Multi-parameter regulation and control deviation resetting system
Through the multi-parameter control deviation reset system, the problems of unclear parameter selection and unreasonable value range in complex multi-parameter systems are solved, precise control and real-time monitoring of the system are achieved, and the stability and operation efficiency of the system are improved.
Patent Information
- Application Number
- CN202510796649.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-16
- Publication Date
- 2025-09-26
AI Technical Summary
Traditional parameter control methods have the problems of unclear parameter selection and unreasonable value range in complex multi-parameter systems, making it difficult to achieve global optimal configuration. In addition, there is a lack of real-time monitoring and automatic parameter reset methods, which causes the system to deviate from expected operation and performance degradation.
A multi-parameter control deviation reset system is adopted, including a parameter space definition module, a particle swarm optimization algorithm module, a deviation detection and evaluation module, and a parameter reset module. By accurately determining the parameter range, real-time monitoring and automatic adjustment of parameters can be carried out to reduce deviations.
It achieves precise control of system parameters, improves stability and adaptability, enables efficient operation in multiple fields, and improves the overall performance and operation and maintenance efficiency of the system.
Smart Images
Figure CN120704128A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of multi-parameter system intelligent control, and in particular to a multi-parameter control deviation resetting system. Background Art
[0002] Complex multi-parameter systems exist in many fields, such as industrial production, intelligent transportation, and energy management. Their operation is influenced by multiple parameters and the relationships between them are complex. Traditional parameter control methods have many flaws. When determining key parameters, they rely on experience and simple analysis, which can easily omit key parameters or include irrelevant parameters, resulting in a lack of accuracy. Setting parameter value ranges does not fully incorporate multiple factors, and unreasonable ranges can easily lead to safety hazards and limit optimization space. Seeking the optimal parameter combination can easily lead to local optimality and fail to achieve the global optimal configuration. In particular, when deviations occur in system operation, the lack of effective real-time monitoring and automatic parameter reset methods makes it difficult to quickly detect and adjust parameters, causing the system to continue to deviate from expectations and exacerbating performance degradation. Summary of the Invention
[0003] In response to the deficiencies of the prior art, the present invention provides a multi-parameter control deviation resetting system, which solves the problem that parameter control is difficult to implement in complex multi-parameter systems to cope with system operation deviations.
[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a multi-parameter control deviation reset system, comprising:
[0005] A parameter space definition module, which is used to determine multiple system-related parameters and set the value range of each parameter;
[0006] a particle swarm optimization algorithm module, the particle swarm optimization algorithm module being connected to the parameter space definition module and performing calculations based on the parameters and value ranges determined by the parameter space definition module to obtain an optimal parameter combination;
[0007] A deviation detection and evaluation module, which is connected to the system operation state and is used to detect the deviation between the actual system operation state and the expected state;
[0008] A parameter reset module is connected to the particle swarm optimization algorithm module and the deviation detection and evaluation module respectively, and is used to reset system parameters according to the optimal parameter combination obtained by the particle swarm optimization algorithm module to reduce system deviation.
[0009] By adopting the above technical solutions, it is possible to accurately determine parameters and ranges, optimize parameter combinations, monitor in real time and automatically correct deviations, improve system stability and adaptability, and be applicable to multiple fields, helping each field to operate efficiently.
[0010] Preferably, the parameters determined by the parameter space definition module are a plurality of adjustable parameters based on the physical characteristics of the system and actual operation requirements.
[0011] Preferably, the particle swarm optimization algorithm module includes a particle initialization submodule, which is used to randomly generate multiple particles within the range of parameter values determined by the parameter space definition module, the position of each particle represents a set of parameter combinations, and each particle has an initial velocity.
[0012] Preferably, the particle swarm optimization algorithm module also includes a fitness evaluation submodule, which is used to design a fitness function based on the operating objectives expected to be achieved by the system, and evaluate the parameter combination represented by each particle through the fitness function to obtain the fitness value corresponding to each particle, and the fitness value is used to measure the influence of each parameter combination on the system performance.
[0013] Preferably, the particle swarm optimization algorithm module also includes an individual and global optimal update submodule, which is used to compare the current fitness value of each particle with its historical optimal fitness value. If the current fitness value is better, the individual historical optimal position of the particle is updated, and the particle with the best fitness value is determined among all particles, and its position is used as the global optimal position. The global optimal position represents the optimal parameter combination found by the current particle swarm.
[0014] Preferably, the particle swarm optimization algorithm module further includes a particle speed and position update submodule, which is used to update the speed and position of the particle according to the current position, speed, individual historical best position and global best position of the particle according to the following formula: i (t+1)=w×v i (t)+c1×rand()×(pbest i -x i (t))+c2×rand()×(gbest-x i (t)), where v i (t) represents the velocity of particle i at time t, x i (t) represents the position of particle i, w is the inertia weight, c1 and c2 are learning factors, rand() is a function that generates random numbers between (0,1), pbest i is the individual best position of particle i, gbest is the global best position of the entire particle swarm; the position update formula is: i (t+1)=x i (t)+v i (t+1), where x i(t+1) represents the position of particle i at time t+1, x i (t) represents the position of particle i at time t, v i (t+1) represents the speed of particle i at time t+1. The speed and position of the particle are updated through continuous iteration until the preset stopping condition is met. The stopping condition is that the change in fitness value is less than the set threshold or the convergence degree of the particle swarm reaches the set standard.
[0015] Preferably, the deviation detection and evaluation module is used to collect the actual operation output data of the system in real time, compare the actual operation output data with the pre-set expected output data, and calculate the deviation value through a preset deviation measurement indicator. The deviation measurement indicator includes but is not limited to the mean square error or the mean absolute error, so as to determine whether the deviation value exceeds the allowable range pre-set by the system.
[0016] Preferably, after receiving the trigger signal issued by the deviation detection and evaluation module indicating that the deviation exceeds the allowable range, the parameter reset module adjusts the current parameters of the system to the parameter combination corresponding to the global optimal position determined by the particle swarm optimization algorithm module, thereby changing the operating state of the system and reducing the system deviation.
[0017] A multi-parameter control deviation reset method, used in the multi-parameter control deviation reset system, comprises the following steps:
[0018] S1. Determine multiple system-related parameters and value ranges through the parameter space definition module;
[0019] S2, using the particle swarm optimization algorithm module to perform particle initialization, fitness evaluation, individual and global optimal updates, and particle velocity and position update operations within the value range to obtain the optimal parameter combination;
[0020] S3, using the deviation detection and evaluation module to detect the deviation between the actual operating state of the system and the expected state in real time;
[0021] S4. When the deviation exceeds the allowable range, the parameter reset module resets the system parameters according to the optimal parameter combination to regulate and reset the system deviation.
[0022] The present invention provides a multi-parameter control deviation reset system. It has the following beneficial effects:
[0023] 1. This invention utilizes a multi-parameter control deviation reset system to precisely control complex systems, monitor and correct deviations in real time to ensure stable operation. This system offers significant advantages in multiple fields, improving energy efficiency. It addresses the difficulty of implementing parameter control to address operational deviations in complex, multi-parameter systems.
[0024] 2. Through the parameter space definition module, the present invention accurately determines multiple controllable parameters related to the system and reasonably sets their value ranges. This clarifies the target and direction for subsequent control operations, improves the pertinence and effectiveness of control, and ensures that the system can operate safely, efficiently, and in accordance with actual needs. This helps to improve the overall stability and reliability of the system and fully utilize the system's optimal performance. This solves the problems of unclear parameter selection and unreasonable parameter value ranges in multi-parameter control.
[0025] 3. The present invention enables the coordinated operation of various submodules, performing calculations based on the given range of the parameter space definition module, and gradually searching for more optimal parameter combinations in complex parameter spaces. Leveraging the advantages of swarm intelligence, the parameter space is comprehensively covered, increasing the probability of finding the globally optimal parameter combination. This allows the system to operate with parameter configurations close to the ideal state, effectively improving the overall operating efficiency and stability of the system, and assisting in the discovery of more optimal parameter settings in multi-domain systems to enhance corresponding performance. This solves the problem of relying on conventional methods to quickly and accurately find the optimal parameter combination in multi-parameter systems.
[0026] 4. By connecting to the system's operating status, collecting data in real time and using deviation measurement indicators for comparative analysis, this invention can accurately monitor system deviations, presenting differences with intuitive numerical values. This provides an accurate basis for regulation and control, facilitating clear judgment and analysis by relevant personnel, helping to improve the efficiency and accuracy of system operation and maintenance, and ensuring the system operates stably as expected. This solves the problem of difficulty in intuitively and accurately understanding the degree of deviation between the actual system operation and the expected state. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is the architecture diagram of the multi-parameter control deviation reset system proposed by the present invention;
[0028] Figure 2 This is a flow chart of the multi-parameter control deviation resetting method proposed in the present invention. DETAILED DESCRIPTION
[0029] The following will clearly and completely describe the technical solution of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0030] Please see the attached Figure 1 , an embodiment of the present invention provides a multi-parameter control deviation reset system, including:
[0031] The parameter space definition module is used to determine multiple system-related parameters and set the value range of each parameter.
[0032] The parameters determined by the parameter space definition module are multiple controllable parameters based on the physical characteristics of the system and the actual operation requirements.
[0033] Specifically, a comprehensive analysis of the target system is conducted, delving deeply into its physical properties. For example, if the system is a chemical reaction device used in industrial production, the various physical phenomena involved in the reaction process must be considered, such as heat transfer, the flow characteristics of substances, and the physical laws governing the interactions between different substances. Based on these physical properties, key factors that are closely related to the system's operating state and can be controlled are identified. These factors serve as preliminary parameter candidates.
[0034] Parameters are further screened and determined based on the system's actual operational requirements. These requirements encompass multiple aspects, such as production efficiency, which considers product output per unit time and quality stability; safety, which ensures that parameters like temperature and pressure remain within safe ranges during system operation to avoid hazardous situations; and cost control, which balances resource consumption. By comprehensively considering these operational requirements, the final set of controllable parameters is determined from the initial candidate list.
[0035] After determining the specific controllable parameters, the range of values for each parameter is set based on the system's physical limits, past operating experience, and industry standards. For example, consider the temperature parameter in a chemical reaction device. From a physical properties perspective, it's important to consider the temperature range within which the reactants can react normally without causing adverse consequences such as decomposition due to excessively high temperatures. From an operational perspective, excessively high temperatures may exceed the equipment's tolerance range, impacting its lifespan, while excessively low temperatures will slow the reaction rate and affect production efficiency. Taking these factors into consideration, a reasonable temperature range is set, such as [100°C, 300°C]. For other parameters, such as pressure and flow rate, similar principles are used to set corresponding ranges based on their corresponding physical properties and actual operational requirements.
[0036] The parameter space definition module accurately identifies multiple controllable system parameters and sets their value ranges appropriately. This clarifies the target and direction for subsequent control operations, improving the pertinence and effectiveness of control. It also ensures that the system can operate safely, efficiently, and in accordance with actual needs, helping to improve the overall stability and reliability of the system and maximize its optimal performance. This solves the problems of unclear parameter selection and unreasonable parameter value ranges in multi-parameter control.
[0037] The particle swarm optimization algorithm module is connected to the parameter space definition module, and performs calculations based on the parameters and value ranges determined by the parameter space definition module to obtain the optimal parameter combination.
[0038] The particle swarm optimization algorithm module includes a particle initialization submodule, which is used to randomly generate multiple particles within the range of parameter values determined by the parameter space definition module. The position of each particle represents a set of parameter combinations, and each particle has an initial velocity.
[0039] The particle swarm optimization algorithm module also includes a fitness evaluation submodule, which is used to design a fitness function based on the system's expected operating goals. The fitness function is used to evaluate the parameter combination represented by each particle to obtain the fitness value corresponding to each particle. The fitness value is used to measure the impact of each parameter combination on system performance.
[0040] The particle swarm optimization algorithm module also includes an individual and global optimal update submodule, which is used to compare the current fitness value of each particle with its historical best fitness value. If the current fitness value is better, the individual historical best position of the particle is updated, and the particle with the best fitness value is determined among all particles, and its position is used as the global best position. The global best position represents the optimal parameter combination found by the current particle swarm.
[0041] The particle swarm optimization algorithm module also includes a particle speed and position update submodule, which is used to update the particle speed and position according to the following formula based on the particle's current position, speed, individual historical best position and global best position: i (t+1)=w×v i (t)+c1×rand()×(pbest i -x i (t))+c2×rand()×(gbest-x i (t)), where v i (t) represents the velocity of particle i at time t, x i (t) represents the position of particle i, w is the inertia weight, c1 and c2 are learning factors, rand() is a function that generates random numbers between (0,1), pbest i is the individual best position of particle i, gbest is the global best position of the entire particle swarm; the position update formula is: i (t+1)=x i (t)+v i (t+1), where x i (t+1) represents the position of particle i at time t+1, x i (t) represents the position of particle i at time t, v i(t+1) represents the speed of particle i at time t+1. The speed and position of the particle are updated through continuous iteration until the preset stopping condition is met. The stopping condition is that the change in fitness value is less than the set threshold or the convergence degree of the particle swarm reaches the set standard.
[0042] Specifically, the connection with the parameter space definition module and the overall operation basis: This module is connected to the parameter space definition module to obtain the system-related parameters determined by it and the corresponding value range information of each parameter. Based on this, operations are carried out in this established multi-dimensional parameter space, aiming to continuously search and explore the parameter combination that can achieve the optimal system performance by simulating group intelligent behaviors such as bird foraging. For example, in an intelligent traffic flow control system, the parameter space definition module determines parameters such as the length of green lights at intersections, the speed limits of various road sections, and their reasonable value ranges. The particle swarm optimization algorithm module then carries out subsequent optimization search work around these parameters.
[0043] Particle Initialization Submodule: The Particle Initialization Submodule operates within the parameter value ranges defined by the Parameter Space Definition Module. It utilizes a random number generation mechanism to randomly generate specific values for each parameter dimension within the corresponding value range. These values, combined, constitute the position of a particle, which in turn represents a specific set of parameter combinations. Each particle is also assigned an initial velocity, also randomly determined within a reasonable range. The values for each dimension represent the speed of change of the corresponding parameter in its initial state. For example, in an intelligent traffic flow control system, the position of a particle might be (green light duration 30 seconds, speed limit 50 km / h on a certain road section), while its initial velocity might be (green light duration increasing by 1 second per second, speed limit increasing by 2 km / h per iteration). This generates numerous particles with different initial states, forming an initial particle swarm.
[0044] Fitness evaluation submodule: The fitness function is designed based on the operational goals that the system expects to achieve. For example, for an intelligent traffic flow control system, the operational goals may be to reduce the average waiting time of vehicles, reduce road congestion, and increase overall traffic flow. In this case, the fitness function can comprehensively consider quantifiable indicators such as the length of vehicles queuing at intersections and the average speed of roads. These indicators are integrated through specific mathematical operations to construct a fitness function that reflects the degree of influence of the parameter combination represented by each particle on the system performance. Afterwards, for the specific parameter combination represented by each particle (i.e., the position of the particle), the corresponding parameters are substituted into the fitness function for calculation to obtain the fitness value corresponding to each particle. For example, if the fitness value calculated after the parameter combination represented by a particle is substituted is higher, it means that the parameter combination is more conducive to achieving traffic flow control goals such as reducing the average waiting time of vehicles under the current settings.
[0045] Individual and Global Optimum Update Submodule: For each particle, this submodule compares its current fitness value with its best fitness value in its historical record. As particles continuously update their positions and participate in fitness evaluations, if the current fitness value is better than the previously recorded best fitness value, it indicates that the current parameter combination is better at improving system performance. In this case, the particle's individual best position is updated and recorded as the current position. Simultaneously, the fitness values of all particles in the entire particle swarm are traversed to identify the particle with the best fitness value, and its position is set as the global best position. This global best position represents the parameter combination found by the current particle swarm after a certain search and considered optimal at the current iteration. For example, in a traffic flow control system, after multiple iterations, if a particle's corresponding green light duration and road speed limit parameter combination is found to maximize traffic congestion relief and has the highest fitness value, then this particle's position becomes the global best position.
[0046] Particle speed and position update submodule: update the formula v according to the established speed i (t+1)=w×v i (t)+c1×rand()×(pbest i -x i (t))+c2×rand()×(gbest-x i (t)) and the position update formula x i (t+1)=x i (t)+v i (t+1) is used to update the particle's speed and position. Among them, the inertia weight w determines the extent to which the particle maintains its original motion trend based on the original speed, the learning factor c1 controls the extent to which the particle approaches its own historical best position, and the learning factor c2 controls the extent to which the particle approaches the global best position. Taking the traffic flow control system as an example, in each iteration, the particle's speed will be adjusted according to factors such as its current speed, the gap between its individual best position and the global best position, and then the particle's position will be updated based on the updated speed, that is, the parameter combination it represents will be updated. The iterative process will continue until the preset stopping condition is met. This stopping condition is specifically that the change in fitness value is less than the set threshold, which means that it is difficult for the particle swarm to find a parameter combination that can significantly improve the system performance, or the convergence degree of the particle swarm reaches the set standard, that is, the distribution of particles in the particle swarm in the parameter space has become stable and basically no longer changes significantly.
[0047] By enabling the coordinated operation of various submodules and performing calculations within the given range of the parameter space definition module, the system can gradually search for optimal parameter combinations within a complex parameter space. Leveraging the advantages of swarm intelligence, the system comprehensively covers the parameter space, increasing the probability of finding the globally optimal parameter combination. This allows the system to operate with parameter configurations close to the ideal state, effectively improving the overall system efficiency and stability. This helps discover optimal parameter settings in multi-domain systems to enhance corresponding performance. This solves the problem of relying on conventional methods to quickly and accurately find the optimal parameter combination in multi-parameter systems.
[0048] The deviation detection and evaluation module is connected to the system operation status and is used to detect the deviation between the actual operation status of the system and the expected status.
[0049] The deviation detection and evaluation module is used to collect the actual operation output data of the system in real time, compare the actual operation output data with the pre-set expected output data, and calculate the deviation value through the preset deviation measurement indicators. The deviation measurement indicators include but are not limited to the mean square error or mean absolute error to determine whether the deviation value exceeds the allowable range preset by the system.
[0050] Specifically, connection with the system operation status and data collection: This module directly establishes a connection with the system operation status, and through various types of adaptive sensors and corresponding data acquisition interfaces, obtains the output data generated by the system during actual operation in real time. For example, on an automated production line, temperature sensors are used to collect real-time temperature data of each key production link, pressure sensors are used to obtain pressure data of the corresponding parts, and counters are used to record the output quantity of products and other types of data information reflecting the system operation status. These sensors work according to different physical principles. For example, temperature sensors are based on the principle that the resistance of a thermistor changes with temperature, converting the physical quantity of temperature into an electrical signal that can be collected and processed. Pressure sensors use the characteristic that the pressure-sensitive element deforms under pressure, causing changes in the electrical signal, to obtain pressure data, etc., to ensure that various relevant data output by the actual operation of the system can be accurately and timely captured.
[0051] Data comparison and application of deviation measurement indicators: After collecting the actual system operation output data, compare it with the expected output data that has been set in advance based on the system design requirements, production goals, etc. The expected output data is the standard value that each operating parameter and the final output result should reach when the system is in an ideal operating state. For example, when the automated production line is in normal production, the expected number of qualified products per unit time, the standard temperature range of each link, the stable pressure value, etc. In order to quantify the degree of difference presented by this comparison, the preset deviation measurement indicator is used to calculate the deviation value. Deviation measurement indicators include but are not limited to mean square error (MSE) and mean absolute error (MAE). Taking mean square error as an example, its calculation formula is, where n represents the number of samples, y represents the number of samples, and y represents the number of samples. i is the i-th observation value in the actual output data, is the i-th observation value in the corresponding expected output data. Through such calculation, the square of the deviation of each corresponding data point is summed and averaged to obtain a value reflecting the overall degree of deviation. The calculation formula of mean absolute error is This measure, which sums and averages the absolute values of the deviations at each data point, is also used to measure the deviation between the actual output and the expected output. By calculating the specific deviation value using these deviation metrics, we can intuitively understand the degree of difference between the actual system operating status and the expected state.
[0052] Deviation range determination: The calculated deviation value is compared with the system's pre-set tolerance range. This tolerance range is determined by comprehensively considering multiple factors, including system stability, safety, production efficiency, and product quality requirements. For example, on an automated production line, the tolerance range for product size deviation may be set within a certain number of millimeters, or the tolerance range for temperature deviation may be set within a certain number of degrees Celsius. If the deviation value exceeds this pre-set tolerance range, it means that the system is not operating as expected, and appropriate adjustments need to be taken.
[0053] By connecting to the system's operating status, collecting data in real time and using deviation metrics for comparative analysis, it can accurately monitor system deviations, presenting differences with intuitive numerical values. This provides an accurate basis for control and facilitates clear judgment and analysis by relevant personnel, helping to improve the efficiency and accuracy of system operation and maintenance, ensuring the system operates stably as expected. This solves the problem of difficulty in intuitively and accurately understanding the degree of deviation between the actual system operation and the expected state.
[0054] The parameter reset module is connected to the particle swarm optimization algorithm module and the deviation detection and evaluation module respectively, and is used to reset the system parameters according to the optimal parameter combination obtained by the particle swarm optimization algorithm module to reduce the system deviation.
[0055] After receiving the trigger signal from the deviation detection and evaluation module indicating that the deviation exceeds the allowable range, the parameter reset module adjusts the current system parameters to the parameter combination corresponding to the global optimal position determined by the particle swarm optimization algorithm module, thereby changing the system's operating state and reducing the system deviation.
[0056] Specifically, the module connection and information exchange mechanism: The parameter reset module connects with the particle swarm optimization module and the deviation detection and evaluation module, establishing a channel for information exchange. On the one hand, it obtains the optimal parameter combination determined by the particle swarm optimization module after multiple rounds of iterative calculations. This is the set of parameter values corresponding to the global optimal position found by the particle swarm during the entire search process. The particle swarm optimization module here comprehensively explores the system's multi-parameter space. Through a series of operations such as particle initialization, fitness evaluation, individual and global optimal updates, and iterative updates of velocity and position, it ultimately identifies the parameter combination that achieves the optimal system performance and passes it to the parameter reset module. On the other hand, the parameter reset module constantly monitors signals from the deviation detection and evaluation module. This module monitors the deviation between the actual system operating state and the expected state in real time. If the calculated deviation value exceeds the system's pre-set tolerance range, it sends a trigger signal to the parameter reset module, notifying it that a parameter adjustment is required.
[0057] Parameter Adjustment and Operational Status Change Process: Upon receiving a trigger signal from the Deviation Detection and Evaluation Module indicating that a deviation has exceeded the allowable range, the Parameter Reset Module initiates the parameter adjustment process. It extracts the parameter combination corresponding to the global optimal position determined by the Particle Swarm Optimization (PSO) module and then replaces each currently used system parameter with the corresponding parameter value from this optimal combination. For example, in an intelligent greenhouse environmental control system, the current system operating parameters are 20°C, 60% humidity, and 800 lux. If the optimal parameter combination corresponding to the global optimal position determined by the PSO module is 22°C, 65% humidity, and 850 lux, the Parameter Reset Module will adjust the system's temperature to 22°C, humidity to 65%, and light intensity to 850 lux, and so on, completing the reset of all relevant parameters. Through such parameter adjustments, the operating state of the system will change accordingly, because each parameter directly affects the physical process within the system, the operating state of the equipment, etc. The new parameter value will guide the system to develop in the direction of reducing deviations, so that it gradually approaches the expected ideal operating state.
[0058] By connecting and interacting with the particle swarm optimization algorithm module and the deviation detection and evaluation module, the system can promptly and accurately reset system parameters based on the optimal parameter combination, prompting the system to quickly return to the expected operating state, ensuring stability and reliability. It can also fully utilize the algorithm's advantages to optimize parameter configuration, further improving overall system performance, such as increasing production efficiency and reducing energy consumption. This solves the problem of the system being unable to automatically and efficiently adjust parameters to restore normal operation after deviations occur.
[0059] Please see the attached Figure 2 A multi-parameter control deviation reset method is used for a multi-parameter control deviation reset system, comprising the following steps:
[0060] S1. Determine multiple system-related parameters and value ranges through the parameter space definition module;
[0061] S2. Use the particle swarm optimization algorithm module to perform particle initialization, fitness evaluation, individual and global optimal updates, and particle velocity and position updates within the value range to obtain the optimal parameter combination;
[0062] S3, using the deviation detection and evaluation module to detect the deviation between the actual operating state of the system and the expected state in real time;
[0063] S4. When the deviation exceeds the allowable range, the parameter reset module resets the system parameters according to the optimal parameter combination to regulate and reset the system deviation.
[0064] The parameter space definition module thoroughly analyzes the system's physical mechanisms, internal connections, and actual operational requirements, identifying key and controllable parameters such as temperature, pressure, and flow. By referencing the system's physical limits, past experience, and industry standards, it precisely sets reasonable value ranges for each parameter, establishing clear boundaries for subsequent adjustments. The particle swarm optimization algorithm module interfaces with the parameter space definition module to perform operations within the defined parameter space. It first randomly generates a swarm of particles, whose positions represent parameter combinations and are assigned initial velocities. It then constructs a fitness function based on the system's desired operational objectives, assessing the impact of each particle's corresponding parameter combination on system performance and updating individual and global optimal positions. It then iterates continuously according to specific velocity and position update formulas until a stopping condition is met, ultimately identifying the optimal parameter combination. The deviation detection and evaluation module, which is connected to the system's operational status in real time, collects actual operational data using various sensors based on physical principles. It then compares this data with the preset expected output data and uses deviation metrics such as mean squared error and mean absolute error to calculate the deviation value and determine whether it exceeds the acceptable range. The parameter reset module monitors the signal of the deviation detection and evaluation module. Once the deviation exceeds the range, it extracts the optimal parameter combination determined by the particle swarm optimization algorithm module, replaces the current system parameters, changes the operating state, and reduces the system deviation, gradually approaching the expected state.
[0065] The multi-parameter control deviation reset system enables precise control of complex systems, real-time monitoring and correction of deviations, and stable system operation. It demonstrates significant advantages in various application areas, such as improving industrial production efficiency and product quality, optimizing traffic flow, and improving energy efficiency, helping the system maintain optimal performance and better achieve its intended goals. This solves the problem of difficulty in implementing parameter control to address system operation deviations in complex multi-parameter systems.
[0066] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.
Claims
1. Multi-parameter control deviation reset system, characterized by: include: A parameter space definition module, which is used to determine multiple system-related parameters and set the value range of each parameter; a particle swarm optimization algorithm module, the particle swarm optimization algorithm module being connected to the parameter space definition module and performing calculations based on the parameters and value ranges determined by the parameter space definition module to obtain an optimal parameter combination; A deviation detection and evaluation module, which is connected to the system operation state and is used to detect the deviation between the actual system operation state and the expected state; A parameter reset module is connected to the particle swarm optimization algorithm module and the deviation detection and evaluation module respectively, and is used to reset system parameters according to the optimal parameter combination obtained by the particle swarm optimization algorithm module to reduce system deviation.
2. The multi-parameter control deviation resetting system according to claim 1, characterized in that: The parameters determined by the parameter space definition module are multiple adjustable parameters based on the physical characteristics of the system and actual operation requirements.
3. The multi-parameter control deviation reset system according to claim 1, characterized in that: The particle swarm optimization algorithm module includes a particle initialization submodule, which is used to randomly generate multiple particles within the range of parameter values determined by the parameter space definition module. The position of each particle represents a set of parameter combinations, and each particle has an initial velocity.
4. The multi-parameter control deviation resetting system according to claim 1, characterized in that: The particle swarm optimization algorithm module also includes a fitness evaluation submodule, which is used to design a fitness function based on the operating objectives expected to be achieved by the system, and evaluate the parameter combination represented by each particle through the fitness function to obtain the fitness value corresponding to each particle. The fitness value is used to measure the impact of each parameter combination on system performance.
5. The multi-parameter control deviation resetting system according to claim 1, characterized in that: The particle swarm optimization algorithm module also includes an individual and global optimal update submodule, which is used to compare the current fitness value of each particle with its historical optimal fitness value. If the current fitness value is better, the individual historical optimal position of the particle is updated, and the particle with the best fitness value is determined among all particles, and its position is used as the global optimal position. The global optimal position represents the optimal parameter combination found by the current particle swarm.
6. The multi-parameter control deviation resetting system according to claim 1, characterized in that: The particle swarm optimization algorithm module also includes a particle speed and position update submodule, which is used to update the particle speed and position according to the following formula based on the particle's current position, speed, individual historical best position and global best position: i (t+1)=w×v i (t)+c1×rand()×(pbest i -x i (t))+c2×rand()×(gbest-x i (t)), where v i (t) represents the velocity of particle i at time t, x i (t) represents the position of particle i, w is the inertia weight, c1 and c2 are learning factors, rand() is a function that generates random numbers between (0,1), pbest i is the individual best position of particle i, gbest is the global best position of the entire particle swarm; the position update formula is: i (t+1)=x i (t)+v i (t+1), where x i (t+1) represents the position of particle i at time t+1, x i (t) represents the position of particle i at time t, v i (t+1) represents the speed of particle i at time t+1. The speed and position of the particle are updated through continuous iteration until the preset stopping condition is met. The stopping condition is that the change in fitness value is less than the set threshold or the convergence degree of the particle swarm reaches the set standard.
7. The multi-parameter control deviation resetting system according to claim 1, characterized in that: The deviation detection and evaluation module is used to collect the actual operation output data of the system in real time, compare the actual operation output data with the pre-set expected output data, and calculate the deviation value using a preset deviation measurement indicator. The deviation measurement indicator includes but is not limited to the mean square error or the mean absolute error, so as to determine whether the deviation value exceeds the allowable range pre-set by the system.
8. The multi-parameter control deviation resetting system according to claim 1, characterized in that: After receiving the trigger signal from the deviation detection and evaluation module indicating that the deviation exceeds the allowable range, the parameter reset module adjusts the current parameters of the system to the parameter combination corresponding to the global optimal position determined by the particle swarm optimization algorithm module, thereby changing the operating state of the system and reducing the system deviation.
9. A multi-parameter control deviation reset method, characterized in that: The multi-parameter control deviation resetting system according to any one of claims 1 to 8 comprises the following steps: S1. Determine multiple system-related parameters and value ranges through the parameter space definition module; S2, using the particle swarm optimization algorithm module to perform particle initialization, fitness evaluation, individual and global optimal updates, and particle velocity and position update operations within the value range to obtain the optimal parameter combination; S3, using the deviation detection and evaluation module to detect the deviation between the actual operating state of the system and the expected state in real time; S4. When the deviation exceeds the allowable range, the parameter reset module resets the system parameters according to the optimal parameter combination to regulate and reset the system deviation.