Feedback adjustment method of sliding mode control system under data interconnection

Abstract

The invention provides a feedback adjustment method of a sliding-mode control system under data interconnection, and relates to the technical field of sliding-mode control, and the method comprises the steps: obtaining feedback monitoring data, carrying out the reliability analysis and evaluation of a data source based on the quality evaluation parameter of a corresponding data source and historical data, and determining a reliability coefficient of the data source; performing screening setting on the data source reliability coefficient by using a preset screening threshold to obtain an effective data source; setting a feedback loop ratio according to the data source reliability coefficient; establishing a control updating relation between the effective data source and the sliding mode surface, and feeding back the effective data source and a feedback loop ratio of the data source to a sliding mode control system; and adjusting the sliding mode surface gain, and generating a control law to control and adjust the sliding mode control system. The sliding mode control system solves the technical problem that a sliding mode control system in the prior art usually adopts a fixed sensor for data input, so that the performance of the control system is influenced by unreliable or inaccurate data, and the control effect is not ideal.

Description

Technical Field

[0001] The invention relates to the technical field of sliding mode control, and in particular to a feedback adjustment method of a sliding mode control system under data interconnection. Background Art

[0002] Sliding mode control (SMC) is a nonlinear control method that is widely used in areas where the system has strong robustness requirements against uncertainties and external disturbances. It designs a sliding surface (SlidingSurface) so that the system state slides on the surface, thereby achieving precise control of the system. The main advantages of sliding mode control lie in its strong robustness and good dynamic performance, especially in the face of model uncertainty, external disturbances and changes in system parameters.

[0003] However, existing sliding mode control systems usually use fixed sensor data input, while ignoring the differences in quality and timeliness of different sensor data. In many application scenarios, the data provided by multiple sensors may be inconsistent due to hardware failure, environmental interference, data loss or delay, etc. Traditional sliding mode control methods usually lack the ability to dynamically evaluate and adaptively adjust the reliability of data sources. Therefore, when processing multi-sensor data, the performance of the system is often affected. Summary of the invention

[0004] The present application provides a feedback adjustment method for a sliding mode control system under data interconnection, aiming to solve the technical problem that the sliding mode control system in the prior art usually adopts fixed sensor data input, ignores the differences in quality and timeliness of different sensor data, causes unreliable or inaccurate data to affect the performance of the control system, and makes the control effect unsatisfactory.

[0005] The present application discloses a feedback adjustment method for a sliding mode control system under data interconnection, the method comprising: obtaining feedback monitoring data, performing reliability analysis and evaluation on the data source based on the quality evaluation parameters and historical data of the data source corresponding to the feedback monitoring data, and determining the data source reliability coefficient; screening and setting the data source reliability coefficient using a preset screening threshold to obtain a valid data source; performing feedback loop setting according to the data source reliability coefficient of the valid data source; establishing a control update relationship between the valid data source and a sliding mode surface, and feeding back the valid data source and its feedback loop of the data source to the sliding mode control system; adjusting the sliding mode surface gain according to the feedback loop ratio, and generating a control law to control and adjust the sliding mode control system.

[0006] One or more technical solutions provided in this application have at least the following beneficial effects: By obtaining feedback monitoring data and performing reliability analysis and evaluation on the data source, the reliability of the data source can be evaluated based on the quality evaluation parameters and historical data of the data source. This process ensures that the data is screened based on its reliability when used, thereby improving the accuracy of the data source. This method can effectively exclude unreliable or low-quality data sources and avoid their negative impact on system control; by using the preset screening threshold to screen the reliability coefficient of the data source, an effective data source is obtained. This screening process ensures that only reliable data sources are selected into the control system, which helps to reduce the impact of noise data on the control system. The dynamic screening mechanism enables the system to select the best input data in real time according to changes in the data source, thereby enhancing the flexibility and adaptability of the system; the feedback loop is set based on the reliability coefficient of the effective data source. The adaptive adjustment of the feedback loop can dynamically change the influence of each data source in the control system according to its quality and reliability. , reliable data sources will have a greater impact on the control system, while the impact of unreliable data sources is effectively reduced. This process ensures that the control system can maintain stable and precise control when the quality of the data source fluctuates; by establishing a control update relationship between the effective data source and the sliding surface, and returning the feedback loop to the sliding mode control system, it is ensured that the system can adjust the sliding surface according to the real-time effective data. This control update mechanism allows the system to dynamically adjust according to the actual data quality, optimize the controller response and stability, and improve the overall performance of the control system; adjust the sliding surface gain according to the feedback loop ratio to generate the control law for control adjustment. This step dynamically adjusts the sliding surface gain so that the control system can maintain a high robustness when facing data sources of different quality, enhances the system's adaptability to changes in data quality, and ensures that the system can make accurate control decisions even in the event of sensor failure or data fluctuations.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flow chart of a feedback adjustment method for a sliding mode control system under data interconnection is provided for an embodiment of the present application.

[0009] Figure 2 A schematic diagram of a data compensation process in a feedback adjustment method of a sliding mode control system under data interconnection is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0010] The embodiments of the present application provide a feedback adjustment method for a sliding mode control system under data interconnection, thereby solving the technical problem that the sliding mode control system in the prior art usually adopts fixed sensor data input, ignores the differences in quality and timeliness of different sensor data, and causes unreliable or inaccurate data to affect the performance of the control system, resulting in unsatisfactory control effect.

[0011] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0012] like Figure 1 As shown, the embodiment of the present application provides a feedback adjustment method for a sliding mode control system under data interconnection, and the method includes: Acquire feedback monitoring data, perform reliability analysis and evaluation on the data source based on the quality evaluation parameters and historical data of the data source corresponding to the feedback monitoring data, and determine the reliability coefficient of the data source.

[0013] Feedback monitoring data is the data collected by sensors or measuring devices in sliding mode control systems. These data are used to monitor the status and performance of the system in real time. They include temperature, pressure, current, speed and other types of data, depending on the application field of the control system.

[0014] Quality evaluation parameters are indicators for evaluating different quality characteristics of data sources, including data accuracy, completeness, timeliness, stability, consistency, etc. These evaluation parameters can reflect the reliability and accuracy of data sources and are the basis for evaluating data sources. Reliability analysis and evaluation of data sources are performed based on quality evaluation parameters. Specifically, corresponding quality evaluation parameters are extracted by feedback monitoring data. For example, for sensor data, its accuracy, error range, etc. can be evaluated. These quality evaluation parameters are analyzed and calculated. Usually, a certain weight is assigned to each parameter to represent its importance in the overall evaluation. A weighted algorithm is used to comprehensively calculate multiple evaluation parameters to obtain a comprehensive quality score for each data source. This score is the reliability evaluation of the data source.

[0015] Historical data refers to the data collected by the system over a period of time in the past. These data can provide clues about the stability of the data source, especially during long-term use, when the data source may experience different performance fluctuations. Reliability analysis is performed based on historical data. Specifically, historical data is analyzed to evaluate the long-term reliability of the data source. Methods include error analysis, time series analysis, stability evaluation, etc. The stability and reliability of the data source are evaluated based on the time series characteristics of these historical data to obtain a stability score, which represents the reliability of the data source in long-term operation.

[0016] Based on the comprehensive analysis results of quality evaluation parameters and historical data, the reliability coefficient of each data source is obtained. This coefficient reflects the credibility of the data source and is used to measure the contribution of the data source to the control system. The reliability coefficient is a value between 0 and 1, where 1 means that the data source is very reliable and 0 means that the data source is completely unreliable. Different data sources can be assigned different reliability coefficients based on their quality evaluation and historical stability results.

[0017] The data source reliability coefficient is screened and set using a preset screening threshold to obtain a valid data source.

[0018] The reliability coefficient of the data source is screened using a preset screening threshold to determine which data sources are valid and which should be excluded. Specifically, the preset screening threshold is preset based on the specific needs and design standards of the system. For example, the threshold is set to 0.5, which means that only data sources with a reliability coefficient greater than or equal to 0.5 are considered credible. The reliability coefficients of all data sources are evaluated and compared with the preset screening threshold. Those data sources with a reliability coefficient higher than the threshold are selected as valid data sources for further control and analysis.

[0019] A feedback loop setting is performed according to the data source reliability coefficient of the effective data source.

[0020] The data source reliability coefficient of the valid data source is used to set the feedback ratio of each valid data source. The feedback ratio refers to adjusting the weight of the control input according to the data validity. For example, if the data provided by a sensor is very accurate and timely, then the feedback effect of the data source on the sliding mode controller may be enhanced; conversely, if the validity of the data source is low, the feedback effect on the control system may be reduced. The feedback ratio of each data source is used to adjust the impact of the data source on the controller output. The larger the feedback ratio of the data source, the greater its impact on the controller decision.

[0021] A control update relationship between the effective data source and the sliding surface is established, and the effective data source and its data source feedback loop are fed back to the sliding mode control system.

[0022] Integrate the effective data source and its calculated feedback loop ratio into the control logic of the sliding mode control system to ensure that changes in the data source can affect the control strategy in real time. Specifically, first determine how to integrate the information of each effective data source (such as sensor readings) and its feedback loop ratio into the calculation model of the control system to ensure that the data can be correctly reflected in the control output. The sliding surface is a mathematical expression used in control system design to describe the error between the system state and the desired state. In this step, adjust the definition of the sliding surface according to the feedback loop ratio of the effective data source, such as adjusting the weight of the error term to reflect the importance and reliability of different data sources.

[0023] The sliding surface gain is adjusted according to the feedback loop ratio, and a control law is generated to control and adjust the sliding mode control system.

[0024] The sliding surface gain determines the speed and intensity of the system's response to deviations. By adjusting these gain parameters, usually proportional, integral, and differential gains, the control response can be optimized based on the reliability of the effective data source. For example, if a data source is very reliable and its corresponding feedback loop ratio is high, then the related control gain can also be set higher to respond more quickly to changes in the data source. The control law is designed based on the adjusted sliding surface, which defines how to calculate the control input that needs to be applied to the system from the current state of the sliding surface, including but not limited to control strategies for linear and nonlinear systems. By establishing a control update relationship between the effective data source and the sliding surface, and adjusting the sliding surface gain based on the feedback loop ratio, it is ensured that the sliding mode control system can dynamically adjust its behavior based on the most reliable and relevant data. Such a strategy helps to improve the accuracy and adaptability of the system and maintain performance and stability under changing environments and conditions.

[0025] Furthermore, if Figure 2 As shown, the obtaining of a valid data source further includes: Determine whether the effective data source meets the control requirements; when the control requirements are not met, perform data compensation according to the effective data source, add the compensation data to the effective data source, and set the reliability coefficient of the compensation data source according to the compensation.

[0026] The control requirement is the minimum number of valid data sources required for system design. Check whether the number of valid data sources reaches the minimum number required for system design. If the data sources are not sufficient to cover all control variables of the system or cannot provide the required data quality, further processing is required.

[0027] When it is determined that the effective data source fails to meet the control requirements, a data compensation mechanism needs to be implemented. Specifically, first clarify the type of data that needs to be compensated, such as temperature, pressure, etc., as well as the quantity and quality that need to be compensated. Generate compensation data by adding additional sensors or using software algorithms. For example, if the data of a certain sensor is unstable, the data of other sensors can be introduced for averaging or weighted processing to enhance the stability of the data. According to the generation method and accuracy of the compensation data, set appropriate reliability coefficients for these data. For example, if the compensation data is obtained by fusion of data from multiple sensors, its reliability coefficient may be higher than that of the data from a single sensor. Integrate the compensation data and its reliability coefficient into the control system to ensure that these data can be used correctly by the system and adjust the corresponding control strategy.

[0028] Furthermore, it also includes: A multi-level screening threshold is set, where the multi-level screening threshold is a multi-level credible interval for screening data sources; the multi-level credible interval is used to perform multi-level matching identification on the reliability coefficient of the data source to obtain the effective level of the data source; a priority factor is configured according to the effective level of the data source, and the priority factor is used to correct the feedback loop ratio, wherein the higher the effective level, the larger the corresponding priority factor.

[0029] According to the reliability requirements of different data sources, multi-level screening thresholds are set. These thresholds are divided into multiple levels and are used to construct multi-level trust intervals. For example, three levels of thresholds can be set, such as high level, medium level, and low level. The threshold corresponding to each level reflects the reliability of the data source, ensuring that the screened data sources can meet different control requirements. These level thresholds form multiple trust intervals, and the reliability coefficient of each data source will be mapped to the corresponding interval. In this way, different control strategies can be adopted according to the reliability of different data sources.

[0030] For each valid data source, its credible interval is determined based on its reliability coefficient. For example, if the reliability coefficient of a data source is 0.7 and the set credible interval is 0.5 to 0.8, the data source is classified as medium reliability. Through the matching process, the validity level of each data source is identified. The higher the validity level of a data source, the more reliable it is and the more valuable information it can provide for the control system.

[0031] According to the effective level of the data source, a priority factor is configured for each data source to correct the feedback loop ratio of the data source to the sliding mode control system. The priority factor is used to indicate the importance of data sources of different levels. Generally speaking, the higher the effective level of the data source, the larger the configured priority factor, which means that high-level data sources will have a greater impact on the feedback loop ratio, while low-level data sources will have a smaller impact. When the priority factors are determined, these factors are applied to the calculation of the feedback loop ratio. The priority factors will correct the feedback loop ratio, so that the reliability of the data source directly affects its contribution to the control system. Specifically, the feedback loop ratio with a higher priority factor will have a stronger impact on the control system, while the feedback loop ratio with a lower priority factor will have a smaller impact. This process ensures that the sliding mode control system can adjust the feedback loop ratio according to the quality of the data source, and can make a reasonable response when facing changes in the quality of different data sources, thereby optimizing the control effect.

[0032] Furthermore, according to the data source reliability coefficient of the effective data source, feedback loop setting is performed, including: The feedback loop calculation formula is: , calculate the feedback loop ratio, where, is the feedback ratio of the i-th data source, is the reliability coefficient of the ith data source, is a set of valid data sources; the feedback loop is corrected using the priority factor, and the correction expression of the feedback loop calculation formula is: , k is the priority factor, is the priority factor of the i-th data source.

[0033] The feedback loop calculation formula is as follows: ,in, represents the feedback loop ratio of the i-th data source at time t. The feedback loop ratio is an indicator used to measure the contribution weight of each data source to the sliding mode control system; is the reliability coefficient of the ith data source at time t, indicating the reliability of the data source. The higher the reliability coefficient, the more reliable the data source, and vice versa. is the set of valid data sources at time t, meaning all valid data sources with sufficient reliability at that time; It is the sum of the reliability coefficients of all valid data sources at time t, reflecting the total reliability of all valid data sources.

[0034] Overall, Describes the proportion of the i-th data source in all valid data sources. The higher the reliability coefficient of the data source, the higher the corresponding feedback loop The larger the value, the greater the contribution of the data source to the control system, and vice versa. This formula helps determine the degree of influence of each data source on the feedback of the sliding mode control system.

[0035] The feedback loop is corrected using the priority factor, and the expression is: ,in, is the priority factor of the ith data source, which indicates the relative priority of the data source among all data sources. The larger the priority factor, the more important the data source is in the feedback calculation. This formula shows that the priority factor is added on the basis of the original calculation of the feedback loop ratio, so as to correct its influence on the sliding mode control system according to the priority of the data source. Specifically, by introducing the priority factor, the proportion of the data source in the feedback loop ratio is adjusted. If a data source has a higher priority factor , its feedback loop ratio will increase accordingly, otherwise it will decrease. The priority factor is usually set according to the importance or reliability of the data source. This method allows the system to flexibly adjust the influence of different data sources to ensure that the most important and reliable data sources have a greater impact on the feedback of the control system, thereby improving the accuracy and stability of the overall system. In short, this correction process further weights and adjusts the feedback loop ratio of each data source through the priority factor to ensure that the data sources that are critical to the control system receive a higher weight, further improving the responsiveness and robustness of the system.

[0036] Furthermore, the reliability of the data source is analyzed and evaluated based on the quality evaluation parameters of the data source corresponding to the feedback monitoring data and historical data to determine the reliability coefficient of the data source, including: According to the quality evaluation parameters of the data source, each parameter is evaluated separately to obtain the evaluation value of the multi-dimensional quality evaluation parameter; the influence of each dimensional quality evaluation parameter on the reliability of the data source is obtained, and the real-time evaluation weight is set; based on the real-time evaluation weight, the evaluation value of the multi-dimensional quality evaluation parameter is weightedly calculated to obtain the quality evaluation result; according to the historical data, the reliability and stability of the data source is evaluated to obtain the time series stability evaluation result; according to the quality evaluation result and the time series stability evaluation result, a comprehensive evaluation is performed to determine the reliability coefficient of the data source.

[0037] In order to comprehensively evaluate the quality of data sources, we first need to determine quality evaluation parameters in multiple dimensions, including but not limited to data accuracy, timeliness, stability, consistency, completeness, etc., and independently evaluate the quality parameters of each dimension and assign scores based on the performance of each parameter. For example, if a data source excels in accuracy, it may be assigned a higher score; if it performs poorly in timeliness, it may receive a lower score.

[0038] Analyze the contribution of different quality parameters to the reliability of the data source. For example, accuracy may have a greater impact on reliability, while timeliness may not be as important as accuracy in some applications. Set the real-time evaluation weight of each parameter based on the importance of each quality parameter. For example, if accuracy is critical to the control system, a higher weight can be assigned to it, while timeliness may have a lower weight. The real-time weight reflects the contribution ratio of each evaluation parameter to the reliability of the data source in the actual control process.

[0039] According to the set real-time weight, the evaluation value of each quality parameter is weighted and summed. The evaluation value of each dimension is combined into a final quality evaluation result through weighted calculation. The weighted calculation result is the quality evaluation result of the data source, which reflects the comprehensive quality level of the data source in different dimensions. The quality evaluation result is a value between 0 and 1. The larger the value, the more reliable the data source.

[0040] Based on historical data, the stability of each data source is evaluated, with special attention paid to time series stability. Time series stability involves the stability, change trend and volatility of the data source over a long period of time. Specifically, time series data analysis is performed, that is, based on historical data, the changes in the data source in the time series are analyzed. For example, for lidar data, the stability and deviation of lidar data in history are evaluated based on past positioning errors. Stability evaluation is performed. For example, in an autonomous driving system, the historical positioning deviation of lidar can be used to evaluate the time series stability of lidar. If the positioning error of lidar does not change much over time, it means that its time series stability is good. If the error is large and unstable, it means that the stability of the data source is poor. Time series volatility is evaluated. For industrial automation systems, the long-term drift of monitoring sensors is also part of time series stability. If the temperature sensor runs stably for a long time and its data volatility is small, then its time series stability is also good. Finally, through time series stability analysis, a time series stability evaluation result is generated, that is, the stability score of the data source in long-term operation.

[0041] Combining the quality evaluation results and the timing stability evaluation results, the final reliability coefficient is calculated using methods such as weighted average. The final reliability coefficient is between 0 and 1. The larger the value, the higher the reliability of the data source. The data source with a higher reliability coefficient will have a greater impact on the feedback loop of the sliding mode control system, ensuring that the system can make more accurate and stable control decisions.

[0042] Furthermore, the reliability and stability of the data source is evaluated based on the historical data to obtain a time series stability evaluation result, including: The error value of the data source is calculated according to the historical data, and the quality stability time zone is divided according to the error value, wherein the divided time zone has a stability evaluation value identifier; the collection time of the feedback monitoring data is used to align with the time of the divided time zone, the stability evaluation value of the feedback monitoring data is identified, and the time series stability evaluation result is obtained.

[0043] For each data source, the error value of its historical data is calculated. The error value refers to the deviation between the actual observation value and the predicted value or expected value. For example, in an autonomous driving system, the error value of GPS positioning data can be obtained by calculating the difference between the actual GPS position and the expected position. Based on these error values, the system analyzes the stability of the data. If the error value fluctuates greatly, it means that the stability of the data source is poor; if the error value is small and the fluctuation is stable, it means that the data source is more reliable.

[0044] According to the change of error value, historical data is divided into different time zones. Each time zone represents the stability level of the data source during that period of time. For example, data with small error value and stable fluctuation can be divided into one time zone, and data with large error value and unstable fluctuation can be divided into another time zone. Each time zone is marked with a stability evaluation value, which indicates the stability of the data source in the time zone. For example, a time zone with high stability may get an evaluation with an identification value of "1", indicating that the data in the time zone is very stable; while an unstable time zone may get an evaluation with an identification value of "0".

[0045] Feedback monitoring data is usually collected at different time points. The timestamp of the real-time collected feedback data is compared with the previously divided time zone to determine which time zone the currently collected feedback data belongs to. For example, if the real-time feedback data of a data source is collected in time zone 2 of the historical data, the data will be compared with the data stability of time zone 2.

[0046] When the collection time is aligned, the stability of the current feedback data is evaluated based on the stability evaluation value of the time zone. If the feedback data is located in a time zone with high stability, it means that its data source performs stably in the time zone; if the feedback data is located in an unstable time zone, a lower evaluation value will be obtained.

[0047] Finally, by aligning and analyzing the feedback data and the time zone stability evaluation values, the time series stability evaluation results of the feedback data are obtained. This result reflects the temporal stability level of the feedback data. For example, if the data source is very stable during a certain period of history, and the real-time data is aligned with this time period, then the time series stability evaluation result is high, indicating that the data source is reliable, otherwise the stability is low.

[0048] Furthermore, it also includes: When the sliding mode control system includes multiple sub-sliding mode controllers, a collaborative control relationship among the multiple sub-sliding mode controllers is obtained; the consistency requirements of interconnected data between the sub-sliding mode controllers are analyzed according to the collaborative control relationship, and feedback delay constraint analysis is performed based on the consistency requirements obtained by the analysis to obtain interconnected data delay compensation constraints; and feedback time control is performed on the interconnected data of each sub-sliding mode controller according to the interconnected data delay compensation constraints.

[0049] Define the collaborative control relationship between each sub-sliding mode controller, which involves multiple sub-controllers jointly implementing a complex task. For example, multiple controllers are responsible for different control variables, such as position, speed, temperature, etc., but they need to coordinate together to ensure the consistency and performance of the entire system. The relationship between different sub-controllers may be independent of each other, or there may be data dependencies. For example, the output of a sub-controller may become the input of another sub-controller. Clarifying these dependencies will help with subsequent delay compensation and data consistency analysis.

[0050] The data consistency requirements between sub-sliding mode controllers include state consistency, control consistency, and time consistency of data exchange between controllers. For example, different sub-controllers may need to obtain the same data at a specific time, or their data need to meet certain timing requirements. Since the data of multiple sub-controllers need to be exchanged through a communication network or bus, delays (such as data transmission delays) may lead to data inconsistency. The impact of such delays on the performance of the control system is evaluated through delay constraint analysis. For example, data transmission delays may cause controller response lags, affecting the stability and response speed of the system. Based on delay analysis, a suitable delay compensation strategy is calculated to ensure data consistency. The compensation strategy includes adjusting the controller update frequency, adding a buffer mechanism, or using technical means such as timestamp synchronization to reduce data inconsistency caused by delays.

[0051] Based on the delay compensation constraint obtained, the data feedback time between each sub-sliding mode controller is adjusted. Specifically, by setting an appropriate feedback time window, it is ensured that each sub-controller can receive the updated data of other sub-controllers in time and make reasonable adjustments to its feedback. If the data delay of some sub-controllers is large, the time of its output feedback can be delayed, or interpolation algorithms and other technologies can be used to compensate for the delay. Through reasonable time control, the coordination of each sub-controller is optimized to ensure that the output of each sub-controller is fed back to other controllers at the right time point to avoid mismatch or outdated data caused by delay. This process effectively improves the coordination between multiple sub-controllers and ensures that the system can maintain good stability and dynamic response when multiple sub-controllers work together.

[0052] Furthermore, the interconnection data consistency requirement analysis between the sub-sliding mode controllers is performed according to the collaborative control relationship, including: Demand analysis is performed from the perspective of state consistency to determine the state variable alignment requirements between sub-sliding mode controllers; demand analysis is performed from the perspective of control consistency to determine the input and output coupling requirements between sub-sliding mode controllers; demand analysis is performed from the perspective of time consistency to determine the exchange time consistency requirements for interconnected data.

[0053] Analyze and determine the state consistency requirements between each sub-sliding mode controller to ensure that their state variables can be aligned during the control process. Specifically, define the state variables, that is, each sub-controller is responsible for controlling specific variables, such as speed, position, temperature, etc. The state variables reflect the current state of the control system and are the core input of the control algorithm. When multiple sub-controllers work together, it is necessary to ensure that the state variables of each controller can be synchronized and coordinated. For example, the output of sub-controller A may be the input of sub-controller B. If the state variables of A and B are not synchronized, B's control decision may be based on inaccurate state settings, thereby affecting the performance of the entire system. At this time, it is necessary to ensure that the state variables of all relevant sub-controllers have consistent values ​​at the same time. For example, if there is a time lag between the state variable of one sub-controller and the input variable of another sub-controller, they must be aligned through timing synchronization or time delay compensation.

[0054] Analyze and determine the control consistency requirements between each sub-sliding mode controller, especially the input and output coupling requirements between them. Specifically, in a multi-sub-controller system, the output of a sub-controller is often used as the input of other sub-controllers. This requires the system to clearly define the coupling relationship between the input and output of each sub-controller. For example, the output of sub-controller A may be used as the input of sub-controller B. In order to ensure the correctness of the control, B needs to obtain the output of A. When there is data coupling between controllers, the output data of A needs to be transmitted to B in a timely manner and must remain accurate. If there is a delay or error in the data of A obtained by B, it may lead to inaccurate control decisions, thereby affecting the stability of the system. Based on the requirements of input and output coupling, it is necessary to ensure that the input and output data between controllers are updated in a timely manner and remain accurate. Especially in real-time control, data delays and loss will lead to increased errors in the control system.

[0055] Analyze and determine the timing requirements for data exchange between each sub-controller to ensure the time consistency of the data. Specifically, when multiple sub-controllers need to exchange data, it is necessary to ensure that the time points of data exchange are consistent. If the data exchange between controllers is not synchronized, it may cause delays or distortion when the data is used. Through analysis, define the specific time points for data exchange. For example, sub-controller A may send data to sub-controller B at a certain moment, and sub-controller B must make control decisions as soon as possible after receiving the data. Any delay or timing inconsistency may affect the response speed of the controller and even cause the control system to be unstable. In order to meet the time consistency requirements, synchronization mechanisms such as timestamps, data caching or coordinated control strategies can be used. These methods ensure that data exchanges between different sub-controllers follow a unified time frame, thereby avoiding control errors caused by time deviations.

[0056] In general, through the above process, the coordination requirements between multi-sub sliding mode controllers are analyzed from three aspects: state consistency, control consistency, and time consistency. Among them, state consistency ensures that the state variables of each sub-controller are aligned in time to avoid erroneous control caused by inconsistent data; control consistency analyzes the coupling relationship between input and output to ensure that the input data of the sub-controller is timely and accurate; time consistency ensures the synchronization of data exchange between controllers, thereby avoiding control problems caused by time delays. The purpose of these steps is to ensure that the sub-controllers in the sliding mode control system can work in coordination and maintain consistency in data exchange and feedback control, thereby improving the robustness and stability of the entire system.

[0057] In summary, the feedback adjustment method of the sliding mode control system under data interconnection provided by the embodiment of the present application has the following technical effects: By obtaining feedback monitoring data and performing reliability analysis and evaluation on the data source, the reliability of the data source can be evaluated based on the quality evaluation parameters and historical data of the data source. This process ensures that the data is screened based on its reliability when used, thereby improving the accuracy of the data source. This method can effectively exclude unreliable or low-quality data sources and avoid their negative impact on system control; by using the preset screening threshold to screen the reliability coefficient of the data source, an effective data source is obtained. This screening process ensures that only reliable data sources are selected into the control system, which helps to reduce the impact of noise data on the control system. The dynamic screening mechanism enables the system to select the best input data in real time according to changes in the data source, thereby enhancing the flexibility and adaptability of the system; the feedback loop is set based on the reliability coefficient of the effective data source. The adaptive adjustment of the feedback loop can dynamically change the influence of each data source in the control system according to its quality and reliability. , reliable data sources will have a greater impact on the control system, while the impact of unreliable data sources is effectively reduced. This process ensures that the control system can maintain stable and precise control when the quality of the data source fluctuates; by establishing a control update relationship between the effective data source and the sliding surface, and returning the feedback loop to the sliding mode control system, it is ensured that the system can adjust the sliding surface according to the real-time effective data. This control update mechanism allows the system to dynamically adjust according to the actual data quality, optimize the controller response and stability, and improve the overall performance of the control system; adjust the sliding surface gain according to the feedback loop ratio to generate the control law for control adjustment. This step dynamically adjusts the sliding surface gain so that the control system can maintain a high robustness when facing data sources of different quality, enhances the system's adaptability to changes in data quality, and ensures that the system can make accurate control decisions even in the event of sensor failure or data fluctuations.

[0058] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A feedback adjustment method for a sliding mode control system under data interconnection, characterized in that: The method comprises: Acquire feedback monitoring data, perform reliability analysis and evaluation on the data source based on the quality evaluation parameters and historical data of the data source corresponding to the feedback monitoring data, and determine the reliability coefficient of the data source; Using a preset screening threshold to screen and set the reliability coefficient of the data source to obtain a valid data source; According to the data source reliability coefficient of the effective data source, feedback loop setting is performed; Establishing a control update relationship between the effective data source and the sliding surface, and feeding back the effective data source and its data source feedback loop to the sliding mode control system; The sliding mode surface gain is adjusted according to the feedback loop ratio, and a control law is generated to control and adjust the sliding mode control system.

2. The feedback adjustment method of the sliding mode control system under data interconnection according to claim 1 is characterized in that: The obtaining of a valid data source further comprises: Determining whether the valid data source meets the control requirements; When the control requirement is not met, data compensation is performed according to the effective data source, the compensation data is added to the effective data source, and a reliability coefficient of the compensation data source is set according to the compensation.

3. The feedback adjustment method of the sliding mode control system under data interconnection according to claim 1 is characterized in that: Also includes: Setting a multi-level screening threshold, wherein the multi-level screening threshold is a multi-level credible interval for screening data sources; Using the multi-level credible intervals to perform multi-level matching and identification on the reliability coefficient of the data source to obtain the effective level of the data source; A priority factor is configured according to the validity level of the data source, and the priority factor is used to correct the feedback loop ratio, wherein a higher validity level corresponds to a larger priority factor.

4. The feedback adjustment method of the sliding mode control system under data interconnection according to claim 3 is characterized in that: According to the data source reliability coefficient of the effective data source, feedback loop setting is performed, including: The feedback loop calculation formula is: , calculate the feedback ratio, where is the feedback ratio of the i-th data source, is the reliability coefficient of the ith data source, is a collection of valid data sources; The feedback loop ratio is corrected using the priority factor, and the correction expression of the feedback loop ratio calculation formula is: , k is the priority factor, is the priority factor of the i-th data source.

5. The feedback adjustment method of the sliding mode control system under data interconnection according to claim 1 is characterized in that: Based on the quality evaluation parameters of the data source corresponding to the feedback monitoring data and historical data, the reliability analysis and evaluation of the data source is performed to determine the reliability coefficient of the data source, including: According to the quality evaluation parameters of the data source, each parameter is evaluated respectively to obtain the evaluation value of the multi-dimensional quality evaluation parameter; Obtain the impact of each dimension's quality evaluation parameters on the reliability of the data source and set real-time evaluation weights; Performing weighted calculation on the evaluation values ​​of the multi-dimensional quality evaluation parameters based on the real-time evaluation weights to obtain a quality evaluation result; Perform reliability and stability evaluation of the data source based on the historical data to obtain a time series stability evaluation result; A comprehensive evaluation is performed based on the quality evaluation results and the time series stability evaluation results to determine the data source reliability coefficient.

6. The feedback adjustment method of the sliding mode control system under data interconnection according to claim 5 is characterized in that: The reliability and stability of the data source is evaluated based on the historical data to obtain a time series stability evaluation result, including: Calculating the error value of the data source according to the historical data, and dividing the quality stability time zone according to the error value, wherein the divided time zone has a stability evaluation value mark; The collection time of the feedback monitoring data is used to align with the time of the divided time zone, identify the stability evaluation value of the feedback monitoring data, and obtain the time series stability evaluation result.

7. The feedback adjustment method of the sliding mode control system under data interconnection according to claim 1 is characterized in that: Also includes: When the sliding mode control system includes a plurality of sub-sliding mode controllers, obtaining a coordinated control relationship among the plurality of sub-sliding mode controllers; According to the collaborative control relationship, the interconnection data consistency requirements between the sub-sliding mode controllers are analyzed, and the feedback delay constraint analysis is performed based on the consistency requirements obtained by the analysis to obtain the interconnection data delay compensation constraint; According to the interconnection data delay compensation constraint, feedback time control is performed on the interconnection data of each sub-sliding mode controller.

8. The feedback adjustment method of the sliding mode control system under data interconnection according to claim 7 is characterized in that: According to the collaborative control relationship, the interconnection data consistency requirements between the sub-sliding mode controllers are analyzed, including: Analyze the requirements from the state consistency and determine the state variable alignment requirements between the sub-sliding mode controllers; Analyze requirements from the perspective of control consistency and determine the coupling requirements between input and output of sub-sliding mode controllers; Analyze the requirements from the perspective of time consistency and determine the exchange time consistency requirements for interconnected data.

Images