within the coordination controller at the current moment is obtained. The adjustment trend degree reflects the recent changes in parameter deviation values ​​and transmission delay duration, while the parameter optimization priority determines the order in which different parameters are optimized. The feedback adjustment amount is a key quantity for optimized control, calculated based on multiple factors.

[0051] Based on the feedback adjustment and actual adjustment values ​​within the collaborative controller at the current moment, the remote collaborative process of process parameters is controlled and optimized using the collaborative controller. By combining the feedback adjustment values ​​with the actual adjustment values, the remote collaborative process of process parameters is dynamically adjusted, thereby achieving optimization of the process parameters.

[0052] Example 1:

[0053] Extracting the fluctuation period of parameter deviation and transmission delay requires combining historical data sequence analysis and feature recognition.

[0054] In extracting the fluctuation cycle of parameter deviation values, the first step is to determine the preset time period. The selection of the preset time period needs to be based on the actual operating rhythm of the remote collaborative process parameter scenario. For example, in a continuous production scenario, it can be determined according to the sampling frequency of the equipment and the production shift. If the sampling frequency is once per minute and needs to cover at least three complete potential fluctuation cycles, the preset time period can be set to 2 hours to ensure that it includes a sufficient number of historical sampling points. The deviation values ​​of all historical sampling moments within this preset time period before the current moment are collected. These deviation values ​​refer to the difference between the actual process parameters and the set standard parameters, covering the deviation of various process parameters such as temperature, pressure, and flow rate at different times. These deviation values ​​are arranged sequentially according to time to form the deviation value sequence for the current moment. Each element in the sequence corresponds to the deviation value of a specific sampling moment, and the order of the elements strictly follows the chronological order, from the earliest sampling moment to the last sampling moment before the current moment.

[0055] When performing time-domain analysis on a deviation value sequence, it is necessary to traverse the deviation value changes at each time point in the sequence. During the time-domain analysis, by observing the fluctuations of the deviation value over time, the volatility characteristics at each time point are identified. These volatility characteristics are reflected in the degree of difference between the deviation value and the deviation values ​​at adjacent time points; the greater the difference, the more significant the volatility characteristic. Among the volatility characteristics at all time points, the time points corresponding to the largest volatility characteristics are selected. These time points are the key nodes where significant changes in the deviation value occur. The time intervals between these key nodes are calculated. For example, if the time points corresponding to the largest volatility characteristics occur at 10 minutes, 30 minutes, and 50 minutes respectively, then the interval between two adjacent time points is 20 minutes. This interval is taken as the short-term deviation value fluctuation period before the current time.

[0056] To extract the fluctuation period of transmission delay, we first collect the transmission delay durations of all historical sampling moments within a preset time period before the current moment. Transmission delay duration refers to the time elapsed from when the process parameters are sent from the transmitter to when they are received and processed by the receiver, including the total time consumed by each step such as data packaging, network transmission, and unpacking verification. These transmission delay durations are arranged in chronological order to form a transmission delay sequence, where each element corresponds to the transmission delay duration of a sampling moment, and the order of arrangement is consistent with the sampling time.

[0057] A sliding window analysis method is used to process the transmission delay sequence. The size of the sliding window can be set according to the historical variation pattern of the transmission delay. If the transmission delay changes frequently, the window size can be appropriately reduced; if the change is relatively gradual, the window size can be appropriately increased. The window slides point by point on the transmission delay sequence in chronological order. After each slide, the transmission delay duration contained within the window is analyzed. The analysis includes the maximum, minimum, and average transmission delay within the window, as well as the change between adjacent time points. These data are used to determine the fluctuation of the transmission delay within the window. After the window has slid through the entire transmission delay sequence, the analysis results of all windows are summarized to identify the time intervals in which the transmission delay duration exhibits regular fluctuations. This interval is the transmission delay fluctuation period in the short period before the current time.

[0058] In practice, the length of the preset time period can be adjusted according to the type of process parameter. For parameters that change rapidly, the preset time period can be shortened to capture recent fluctuation patterns more promptly; for parameters that change slowly, the preset time period can be extended to ensure that the complete fluctuation cycle is included. In time-domain analysis, the judgment of the maximum fluctuation feature needs to be combined with the specific parameter characteristics. If the parameter has high stability requirements, even small fluctuations may be considered significant features; if the parameter allows for a certain range of fluctuations, a relatively lenient judgment standard needs to be set. In sliding window analysis, the sliding step size of the window can be consistent with the sampling interval to ensure that each sampling point is covered by multiple windows, thereby more comprehensively reflecting the fluctuation trend of transmission delay. Through the above process, the fluctuation cycle of deviation value and transmission delay duration can be extracted separately, providing basic data for the subsequent construction of coordination degree and delay impact degree.

[0059] When extracting fluctuation periods, it is crucial to ensure the completeness of historical sampling data. If data is missing for some sampling moments within a preset time period, data from adjacent moments can be interpolated to supplement the missing data. The supplementation method should be determined based on the data's trend to avoid deviations in fluctuation period extraction due to missing data. For deviation value sequences, if multiple time points with similar fluctuation characteristics appear in the time-domain analysis, the time interval with the highest frequency can be selected as the fluctuation period. For transmission delay sequences, if sliding window analysis yields multiple possible fluctuation periods, the period that best matches the overall trend can be selected for filtering based on transmission delay changes over a longer time range. Furthermore, fluctuation periods are not fixed and can change with variations in process conditions and network environment. Therefore, in practical applications, it is necessary to periodically re-extract fluctuation periods to adapt to new changes.

[0060] Example 2:

[0061] The construction of the degree of coordination needs to be based on the sequence analysis within the fluctuation period of the deviation value. After obtaining the fluctuation period of each deviation value, all deviation values ​​within each period are arranged sequentially from earliest to latest time, forming a deviation value period sequence corresponding to each fluctuation period. The number of elements in each sequence is consistent with the number of samplings within that fluctuation period, and the order of the elements strictly corresponds to the actual sampling time order, arranged from the deviation value at the beginning of the period to the deviation value at the end of the period. Correspondingly, for the transmission delay duration, after determining each delay fluctuation period, the transmission delay duration within each period is also arranged in ascending time order, forming a delay period sequence for each transmission delay fluctuation period. The structure of the sequence is consistent with the deviation value period sequence, ensuring that the transmission delay duration at each time point is in the corresponding position in the sequence.

[0062] For each deviation value periodic sequence, further processing is required to obtain its fluctuation characteristics. Specifically, the first-order change sequence of the sequence is calculated, which reflects the change in deviation values ​​by the difference between the deviation values ​​at two adjacent time points in the sequence. For example, if a deviation value periodic sequence contains deviation values ​​that are measurements taken at different times, then each element in the first-order change sequence is the difference between the deviation value at the later time point and the deviation value at the previous time point. In this way, the original deviation value sequence is transformed into a sequence reflecting the changing trend, thus more intuitively showing the dynamic changes of deviation values ​​within the fluctuation period.

[0063] After obtaining the first-order change sequence, the non-zero elements are statistically analyzed. During this process, each element in the first-order change sequence is examined one by one, and the number of all non-zero elements is recorded; this number represents the fluctuation frequency. The fluctuation frequency reflects the number of times the deviation value changes within a fluctuation period; the higher the frequency, the more frequently the deviation value changes within that period. Simultaneously, the average absolute value of all non-zero elements in the first-order change sequence is calculated. This involves summing the absolute values ​​of each non-zero element and then dividing by the total number of non-zero elements; the result is the average fluctuation amplitude. The average fluctuation amplitude reflects the average magnitude of each change in the deviation value; the larger the value, the more significant the magnitude of a single change. Multiplying the fluctuation frequency by the average fluctuation amplitude yields the degree of coordination within the fluctuation period of the deviation value. This value comprehensively reflects the activity level of the deviation value within the fluctuation period.

[0064] The construction of the delay impact degree is also based on the delay cycle sequence. For each delay cycle sequence, its first-order change sequence is calculated in the same way as the first-order change sequence of the deviation value cycle sequence, that is, the sequence is constructed by the difference in transmission delay duration between two adjacent time points. Unlike the calculation of the coordination degree, the calculation of the delay impact degree needs to consider the absolute values ​​of all elements in the first-order change sequence, regardless of whether these elements are zero. Specifically, the absolute values ​​of each element in the first-order change sequence are added together and then divided by the total number of elements in the sequence. The average value obtained is the delay impact degree of that transmission delay fluctuation cycle. This value comprehensively reflects the overall fluctuation amplitude of the transmission delay duration within the fluctuation cycle, covering the changes at all time points, including those moments when the change is zero, thus fully reflecting the impact of transmission delay duration fluctuations on the remote coordination process.

[0065] In practice, the lengths of the deviation value periodic sequence and the delay periodic sequence may differ depending on the fluctuation period. The specific composition of the sequence needs to be determined based on the actual number of samples within each period. The length of the first-order change sequence is always one element less than the original periodic sequence due to the way the difference between adjacent elements is calculated. When counting the number of non-zero elements, the boundary between zero and non-zero must be carefully distinguished to avoid statistical errors caused by approximating small values. When calculating the average fluctuation amplitude and delay impact, all relevant elements must be included in the calculation, including those with small values, to ensure that the results accurately reflect the actual fluctuation situation. Furthermore, for different remote collaboration scenarios with different process parameters, the range of deviation values ​​and transmission delay durations may differ, but the process for constructing the collaboration degree and delay impact remains consistent; only appropriate processing based on the characteristics of the actual data is required.

[0066] Example 3:

[0067] Determining the degree of remote collaboration difference of parameters in the current remote collaboration system requires basing it on the collaboration degree sequence and the delay impact degree sequence, combined with the correlation level of the two and their respective average states.

[0068] Collect the coherence degree corresponding to all deviation value fluctuation periods before the current moment, and arrange these coherence degrees in chronological order of their corresponding fluctuation periods to form a coherence degree sequence. For example, if there are three deviation value fluctuation periods, corresponding to coherence degrees C1, C2, and C3 respectively, and the chronological order of the three periods is the first period, the second period, and the third period, then the coherence degree sequence is [C1, C2, C3]. Similarly, collect the delay impact degree corresponding to all transmission delay fluctuation periods before the current moment, and arrange them in chronological order of their corresponding fluctuation periods to form a delay impact degree sequence. If there are three transmission delay fluctuation periods, corresponding to delay impact degrees D1, D2, and D3 respectively, and the chronological order is consistent with the above deviation value fluctuation periods, then the delay impact degree sequence is [D1, D2, D3].

[0069] Analyze the correlation level between the synergy series and the delayed influence series. The correlation level can be analyzed by observing whether the trends of the two series change over time are consistent. For example, when the value in the synergy series increases, observe whether the value in the delayed influence series increases, decreases, or remains stable to determine whether their trends are synchronized. If the values ​​of the two series increase or decrease simultaneously for most of the time period, the correlation level is high; if the trends frequently show opposites, the correlation level is low. Furthermore, comparing the magnitude of the changes in the values ​​of the two series can also aid in the judgment; if a large change in one series is often accompanied by a large change in the other, it further demonstrates a high correlation level.

[0070] Calculate the average state of elements within the coordination degree sequence. This is done by summing the values ​​of all elements in the coordination degree sequence and then dividing by the total number of elements. The result is the average state of coordination degree. This value reflects the overall level of coordination degree over all deviation fluctuation periods prior to the current time. Similarly, calculate the average state of elements within the delay impact degree sequence. This is done by summing the values ​​of all elements in the delay impact degree sequence and then dividing by the total number of elements. The result is the average state of delay impact degree, which reflects the overall level of delay impact degree over all transmission delay fluctuation periods prior to the current time.

[0071] After obtaining the average states of correlation level, coordination degree, and delay impact degree, the degree of remote coordination difference is determined by combining these three aspects of information. Specifically, if the correlation level is high, it indicates that the trends of coordination degree and delay impact degree are relatively consistent. In this case, the average states of both need to be considered. If the average states of coordination degree and delay impact degree are both at low levels, it indicates that the activity of parameter deviation values ​​and the fluctuation of transmission delay time are relatively small overall, and the degree of remote coordination difference is relatively low. If the average states of both are at high levels, the degree of remote coordination difference is relatively high. If the correlation level is low, it indicates that the trends of coordination degree and delay impact degree are inconsistent. In this case, the impact of the average states of both on the degree of remote coordination difference needs to be considered separately. For example, when the average state of coordination degree is high and the average state of delay impact degree is low, the degree of remote coordination difference is mainly determined by coordination degree, and the overall performance is moderate. When the average state of delay impact degree is high and the average state of coordination degree is low, the degree of remote coordination difference is mainly determined by delay impact degree, and the overall performance may also be moderate.

[0072] In practical processing, the lengths of the coordination degree sequence and the delay impact degree sequence must be consistent. If the sequence lengths are inconsistent due to differences in the number of fluctuation periods of the deviation value and the transmission delay, they can be adjusted by extracting the same number of recent period data to ensure effective correlation analysis. The judgment of the correlation level needs to be based on a sufficient amount of period data. If the sequence length is too short, the judgment result may be unreliable. Therefore, in practical applications, it is necessary to ensure that the number of collected fluctuation periods reaches a certain scale. The average state calculation of coordination degree and delay impact degree must include all elements in the sequence, without omitting any fluctuation period data, to ensure that the average state can truly reflect the overall level. Furthermore, in remote coordination scenarios with different process parameters, the numerical ranges of coordination degree and delay impact degree may differ. Therefore, when judging the degree of difference in remote coordination, a comprehensive analysis based on the parameter characteristics of the specific scenario is necessary to avoid judgment bias due to differences in numerical ranges. Through the above process, the correlation between coordination degree and delay impact degree and the overall level can be comprehensively considered, thereby reasonably determining the degree of remote coordination difference of parameters in the remote coordination system at the current moment.

[0073] The degree of difference in remote collaboration can also be determined in the following ways:

[0074] Remote collaboration difference level = (average collaboration level × average delay impact level) × correlation level coefficient

[0075] Among them, the average state of synergy is the average value of all elements in the synergy sequence; the average state of delay influence is the average value of all elements in the delay influence sequence; the correlation level coefficient is a coefficient set according to the correlation level between the synergy sequence and the delay influence sequence. The higher the correlation level, the larger the correlation level coefficient, and vice versa.

[0076] Example 4:

[0077] To construct the adjustment trend of parameters at the current moment, we need to start with the changes in parameter deviation and transmission delay at adjacent sampling moments, gradually analyze the trend characteristics of both, and perform comprehensive calculations.

[0078] First, determine the adjacent sampling times of the current moment. Adjacent sampling times refer to the multiple moments with the closest time interval to the current moment. The specific number can be determined based on the sampling frequency and rate of change of the process parameters. For example, if the sampling frequency is once every 5 minutes and the parameter changes frequently, three moments with intervals of 5 minutes, 10 minutes, and 15 minutes from the current moment can be selected as adjacent sampling times. If the parameter changes slowly, five adjacent sampling times can be selected to more comprehensively capture the trend of change. These adjacent sampling times must be arranged strictly in chronological order, ensuring that the time intervals from the current moment increase sequentially and are all before the current moment.

[0079] Collect the parameter deviation values ​​and transmission delay durations at these adjacent sampling times. The parameter deviation value refers to the difference between the actual process parameter and the standard parameter at each adjacent time. For example, if the standard value of a temperature parameter is 80℃, and the actual values ​​at adjacent sampling times are 82℃, 81℃, and 79℃, then the corresponding deviation values ​​are 2℃, 1℃, and -1℃. The transmission delay duration refers to the time consumed during the remote transmission of the process parameter at each adjacent time. For example, the transmission delay durations at adjacent sampling times might be 0.3 seconds, 0.4 seconds, and 0.5 seconds, respectively. Arrange these parameter deviation values ​​in ascending order of time, that is, from the earliest adjacent sampling time to the most recent adjacent sampling time, to form a deviation value trend sequence for the current time. For example, the deviation values ​​of the above three adjacent times arranged in ascending order of time are [2℃, 1℃, -1℃], forming a deviation value trend sequence. Similarly, arrange the transmission delay durations in ascending order of time, such as [0.3 seconds, 0.4 seconds, 0.5 seconds], to form a delay trend sequence.

[0080] The first-order change sequence of the deviation value trend sequence is calculated. The first-order change sequence is obtained by the difference between each element and the previous element in the sequence. For example, the first-order change sequence of the deviation value trend sequence [2℃, 1℃, -1℃] is [1℃-2℃, -1℃-1℃], or [-1℃, -2℃]. Negative values ​​in this first-order change sequence are analyzed; negative values ​​indicate that the deviation value at a later time point has decreased compared to the previous time point. The sum of the absolute values ​​of these negative values ​​is calculated. In the example above, both elements of the first-order change sequence are negative, with absolute values ​​of 1℃ and 2℃ respectively, totaling 3℃. This sum represents the downward trend of the deviation value at the current time point. If there are positive values ​​or zero values ​​in the first-order change sequence, positive values ​​indicate an increase in the deviation value, and zero values ​​indicate no change. These values ​​are not included in the calculation of the downward trend of the deviation value; only the sum of the absolute values ​​of the negative values ​​is considered.

[0081] For a delay trend sequence, its first-order change sequence is also calculated. For example, the first-order change sequence of the delay trend sequence [0.3 seconds, 0.4 seconds, 0.5 seconds] is [0.4 seconds - 0.3 seconds, 0.5 seconds - 0.4 seconds], or [0.1 seconds, 0.1 seconds]. Analyze the positive values ​​in this first-order change sequence; positive values ​​indicate that the transmission delay at the next moment has increased compared to the previous moment. Sum these positive values; in the example above, both elements are positive, and the sum is 0.2 seconds. This sum represents the delay increase trend at the current moment. If there are negative values ​​or zero in the first-order change sequence, negative values ​​indicate a decrease in transmission delay, and zero indicates no change. These values ​​are not included in the calculation of the delay increase trend; only the sum of positive values ​​is considered.

[0082] Multiplying the current deviation's downward trend by the current delay's upward trend yields the adjustment trend degree of the parameter at that moment. For example, in the above case, the deviation's downward trend is 3°C, and the delay's upward trend is 0.2 seconds; multiplying them gives 0.6 (°C·second), which is the adjustment trend degree at that moment. If the first-order change sequence of the deviation trend sequence contains no negative values, meaning the deviation does not show a downward trend, then the deviation's downward trend is 0, and the adjustment trend degree is also 0. If the first-order change sequence of the transmission delay trend sequence contains no positive values, meaning the transmission delay duration does not show an upward trend, then the delay's upward trend is 0, and the adjustment trend degree is also 0.

[0083] In practice, the number of adjacent sampling times can be flexibly adjusted according to the stability of the process parameters. For parameters with high stability and slow changes, the number of adjacent sampling times can be appropriately reduced to avoid introducing too much redundant information; for parameters with low stability and frequent changes, the number of adjacent sampling times needs to be increased to more accurately capture trend changes. When collecting data, it is necessary to ensure that the parameter deviation values ​​and transmission delay durations of each adjacent sampling time are accurate. If there is a significant anomaly in the data at a certain moment, such as a sudden deviation value exceeding the normal range or a sharp increase in transmission delay duration, the data at that moment can be temporarily excluded, or reasonable data from adjacent moments can be used to supplement it, so as to avoid the outlier interfering with the trend analysis.

[0084] When constructing the deviation value trend series and the delayed trend series, it is essential to strictly adhere to the ascending chronological order to ensure that the order of elements in the series matches the actual chronological order. Otherwise, errors may occur in the calculation of the first-order change series. When calculating the first-order change series, the difference of each element must be carefully checked to avoid affecting the subsequent trend quantity statistics due to calculation errors. When calculating the downward trend quantity of the deviation value and the upward trend quantity of the delayed trend, it is crucial to clearly distinguish between positive, negative, and zero values ​​to ensure that only elements meeting the criteria are included in the calculation.

[0085] Example 5:

[0086] The calculation of the parameter optimization priority of the remote collaborative system at the current moment needs to be based on the degree of remote collaborative difference and the adjustment trend. The degree of remote collaborative difference of the parameters in the remote collaborative system at the current moment is obtained, which is the result of comprehensively considering the correlation level and average state of the collaborative degree and delay impact of previous fluctuation cycles. Simultaneously, the adjustment trend of the parameters at the current moment is obtained, which is obtained by multiplying the decreasing trend of parameter deviation values ​​between adjacent sampling moments by the increasing trend of transmission delay. Multiplying these two values ​​yields a product value that comprehensively reflects the current parameter collaborative state and its changing trend.

[0087] The product value is mapped using an exponential function. The choice of exponential function depends on the specific scenario of parameter optimization. For example, the natural exponential function can be used, as it amplifies changes in numerical values ​​and more clearly distinguishes the priority differences corresponding to different product values. During the mapping process, the product value serves as the input to the exponential function, and the output value obtained after function calculation represents the parameter optimization priority at the current moment. The magnitude of this value reflects the urgency of the parameter's need for optimization; the larger the value, the higher the priority for addressing the problem with that parameter during remote collaboration.

[0088] During the calculation, if the degree of difference in remote collaboration or the degree of adjustment trend are zero, then their product is also zero. After the exponential function mapping, the parameter optimization priority will remain at a baseline level, indicating that the current parameter state is relatively stable and does not require priority optimization. If both are large values, their product will be larger, and after the exponential function mapping, the parameter optimization priority will increase significantly, indicating that the parameter needs to be optimized and adjusted immediately to improve the remote collaboration effect.

[0089] The calculation of the feedback adjustment amount within the collaborative controller at the current moment needs to consider multiple factors. First, a baseline adjustment amount must be determined within the collaborative controller. This baseline adjustment amount is an initial adjustment value preset by the collaborative controller based on the basic characteristics of the process parameters and the general requirements of remote collaboration. Its magnitude is related to factors such as the type of parameter and the scale of the remote collaboration system. For example, for temperature-related process parameters, the baseline adjustment amount may be set to a relatively small value to avoid excessive adjustment affecting process stability; for flow-related process parameters, the baseline adjustment amount may be relatively large to accommodate their rapid change characteristics.

[0090] The calculation focuses on the difference in parameter optimization priority, specifically the difference between the current parameter optimization priority and the priority at the previous sampling time. If the current priority is higher than the previous priority, the difference is positive, indicating an increasing urgency for parameter optimization; conversely, if the current priority is lower than the previous priority, the difference is negative, indicating a decreasing urgency for parameter optimization. This difference reflects the changing trend of the parameter state and serves as a crucial basis for adjusting the feedback adjustment amount.

[0091] A preset adjustment threshold is determined, which is set based on the allowable adjustment fluctuation range during remote collaboration of process parameters. The size of the threshold needs to balance the flexibility of adjustment with the stability of the system. If the threshold is set too high, the change in the feedback adjustment amount may be too drastic, affecting the stability of the system; if the threshold is set too low, the feedback adjustment amount may not be able to adapt to rapid changes in parameter status, resulting in poor optimization effect.

[0092] After obtaining the difference between the baseline adjustment amount, the parameter optimization priority, and the preset adjustment amount threshold, the feedback adjustment amount at the current moment is calculated by combining these three factors. The specific method can be determined according to the actual situation. For example, when the difference is positive and less than the preset adjustment amount threshold, a value proportional to the difference can be added to the baseline adjustment amount to moderately increase the feedback adjustment amount; when the difference is positive and greater than or equal to the preset adjustment amount threshold, the feedback adjustment amount can be directly adjusted to the sum of the baseline adjustment amount and the preset adjustment amount threshold to avoid excessive adjustment; when the difference is negative, a value proportional to the absolute value of the difference can be subtracted from the baseline adjustment amount to moderately decrease the feedback adjustment amount, but it is necessary to ensure that the adjusted feedback adjustment amount is not lower than zero to ensure the effectiveness of the adjustment.

[0093] In practical applications, the calculation of the feedback adjustment amount needs to be flexibly adjusted based on the specific circumstances of the parameter optimization priority difference. For example, when the difference is positive and small, it indicates that although the parameter state has deteriorated, the degree is minor, and a small increase in the feedback adjustment amount is sufficient. When the difference is positive and large, it indicates that the parameter state has deteriorated significantly, and a larger increase in the feedback adjustment amount is needed to quickly improve the state. When the difference is negative, it indicates that the parameter state is improving, and the feedback adjustment amount can be appropriately reduced to save adjustment resources and maintain system stability.

[0094] In different remote collaborative scenarios involving process parameters, the settings for baseline adjustment amounts and preset adjustment thresholds vary and need to be configured according to the actual needs of the specific scenario. For example, in precision manufacturing processes with extremely high stability requirements, both the baseline adjustment amount and the preset adjustment threshold are set relatively small to ensure the smoothness of the adjustment process; while in chemical production processes with high response speed requirements, the baseline adjustment amount and the preset adjustment threshold may be set relatively large to achieve rapid adjustment.

[0095] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0096] 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 method for remote collaborative optimization of process parameters, characterized in that, Includes the following steps: Obtain the deviation values ​​of parameters and the transmission delay during the remote collaboration of process parameters; Time-domain analysis was performed on the deviation value and transmission delay duration to extract the fluctuation period of the deviation value and transmission delay duration. By analyzing the magnitude of frequent changes in the deviation values ​​of parameters in the remote collaborative system and the magnitude of fluctuations in the transmission delay duration over time within each fluctuation cycle, the degree of collaboration and the degree of delay impact are constructed respectively. Based on the correlation between the degree of coordination and the degree of delay impact of each fluctuation cycle before the current moment, as well as the average state of the degree of coordination and the average state of the degree of delay impact, the degree of remote coordination difference of parameters in the remote coordination system at the current moment is obtained. Analyze the decreasing trend of parameter deviation values ​​and the increasing trend of transmission delay time at multiple adjacent sampling times at the current time to construct the adjustment trend degree of parameters at the current time. Combined with the degree of remote collaboration difference of parameters in the remote collaboration system, obtain the parameter optimization priority of the remote collaboration system at the current time. Utilize the change difference of parameter optimization priority and the baseline adjustment amount and preset adjustment amount threshold in the collaboration controller at the current time to obtain the feedback adjustment amount in the collaboration controller at the current time. Based on the feedback adjustment and actual adjustment in the collaborative controller at the current moment, the remote collaborative process of process parameters is controlled and optimized using the collaborative controller. The construction of the adjustment trend degree of the parameters at the current moment includes: The multiple times with the closest time interval to the current time are all taken as the adjacent sampling times of the current time. Based on the changes in parameter deviation values ​​and transmission delay duration of the multiple adjacent sampling times, the decreasing trend of the deviation value and the increasing trend of the delay at the current time are obtained. The product of the current deviation value's downward trend and the delayed upward trend is determined as the degree of adjustment trend in the parameters at the current moment. The current deviation value's downward trend and delayed upward trend further include: The parameter deviation values ​​and transmission delay durations at multiple adjacent sampling times are arranged in ascending order of time to form the deviation value trend sequence and delay trend sequence at the current time. The first-order change sequences of the deviation value trend sequence and delay trend sequence at the current time are calculated respectively. The sum of the absolute values ​​of all negative values ​​in the first-order change sequence of the deviation value trend sequence is taken as the current deviation value downward trend amount, and the sum of all positive values ​​in the first-order change sequence of the delayed trend sequence is taken as the current delayed upward trend amount.

2. The method for remote collaborative optimization of process parameters as described in claim 1, characterized in that, The extraction of the fluctuation period of the parameter deviation value and the transmission delay duration further includes: Obtain the sequence of deviation values ​​and transmission delay durations of all historical sampling times within a preset time period before the current time, arranged in chronological order, and use it as the deviation value sequence and transmission delay sequence for the current time. A time-domain analysis is performed on the deviation value sequence at the current moment to obtain the fluctuation characteristics of all time points in the time domain. The interval of the time point corresponding to the maximum fluctuation characteristic is taken as the deviation value fluctuation period in the short period before the current moment. Correspondingly, for the transmission delay sequence, a sliding window analysis is used to obtain the transmission delay fluctuation period in the short period before the current moment.

3. The method for remote collaborative optimization of process parameters as described in claim 1, characterized in that, The construction of the degree of synergy further includes: The deviation values ​​within each deviation value fluctuation period before the current time are arranged in ascending order of time to form the deviation value fluctuation period sequence of each deviation value fluctuation period. Correspondingly, for the transmission delay duration within each delay fluctuation period, the delay period sequence of each transmission delay fluctuation period is obtained. For each deviation value periodic sequence, obtain the first-order change sequence of the deviation value periodic sequence, count the number of all non-zero elements in the first-order change sequence, and record it as the fluctuation frequency. Calculate the average value of the absolute values ​​of all non-zero elements in the first-order change sequence, and record it as the average fluctuation amplitude. The product of the fluctuation frequency and the average fluctuation amplitude is used as the degree of coordination within each deviation value fluctuation period.

4. The method for remote collaborative optimization of process parameters as described in claim 3, characterized in that, The construction of the delay impact further includes: For each delay period sequence, the average of the absolute values ​​of all elements in the first-order change sequence of the delay period sequence is calculated as the delay influence degree of each transmission delay fluctuation period.

5. The method for remote collaborative optimization of process parameters as described in claim 4, characterized in that, The process of determining the degree of remote collaboration difference of parameters in the remote collaboration system at the current moment is as follows: combining the correlation level between the collaboration degree sequence and the delay impact degree sequence at the current moment, as well as the average state of the elements in the collaboration degree sequence and the average state of the elements in the delay impact degree sequence, wherein the collaboration degree of all deviation value fluctuation cycles is arranged in ascending order of time to form the collaboration degree sequence at the current moment, and the delay impact degree of all transmission delay fluctuation cycles is arranged in ascending order of time to form the delay impact degree sequence at the current moment.

6. The method for remote collaborative optimization of process parameters as described in claim 1, characterized in that, The calculation process for the parameter optimization priority of the remote collaboration system at the current moment is as follows: the parameter optimization priority at the current moment is obtained by mapping the product of the degree of remote collaboration difference of the parameters in the remote collaboration system at the current moment and the degree of adjustment trend of the parameters at the current moment through an exponential function.

7. The method for remote collaborative optimization of process parameters as described in claim 1, characterized in that, The calculation process of the feedback adjustment amount in the current moment's collaborative controller is as follows: it is determined by combining the baseline adjustment amount in the current moment's collaborative controller, the difference between the parameter optimization priority at the current moment and the parameter optimization priority at the previous sampling moment, and the preset adjustment amount threshold.

8. A remote collaborative optimization system for process parameters, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-7.

Images