An adaptive water tank temperature control method based on monte carlo simulation

Abstract

The application belongs to the technical field of household appliances, and discloses a self-adaptive water tank temperature control method based on Monte Carlo simulation. The application predicts a future heat dissipation disturbance path through Monte Carlo simulation, combines an optimization algorithm to solve an optimal power sequence, and realizes self-adaptive control on the water tank temperature. The application can online self-adaptively adjust control parameters, realizes high-precision temperature control within a range of ±0.2℃ under the premise of not depending on a physical model, automatically finds a globally optimal control configuration, and improves control efficiency and energy efficiency.

Description

Technical Field

[0001] This invention relates primarily to the field of household appliance technology, and in particular to an adaptive water tank temperature control method based on Monte Carlo simulation. Background Technology

[0002] Currently, the temperature control technologies widely used in the home appliance and scientific research fields are traditional PID temperature control (proportional-integral-derivative) and model predictive control (MPC). Traditional PID temperature control has been widely adopted due to its simple structure and ease of implementation; however, its control performance is highly dependent on an accurate system model. In real-world temperature control scenarios, the heat dissipation rate fluctuates randomly with changes in wind speed and room temperature. Under such circumstances, traditional PID controllers may experience overshoot, oscillation, or control failure due to model mismatch. Furthermore, the control parameters of traditional PID controllers often rely on manual tuning, which is time-consuming and difficult to adapt to dynamic environments. Model predictive control (MPC) can handle multivariable coupling and process constraints; however, it also highly depends on an accurate system model. When the thermal conductivity and ambient temperature are uncertain in the actual environment, the output of MPC will deviate significantly from the actual thermal behavior, leading to accumulated prediction errors and decreased control performance.

[0003] In summary, traditional PID temperature control technology and model predictive control have the following drawbacks: (1) they cannot cope with the randomness and nonlinearity of heat dissipation rate; (2) the parameters need to be manually tuned and lack automation and global optimization capabilities; and (3) they lack quantitative evaluation and closed-loop optimization mechanisms for controller performance. Summary of the Invention

[0004] To address the problems of existing temperature control technologies, such as inability to handle random heat dissipation disturbances, reliance on manual parameter tuning, and low control effort, this invention provides an adaptive water tank temperature control method based on Monte Carlo simulation. Under conditions of random changes in the environmental heat dissipation rate, it utilizes Monte Carlo uncertainty modeling to achieve online adaptive adjustment of control parameters; it achieves high-precision temperature control within a range of ±0.2℃ without relying on a physical model; and it automatically seeks the globally optimal control configuration, improving control efficiency and energy efficiency.

[0005] To achieve the above objectives, this invention provides an adaptive water tank temperature control method based on Monte Carlo simulation, comprising the following steps:

[0006] Step 1: Construct the heat balance equation;

[0007] Step 2: Establish the Monte Carlo perturbation model:

[0008] Step 2.1: Initialize the Monte Carlo perturbation model parameters, including the number of Monte Carlo simulations. Prediction time domain ;

[0009] Step 2.2: Sample the disturbance value of the heat dissipation power. The disturbance value of the heat dissipation power is within the range of... ,in, Let be the perturbation value of the heat dissipation power at time t in the i-th sampling. The standard deviation of the disturbance. The Gaussian distribution followed by the perturbation value of heat dissipation power;

[0010] Step 2.3: Calculate the actual heat dissipation power using the Monte Carlo perturbation model. The formula for the actual heat dissipation power is: ,in, Let be the actual heat dissipation power at time t in the i-th sampling. As the reference heat dissipation power, Let be the perturbation value of the heat dissipation power at time t in the i-th sampling;

[0011] Step 3: Establish and optimize the objective function To minimize the objective function, where Let be the temperature at time t in the i-th simulation. To simulate the start time, The time for the simulation to end, For the target temperature, For power control sequence, Let be the objective function. The heating power of the j-th control segment is... The heating power of the (j+1)th control segment, The number of segments, This is the temperature oscillation penalty coefficient, used to balance temperature control accuracy and power stability;

[0012] Step 4: Control the segment interval Number of segments and power control sequence Implement a loop in the Monte Carlo perturbation model;

[0013] Step 5: In each Update the standard deviation of the disturbance after each control cycle. Complete online learning to achieve temperature control. It is a constant.

[0014] Furthermore, in step 4, the segment interval is controlled. Number of segments and power control sequence Implementing a loop to control the Monte Carlo perturbation model includes the following steps:

[0015] Step 4.1: Read the current state of the Monte Carlo perturbation model and obtain the water tank temperature. and simulation start time ;

[0016] Step 4.2: Generate the perturbation path: in the prediction time domain Intrinsic generation and Monte Carlo simulation times Equal number , This represents the actual heat dissipation power at time t in the i-th sampling.

[0017] Step 4.3: Optimize the power control sequence For simulation number i, solve Make the objective function Take the minimum value. To make the objective function The power control sequence with the smallest value;

[0018] Step 4.4: Output the first stage power: The first paragraph As a control command;

[0019] Step 4.5: Online Adjustment: If power fluctuations are detected for h consecutive control cycles that are greater than or equal to 5% of the average heating power, the temperature oscillation penalty coefficient is increased. The average heating power is The average power of each execution, where h is a constant.

[0020] Furthermore, in step 5, the standard deviation of the disturbance... Represented as ,in, The number of control cycles involved in the calculation. This represents the actual heat dissipation power during the k-th control cycle. In order to be in Average heat dissipation power during each control cycle.

[0021] Furthermore, the heat balance equation in step 1 is: ,in, For system heat capacity, For water tank temperature, For heating power, For randomly varying heat dissipation power, This represents the rate of change of the water tank temperature.

[0022] Beneficial Effects: Compared to existing technologies, this invention provides an adaptive water tank temperature control method based on Monte Carlo simulation. It predicts future heat dissipation disturbance paths through Monte Carlo simulation and combines this with an optimization algorithm to solve for the optimal power sequence, thereby achieving adaptive control of the water tank temperature. This invention can adaptively adjust control parameters online; achieve high-precision temperature control within a range of ±0.2℃ without relying on a physical model; and automatically find the globally optimal control configuration, improving control efficiency and energy efficiency. Attached Figure Description

[0023] Figure 1 This is a flowchart of an adaptive water tank temperature control method based on Monte Carlo simulation, which is involved in an embodiment of the present invention. Detailed Implementation

[0024] like Figure 1 As shown, this invention provides an adaptive water tank temperature control method based on Monte Carlo simulation. This scheme predicts future heat dissipation disturbance paths through Monte Carlo simulation and solves the optimal power sequence using optimization algorithms to achieve adaptive control of the water tank temperature. Key innovations include: (1) Disturbance modeling: using Gaussian processes or historical data to generate random heat dissipation paths and quantify uncertainties. (2) Multi-objective optimization: objective function Simultaneously minimizing temperature error and suppressing power fluctuations, by applying a temperature oscillation penalty coefficient. Balancing the two. (3) Online learning: dynamically updating the standard deviation of the disturbance. Adapt to environmental changes.

[0025] Example 1: This embodiment of the invention provides an adaptive water tank temperature control method based on Monte Carlo simulation, comprising the following steps:

[0026] Step 1: Construct the heat balance equation, which is as follows: ,in, For system heat capacity, For water tank temperature, For heating power, For randomly varying heat dissipation power, This represents the rate of change of the water tank temperature, and can also be understood as the derivative of the water tank with respect to time.

[0027] Step 2: Establish the Monte Carlo perturbation model:

[0028] Step 2.1: Initialize the Monte Carlo perturbation model parameters, including the number of Monte Carlo simulations. Prediction time domain ;

[0029] Step 2.2: Sample the disturbance value of the heat dissipation power. The disturbance value of the heat dissipation power is within the range of... ,in, Let be the perturbation value of the heat dissipation power at time t in the i-th sampling. The standard deviation of the disturbance. The Gaussian distribution followed by the perturbation value of heat dissipation power;

[0030] Step 2.3: Calculate the actual heat dissipation power using the Monte Carlo perturbation model. The formula for the actual heat dissipation power is: ,in, Let be the actual heat dissipation power at time t in the i-th sampling. As the reference heat dissipation power, Let be the perturbation value of the heat dissipation power at time t in the i-th sampling.

[0031] Step 3: Establish and optimize the objective function To minimize the objective function, where Let be the temperature at time t in the i-th simulation. To simulate the start time, The simulation ends at time. For the target temperature, For power control sequence, Let be the objective function. The heating power of the j-th control segment is... The heating power of the (j+1)th control segment, The number of segments, This is the temperature oscillation penalty coefficient, used to balance temperature control accuracy and power stability.

[0032] Step 4: Control the segment interval Number of segments and power control sequence Implement a loop in the Monte Carlo perturbation model;

[0033] Step 4.1: Read the current state of the Monte Carlo perturbation model and obtain the water tank temperature. and simulation start time ;

[0034] Step 4.2: Generate the perturbation path: in the prediction time domain Intrinsic generation and Monte Carlo simulation times Equal number , This represents the actual heat dissipation power at time t in the i-th sampling.

[0035] Step 4.3: Optimize the power control sequence For simulation number i, solve Make the objective function Take the minimum value. To make the objective function The power control sequence with the smallest value;

[0036] Step 4.4: Output the first stage power: The first paragraph As a control command;

[0037] Step 4.5: Online Adjustment: If power fluctuations are detected for h consecutive control cycles that are greater than or equal to 5% of the average heating power, the temperature oscillation penalty coefficient is increased. The average heating power is The average power of each execution, where h is a constant.

[0038] Step 5: In each Update the standard deviation of the disturbance after each control cycle. Complete online learning to achieve temperature control. It is a constant. The standard deviation of the disturbance in step 5. Represented as ,in, The number of control cycles involved in the calculation. This represents the actual heat dissipation power during the k-th control cycle. In order to be in Average heat dissipation power during each control cycle.

[0039] Example 2: This example is basically the same as Example 1, except that,

[0040] This invention provides an adaptive water tank temperature control method based on Monte Carlo simulation, comprising the following steps:

[0041] (1) System parameter settings: system heat capacity initial temperature Target temperature The heat balance equation is .

[0042] (2) Initialize the Monte Carlo perturbation model parameters:

[0043] Monte Carlo simulation times 100 times;

[0044] Prediction Time Domain It lasts for 30 seconds;

[0045] Reference heat dissipation power It is 100W;

[0046] The formula for actual heat dissipation power is: .

[0047] (3) Establish and optimize the objective function To minimize the objective function, where Let be the temperature at time t in the i-th simulation. To simulate the start time, For power control sequence, Let be the objective function. The heating power of the j-th control segment is... The heating power of the (j+1)th control segment, This is the temperature oscillation penalty coefficient, used to balance temperature control accuracy and power stability;

[0048] Segmented intervals 1 second; number of segments 30; power control sequence Power range .

[0049] (4) Control segment interval Number of segments and power control sequence Implement a loop in the Monte Carlo perturbation model;

[0050] Step 4.1: Read the current state of the Monte Carlo perturbation model and obtain the water tank temperature. and simulation start time ;

[0051] Step 4.2: Generate the perturbation path: in the prediction time domain Intrinsic generation and Monte Carlo simulation times Equal number , This represents the actual heat dissipation power at time t in the i-th sampling.

[0052] Step 4.3: Optimize the power control sequence For simulation number i, solve Make the objective function Take the minimum value. To make the objective function The power control sequence with the smallest value;

[0053] Step 4.4: Output the first stage power: The first paragraph As a control command;

[0054] Step 4.5: Online Adjustment: If power fluctuations are detected for h consecutive control cycles that are greater than or equal to 5% of the average heating power, the temperature oscillation penalty coefficient is increased. The average heating power is The average power of each execution, h is a constant. In this embodiment, the value of h is set manually and can be less than or equal to... . There are no fixed requirements or methods for improvement; they need to be tailored to the specific usage scenario.

[0055] (5) After every 10 control cycles, update the standard deviation of the disturbance based on historical samples. Complete online learning to achieve temperature control.

[0056] This invention provides an adaptive water tank temperature control method based on Monte Carlo simulation. It predicts future heat dissipation disturbance paths using Monte Carlo simulation and combines this with an optimization algorithm to solve for the optimal power sequence, thereby achieving adaptive control of the water tank temperature. This invention can adaptively adjust control parameters online; achieve high-precision temperature control within a range of ±0.2℃ without relying on a physical model; and automatically find the globally optimal control configuration, improving control efficiency and energy efficiency.

[0057] Finally, it should be noted that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. However, any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive water tank temperature control method based on Monte Carlo simulation, characterized in that, Includes the following steps: Step 1: Construct the heat balance equation; Step 2: Establish the Monte Carlo perturbation model: Step 2.1: Initialize the Monte Carlo perturbation model parameters, including the number of Monte Carlo simulations. Prediction time domain ; Step 2.2: Sample the disturbance value of the heat dissipation power. The disturbance value of the heat dissipation power is within the range of... ,in, Let be the perturbation value of the heat dissipation power at time t in the i-th sampling. The standard deviation of the disturbance. The Gaussian distribution followed by the perturbation value of heat dissipation power; Step 2.3: Calculate the actual heat dissipation power using the Monte Carlo perturbation model. The formula for the actual heat dissipation power is: ,in, Let be the actual heat dissipation power at time t in the i-th sampling. As the reference heat dissipation power, Let be the perturbation value of the heat dissipation power at time t in the i-th sampling; Step 3: Establish and optimize the objective function To minimize the objective function, where Let be the temperature at time t in the i-th simulation. To simulate the start time, The simulation ends at time. For the target temperature, For power control sequence, Let be the objective function. The heating power of the j-th control segment is... The heating power of the (j+1)th control segment, The number of segments, This is the temperature oscillation penalty coefficient, used to balance temperature control accuracy and power stability; Step 4: Control the segment interval Number of segments and power control sequence Implement a loop in the Monte Carlo perturbation model; Step 5: In each Update the standard deviation of the disturbance after each control cycle. Complete online learning to achieve temperature control. It is a constant.

2. The adaptive water tank temperature control method based on Monte Carlo simulation according to claim 1, characterized in that, In step 4, the segment interval is controlled. Number of segments and power control sequence Implementing a loop to control the Monte Carlo perturbation model includes the following steps: Step 4.1: Read the current state of the Monte Carlo perturbation model and obtain the water tank temperature. and simulation start time ; Step 4.2: Generate the perturbation path: in the prediction time domain Intrinsic generation and Monte Carlo simulation times Equal number , This represents the actual heat dissipation power at time t in the i-th sampling. Step 4.3: Optimize the power control sequence For simulation number i, solve Make the objective function Take the minimum value. To make the objective function The power control sequence with the smallest value; Step 4.4: Output the first stage power: The first paragraph As a control command; Step 4.5: Online Adjustment: If power fluctuations are detected for h consecutive control cycles that are greater than or equal to 5% of the average heating power, the temperature oscillation penalty coefficient is increased. The average heating power is The average power of each execution, where h is a constant.

3. The adaptive water tank temperature control method based on Monte Carlo simulation according to claim 1, characterized in that, Standard deviation of disturbance in step 5 Represented as ,in, The number of control cycles involved in the calculation. This represents the actual heat dissipation power during the k-th control cycle. In order to be in Average heat dissipation power during each control cycle.

4. The adaptive water tank temperature control method based on Monte Carlo simulation according to claim 1, characterized in that, The heat balance equation in step 1 is: ,in, For the system heat capacity, For water tank temperature, For heating power, For randomly varying heat dissipation power, This represents the rate of change of the water tank temperature.

Images