Full-performance automatic detection process optimization method and system

Abstract

The invention discloses a full-performance automatic detection process optimization method and system, and relates to the technical field of automatic detection optimization, and the method comprises the steps: distributing the task weight of each data collection unit of a detection system through a multi-layer adaptive interactive data collection optimization algorithm; dynamically adjusting the acquisition frequency of the acquisition unit according to the adjusted weight value; constructing a state vector of the system by using a recursive hierarchical feedback and dynamic prediction scheduling algorithm, and generating a feedback signal; and dynamically adjusting the priority and the task allocation quantity of each task based on the feedback signal. Through adaptive task allocation and dynamic acquisition frequency adjustment, the data acquisition efficiency is effectively improved, the problems of data loss and delay in a traditional method are solved, and reasonable allocation and high acquisition precision of system resources are ensured. The task priority is further optimized through a recursive hierarchical feedback and dynamic prediction scheduling algorithm, the load peak pressure is relieved, and the process continuity is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automated detection optimization, and in particular to a method and system for optimizing a full-performance automated detection process. Background Art

[0002] In modern industrial production, logistics management, intelligent detection and other automation systems, the requirements for detection tasks are becoming increasingly diverse and complex. The detection system not only needs to quickly process a large number of detection tasks, but also needs to maintain high precision and high stability under complex conditions. With the improvement of the level of industrial automation, various production detection tasks require the system to have flexible task scheduling capabilities and high adaptability to adapt to the ever-changing production environment and real-time detection needs. In recent years, the application of data acquisition and task scheduling technology in the field of automated detection has gradually increased, but due to the diversity of data source types, different task priorities and dynamic changes in load, the existing detection system still needs to be further improved and optimized in terms of multi-task parallel processing, real-time acquisition frequency adjustment and dynamic scheduling of task priorities.

[0003] Especially in multi-level automated detection systems, the rapid changes in real-time priorities and task resource requirements of different detection tasks pose great challenges to the detection system. The acquisition unit in the detection system must not only meet the real-time data collection under high-load tasks, but also dynamically adjust the acquisition frequency and task priority to ensure high accuracy and efficiency of detection in complex scenarios. In addition, automated detection systems usually need to provide feedback and optimize the system status while completing the current task to further improve the system's resource utilization efficiency and operational stability. This requirement requires the detection system to have a multi-level feedback mechanism and recursive control capabilities, so that the system can optimize itself through the feedback mechanism while executing the detection task.

[0004] There are at least the following technical problems in the existing technology: the existing technology is unable to effectively allocate resources when the task volume increases or the load changes, which makes the detection process prone to congestion or even bottlenecks when processing complex data, thereby affecting the detection efficiency and the accuracy of the detection results; it is unable to achieve real-time priority adjustment of tasks at all levels, which easily leads to waste of resources or task backlogs, causing the stability of the system in high-load or multi-task scenarios to be greatly reduced; it is unable to optimize the process autonomously and lacks adaptability to complex environments, thereby affecting the stability and automation level of the detection system. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is: how to solve the technical problems existing in the prior art, such as unreasonable resource allocation, delayed task priority adjustment, low detection efficiency and insufficient adaptive capability.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for optimizing the full performance automated detection process, which comprises the following steps:

[0008] Through the multi-layer adaptive interactive data acquisition optimization algorithm, the task weight of each data acquisition unit of the detection system is allocated;

[0009] According to the adjusted weight value, dynamically adjust the collection frequency of the collection unit;

[0010] Use recursive hierarchical feedback and dynamic predictive scheduling algorithms to build the system's state vector and generate feedback signals;

[0011] Dynamically adjust the priority and allocation of each task based on feedback signals.

[0012] As a preferred solution of the full-performance automated detection process optimization method described in the present invention, wherein: the task weights of each data acquisition unit of the detection system are allocated through a multi-layer adaptive interactive data acquisition optimization algorithm, including:

[0013] By allocating tasks and adjusting priorities of each data acquisition unit in the full-performance automated detection system, the initial acquisition weight of each unit is determined, and the initial task weight of each acquisition unit is set as a multidimensional matrix The weight of the jth unit in the i-th layer is expressed as:

[0014]

[0015] in, represents the initial weight value of the jth unit in the i-th layer; R i is the data collection priority of the i-th layer, indicating the priority of the i-th layer in the overall collection task; T ij is the acquisition time allocation ratio, which indicates the time ratio of task execution in the current acquisition layer and unit; M is the total number of layers in the system.

[0016] As a preferred solution of the full-performance automated detection process optimization method described in the present invention, wherein:: dynamically adjusting the acquisition frequency of the acquisition unit according to the adjusted weight value includes:

[0017] As the data collection task progresses, each layer of collection units dynamically adjusts the weight value according to the real-time feedback to meet the dynamic needs of the detection task. The layers interact with each other to form a feedback correction matrix, which is expressed as:

[0018]

[0019] in, Represents the weight matrix in the current nth acquisition task; is the amount of feedback data from the kth unit in the i-th layer in the n-th acquisition, θ ij and θ ik are the phases of the jth and kth units in the i-th layer, γ is the feedback correction coefficient, is the time derivative of the task weight, and N is the total number of units in the i-th layer.

[0020] As a preferred solution of the method for optimizing the full-performance automated detection process described in the present invention, the dynamically adjusting the acquisition frequency of the acquisition unit according to the adjusted weight value also includes:

[0021] After completing the feedback correction of the weight, each acquisition unit adjusts the acquisition frequency. The frequency adjustment is based on the corrected weight matrix and the current task load, and the expression is:

[0022]

[0023] Among them, f ij (t) is the real-time acquisition frequency of the jth unit in the i-th layer, f base is the system's base acquisition frequency, δ and is the adjustment coefficient, and the task load value P of the jth unit in the i-th layer at time t ij (t) and the average load value P avg The difference of (t).

[0024] As a preferred solution of the full performance automatic detection process optimization method described in the present invention, wherein: the state vector of the system is constructed by using recursive hierarchical feedback and dynamic prediction scheduling algorithm to generate feedback signals, including:

[0025] The real-time task completion rate, task queue length, processing capacity and bottleneck prediction coefficient constitute the input state vector of the recursive hierarchical feedback algorithm. The state vector serves as the basis of the recursive hierarchical feedback. After feedback calculation, the feedback signal is generated to complete dynamic priority scheduling and task allocation.

[0026] Define the state vector for each layer, the expression is:

[0027] S i (t) = [C i (t),Q i (t),P i (t),Λ i (t)])

[0028] Among them, C i (t) represents the task completion rate, Q i (t) represents the length of the task queue, P i(t) represents the processing capacity of the current layer, Λ i (t) is the bottleneck prediction coefficient;

[0029] The state vector monitors the operating status of each layer, and the expression is:

[0030]

[0031] Among them, η is the adjustment coefficient of the completion rate, is the time decay factor;

[0032] The recursive hierarchical feedback function integrates the state vectors of different levels to form a system feedback signal to adjust the priority of the current layer. The recursive hierarchical feedback function generates a function for the priority of each layer of tasks based on the derivative and integral operations of the state vector. The expression is:

[0033]

[0034] Among them, F i (t) is the feedback signal of the i-th layer at time t, It is The state vector of the layer, τ is the time-integrated variable.

[0035] As a preferred solution of the full-performance automated detection process optimization method described in the present invention, the method of dynamically adjusting the priority and task allocation of each task based on the feedback signal includes:

[0036] After the recursive feedback signal is generated, the next priority will be dynamically scheduled in combination with the dynamic prediction scheduling model. The next task priority is calculated based on the feedback signal and the bottleneck prediction coefficient. The expression is:

[0037]

[0038] in, is the priority of the next step of the i-th layer, σ is the adjustment coefficient of the control priority, μ is the feedback intensity adjustment coefficient, is the bottleneck adjustment coefficient, is the state adjustment coefficient;

[0039] By nonlinearly activating the feedback signal and combining the bottleneck prediction coefficient to schedule the next step priority, the appropriate priority adjustment of the task under different loads can be completed.

[0040] As a preferred solution of the full-performance automated detection process optimization method described in the present invention, the method of dynamically adjusting the priority and task allocation of each task based on the feedback signal also includes:

[0041] After the priority is determined, the real-time load is balanced by dynamically adjusting the task allocation. The task allocation is based on the priority and bottleneck prediction, and the expression is:

[0042]

[0043] Among them, A i (t) is the task allocation of the i-th layer at time t, A base is the task baseline allocation, θ is the task adjustment coefficient, is the coefficient used to control the adjustment of attenuation;

[0044] When the task allocation A of each layer is calculated i (t), the assigned value is applied to the data acquisition units and detection modules of each layer.

[0045] If the task allocation of a certain layer is lower than the preset threshold, the number of collection unit activities of the current layer will be reduced and the collection frequency will be reduced, otherwise it will be increased;

[0046] After each layer of task is completed, the generated data is immediately fed back to the control center to evaluate the detection accuracy and efficiency in real time;

[0047] Periodically review task allocation based on test data and task completion status to form a new round of task allocation strategy and optimize the allocation amount.

[0048] Another object of the present invention is to provide a full-performance automated detection process optimization system, which can achieve efficient allocation of detection tasks and dynamic optimization of resources through multi-layer adaptive interactive data acquisition optimization, weight feedback dynamic adjustment, recursive hierarchical feedback mechanism and dynamic task priority scheduling, solving the problems of unreasonable resource allocation, inability to adjust task priorities in real time, unbalanced system load and insufficient adaptability in the prior art.

[0049] In order to solve the above technical problems, the present invention provides the following technical solutions: a full-performance automated detection process optimization system, comprising: a data acquisition optimization module, a weight feedback and acquisition frequency dynamic adjustment module, a detection module, a dynamic scheduling and task allocation module;

[0050] The data acquisition optimization module allocates and optimizes the task weights of data acquisition units through a multi-layer adaptive interactive data acquisition optimization algorithm. According to the acquisition level and task priority, the task weights are initially set and a multi-dimensional matrix is ​​established to optimize the task importance of each data acquisition unit, and the acquisition time is allocated based on the task allocation weights.

[0051] The weight feedback and acquisition frequency dynamic adjustment module dynamically corrects the weight value of the data acquisition unit according to the task feedback and adjusts the acquisition frequency of each unit;

[0052] Collect real-time feedback data, generate feedback correction matrix, dynamically adjust weight value, ensure weight matches task execution, dynamically adjust collection frequency of each data collection unit according to the corrected weight value and current task load, and balance system resource usage;

[0053] The detection module performs specific detection tasks and feeds back the detection data to the system control center in real time to form a data closed loop;

[0054] Receive tasks assigned by the system, complete data collection and detection, generate detection results, and feed back the detection results and task execution status to the control center in real time. After completing the task, the detection module evaluates the detection accuracy and execution efficiency of the system to form an optimization basis;

[0055] The dynamic scheduling and task allocation module dynamically adjusts the system's task priority and task allocation amount through recursive hierarchical feedback and dynamic prediction scheduling algorithm to optimize the task scheduling strategy;

[0056] Feedback signals are generated using parameters such as task completion rate, queue length, processing capacity, and bottleneck prediction coefficient to reflect the system status. Based on the feedback signals, the priorities of tasks at each level are calculated, and the execution order of tasks is dynamically adjusted in combination with bottleneck prediction. The task volume is dynamically allocated according to task priority and load conditions to balance the real-time load of the system.

[0057] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for optimizing a full-performance automated detection process are implemented.

[0058] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for optimizing a full-performance automated detection process.

[0059] The beneficial effects of the present invention are as follows: by introducing an adaptive interactive task allocation algorithm between multiple layers of data acquisition units, the acquisition frequency and task weight of each acquisition unit can be dynamically adjusted to adapt to different loads and real-time task requirements. This dynamic adjustment mechanism effectively improves data acquisition efficiency, ensures reasonable allocation of system resources, and avoids data loss and acquisition delay problems in traditional methods. At the same time, the acquisition frequency is adaptively adjusted to ensure that each acquisition unit can maintain high acquisition accuracy under different load conditions, thereby improving the data accuracy of the system.

[0060] Through the closed-loop control mechanism of task allocation and feedback regulation, the system can adaptively adjust the task allocation of each layer according to the real-time feedback, forming a dynamic optimization of task allocation. During the detection task, the system redistributes the task priority according to the real-time completion rate, task queue length and processing capacity of each layer, ensuring that resources are tilted towards key detection tasks and achieving load balancing between different task layers. This mechanism effectively solves the bottleneck problem of traditional systems under load imbalance, and significantly improves the system's working stability and load scheduling efficiency in complex environments.

[0061] The present invention forms a multi-layer data interactive feedback and priority control mechanism through a recursive hierarchical feedback algorithm and a dynamic prediction scheduling algorithm, which can timely detect the bottleneck conditions of each layer and adjust the priority. The system uses the dual functions of real-time monitoring and feedback signals to dynamically predict the bottleneck possibility of tasks at each layer, and adjust the task scheduling order according to the bottleneck prediction results, thereby reducing the pressure during peak load periods and further optimizing the detection process. By predicting and regulating the system bottleneck state in advance, the present invention effectively improves the system processing capacity, ensures the smooth connection of the detection process between tasks at each layer, and ensures the continuity and real-time nature of the detection tasks.

[0062] Through multi-dimensional state vector calculation and recursive feedback algorithm, the present invention realizes intelligent control of the system, and each layer of tasks can be automatically adjusted based on dynamic feedback. Task allocation will be automatically reviewed after completing a round of detection. The system will automatically generate a new round of task strategies based on the completion rate of each layer of tasks, detection accuracy and system load conditions to achieve task allocation optimization within the detection cycle. This intelligent mechanism makes the system adaptive. Even in complex or sudden load environments, the system can still adjust the allocation of detection resources in real time, reduce manual intervention, and improve detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0064] Figure 1 An overall flow chart of a method for optimizing a full-performance automated detection process provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0066] Example 1, reference Figure 1 As an embodiment of the present invention, a method for optimizing a full-performance automated detection process is provided, comprising:

[0067] S1. Through the multi-layer adaptive interactive data acquisition optimization algorithm, the task allocation of each data acquisition unit in the system is ensured to be rational. Through the setting of the initial task weight matrix and the weight correction based on feedback, the real-time acquisition frequency is adaptively adjusted, and the multi-dimensional task operation data in the detection system is collected according to the real-time acquisition frequency;

[0068] In order to solve the technical problems of existing automated detection systems in data acquisition efficiency, real-time processing capabilities and task scheduling accuracy, a multi-layer adaptive interactive data acquisition optimization algorithm and a recursive hierarchical feedback and dynamic prediction scheduling algorithm were introduced to achieve closed-loop control and adaptive optimization of data acquisition and task scheduling, ensuring that the full-performance automated detection system can operate efficiently and stably under various load conditions.

[0069] By allocating tasks and adjusting priorities of each data acquisition unit in the full-performance automated detection system, the initial acquisition weight of each unit is determined to meet the needs of the current detection task. In order to achieve reasonable allocation of acquisition resources, the system sets the initial task weight of each acquisition unit as a multidimensional matrix That is, the weight of the jth unit in the i-th layer. The weight calculation formula is as follows:

[0070]

[0071] in, represents the initial weight value of the jth unit in the i-th layer; R i is the data collection priority of the i-th layer, indicating the priority of the i-th layer in the overall collection task; T ij is the proportion of collection time allocation, indicating the time proportion of task execution in the current collection layer and unit; M is the total number of layers in the system. The adjustment coefficients α and β are used to control the initial range and offset of the weight to adapt to the different load requirements of the collection task. By combining the calculation of task requirements and collection priority, the system can reasonably allocate the initial collection weight of each unit, achieve uniform distribution of task load when the system starts, and avoid data loss or overload due to unreasonable distribution of collection tasks.

[0072] As the data collection task progresses, each layer of collection units gradually corrects the weights based on real-time feedback to adapt to the dynamic needs of the detection task; each layer interacts with the feedback data to form a feedback correction matrix, thereby dynamically adjusting the weight value. The correction process of this step is based on the following formula:

[0073]

[0074] in, Represents the weight matrix in the current nth acquisition task; is the amount of feedback data from the kth unit in the i-th layer in the n-th acquisition, dynamically reflecting the status of the current task execution; θ ij and θ ik are the phases of the jth and kth units in the i-th layer, respectively, and are used to calculate the phase difference between adjacent units; γ is the feedback correction coefficient, which is used to control the strength of the feedback correction; is the time derivative of the task weight, indicating the rate of change of the weight over time; N is the total number of units in the i-th layer. Through feedback interaction, the weight of each acquisition unit can be adjusted in real time with the task, so that the system can adapt to the changing data volume and different task requirements. Weight correction makes the data acquisition load more balanced, which helps to improve the acquisition efficiency of the overall system and avoid data loss caused by overload or insufficient acquisition of a certain layer.

[0075] After completing the feedback correction of the weights, each acquisition unit also needs to adjust its acquisition frequency to enable efficient data acquisition under real-time task load. The frequency adjustment is based on the corrected weight matrix and the current task load, and is achieved through the following nonlinear formula:

[0076]

[0077] Among them, f ij (t) is the real-time acquisition frequency of the jth unit in the i-th layer, which ensures that different acquisition units can complete the acquisition task at an adaptive speed; f base is the system's benchmark acquisition frequency, which is used to define the benchmark acquisition speed; δ and κ are adjustment coefficients that control the fluctuation amplitude of the acquisition frequency, based on the modified weight matrix; the task load value P of the jth unit in the i-th layer at time t ij (t) and the average load value P avg The difference of (t) reflects the pressure level of the unit task. By calculating the real-time acquisition frequency, the dynamic adaptive adjustment of the data acquisition frequency is realized, ensuring that the acquisition unit can adapt to the real-time load and achieve efficient and balanced data acquisition.

[0078] The multi-dimensional task operation data in the detection system is collected according to the real-time collection frequency, including the current completion rate of each task (used to determine the real-time progress of the task), the length of each layer of task queue (indicating the system load and the number of waiting tasks), the current processing capacity of the system (reflecting the resource usage of the system in task execution, that is, the current processing efficiency of the system), and the bottleneck prediction coefficient of each layer (used to identify potential performance bottlenecks of the system in advance).

[0079] S2. The collected data is used for recursive hierarchical feedback and dynamic prediction scheduling algorithms to construct the state vector of each layer and generate feedback signals. According to the feedback signals and bottleneck predictions, the priority and task allocation of each layer are dynamically adjusted. By dynamically adjusting the task allocation, the real-time load of the system is balanced and the system detection process is optimized.

[0080] The collected data is used for subsequent recursive hierarchical feedback and dynamic prediction scheduling algorithms. The real-time task completion rate, task queue length, processing capacity and bottleneck prediction coefficient will constitute the input state vector of the recursive hierarchical feedback algorithm. The state vector serves as the basis of recursive hierarchical feedback. After feedback calculation, a feedback signal is generated to further guide dynamic priority scheduling and task allocation.

[0081] According to the feedback results of data collection, the detection process is optimized in real time to ensure that the system's detection tasks can run efficiently under different load conditions. First, the state vector S is defined for each layer. i (t) = [C i (t),Q i (t),P i (t),Λ i (t)]), where C i (t) represents the task completion rate, Q i (t) represents the length of the task queue, P i (t) represents the processing capacity of the current layer, Λ i (t) is the bottleneck prediction coefficient. The state vector is used to monitor the operating status of each layer and synthesize various parameters through the following formula:

[0082]

[0083] Among them, η is the adjustment coefficient of the completion rate, which controls the completion rate weight of the state vector; is the time decay factor. The state vector calculation formula calculates the ratio of completion rate to processing capacity, and combines the logarithmic function to describe the impact of the bottleneck coefficient under dynamic conditions, and adds the time decay factor in the time dimension to reflect the weakening of the past state on the current scheduling. Through the calculation of the state vector, the system can capture the current task completion status and the possibility of future bottlenecks from multiple dimensions, providing effective input for the feedback algorithm.

[0084] The recursive hierarchical feedback function is used to integrate the state vectors of different levels to form a system feedback signal and adjust the priority of the current layer. The recursive hierarchical feedback function is implemented based on the derivative and integral operations of the state vector, and the priority generation function for each layer of tasks is as follows:

[0085]

[0086] Among them, F i (t) is the feedback signal of the i-th layer at time t. The feedback signal is generated by the differential and cumulative feedback signals of the state changes of each layer, including the feedback contributions of all layers from the 1st layer to the i-th layer, which is used to adjust the task priority of each layer; It is The state vector of the layer, The recursive feedback coefficient ξ is used to control the recursive feedback strength to enhance or weaken the cumulative effect of recursive feedback; τ is the time integral variable. The recursive feedback function not only contains the information of the current level, but also accumulates the results of the upper level feedback to form a real-time dynamic system feedback, so that each layer can integrate the historical state and current data changes, effectively deal with the workload between different levels, and ensure the balance of system scheduling.

[0087] After the recursive feedback signal is generated, the system will dynamically schedule the next priority in combination with the dynamic prediction scheduling model, and calculate the next task priority based on the feedback signal and the bottleneck prediction coefficient. The formula is as follows:

[0088]

[0089] in, is the priority of the next step of the i-th layer, indicating the priority weight of the next scheduled task; σ is the adjustment coefficient of the control priority; μ is the feedback strength adjustment coefficient, which is used to adjust the sensitivity of the feedback signal; is the bottleneck adjustment coefficient, which controls the impact of the bottleneck coefficient on the priority; It is the state adjustment coefficient, which controls the weight of the state vector on the bottleneck. By nonlinearly activating the feedback signal and combining the bottleneck prediction coefficient to schedule the next step priority, the output next step priority has adaptive capabilities under the current load of each layer. By integrating the feedback signal and the bottleneck prediction, it ensures that the task can get appropriate priority adjustment under different loads.

[0090] After the priority is determined, the system balances the real-time load of the system by dynamically adjusting the task allocation. The task allocation is based on the priority and bottleneck prediction and is defined as:

[0091]

[0092] Among them, A i(t) is the amount of tasks assigned to the i-th layer at time t; A base is the task baseline allocation; θ is the task adjustment coefficient; is the coefficient used to control the adjustment of attenuation.

[0093] When the task allocation A of each layer is calculated i (t), the system applies the allocation value to the data collection units and detection modules of each layer. Each layer starts or pauses the corresponding collection unit according to the size of the task allocation, and adjusts the detection frequency and collection depth to ensure that the workload of each unit meets the overall needs of the current system. For example, if the task allocation of a certain layer is lower than the preset threshold, the system will reduce the number of collection unit activities of this layer and reduce the collection frequency to ensure that system resources are concentrated on detection tasks with higher priority, so that task allocation not only depends on the initial allocation, but also makes detailed adjustments according to the real-time completion of the detection task and the load conditions of each layer.

[0094] After each layer of tasks is completed, the generated data will be immediately fed back to the system control center for real-time evaluation of the detection accuracy and efficiency of the entire system. The system will periodically review the task allocation based on the detection data and task completion status, form a new round of task allocation strategy, and further optimize the allocation amount to ensure that the task allocation of the next detection cycle is more accurate and efficient.

[0095] Embodiment 2 is an embodiment of the present invention, which provides a system for a full-performance automated detection process optimization method, including: a data acquisition optimization module, a weight feedback and acquisition frequency dynamic adjustment module, a detection module, and a dynamic scheduling and task allocation module;

[0096] The data acquisition optimization module allocates and optimizes the task weights of data acquisition units through a multi-layer adaptive interactive data acquisition optimization algorithm. According to the acquisition level and task priority, the task weights are initially set and a multi-dimensional matrix is ​​established to optimize the task importance of each data acquisition unit, and the acquisition time is allocated based on the task allocation weights.

[0097] The weight feedback and acquisition frequency dynamic adjustment module dynamically corrects the weight value of the data acquisition unit according to the task feedback and adjusts the acquisition frequency of each unit;

[0098] Collect real-time feedback data, generate feedback correction matrix, dynamically adjust weight value, ensure weight matches task execution, dynamically adjust collection frequency of each data collection unit according to the corrected weight value and current task load, and balance system resource usage;

[0099] The detection module performs specific detection tasks and feeds back the detection data to the system control center in real time to form a data closed loop;

[0100] Receive tasks assigned by the system, complete data collection and detection, generate detection results, and feed back the detection results and task execution status to the control center in real time. After completing the task, the detection module evaluates the detection accuracy and execution efficiency of the system to form an optimization basis;

[0101] The dynamic scheduling and task allocation module dynamically adjusts the system's task priority and task allocation amount through recursive hierarchical feedback and dynamic prediction scheduling algorithm to optimize the task scheduling strategy;

[0102] Feedback signals are generated using parameters such as task completion rate, queue length, processing capacity, and bottleneck prediction coefficient to reflect the system status. Based on the feedback signals, the priorities of tasks at each level are calculated, and the execution order of tasks is dynamically adjusted in combination with bottleneck prediction. The task volume is dynamically allocated according to task priority and load conditions to balance the real-time load of the system.

[0103] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0104] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0105] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0106] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0107] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for optimizing a full-performance automated detection process, characterized in that: include: Through the multi-layer adaptive interactive data acquisition optimization algorithm, the task weight of each data acquisition unit of the detection system is allocated; According to the adjusted weight value, dynamically adjust the collection frequency of the collection unit; Use recursive hierarchical feedback and dynamic predictive scheduling algorithms to build the system's state vector and generate feedback signals; Dynamically adjust the priority and allocation of each task based on feedback signals.

2. A method for optimizing a full-performance automated testing process as claimed in claim 1, characterized in that: The multi-layer adaptive interactive data acquisition optimization algorithm is used to allocate the task weights of each data acquisition unit of the detection system, including: By allocating tasks and adjusting priorities of each data acquisition unit in the full-performance automated detection system, the initial acquisition weight of each unit is determined, and the initial task weight of each acquisition unit is set as a multidimensional matrix The weight of the jth unit in the i-th layer is expressed as: in, represents the initial weight value of the jth unit in the i-th layer; R i is the data collection priority of the i-th layer, indicating the priority of the i-th layer in the overall collection task; T ij is the acquisition time allocation ratio, which indicates the time ratio of task execution in the current acquisition layer and unit; M is the total number of layers in the system.

3. A method for optimizing a full-performance automated testing process as claimed in claim 2, characterized in that: The dynamically adjusting the acquisition frequency of the acquisition unit according to the adjusted weight value includes: As the data collection task progresses, each layer of collection units dynamically adjusts the weight value according to the real-time feedback to meet the dynamic needs of the detection task. The layers interact with each other to form a feedback correction matrix, which is expressed as: in, Represents the weight matrix in the current nth acquisition task; is the amount of feedback data from the kth unit in the i-th layer in the n-th acquisition, θ ij and θ ik are the phases of the jth and kth units in the i-th layer, γ is the feedback correction coefficient, is the time derivative of the task weight, and N is the total number of units in the i-th layer.

4. A method for optimizing a full-performance automated testing process as claimed in claim 3, characterized in that: The dynamically adjusting the acquisition frequency of the acquisition unit according to the adjusted weight value also includes: After completing the feedback correction of the weight, each acquisition unit adjusts the acquisition frequency. The frequency adjustment is based on the corrected weight matrix and the current task load, and the expression is: Among them, f ij (t) is the real-time acquisition frequency of the jth unit in the i-th layer, f base is the system's baseline acquisition frequency, δ and κ are adjustment coefficients, and the task load value P of the jth unit in the i-th layer at time t ij (t) and the average load value P avg The difference of (t).

5. A method for optimizing a full-performance automated testing process as claimed in claim 4, characterized in that: The method of using recursive hierarchical feedback and dynamic prediction scheduling algorithm to construct the state vector of the system and generate feedback signals includes: The real-time task completion rate, task queue length, processing capacity and bottleneck prediction coefficient constitute the input state vector of the recursive hierarchical feedback algorithm. The state vector serves as the basis of the recursive hierarchical feedback. After feedback calculation, the feedback signal is generated to complete dynamic priority scheduling and task allocation. Define the state vector for each layer, the expression is: S i (t)=[C i (t),Q i (t),P i (t),Λ i (t)]) Among them, C i (t) represents the task completion rate, Q i (t) represents the length of the task queue, P i (t) represents the processing capacity of the current layer, Λ i (t) is the bottleneck prediction coefficient; The state vector monitors the operating status of each layer, and the expression is: Among them, η is the adjustment coefficient of the completion rate, is the time decay factor; The recursive hierarchical feedback function integrates the state vectors of different levels to form a system feedback signal to adjust the priority of the current layer. The recursive hierarchical feedback function generates a function for the priority of each layer of tasks based on the derivative and integral operations of the state vector. The expression is: Among them, F i (t) is the feedback signal of the i-th layer at time t, It is The state vector of the layer, τ is the time-integrated variable.

6. A method for optimizing a full-performance automated testing process as claimed in claim 5, characterized in that: The dynamically adjusting the priority of each task and the amount of task allocation based on the feedback signal includes: After the recursive feedback signal is generated, the next priority will be dynamically scheduled in combination with the dynamic prediction scheduling model. The next task priority is calculated based on the feedback signal and the bottleneck prediction coefficient. The expression is: in, is the priority of the next step of the i-th layer, σ is the adjustment coefficient of the control priority, μ is the feedback intensity adjustment coefficient, is the bottleneck adjustment coefficient, is the state adjustment coefficient; By nonlinearly activating the feedback signal and combining the bottleneck prediction coefficient to schedule the next step priority, the appropriate priority adjustment of the task under different loads can be completed.

7. A method for optimizing a full-performance automated testing process as claimed in claim 6, characterized in that: The dynamically adjusting the priority of each task and the amount of task allocation based on the feedback signal also includes: After the priority is determined, the real-time load is balanced by dynamically adjusting the task allocation. The task allocation is based on the priority and bottleneck prediction, and the expression is: Among them, A i (t) is the task allocation of the i-th layer at time t, A base is the task baseline allocation, θ is the task adjustment coefficient, is the coefficient used to control the adjustment of attenuation; When the task allocation A of each layer is calculated i (t), the assigned value is applied to the data acquisition units and detection modules of each layer. If the task allocation of a certain layer is lower than the preset threshold, the number of collection unit activities of the current layer will be reduced and the collection frequency will be reduced, otherwise it will be increased; After each layer of task is completed, the generated data is immediately fed back to the control center to evaluate the detection accuracy and efficiency in real time; Periodically review task allocation based on test data and task completion status to form a new round of task allocation strategy and optimize the allocation amount.

8. A system using a full performance automated testing process optimization method as claimed in any one of claims 1 to 7, characterized in that: It includes data collection optimization module, weight feedback and collection frequency dynamic adjustment module, detection module, dynamic scheduling and task allocation module; The data acquisition optimization module allocates and optimizes the task weights of data acquisition units through a multi-layer adaptive interactive data acquisition optimization algorithm. According to the acquisition level and task priority, the task weights are initially set and a multi-dimensional matrix is ​​established to optimize the task importance of each data acquisition unit, and the acquisition time is allocated based on the task allocation weights. The weight feedback and acquisition frequency dynamic adjustment module dynamically corrects the weight value of the data acquisition unit according to the task feedback and adjusts the acquisition frequency of each unit; Collect real-time feedback data, generate feedback correction matrix, dynamically adjust weight value, ensure weight matches task execution, dynamically adjust collection frequency of each data collection unit according to the corrected weight value and current task load, and balance system resource usage; The detection module performs specific detection tasks and feeds back the detection data to the system control center in real time to form a data closed loop; Receive tasks assigned by the system, complete data collection and detection, generate detection results, and feed back the detection results and task execution status to the control center in real time. After completing the task, the detection module evaluates the detection accuracy and execution efficiency of the system to form an optimization basis; The dynamic scheduling and task allocation module dynamically adjusts the system's task priority and task allocation amount through recursive hierarchical feedback and dynamic prediction scheduling algorithm to optimize the task scheduling strategy; Feedback signals are generated using parameters such as task completion rate, queue length, processing capacity, and bottleneck prediction coefficient to reflect the system status. Based on the feedback signals, the priorities of tasks at each level are calculated, and the execution order of tasks is dynamically adjusted in combination with bottleneck prediction. The task volume is dynamically allocated according to task priority and load conditions to balance the real-time load of the system.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a full-performance automated detection process optimization method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a full-performance automated detection process optimization method described in any one of claims 1 to 7 are implemented.

Images