A meter reading scheduling strategy simulation verification method and system based on configurable component and queuing theory fusion

Abstract

The application provides a meter reading scheduling strategy simulation verification method and system based on configurable components and queuing theory fusion, which comprises the following steps: configuring simulation scene parameters such as meter file, collection task, queuing theory model elements and the like; converting Excel configuration into executable test cases through a DSL parser; simulating the task arrival, queuing, service and discarding process based on an M / M / m / B queuing theory model, and calculating the carrier utilization rate rho, the average queue length Lq, the average waiting time Wq and the task loss probability Ploss in real time; capturing meter reading response data and queuing state data, drawing actual and theoretical collection success rate curves and superimposedly displaying them; and outputting the scheduling strategy advantage grade according to the actual collection success rate, the steady-state performance index and Ploss. The application realizes the convenient configuration and script automatic generation of a simulation scene, the real-time comparison of theoretical prediction and simulation verification, and the quantitative evaluation of a scheduling strategy.

Description

Technical Field

[0001] This invention relates to the field of electricity information acquisition simulation, and in particular to a simulation verification system and method for meter reading scheduling strategy based on the integration of configurable components and queuing theory. Background Technology

[0002] With the rapid advancement of smart grid construction and the continuous expansion of electricity information collection systems, the complexity of meter reading services is increasing. Traditional meter reading and scheduling strategies typically rely on verification in the actual field environment, that is, deploying concentrators, collectors, and smart meters in real distribution areas, and evaluating the meter reading success rate and the effectiveness of the scheduling algorithm through long-term field operation.

[0003] However, this verification method based on the actual environment has obvious shortcomings: First, field testing requires a lot of manpower and resources, has a long deployment cycle, and high testing costs; second, the field environment is uncontrollable and it is difficult to reproduce specific abnormal scenarios (such as communication signal interference, meter failure, server resource overload, concentrator offline, etc.), resulting in an insufficient performance evaluation of the meter reading scheduling strategy under extreme or abnormal conditions.

[0004] Existing simulation testing systems are mostly dedicated testing tools for single devices or single communication protocols. Their test scenarios are fixed and lack flexible configurability. In particular, they ignore the impact of the simulation environment's own resource limitations (such as the number of concurrent carrier communication channels and task queuing capacity) on the execution effect of meter reading scheduling strategies. Furthermore, they do not realize convenient configuration of test parameters and automatic generation of test scripts, resulting in a large deviation between simulation results and performance in the real deployment environment. They cannot meet the verification needs of diverse meter reading scheduling strategies.

[0005] Queuing theory, as an important mathematical tool for studying stochastic service systems, has been widely applied in fields such as communication network bandwidth prediction and data center resource allocation. For example, CN103632209A discloses a method for predicting the bandwidth of intelligent power distribution service data transmission based on queuing theory, which solves the predicted bandwidth through QoS parameter mapping and queuing theory modeling; CN114050984B proposes a method for predicting the communication bandwidth of intelligent power distribution service in smart parks, which uses the MMPP / m / c / n queuing system to establish a QoS parameter mapping model. However, existing technologies mainly focus on bandwidth prediction and communication performance analysis, treating queuing theory as an independent analysis tool rather than embedding it as a configurable component into the simulation system itself. This prevents the implementation of configuring task scheme parameters in Excel spreadsheet format, generating executable test cases through the acquisition of professional language (DSL) parsing, and dynamically adjusting queuing model parameters during simulation, as well as comparing theoretical predictions with actual simulation results in real time.

[0006] Therefore, how to provide a meter reading scheduling strategy simulation verification scheme that deeply integrates the M / M / m / B queuing theory model as a configurable component, supports parameter configuration via Excel and test case generation via DSL parsing, enables bidirectional comparison between theoretical prediction and simulation verification, and can quickly construct diverse test scenarios has become an urgent technical problem to be solved in this field. Summary of the Invention

[0007] This invention addresses the technical deficiencies in existing meter reading scheduling strategy verification technologies by proposing a simulation verification method and system for meter reading scheduling strategies based on the integration of configurable components and queuing theory.

[0008] The technical solution of this invention is: In a first aspect, the present invention provides a simulation verification method for a meter reading scheduling strategy based on the fusion of configurable components and queuing theory, the method comprising the following steps: Step S1: Perform the configuration operation of simulation scenario parameters. The simulation scenario parameters include meter file information, data acquisition task parameters, data acquisition scheme, and queuing theory model elements. The queuing theory model elements include task arrival rate λ, task service rate μ per unit time, number of concurrent carrier communication channels m, and maximum task queuing length B allowed by the system. Step S2: Run the professional language DSL parser, extract the simulation scenario parameters and meter reading execution steps configured in Excel spreadsheet format through the DSL parser, convert them into executable logical steps of the simulation system, and construct test cases; Step S3: Load the test cases and start the simulation process. Simulate the entire process of meter reading task arrival, queuing, channel service, and queue overflow discard based on the M / M / m / B queuing theory model. During the simulation, calculate the theoretical steady-state performance index of the system and the probability of task loss in real time. The steady-state performance indicators include carrier utilization ρ, average task queue length Lq, and average task waiting time Wq. Step S4: Capture meter reading response data and queuing status data in real time during the simulation execution process, and synchronize the captured data to the dynamic buffer. Step S5: Calculate and plot the actual acquisition success rate curve in real time based on the meter reading response data, and generate the theoretical acquisition success rate curve based on the task service rate μ per unit time and the M / M / m / B queuing theory model at the current simulation time; superimpose the actual acquisition success rate curve and the theoretical acquisition success rate curve on the same coordinate system, and calculate the numerical deviation of the two curves at the sampling point. Step S6: Determine whether the simulation should terminate if the simulation duration exceeds a preset threshold or the task queue is empty and no new tasks arrive. If it does not terminate, return to S3. If the simulation meets the termination conditions, call the evaluation algorithm to combine the actual acquisition success rate, steady-state performance indicators, and task loss probability. Output the ranking of the scheduling strategy's merits.

[0009] Further, in step S1, the queuing theory model elements are constructed by abstracting the physical characteristics of carrier communication into an M / M / m / B model; where the task arrival rate λ represents the number of meter reading tasks arriving per unit time; the task service rate μ represents the average number of tasks that a single carrier communication channel can handle per unit time; the number of concurrent carrier communication channels m represents the total number of carrier communication channels that can work in parallel in the system; and the maximum task queuing length B represents the maximum number of tasks that the system can accommodate.

[0010] Furthermore, step S3, which simulates the entire process of meter reading task arrival, queuing, channel service, and queue overflow discarding based on the M / M / m / B queuing theory model, specifically includes: When a new meter reading task arrives, record the arrival time of the task and count the total number of tasks n in the current simulation system, including the number of tasks being processed and those being queued. If n < m, the newly arrived meter reading task is immediately assigned to an idle carrier communication channel for service processing, and the service time follows an exponential distribution. If m≤n<B, the newly arrived meter reading task enters the queuing queue and waits for channel resources in the first-in-first-out order. If n=B, the queue is determined to have reached its saturation capacity, triggering the task discard logic and synchronously recording the task loss event.

[0011] Furthermore, in step S3, the carrier utilization rate ρ, the average task queue length Lq, the average task waiting time Wq, and the task loss probability are... The following formulas are used respectively: Carrier utilization: ; Average task queue length: ; Average task wait time: ; Probability of task loss: ; Where k represents the summation variable, which is used to iterate through the different states of the system and calculate the sum of probabilities of the first m states; Pn represents the probability that there are exactly n tasks when the system is in a steady state, including the total number of tasks being served and those waiting in the queue.

[0012] Furthermore, in step S4, the meter reading response data includes single task response time, meter reading success rate, and meter reading exception response data; the queuing status data includes real-time queue length, task queuing time, queue status distribution data, and queue blocking data. Among them, the actual average task queue length Lq is obtained from the real-time queue length statistics, the actual average task waiting time Wq is obtained from the task queuing time statistics, and the actual carrier utilization ρ is obtained by combining the real-time queue length and the task queuing time statistics. The actual carrier utilization rate ρ, actual average task queue length Lq, and actual average task waiting time Wq obtained from the statistics are compared with the theoretical steady-state indices of the queuing model: carrier utilization rate ρ, average task queue length Lq, and average task waiting time Wq. When the relative deviation is controlled within 10%, it is determined that the theoretical simulation of the M / M / m / B queuing model is consistent with the actual working conditions on site, and the model has engineering usability; queue state distribution data and queue blocking data are used to help characterize queue operation characteristics and blocking abnormal states.

[0013] Furthermore, in step S5, The real-time calculation and plotting of the actual acquisition success rate curve includes: plotting the actual acquisition success rate curve dynamically by plotting points in a coordinate system with time as the horizontal axis and the cumulative acquisition success rate as the vertical axis. The generated theoretical acquisition success rate curve includes: based on the task loss probability of the system calculated in real time during the simulation operation in step S3. According to the formula: Theoretical success rate of data collection = 1 - The theoretical cumulative acquisition success rate is calculated in real time and a theoretical acquisition success rate prediction curve is plotted. The theoretical acquisition success rate prediction curve is also plotted with time on the horizontal axis and the theoretical cumulative acquisition success rate on the vertical axis. If the numerical deviation exceeds a preset threshold, it is determined that the current scheduling strategy has room for optimization in resource allocation.

[0014] Secondly, this invention provides a simulation verification system for a meter reading scheduling strategy based on the fusion of configurable components and queuing theory, the system comprising: A configurable functional component layer is used to define meter files, data acquisition scheme logic, and queuing theory model elements. The queuing theory model elements include task arrival rate λ, task service rate μ per unit time, number of concurrent carrier communication channels m, and maximum task queue length B. The scenario building module integrates a DSL parser, which is used to assemble the configuration parameters generated by the configurable functional component layer into test cases with timing logic. The simulation execution module is used to load the test cases and drive the M / M / m / B queuing theory model to simulate the execution process of the meter reading task under limited physical resources. The dynamic visualization module is used to dynamically compare and display the actual acquisition success rate data during the simulation process with the theoretical acquisition success rate data calculated based on the M / M / m / B queuing theory model using a hyperbola.

[0015] Furthermore, the simulation execution module employs a multi-threaded parallel processing mechanism to simulate m concurrent channels, with each channel being abstracted into an independent service thread; When a task is retrieved from the queue and assigned to the service thread, the simulation execution module dynamically calculates the simulation processing time of the task based on the configured task service rate μ, and releases the thread resources after the time expires.

[0016] Furthermore, the dynamic visualization module is equipped with an indicator dashboard to display the fluctuation of steady-state performance indicators in real time, including carrier utilization ρ, average task queue length Lq, and average task waiting time Wq.

[0017] Furthermore, the configurable functional component layer also includes an abnormal situation configuration component, which is used to dynamically inject communication interference variables, device offline variables, or message corruption variables during the simulation process.

[0018] Compared with existing technologies, the meter reading scheduling strategy simulation verification system and method based on the fusion of configurable components and queuing theory provided by this invention have the following beneficial effects: This invention overcomes the limitation of existing technologies that treat queuing theory merely as an independent analysis tool. By setting up a queuing theory configuration component, it embeds multi-server finite-capacity queuing models such as M / M / m / B as configurable functional components into the simulation system. Users can configure queuing theory parameters such as task arrival rate λ, task service rate μ, number of concurrent carrier communication channels m, and maximum task queue length B through a graphical interface, achieving real-time interaction between the theoretical model and the simulation process.

[0019] This invention uses a dynamic visualization module to plot the actual acquisition success rate curve and the theoretical acquisition success rate curve calculated based on the M / M / m / B queuing theory model in real time within the same coordinate system, and calculates the numerical deviation between the two curves at the sampling points. When the actual curve deviates from the theoretical curve by more than a preset threshold, it can be quickly determined that there is room for optimization in resource allocation of the current scheduling strategy. This realizes a two-way verification mechanism of "theory guiding simulation and simulation verifying theory," which greatly improves the accuracy of problem diagnosis.

[0020] This invention maps physical resource constraints (number of concurrent carrier communication channels m, task service rate μ, etc.) to queuing theory model parameters through configurable functional components. Based on an M / M / m / B multi-server finite-capacity queuing theory model, it simulates the entire process of meter reading task arrival, queuing, channel service, and queue overflow discarding. When a task arrives, the system dynamically determines the following based on the current total number of tasks n: if n < m, channel service is immediately allocated; if m ≤ n < B, the task enters the queue; and if n = B, task discard is triggered. Compared to the simple threshold-triggered method in existing technologies, the queuing theory modeling of this invention can more realistically reflect resource contention, task queuing, and buffer overflow behaviors in actual deployment environments.

[0021] This invention uses a dynamic visualization module to display the fluctuations of steady-state performance indicators such as carrier utilization ρ, average task queue length Lq, and average task waiting time Wq in real time. Testers can intuitively observe the queuing operation characteristics of the system under different load conditions, and realize a quantitative evaluation of the merits of various meter reading scheduling strategies.

[0022] This invention employs software simulation based on queuing theory to replace traditional on-site meter reading tests, eliminating the need for deploying large amounts of physical equipment and server resources. Furthermore, steady-state analysis based on queuing theory can predict system performance over long periods in a short time, significantly shortening the testing cycle and reducing overall R&D and testing costs.

[0023] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0024] The above and other objects, features and advantages of the present invention will become more apparent from the more detailed description of exemplary embodiments of the invention in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the invention.

[0025] Figure 1 This is a schematic diagram of the simulation verification system structure for a meter reading scheduling strategy based on the fusion of configurable components and queuing theory, provided in an embodiment of the present invention.

[0026] Figure 2 This is a flowchart illustrating a simulation verification method for a meter reading scheduling strategy based on the fusion of configurable components and queuing theory, provided in an embodiment of the present invention.

[0027] Figure 3 This is a schematic diagram of the state transition of the M / M / m / B queuing model in an embodiment of the present invention.

[0028] Among them, 100 is the configurable functional component layer; 110 is the scenario building module; 120 is the test case library; 130 is the simulation execution module; and 140 is the dynamic visualization module. Detailed Implementation

[0029] Preferred embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0030] like Figures 1-3 This invention provides a simulation and verification method for meter reading scheduling strategies based on the fusion of configurable components and queuing theory. This method abstracts the physical characteristics of carrier communication into a queuing theory model and combines it with a configurable, component-based simulation architecture to achieve accurate simulation and verification of meter reading scheduling strategies. The method includes the following steps: Step S1: Perform the configuration operation of simulation scenario parameters. The simulation scenario parameters include meter file information, data acquisition task parameters, data acquisition scheme, and queuing theory model elements. The queuing theory model elements include task arrival rate λ, task service rate μ per unit time, number of concurrent carrier communication channels m, and maximum task queuing length B allowed by the system. Specifically, in step S1, the queuing theory model elements are constructed by abstracting the physical characteristics of carrier communication into an M / M / m / B model; where the task arrival rate λ represents the number of meter reading tasks arriving per unit time; the task service rate μ represents the average number of tasks that a single carrier communication channel can handle per unit time; the number of concurrent carrier communication channels m represents the total number of carrier communication channels that can work in parallel in the system; and the maximum task queuing length B represents the maximum number of tasks that the system can accommodate.

[0031] Step S2: Run the professional language DSL parser, extract the simulation scenario parameters and meter reading execution steps configured in Excel spreadsheet format through the DSL parser, convert them into executable logical steps of the simulation system, and construct test cases; Step S3: Load the test cases and start the simulation process. Simulate the entire process of meter reading task arrival, queuing, channel service, and queue overflow discard based on the M / M / m / B queuing theory model. During the simulation, calculate the theoretical steady-state performance index of the system and the probability of task loss in real time. The steady-state performance indicators include carrier utilization ρ, average task queue length Lq, and average task waiting time Wq. Specifically, step S3 simulates the entire process of meter reading task arrival, queuing, channel service, and queue overflow discard based on the M / M / m / B queuing theory model, including: When a new meter reading task arrives, record the arrival time of the task and count the total number of tasks n in the current simulation system, including the number of tasks being processed and those being queued. If n < m, the newly arrived meter reading task is immediately assigned to an idle carrier communication channel for service processing, and the service time follows an exponential distribution. If m≤n<B, the newly arrived meter reading task enters the queuing queue and waits for channel resources in the first-in-first-out order. If n=B, the queue is determined to have reached its saturation capacity, triggering the task discard logic and synchronously recording the task loss event.

[0032] Specifically, in step S3, the carrier utilization rate ρ, the average task queue length Lq, the average task waiting time Wq, and the task loss probability are... The following formulas are used respectively: Carrier utilization: ; Average task queue length: ; Average task wait time: ; Probability of task loss: ; Where k represents the summation variable, which is used to iterate through the different states of the system and calculate the sum of probabilities of the first m states; Pn represents the probability that there are exactly n tasks when the system is in a steady state, including the total number of tasks being served and those waiting in the queue.

[0033] In one example, the task service rate μ represents the number of meter reading tasks that can be processed per unit time on a single carrier channel; the task arrival rate λ and the task service rate μ change dynamically with the meter reading scenario and carrier channel status; the calculated average task queue length... Used to characterize the degree of system congestion, average task waiting time Used to characterize the real-time performance of meter reading, both provide a quantitative basis for the performance evaluation of meter reading scheduling strategies, and form a data support relationship with the meter reading scheduling strategy verification in step S5 and the strategy evaluation in step S6.

[0034] Step S4: Capture meter reading response data and queuing status data in real time during the simulation execution process, and synchronize the captured data to the dynamic buffer. Specifically, in step S4, the meter reading response data includes single task response time, meter reading success rate, and meter reading abnormal response data; the queuing status data includes real-time queue length, task queuing time, queue status distribution data, and queue blocking data. Among them, the actual average task queue length Lq is obtained from the real-time queue length statistics, the actual average task waiting time Wq is obtained from the task queuing time statistics, and the actual carrier utilization ρ is obtained by combining the real-time queue length and the task queuing time statistics. The actual carrier utilization rate ρ, actual average task queue length Lq, and actual average task waiting time Wq obtained from the statistics are compared with the theoretical steady-state indices of the queuing model: carrier utilization rate ρ, average task queue length Lq, and average task waiting time Wq. When the relative deviation is controlled within 10%, it is determined that the theoretical simulation of the M / M / m / B queuing model is consistent with the actual working conditions on site, and the model has engineering usability; queue state distribution data and queue blocking data are used to help characterize queue operation characteristics and blocking abnormal states.

[0035] Step S5: Calculate and plot the actual acquisition success rate curve in real time based on the meter reading response data, and generate the theoretical acquisition success rate curve based on the task service rate μ per unit time and the M / M / m / B queuing theory model at the current simulation time; superimpose the actual acquisition success rate curve and the theoretical acquisition success rate curve on the same coordinate system, and calculate the numerical deviation of the two curves at the sampling point. Specifically, further, in step S5, The real-time calculation and plotting of the actual acquisition success rate curve includes: plotting the actual acquisition success rate curve dynamically by plotting points in a coordinate system with time as the horizontal axis and the cumulative acquisition success rate as the vertical axis. The generated theoretical acquisition success rate curve includes: based on the task loss probability of the system calculated in real time during the simulation operation in step S3. According to the formula: Theoretical success rate of data collection = 1 - The theoretical cumulative acquisition success rate is calculated in real time and a theoretical acquisition success rate prediction curve is plotted. The theoretical acquisition success rate prediction curve is also plotted with time on the horizontal axis and the theoretical cumulative acquisition success rate on the vertical axis. If the numerical deviation exceeds a preset threshold, it is determined that there is room for optimization in the current scheduling strategy in resource allocation.

[0036] In one example, when a new anomaly is injected during the simulation, resource overload occurs, or scheduling parameters are adjusted, the deviation between the actual acquisition success rate curve and the theoretical acquisition success rate curve changes accordingly, and a visual label is generated on the curve graph.

[0037] Step S6: Determine whether the simulation should terminate if the simulation duration exceeds a preset threshold or the task queue is empty and no new tasks arrive. If it does not terminate, return to S3. If the simulation meets the termination conditions, call the evaluation algorithm to combine the actual acquisition success rate, steady-state performance indicators, and task loss probability. Output the ranking of the scheduling strategy's merits.

[0038] This invention also provides a simulation verification system for a meter reading scheduling strategy based on the integration of configurable components and queuing theory. The system includes modules that correspond one-to-one with the steps of the above method, used to implement all or part of the process of the method. Each module can be implemented by hardware, software, or a combination of both. The system includes: The configurable functional component layer 100 is used to define meter files, data acquisition scheme logic, and queuing theory model elements. The queuing theory model elements include task arrival rate λ, task service rate μ per unit time, number of concurrent carrier communication channels m, and maximum task queue length B. The scenario building module 110 integrates a DSL parser, which is used to assemble the configuration parameters generated by the configurable functional component layer 100 into test cases with timing logic, and build a test case library 120. Simulation execution module 130 is used to load the test cases and drive the M / M / m / B queuing theory model to simulate the execution process of meter reading tasks under physical resource constraints. The dynamic visualization module 140 is used to dynamically compare and display the actual acquisition success rate data during the simulation process with the theoretical acquisition success rate data calculated based on the M / M / m / B queuing theory model using a hyperbola.

[0039] Furthermore, the simulation execution module 130 adopts a multi-threaded parallel processing mechanism to simulate m concurrent channels, and each channel is abstracted as an independent service thread; When a task is retrieved from the queue and assigned to the service thread, the simulation execution module 130 dynamically calculates the simulation processing time of the task based on the configured task service rate μ, and releases the thread resources after the time expires.

[0040] Furthermore, the dynamic visualization module 140 is equipped with an indicator dashboard to display the fluctuation of steady-state performance indicators in real time, including carrier utilization ρ, average task queue length Lq, and average task waiting time Wq.

[0041] Furthermore, the configurable functional component layer 100 also includes an abnormal situation configuration component, which is used to dynamically inject communication interference variables, device offline variables, or message corruption variables during the simulation process.

[0042] Example 1: Standard operating condition simulation verification based on the M / M / m / B model This embodiment aims to verify the baseline performance of the meter reading scheduling strategy simulation system based on the M / M / m / B queuing model, and simultaneously verify the consistency between the theoretical calculation results and the simulation and actual test results of queuing state indicators such as carrier utilization, average queue length, and task loss probability.

[0043] Configure the following parameters using the queuing theory configuration component: Queuing model selection: M / M / m / B model Arrival process: Poisson distribution, arrival rate λ = 80 data items / second Service process: exponential distribution, service rate μ = 10 data items / second (average processing time of 0.1 seconds per carrier communication channel). Number of concurrent carrier communication channels: m = 10 (corresponding to a carrier module with a concurrent capacity of 10 channels) System capacity: B = 30, meaning a maximum of 20 tasks are allowed to queue (10 currently being served + 20 in the queue). Resource configuration components are set as follows: the carrier module supports 10 concurrent channels, CPU quota of 5 cores, memory of 8GB, and network bandwidth of 200Mbps.

[0044] According to queuing theory formulas, system utilization... = 80 / (10×10) = 0.8. Theoretical average queue length. The theoretical probability of task loss can be calculated using the M / M / m / B formula, which is approximately 0.5%.

[0045] After the simulation execution engine runs, the dynamic visualization module plots the success rate curve in real time and overlays it with the theoretical prediction curve. The results show that the actual simulation curve and the theoretical prediction curve are in high agreement, verifying the accuracy of the simulation model.

[0046] After the simulation execution engine runs, the dynamic visualization module, on the one hand, plots the actual acquisition success rate curve in real time and overlays it with the theoretical predicted success rate curve, calculating the numerical deviation of each sampling point; on the other hand, for three types of queuing state indicators—carrier utilization, average task queue length, and task loss probability—the module uses the theoretical steady-state values ​​obtained from the M / M / m / B model as a benchmark reference line, overlaying them with the measured time-series statistical curves of the queuing state acquired in real time during the simulation process, quantifying the numerical deviation of the measured data relative to the theoretical steady-state benchmark over the entire simulation cycle. The results show that the actual simulation acquisition success rate curve and the theoretical predicted curve are highly consistent, and the overall deviation of the measured values ​​of queuing-related steady-state indicators relative to the theoretical steady-state benchmark is less than 1%, thus doubly verifying the accuracy of the simulation model from two dimensions: service success rate and system queuing operation status.

[0047] Example 2: Queuing Behavior Analysis under Resource Overload Scenarios This embodiment aims to verify the queuing behavior and task loss characteristics of the system under resource overload conditions.

[0048] Based on Example 1, the service rate is gradually reduced using the resource configuration component. (Simulating a decrease in CPU processing power), gradually decreasing from 10 data items / second (average processing time 0.1 seconds) to 4 data items / second (average processing time 0.25 seconds). According to the M / M / m / B model, when When reduced, system utilization Increase, average queue length Increase the probability of task loss. rise.

[0049] Simulation results show: When μ = 10 data items / second, ρ = 0.8, the theoretical task loss probability is approximately 0.5%, and the actual task loss probability is approximately 0.6%. When μ = 8 data items / second, ρ = 1.0, the theoretical task loss probability is approximately 5.6%, and the actual task loss probability is approximately 5.9%. When μ = 6.7 data items / second, ρ = 1.194, the theoretical task loss probability is approximately 12.3%, and the actual task loss probability is approximately 12.8%. When μ = 5 data items / second, ρ = 1.6, the theoretical task loss probability is approximately 28.4%, and the actual task loss probability is approximately 29.1%. When μ = 4 data items / second, ρ = 2.0, the theoretical task loss probability is approximately 42.3%, and the actual task loss probability is approximately 43.0%. The dynamic visualization module, on the one hand, marks the theoretical and actual task loss probabilities for each operating condition on the success rate curve and calculates the deviation; on the other hand, it overlays and displays the theoretical steady-state baselines of carrier utilization, average queue length, and average task waiting time with the measured time-series curves, quantifying the fitting deviation in stages. In the initial stage of load switching, the system is in a transient transition state, and the indicator deviations are relatively large; after the system converges to a steady state, the relative deviations between the theoretical and measured values ​​of various queuing indicators are generally controlled within 2%. The curves synchronously mark the theoretical loss points for each operating condition, allowing for a direct comparison of the loss probability deviations, fully demonstrating that this invention can accurately reproduce the queuing behavior of the meter reading system under overload scenarios and possesses reliable quantitative evaluation capabilities.

[0050] Example 3: Queuing Theory Modeling of Multi-Priority Tasks This embodiment aims to verify the processing characteristics of multi-priority tasks in a queuing system, and uses a priority extension model based on M / M / m / B for simulation verification.

[0051] Configure four different types of meter reading tasks using the task scheme component, and set the corresponding priority parameters:

[0052] The specific configuration is as follows: 1-Minute Curve Data Collection Task: Collect curve data every 1 minute, involving 15 key monitoring meters, with a priority set to 4 (lowest), and a success rate of no less than 80%. 15-minute curve task: Collect curve data every 15 minutes, involving 100 meters, with a priority of 3 and a 100% success rate required; 1-Hour Curve Data Collection Task: Collect curve data every hour, involving 400 meters, with a priority of 2 and a 100% success rate required; Daily freeze task: Collect daily freeze data at midnight, involving 400 meters, with priority set to 1 (highest), and a 100% success rate required.

[0053] The queuing theory configuration component adopts the M / M / m / B extended model with priority queues, and is configured as follows: Arrival rate distribution: The arrival rate of each priority task is set according to the table above. The total arrival rate λ_total = 0.75 + 0.556 + 0.889 + 0.056 ≈ 2.25 data items / second; Service rate: μ = 10 data items / second (average processing time of 0.1 seconds per channel); Number of concurrent carrier communication channels: m = 10; System capacity: B = 40; Priority scheduling strategy: High-priority tasks can preempt service resources from low-priority tasks, and tasks of the same priority are queued in FIFO order.

[0054] The simulation execution engine runs for 24 hours, and the dynamic visualization module plots the success rate curves of the four types of tasks, and overlays the theoretical prediction curves based on priority queuing theory.

[0055] Simulation results show: The daily freeze task (priority 1) can still maintain a 100% success rate during periods of resource scarcity, consistent with theoretical predictions; The 1-hour curve task (priority 2) maintained a 99.9% success rate for most of the time, with only brief fluctuations during peak system resource usage periods (such as the concentrated execution of daily freezes at midnight), ultimately achieving a success rate of 99.5%. The average success rate of the 15-minute curve task (priority 3) is 99.6%, which drops to around 98.8% when resource competition is intense; The average success rate of the 1-minute curve task (priority 4) is 92.5%, and the success rate is 85.3% during periods of resource overload, still meeting the 80% success rate requirement.

[0056] This embodiment verifies the resource allocation effect of the priority-based queuing scheduling strategy among tasks of different priorities, proving that the present invention can effectively simulate the queuing behavior and success rate changes of multi-priority tasks in a resource-competitive environment, providing a quantitative basis for the parameter optimization of the scheduling strategy. Simultaneously, the accuracy of the queuing theory model is verified by comparing theoretical prediction curves with actual simulation curves. Furthermore, for four types of priority tasks, the theoretical steady-state benchmarks for average task waiting time, average queue length, and channel utilization are superimposed and compared with measured curves, and the deviations are quantified. The overall model fitting deviation is controllable in multi-priority scenarios.

[0057] Example 4: Simulation of Active Cache Management Mechanism This embodiment aims to verify the impact of the Active Queue Management strategy on the performance of the meter reading simulation system.

[0058] Parameter settings are completed through the threshold management configuration unit of the queuing theory configuration component: The basic model uses M / M / m / B, with the number of concurrent carrier channels m=10 and the total system capacity B=50 (queue depth 40). The active task loss threshold γ=30, and the simulation logic is configured as follows: when the queue length exceeds γ, newly arriving tasks are dropped with probability. It was intentionally discarded (i.e., the task was lost); Set up a comparison group: No active management strategy; new tasks are only discarded when the queue is full.

[0059] The resource configuration component has a fixed service rate μ=10 data items / second, and gradually increases the arrival rate λ from 30 to 100 data items / second.

[0060] Simulation results show: Low load phase (λ=30-50): =0.3-0.5, the performance of the two groups is similar, the active management group has a slight increase in latency; Medium load phase (λ=60-80): =0.6-0.8, the proactive management group avoided queue congestion by reducing early task losses, and the success rate of high-priority tasks remained above 96%; High load phase (λ=90-100): =0.9-1.0, the queue length of the active management group is more stable, the jitter is less, and the overall throughput is higher than that of the comparison group; The dynamic visualization module displays the queue length distribution in real time as a heatmap. It also overlays theoretical steady-state baselines and measured time-series curves on the carrier utilization, average queue length, average task latency, and task loss probability of both schemes, quantifying the fitting deviation to visually demonstrate the differentiated impact of the two caching strategies on system performance. The theoretical and measured deviations of the queuing metrics remain low for both schemes, validating the reliability of the quantitative evaluation of the queuing model's adaptability to active cache management scenarios.

[0061] Example 5: Simulation of Burst Traffic under the MMPP / m / c / n Model This embodiment aims to verify the impact of sudden business on the meter reading system, and adopts the MMPP / m / c / n (Markov Modulated Poisson Process) queuing model.

[0062] Select the MMPP / m / c / n model through the queuing theory configuration component, and set: Two-state MMPP: State 1 (normal) arrival rate Data items per second, arrival rate of state 2 (burst) Data item / second State transition probability: p 12 =0.01, p 21 =0.1 (Enters a sudden state on average every 100 seconds, lasting for 10 seconds) Number of concurrent carrier communication channels: m=8 System capacity: c=40 Service rate: μ = 15 data items / second (average processing time approximately 0.067 seconds) The simulation execution engine ran for 60 minutes, simulating periodic sudden business scenarios. The dynamic visualization module showed that during the sudden state, the system utilization instantly increased to ρ>0.8, the queue length increased rapidly, and some tasks were dropped due to full cache.

[0063] The dynamic visualization module synchronously outputs the MMPP theoretical benchmark and measured time-series curves corresponding to carrier utilization, instantaneous queue length, and task loss probability. Except for the acquisition success rate, it calculates the fitting deviation of various queuing indicators for each sampling point. During the stable operation phase of the system, the theoretical and measured values ​​are highly consistent. When the service state changes, the system is in a transient transition period, and the indicators show short-term small deviations. After the system converges to a steady state, the two sets of data converge again. This phenomenon is consistent with the transient evolution law of queuing systems.

[0064] This embodiment verifies the simulation and adaptation capability of the present invention for sudden meter reading services by relying on the dual-dimensional theory of queuing status indicators and data collection success rate and the comparison results of actual measurements.

[0065] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A simulation verification method for meter reading scheduling strategy based on the integration of configurable components and queuing theory, characterized in that, The method includes the following steps: Step S1: Perform the configuration operation of simulation scenario parameters. The simulation scenario parameters include meter file information, data acquisition task parameters, data acquisition scheme, and queuing theory model elements. The queuing theory model elements include task arrival rate λ, task service rate μ per unit time, number of concurrent carrier communication channels m, and maximum task queuing length B allowed by the system. Step S2: Run the professional language DSL parser, extract the simulation scenario parameters and meter reading execution steps configured in Excel spreadsheet format through the DSL parser, convert them into executable logical steps of the simulation system, and construct test cases; Step S3: Load the test cases and start the simulation process. Simulate the entire process of meter reading task arrival, queuing, channel service, and queue overflow discard based on the M / M / m / B queuing theory model. During the simulation, calculate the theoretical steady-state performance index of the system and the probability of task loss in real time. The steady-state performance indicators include carrier utilization ρ, average task queue length Lq, and average task waiting time Wq. Step S4: Capture meter reading response data and queuing status data in real time during the simulation execution process, and synchronize the captured data to the dynamic buffer. Step S5: Calculate and plot the actual acquisition success rate curve in real time based on the meter reading response data, and generate the theoretical acquisition success rate curve based on the task service rate μ per unit time and the M / M / m / B queuing theory model at the current simulation time; superimpose the actual acquisition success rate curve and the theoretical acquisition success rate curve on the same coordinate system, and calculate the numerical deviation of the two curves at the sampling point. Step S6: Determine whether the simulation should terminate if the simulation duration exceeds a preset threshold or the task queue is empty and no new tasks arrive. If it does not terminate, return to S3. If the simulation meets the termination conditions, call the evaluation algorithm to combine the actual acquisition success rate, steady-state performance indicators, and task loss probability. Output the ranking of the scheduling strategy's merits.

2. The simulation verification method for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 1, characterized in that... In step S1, the queuing theory model elements are constructed by abstracting the physical characteristics of carrier communication into an M / M / m / B model; where the task arrival rate λ represents the number of meter reading tasks arriving per unit time; the task service rate μ represents the average number of tasks that a single carrier communication channel can handle per unit time; the number of concurrent carrier communication channels m represents the total number of carrier communication channels that can work in parallel in the system; and the maximum task queuing length B represents the maximum number of tasks that the system can accommodate.

3. The simulation verification method for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 1, characterized in that... In step S3, the simulation of the entire process of meter reading task arrival, queuing, channel service, and queue overflow discard based on the M / M / m / B queuing theory model specifically includes: When a new meter reading task arrives, record the arrival time of the task and count the total number of tasks n in the current simulation system, including the number of tasks being processed and those being queued. If n < m, the newly arrived meter reading task is immediately assigned to an idle carrier communication channel for service processing, and the service time follows an exponential distribution. If m≤n<B, the newly arrived meter reading task enters the queuing queue and waits for channel resources in the first-in-first-out order. If n=B, the queue is determined to have reached its saturation capacity, triggering the task discard logic and synchronously recording the task loss event.

4. The simulation verification method for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 3, characterized in that... In step S3, the carrier utilization rate ρ, the average task queue length Lq, the average task waiting time Wq, and the task loss probability are... The following formulas are used respectively: Carrier utilization: ; Average task queue length: ; Average task wait time: ; Task loss probability: ; Where k represents the summation variable, which is used to iterate through the different states of the system and calculate the sum of probabilities of the first m states; Pn represents the probability that there are exactly n tasks when the system is in a steady state, including the total number of tasks being served and those waiting in the queue.

5. The simulation verification method for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 3, characterized in that... In step S4, the meter reading response data includes single task response time, meter reading success rate, and meter reading abnormal response data; the queuing status data includes real-time queue length, task queuing time, queue status distribution data, and queue blocking data. Among them, the actual average task queue length Lq is obtained from the real-time queue length statistics, the actual average task waiting time Wq is obtained from the task queuing time statistics, and the actual carrier utilization ρ is obtained by combining the real-time queue length and the task queuing time statistics. The actual carrier utilization rate ρ, actual average task queue length Lq, and actual average task waiting time Wq obtained from the statistics are compared with the theoretical steady-state indices of the queuing model: carrier utilization rate ρ, average task queue length Lq, and average task waiting time Wq. When the relative deviation is controlled within 10%, it is determined that the theoretical simulation of the M / M / m / B queuing model is consistent with the actual working conditions on site, and the model has engineering usability; queue state distribution data and queue blocking data are used to help characterize queue operation characteristics and blocking abnormal states.

6. The simulation verification method for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 1, characterized in that... In step S5, The real-time calculation and plotting of the actual acquisition success rate curve includes: plotting the actual acquisition success rate curve dynamically by plotting points in a coordinate system with time as the horizontal axis and the cumulative acquisition success rate as the vertical axis. The generated theoretical acquisition success rate curve includes: based on the task loss probability of the system calculated in real time during the simulation operation in step S3. According to the formula: Theoretical success rate of data collection = 1 - The theoretical cumulative acquisition success rate is calculated in real time and a theoretical acquisition success rate prediction curve is plotted. The theoretical acquisition success rate prediction curve is also plotted with time on the horizontal axis and the theoretical cumulative acquisition success rate on the vertical axis. If the numerical deviation exceeds a preset threshold, it is determined that the current scheduling strategy has room for optimization in resource allocation.

7. A simulation verification system for meter reading scheduling strategies based on the integration of configurable components and queuing theory, characterized in that, The system includes: A configurable functional component layer (100) is used to define meter files, data acquisition scheme logic and queuing theory model elements, wherein the queuing theory model elements include task arrival rate λ, task service rate μ per unit time, number of concurrent carrier communication channels m and maximum task queue length B. The scenario building module (110) integrates a DSL parser, which is used to assemble the configuration parameters generated by the configurable functional component layer (100) into test cases with timing logic; The simulation execution module (130) is used to load the test cases and drive the M / M / m / B queuing theory model to simulate the execution process of the meter reading task under limited physical resources. The dynamic visualization module (140) is used to dynamically compare and display the actual acquisition success rate data during the simulation process with the theoretical acquisition success rate data calculated based on the M / M / m / B queuing theory model using a hyperbola.

8. The simulation verification system for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 7, characterized in that, The simulation execution module (130) uses a multi-threaded parallel processing mechanism to simulate m concurrent channels, and each channel is abstracted as an independent service thread; When a task is taken out of the queue and assigned to the service thread, the simulation execution module (130) dynamically calculates the simulation processing time of the task according to the configured task service rate μ, and releases the thread resources after the time expires.

9. The simulation verification system for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 7, characterized in that, The dynamic visualization module (140) is equipped with an indicator dashboard to display the fluctuation of steady-state performance indicators in real time, including carrier utilization ρ, average task queue length Lq, and average task waiting time Wq.

10. The simulation verification system for meter reading scheduling strategy based on the fusion of configurable components and queuing theory as described in claim 7, characterized in that, The configurable functional component layer (100) also includes an abnormal situation configuration component, which is used to dynamically inject communication interference variables, device offline variables or message corruption variables during the simulation process.

Images