A method and system for regulating the air volume of a heating air outlet fan

Abstract

The present application belongs to the field of HVAC intelligent control technology, and particularly relates to a heating air outlet fan air volume regulation method and system, comprising: obtaining indoor set temperature and current temperature, and calculating current temperature deviation and temperature deviation change rate; calculating heating space thermal response characteristic parameters from logs, i.e. the weighted average time length of temperature deviation from the first positive deviation threshold to the second positive deviation threshold; determining the basic interval by a nonlinear function, and constructing a membership function by scaling the boundary according to the temperature deviation change rate, expanding for the same sign and contracting for the different sign; determining the weight by querying a two-dimensional rule weight mapping table in combination with parameters, temperature deviation and temperature deviation change rate; after determining the upper limit of the air volume domain by outdoor environment temperature and heat source temperature, generating an air volume control instruction by fuzzy reasoning and defuzzification; the present application ensures that the heating air outlet fan can maintain high-precision tracking performance, cold air prevention safety performance and energy saving effect under complex house types and variable climate.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for heating, ventilation, and air conditioning (HVAC). More specifically, this invention relates to a method and system for regulating the airflow of a heating fan. Background Technology

[0002] With increasing demands for improved indoor thermal comfort in buildings, the airflow regulation performance of heating fans, as key terminal heating devices, directly impacts the accuracy of indoor temperature control and user comfort. In heated spaces with different functions, such as residences and offices, the airflow output of the fans needs to be adjusted in real-time based on actual heat load changes to achieve and maintain the set temperature. However, differences in building structure, area, orientation, and thermal insulation performance of the building envelope lead to drastically different thermal response characteristics in various heated spaces, posing a challenge to the universality and effectiveness of airflow regulation strategies. Heating fan airflow regulation refers to the process of calculating and controlling the output airflow of the fans in real-time based on temperature deviations and relevant environmental parameters.

[0003] Currently, most common heating fan exhaust systems employ conventional control strategies based on fixed parameters or basic fuzzy control strategies. These strategies primarily adjust the airflow based on the single difference between the current indoor temperature and the set temperature. The membership function boundary intervals and fuzzy rule weights of the fuzzy control component are typically set to fixed values ​​at the factory or during initial commissioning and remain unchanged during subsequent use. Furthermore, this control strategy does not consider outdoor ambient temperature and heating status as feedforward constraints when determining the output airflow range.

[0004] The main drawbacks of existing technologies in practical applications are: conventional adjustment methods do not fully utilize the response patterns formed by each room during historical heating cycles, making it difficult to adapt to the thermal inertia of different spaces. Existing fuzzy control-type airflow regulation systems also suffer from insufficient flexibility in actual operation; their membership function boundary intervals and fuzzy rule weights are usually fixed at the factory or during initial commissioning, lacking a mechanism for real-time correction based on temperature deviations and their trends. Furthermore, existing control strategies typically do not adequately integrate with outdoor ambient temperature and heating status, making it difficult to promptly remove airflow restrictions to ensure heating demand under extreme low-temperature conditions, while under normal weather conditions, the lack of an upper limit constraint on the airflow domain may result in ineffective high-airflow operation of the fans. Summary of the Invention

[0005] To address the technical problems of existing heating fan airflow regulation strategies being unable to adapt to different spatial thermal inertia, having poor flexibility due to fixed fuzzy rules, and failing to fully integrate outdoor environment and heating status for comprehensive constraints, resulting in low temperature control accuracy, easy generation of cold air blowing or ineffective energy consumption, this invention provides a heating fan airflow regulation method and system.

[0006] In a first aspect, the present invention provides a method for adjusting the air volume of a heating exhaust fan, comprising: S1: acquiring the indoor set temperature, the current temperature, the outdoor ambient temperature, and the heat source temperature; calculating the current temperature deviation and the rate of change of the temperature deviation; calculating the thermal response characteristic parameters of the heating space based on historical temperature log data, wherein the thermal response characteristic parameters of the heating space are the weighted average duration during which the current temperature deviation decreases from a first positive deviation threshold to a second positive deviation threshold in a historical heating cycle; S2: inputting the thermal response characteristic parameters of the heating space into a preset nonlinear function to calculate the basic distribution interval of the fuzzy subset of the temperature deviation; calculating an adjustment coefficient based on the rate of change of the temperature deviation to scale and adjust the boundary points of the basic distribution interval; and dynamically constructing the membership function of the current temperature deviation. The function calls the preset membership function of the temperature deviation change rate; where, when the current temperature deviation and the temperature deviation change rate have the same sign, the adjustment coefficient is used to expand the boundary points; when they have opposite signs, the adjustment coefficient is used to shrink the boundary points; S3: Using the thermal response characteristic parameters of the heating space and the current temperature deviation as input, the two-dimensional rule weight mapping table is queried, and the target weight coefficient of the corresponding fuzzy rule is selected in combination with the temperature deviation change rate; the upper limit of the universe of discourse of the output air volume fuzzy subset is determined according to the outdoor ambient temperature and the heat source temperature, so as to form different upper limits of air volume under low temperature heating, conventional heating and insufficient heat source temperature conditions; using the membership function, target weight coefficient and upper limit of the universe of discourse, fuzzy inference and defuzzification calculation are performed to generate air volume control instructions.

[0007] By employing the above technical solutions, this invention extracts thermal response characteristic parameters using real historical single-heating time, enabling the system to naturally perceive and assess the inherent thermal inertia of the current heating space. Simultaneously, it dynamically expands or contracts the fuzzy membership boundary using nonlinear functions and the sign of the deviation change rate, matching dynamic weights in a two-dimensional mapping table. This breaks the deadlock of traditional fuzzy parameters being fixed at the factory, allowing the system to expand the boundary to suppress overshoot when the temperature diverges and contract the boundary to enhance approximation sensitivity when the temperature converges. Furthermore, by introducing outdoor environment and heat source temperature to dynamically truncate the upper limit of the airflow domain in three levels, a feedforward constraint barrier independent of feedback control is constructed. These specific technical features are deeply integrated, systematically overcoming the shortcomings of existing technologies such as poor adaptability to different spatial thermal inertia, rigid control, and lack of environmental safety coordination. This enables high-precision, cold-wind-proof, and energy-saving temperature tracking control under various climates and apartment types.

[0008] Preferably, the step of calculating the thermal response characteristic parameters of the heating space based on historical temperature log data includes: extracting historical temperature log data for the most recent preset number of heating cycles; obtaining a first timestamp in each heating cycle when the current temperature deviation reaches the first positive deviation threshold, and a second timestamp when it drops to the second positive deviation threshold, and using the difference between the second timestamp and the first timestamp as the time consumed for a single heating cycle; wherein the first positive deviation threshold is greater than the second positive deviation threshold; and multiplying the time consumed for a single heating cycle in each heating cycle by the corresponding attenuation weight and summing them according to the time order to obtain the thermal response characteristic parameters of the heating space.

[0009] Preferably, the step of multiplying the single heating time of each heating cycle by the corresponding attenuation weight and summing them according to the time sequence to obtain the heating space thermal response characteristic parameter further includes: when there is a cycle in the most recent preset number of heating cycles that does not completely cross the first positive deviation threshold and the second positive deviation threshold, the corresponding cycle is removed as an invalid cycle; when the number of valid cycles is less than the preset number, the attenuation weight corresponding to the valid cycle is normalized, and the weighted average of the existing valid cycles is calculated as the heating space thermal response characteristic parameter; if there is no valid cycle, the heating space thermal response characteristic parameter is set as a preset reference duration.

[0010] Preferably, the step of inputting the thermal response characteristic parameters of the heating space into a preset nonlinear function to calculate the basic distribution interval of the fuzzy subset of temperature deviation includes: obtaining the initial value of the core fixed point coordinates of the built-in basic universe of discourse; calculating the natural logarithm of the ratio of the thermal response characteristic parameters of the heating space to the preset reference time by a preset expansion coefficient, and adding a preset compensation offset constant to obtain the calculated value of the interval radius; taking the maximum value between the calculated value of the interval radius and the preset positive lower limit radius value as the final interval radius increment; adding and subtracting the final interval radius increment from the initial value of the core fixed point coordinates of the basic universe of discourse to obtain the boundary coordinates of each fuzzy subset as the basic distribution interval.

[0011] By employing the above technical solution, this invention utilizes the nonlinear characteristics of the natural logarithm function to calculate the interval radius increment, coupled with a positive lower limit radius protection. This mapping method ensures that the range of the fuzzy subset can be smoothly widened under long thermal response conditions, suppressing control oscillations; at the same time, it eliminates boundary inversion or universe of discourse coverage gaps caused by calculating negative values, ensuring the security of the algorithm logic.

[0012] Preferably, the step of calculating the adjustment coefficient based on the temperature deviation change rate to scale and adjust the boundary points of the basic distribution interval includes: when the product of the current temperature deviation and the temperature deviation change rate is greater than 0, using the sum of the product of a first preset coefficient and the absolute value of the temperature deviation change rate plus 1 as the adjustment coefficient to expand the boundary points of the basic distribution interval; when the product of the current temperature deviation and the temperature deviation change rate is less than 0, comparing the difference between 1 and the product of the first preset coefficient and the absolute value of the temperature deviation change rate with a lower threshold, taking the maximum value as the adjustment coefficient to shrink the boundary points of the basic distribution interval; when the product of the current temperature deviation and the temperature deviation change rate is equal to 0, setting the adjustment coefficient to 1; and using the adjustment coefficient to perform product scaling processing on the difference between the boundary coordinates of the basic distribution interval and the initial values ​​of the corresponding core fixed point coordinates.

[0013] By adopting the above technical solution, this invention performs scaling adjustment based on the positive or negative product of deviation and rate of change, and employs a differential scaling mechanism relative to the core point. This achieves smooth output when the deviation diverges and improved sensitivity during convergence, and geometrically avoids subset center drift caused by overall scaling of non-zero core points, thus improving the stability of the control process.

[0014] Preferably, the step of querying the two-dimensional rule weight mapping table and selecting the target weight coefficient of the corresponding fuzzy rule in conjunction with the temperature deviation change rate includes: matching the corresponding row coordinates according to the absolute value of the current temperature deviation, and matching the corresponding column coordinates according to the thermal response characteristic parameters of the heating space; querying the two-dimensional rule weight mapping table to obtain the corresponding rule weight vector; and extracting the corresponding coefficient from the rule weight vector as the target weight coefficient according to the fuzzy level with the largest membership degree of the temperature deviation change rate.

[0015] By adopting the above technical solution, this invention uses a two-dimensional rule-based weight mapping table to orthogonally matrix-fuse transient deviation states with the static thermal inertia characteristics of buildings. This enables the system to accurately select matching activation weights for complex cross-conditions with extremely low computational overhead, solving the problem of slow or overly sensitive adjustment response of conventional fuzzy control under extreme apartment layouts.

[0016] Preferably, determining the upper limit of the universe of discourse for the fuzzy subset of output airflow based on the outdoor ambient temperature and the heat source temperature includes: if the heat source temperature is lower than a preset anti-cold airflow threshold, then setting the upper limit of the universe of discourse to a preset low-level airflow value; if the heat source temperature is not lower than the preset anti-cold airflow threshold and the outdoor ambient temperature is lower than a preset low-temperature threshold, then setting the upper limit of the universe of discourse to the rated maximum airflow of the device; if the heat source temperature is not lower than the preset anti-cold airflow threshold and the outdoor ambient temperature is not lower than the preset low-temperature threshold, then setting the upper limit of the universe of discourse to a limit value of the rated maximum airflow of the device multiplied by a preset derating factor.

[0017] By adopting the above technical solution, this invention dynamically truncates the upper limit of the domain of discourse based on multi-temperature zone thresholds. This constructs a feedforward constraint at the logical level, which can not only block the direct blowing of cold air when the heat source is insufficient, but also unlock the power in severe cold and limit the speed operation in normal weather, thus achieving a multi-dimensional balance between heating guarantee, energy saving and noise reduction, and human comfort.

[0018] Preferably, the step of generating airflow control instructions by performing fuzzy inference and defuzzification calculations using membership functions, target weight coefficients, and upper limits of the universe of discourse includes: during the fuzzy inference process, multiplying the target weight coefficients by the derived membership degrees of the corresponding fuzzy rules, and taking the smaller value between the multiplication result and 1 to perform upper limit truncation processing to obtain weighted rule membership degrees; aggregating each fuzzy rule using the maximum value finding method to obtain an aggregated output airflow fuzzy subset; performing defuzzification calculations on the aggregated output airflow fuzzy subset based on the centroid method, generating discrete feature reference values ​​between 0 and the upper limit of the universe of discourse, calculating the sum of the products of all discrete feature reference values ​​and their corresponding membership degrees in the aggregated output airflow fuzzy subset, and dividing by the sum of all membership degrees to generate the airflow control instructions.

[0019] Preferably, after generating the air volume control command, the method further includes: converting the air volume control command into a PWM duty cycle command based on a preset mapping relationship; and sending the PWM duty cycle command to a hardware communication module to control the operating frequency of the blower motor.

[0020] Secondly, the present invention provides a heating fan air volume adjustment system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned heating fan air volume adjustment method is implemented.

[0021] By adopting the above technical solution, a computer program for regulating the air volume of a heating exhaust fan is generated and stored in a memory so that it can be loaded and executed by a processor. A terminal device can then be made based on the memory and the processor for convenient use.

[0022] The technical solution of the present invention has the following beneficial technical effects: This invention extracts spatial thermal response characteristic parameters from historical temperature log data, accurately reflecting the objective differences in insulation performance, spatial structure, and thermal inertia of different heating environments, thereby improving the adaptability of airflow control logic to actual operating scenarios. By pre-setting a nonlinear function and combining it with the temperature deviation change rate, the system achieves adaptive dynamic adjustment of the fuzzy subset boundary of the temperature deviation. When the current temperature deviation shows an increasing trend, the boundary is expanded to enhance the system's smooth adjustment capability; when the temperature deviation shows a converging trend, the boundary is contracted to improve the real-time sensitivity of the control system and reduce the risk of temperature overshoot.

[0023] Meanwhile, by querying a two-dimensional rule weight mapping table, the target weight coefficients of fuzzy rules are dynamically allocated, improving the matching accuracy between fuzzy inference results and the current complex operating conditions. By performing upper limit truncation on the weighted rule membership degree, the occurrence of membership degree values ​​exceeding one due to weight amplification is effectively avoided, thus ensuring the effectiveness of fuzzy set operations and aggregation processes from a mathematical logic perspective. Furthermore, the upper limit of the universe of discourse of the output airflow fuzzy subset is deeply correlated and coordinated with the outdoor ambient temperature and heat source temperature. In frigid climates where the heat source meets heating requirements, the system opens the equipment's rated maximum airflow to fully guarantee indoor heating needs; under normal operating conditions, the upper limit of the universe of discourse is reasonably limited to reduce system energy consumption; if the heat source temperature is detected to be insufficient, the exhaust fan is forcibly restricted to operate in a preset low-speed airflow mode, thereby physically eliminating the impact of direct cold air blowing on user comfort. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating a method for adjusting the air volume of a heating exhaust fan according to the present invention; Figure 2 This is a schematic diagram of the current temperature deviation fuzzy subset membership function distribution; Figure 3 This is a schematic diagram of the upper limit feedforward constraint of the universe of discourse of the fuzzy subset of output air volume. Detailed Implementation

[0025] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0026] This invention discloses a method for adjusting the air volume of a heating exhaust fan, referring to... Figure 1 This includes steps S1-S3: S1: Obtain environmental parameters and calculate thermal response characteristic parameters.

[0027] In an optional embodiment, the indoor set temperature, current temperature, outdoor ambient temperature and heat source temperature are obtained, the current temperature deviation and temperature deviation change rate are calculated, and the heating space thermal response characteristic parameters are calculated based on historical temperature log data. The heating space thermal response characteristic parameters are the weighted average time during which the current temperature deviation decreases from a first positive deviation threshold to a second positive deviation threshold in the historical heating cycle.

[0028] In practice, the system first reads data from indoor temperature sensors located within the heated space via the heating equipment controller to obtain the current temperature; simultaneously, it reads data from outdoor temperature sensors located outdoors or on the air intake side to obtain the outdoor ambient temperature. At the same time, the system reads temperature sensor data from the heat source heat exchanger, heating coil, PTC heating element, or hot water heat exchange module as the heat source temperature, and then reads the indoor set temperature from the human-machine interface module of the wall thermostat, equipment control panel, or remote control terminal. Subsequently, the system subtracts the current temperature from the indoor set temperature to obtain the current temperature deviation, and calculates the difference in current temperature deviation between two adjacent sampling periods, dividing it by the sampling time interval to obtain the temperature deviation change rate. If the system is in its first power-on or cold-start state, since there is no historical temperature deviation from the previous sampling period in the cache, the system directly sets the initial value of the temperature deviation change rate to zero.

[0029] Based on this, the system extracts historical temperature log data from the most recent preset number of heating cycles through local storage, edge controller database, or device operation log cache. From this data, it filters and extracts the time sequence during which the current temperature deviation decreases from a first positive deviation threshold to a second positive deviation threshold in each cycle. Next, the system applies an exponential moving average algorithm to the time sequence to calculate a weighted average duration. This algorithm assigns a greater smoothing weight to cycles closer to the current time, thereby calculating parameters representing the thermal response characteristics of the heating space. These parameters reflect the differences in insulation performance, spatial structure, and thermal inertia of different heating environments, enabling subsequent airflow control logic to better align with actual operating scenarios.

[0030] In an optional embodiment, the specific process of calculating the thermal response characteristic parameters of the heating space based on historical temperature log data includes: the system extracts historical temperature log data for the most recent preset number of heating cycles; obtains the first timestamp when the current temperature deviation reaches the first positive deviation threshold and the second timestamp when it drops to the second positive deviation threshold in each heating cycle, and uses the difference between the second timestamp and the first timestamp as the time consumed for a single heating cycle; wherein the first positive deviation threshold is greater than the second positive deviation threshold; and multiplies the time consumed for a single heating cycle in each heating cycle by the corresponding attenuation weight and sums them up according to the time order to obtain the thermal response characteristic parameters of the heating space.

[0031] More specifically, preferably, the preset quantity can be 5, meaning the system retrieves historical temperature log data of the target heating device for the most recent 5 complete heating cycles through the device's local storage, edge controller database, or operation log cache. The first positive deviation threshold ranges from 1.5℃ to 2.5℃, and the second positive deviation threshold ranges from 0.2℃ to 0.8℃. Preferably, the first positive deviation threshold can be set to 2℃, and the second positive deviation threshold can be set to 0.5℃. The determination of these thresholds is based on the fact that the human body's sensitivity to fluctuations in ambient temperature is typically around 0.5℃, while the initial temperature difference in conventionally heated spaces is generally above 2℃. Using this range as the thermal response evaluation segment can most accurately reflect the room's temperature rise characteristics within the human body's sensitive zone. The system iterates through the log data to determine the first timestamp when the current temperature deviation first reaches 2℃ in each heating cycle, for example, 08:00:00 on December 1, 2023, and the second timestamp when the system reduces the temperature deviation to 0.5℃, for example, 08:04:00 on December 1, 2023. The difference between these two timestamps on the timeline yields the time required for a single temperature rise in that cycle, which is 240 seconds in this example.

[0032] Assume the time taken for a single temperature increase in the last 5 cycles is recorded in chronological order from most recent to oldest. , , , , .

[0033] After obtaining this set of single-cycle heating time sequences, the system uses an exponentially weighted moving average algorithm to calculate parameters representing the thermal response characteristics of the heating space. In one implementation, the decay weights for the first to fifth periods can be 0.4, 0.3, 0.15, 0.1, and 0.05, respectively. The selection of these decay weights is based on a historical data forgetting factor fitting curve using the least squares method, which ensures that the cumulative weight of the most recent three periods reaches 0.85, thereby filtering out single, accidental fluctuations while maximizing the retention of the impact of recent environmental changes on thermal inertia.

[0034] Specifically, the thermal response characteristic parameters of the heated space satisfy the following relationship:

[0035] in, Indicates the thermal response characteristics of the heated space; to These represent the time taken for a single heating cycle in the five most recent heating cycles, extracted in chronological order from most recent to oldest.

[0036] Substitute the previously extracted example values ​​into the calculation, parameters This weighting mechanism logically assigns a larger weighting coefficient to the most recent heating cycle, thus representing the impact of recent window and door opening / closing status, outdoor wind speed changes, changes in the number of people in the heated space, or changes in the building envelope's heat load on the current thermal inertia of the heated space. The system will output parameters... Stored in the cache as an independent variable that determines the system's adaptability in the fuzzy control strategy.

[0037] In an optional embodiment, the process of multiplying the single heating time of each heating cycle by the corresponding attenuation weight and summing them according to the time sequence further includes: when there is a cycle in the most recent preset number of heating cycles that has not fully crossed the first positive deviation threshold and the second positive deviation threshold, the system will remove the corresponding cycle as an invalid cycle; when the number of valid cycles is less than the preset number, the system will normalize the attenuation weight corresponding to the valid cycle and calculate the weighted average of the existing valid cycles as the heating space thermal response characteristic parameter; if there is no valid cycle, the system will set the heating space thermal response characteristic parameter to a preset reference duration.

[0038] More specifically, when there are fewer than 5 valid heating cycles in the historical logs, the system only performs weighted calculations on valid cycles that complete the decline process from the first positive deviation threshold to the second positive deviation threshold. The specific logic is as follows: the original attenuation weights corresponding to invalid cycles are removed, and the original attenuation weights of the remaining valid cycles are normalized by summing them to ensure that the weights always match the time span of the historical data from the present. For example, if only the two most recent valid cycles exist, the system renormalizes the corresponding attenuation weights of 0.4 and 0.3 to... and For example, if the effective period is not continuous in the time series, such as if only the 1st, 3rd, and 5th periods are effective periods in the last 5 periods, and the data for the 2nd and 4th periods is incomplete, the system extracts the original decay weights of 0.4, 0.15, and 0.05 for the corresponding time positions, divides them by the sum of these three weights, which is 0.6, to obtain the normalized new weights of 0.667, 0.25, and 0.083, and then calculates the parameters based on the normalized weights. If no valid period exists, a preset baseline duration will be used. As a parameter The initial value. Preferably, a preset base duration. A value of 180s can be used, representing the ideal thermal response time of the target heated space obtained under a standard constant temperature test environment. The reason for designing the above-mentioned invalid cycle elimination and weighted normalization mechanism is that, in actual operation, some heating cycles may not fully experience the temperature rise process from the first positive deviation threshold to the second positive deviation threshold due to reasons such as users shutting down the system midway, doors and windows being open for extended periods, or sensor communication interruptions. If these incomplete cycles are included in the weighted calculation without differentiation, it will lead to errors in the thermal response characteristic parameters. The thermal inertia is severely distorted, causing the membership function boundaries and rule weights of subsequent fuzzy control strategies to deviate from the actual thermal inertia characteristics. Through the above processing, the system ensures that the thermal response characteristic parameters are always calculated based on valid and complete heating cycle data, thereby guaranteeing the reliability of adaptive control.

[0039] S2: Construct a membership function based on adaptive adjustment of deviation trend.

[0040] In an optional embodiment, the thermal response characteristic parameters of the heating space are input into a preset nonlinear function to calculate the basic distribution interval of the fuzzy subset of temperature deviation. An adjustment coefficient is calculated based on the rate of change of temperature deviation to scale and adjust the boundary points of the basic distribution interval. The membership function of the current temperature deviation is dynamically constructed, and the membership function of the preset rate of change of temperature deviation is called. When the current temperature deviation and the rate of change of temperature deviation have the same sign, the adjustment coefficient is used to expand the boundary points; when they have opposite signs, the adjustment coefficient is used to shrink the boundary points.

[0041] In the specific execution process, the system first initializes the fuzzy control variables of the current temperature deviation and the rate of change of temperature deviation. In this embodiment, the preset nonlinear function used to determine the basic distribution interval of the fuzzy subset of the current temperature deviation is preferably a natural logarithm analytical function protected by a lower bound. It is understood that, in order to achieve the technical effect of smoothly widening the range of the fuzzy subset under long thermal response conditions, the preset nonlinear function is not limited to the natural logarithm function. Those skilled in the art can also use other nonlinear functions with similar convergent smoothness and decreasing slope mathematical properties, such as square root functions or fractional power functions. Taking the optimal natural logarithm analytical function as an example, specifically, the system will use a weighted average duration parameter The calculated value of the interval radius is used as the independent variable.

[0042] Specifically, the calculated value of the interval radius satisfies the following relationship:

[0043] in, This represents the calculated value of the interval radius; Indicates the preset expansion coefficient; Indicates the thermal response characteristics of the heated space; Indicates the preset baseline duration; This represents the preset compensation offset constant.

[0044] The system then compares the calculated radius of the interval with the preset positive lower limit radius value. Compare the two values ​​and take the maximum value as the final interval radius increment. .

[0045] Positive lower limit radius value This is used to prevent the interval radius increment from being negative or too small, thereby avoiding the inversion of left and right boundaries and the appearance of gaps without membership coverage between adjacent fuzzy subsets.

[0046] The seven fuzzy subsets include negative large, negative medium, negative small, zero, positive small, positive medium, and positive large.

[0047] To avoid large temperature difference inputs exceeding the coverage of the fuzzy universe, the negative large fuzzy subset and the positive large fuzzy subset adopt shoulder-shaped membership functions. The negative large fuzzy subset covers the current temperature deviation input less than the leftmost boundary, and the positive large fuzzy subset covers the current temperature deviation input greater than the rightmost boundary.

[0048] For the rate of change of temperature deviation, the system presets seven fuzzy subsets of the rate of change, namely negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. Each fuzzy subset of the rate of change is constructed using a fixed triangle membership function, and the core fixed point coordinates are calibrated according to the preset sampling period and the thermal response range of the heating space, for example, set to [-0.06, -0.04, -0.02, 0, 0.02, 0.04, 0.06]℃ / s.

[0049] Subsequently, the system extracts the values ​​of the current temperature deviation and the rate of change of temperature deviation, and extracts the sign characteristics of both.

[0050] When the product of the two is greater than 0, the system determines that they have the same sign and calculates the adjustment coefficient.

[0051] Specifically, the adjustment coefficients satisfy the following relationship:

[0052] in, Indicates the adjustment factor; This represents a preset coefficient with the dimension of time per temperature. This represents the absolute value of the rate of change of temperature deviation. This adjustment factor is greater than 1 and is used to expand the boundary points.

[0053] When the product of the two is less than 0, the system determines that they are opposite signs and calculates the adjustment coefficient.

[0054] Specifically, the adjustment coefficients satisfy the following relationship:

[0055] in, Indicates the adjustment factor; This represents the function that takes the maximum value. This represents a preset coefficient with the dimension of time per temperature. Represents the absolute value of the rate of change of temperature deviation; This represents the lower threshold valued at 0. The adjustment factor is less than 1 and is used to shrink boundary points.

[0056] When the product of the two equals 0, the system determines that the current temperature deviation or the rate of change of temperature deviation is in a critical state, and does not classify it as an opposite-sign contraction state, adjusting the coefficient accordingly. Set the value to 1, keeping the boundary points of the current basic distribution interval unchanged.

[0057] When adjusting boundary points, the system employs an offset scaling method relative to the coordinates of their respective core fixed points. This involves first calculating the difference between the boundary coordinates and the corresponding core fixed point coordinates, then multiplying this difference by an adjustment coefficient, and finally adding it to the core fixed point coordinates to obtain the adjusted boundary coordinates. This is different from directly multiplying the boundary point coordinates relative to the global zero point by the adjustment coefficient. This method ensures that each fuzzy subset is scaled around its own core fixed point coordinates, preventing fuzzy subset center drift caused by global multiplicative scaling of non-zero core points. Finally, based on the adjusted boundary point coordinates, the system calls the triangular membership function calculation program embedded in the controller to generate a triangular piecewise variable membership function corresponding to the current temperature deviation.

[0058] In an optional embodiment, the specific process of inputting the thermal response characteristic parameters of the heating space into a preset nonlinear function to calculate the basic distribution interval of the fuzzy subset of temperature deviation includes: the system obtains the initial value of the core fixed point coordinates of the built-in basic universe of discourse; calculates the preset expansion coefficient multiplied by the natural logarithm of the ratio of the thermal response characteristic parameters of the heating space to the preset reference duration, and adds a preset compensation offset constant to obtain the calculated value of the interval radius; takes the maximum value between the calculated value of the interval radius and the preset positive lower limit radius value as the final interval radius increment; adds and subtracts the final interval radius increment from the initial value of the core fixed point coordinates of the basic universe of discourse to obtain the boundary coordinates of each fuzzy subset as the basic distribution interval.

[0059] More specifically, the system pre-configures the initial values ​​of the core point coordinates of the basic universe of discourse representing the current temperature deviation fuzzy universe of discourse in the controller's read-only memory. These core point coordinates correspond to the center positions of the membership functions of seven fuzzy subsets: NB (negative-large), NM (negative-medium), NS (negative-small), ZO (zero), PS (positive-small), PM (positive-medium), and PB (positive-large). In one embodiment, the number of core point coordinates is seven, and their initial coordinate values ​​can be, for example, labeled as follows: , , , , , , The unit is ℃.

[0060] The system obtains parameters representing the thermal response characteristics of the heated space. As the independent variable, the increment of the interval radius is calculated using the built-in analytical formula for the natural logarithm. This analytical formula is: The expansion coefficient For the preset coefficients with temperature dimensions, compensate for the offset constant. This is a preset constant with the dimension of temperature. Preferably, the reference duration... A value of 180s can be used, representing the ideal thermal response time of the target heating space obtained under standard constant temperature test conditions; expansion coefficient The value can be 1.62℃; compensation offset constant. A value of 0.2℃ is acceptable. The above constant was determined by conducting step response tests on rooms with different insulation levels, using a genetic algorithm to calibrate the data under multi-objective optimization conditions, with overshoot less than 0.5℃ and the shortest settling time. Then, the system performs lower limit protection processing.

[0061] Specifically, the final interval radius increment satisfies the following relationship:

[0062] in, Indicates the final interval radius increment; This represents the function that takes the maximum value. This represents the calculated value of the interval radius; This represents the preset positive lower limit radius value.

[0063] Among them, the lower limit radius value Set to a positive number greater than half the distance between adjacent core fixed-point coordinates, for example, preferably 0.7℃. If Below approximately 159 seconds Negative value, or lead to The temperature is only 0.2℃, and the system uses... As the final interval radius increment This avoids inverted left and right boundaries and gaps in coverage between adjacent fuzzy subsets.

[0064] Assuming the input environment parameters If the time is 249 seconds, then the ratio of the variables is approximately 1.3833. The system uses the floating-point unit to execute the above logarithmic formula to calculate the increment of the interval radius. Because this value is greater than Therefore, the final interval radius increment The temperature is still set at 0.726℃. A comparison is made between the initial baseline membership function and the membership function distribution after this incremental expansion. Figure 2 As shown.

[0065] In obtaining the increment Subsequently, the system performs bidirectional translation operations on the initial values ​​of the aforementioned seven core fixed-point coordinates to obtain the basic base boundaries of each triangular fuzzy subset distribution. Taking the fuzzy subset with the core point being the center deviation of PM and an initial center coordinate value of 2℃ as an example, the system calculates that the left boundary point of its basic distribution interval is updated to 1.274℃, and the right boundary point is updated to 2.726℃. This nonlinear expansion step, through the smoothing characteristics of the logarithmic function, ensures that the coverage range of each fuzzy subset can be appropriately widened when the spatial thermal response time is long, thereby suppressing the heating temperature overshoot and high-frequency oscillation phenomenon caused by response lag. Conversely, if this nonlinear interval adjustment based on thermal response characteristics is not performed and a fixed domain boundary is used, in scenarios with large thermal inertia in the heating space, the fixed fuzzy subset coverage range will be too narrow, resulting in insufficient sensitivity of the controller to temperature deviation. This will cause the system to output a small air volume even when the temperature has significantly deviated from the set value, leading to slow heating and decreased comfort. In scenarios with small thermal inertia, a fixed coverage range that is too wide may lead to over-adjustment and temperature overshoot.

[0066] In an optional embodiment, the specific process of calculating the adjustment coefficient based on the temperature deviation change rate to scale and adjust the boundary points of the basic distribution interval includes: when the product of the current temperature deviation and the temperature deviation change rate is greater than 0, the system uses the sum of the product of the first preset coefficient and the absolute value of the temperature deviation change rate plus 1 as the adjustment coefficient to expand the boundary points of the basic distribution interval; when the product of the current temperature deviation and the temperature deviation change rate is less than 0, the system compares the difference between 1 and the product of the first preset coefficient and the absolute value of the temperature deviation change rate with a lower threshold, takes the maximum value as the adjustment coefficient, and shrinks the boundary points of the basic distribution interval; when the product of the current temperature deviation and the temperature deviation change rate is equal to 0, the system sets the adjustment coefficient to 1; and uses the adjustment coefficient to perform product scaling processing on the difference between the boundary coordinates of the basic distribution interval and the initial values ​​of the corresponding core fixed point coordinates.

[0067] More specifically, the system collects sensor feedback data at set sampling time intervals. In one implementation, the sampling time interval can be configured, for example, to be interrupted by a timer every 10 seconds. Assume the system reads the current temperature deviation... The temperature deviation is 1.5℃, while the temperature deviation from the previous sampling period retrieved from the controller cache is... The temperature is 1.3℃. At this point, the system performs differential calculations to obtain the rate of change of temperature deviation. .

[0068] The system determines the current temperature deviation. With rate of change Is the product greater than zero? In this example, the product is... The system determines that both values ​​have the same sign, indicating that the current temperature deviation is diverging away from the target setpoint. Therefore, the system proceeds according to... Calculate the adjustment coefficient. Preferably, the first preset coefficient... This is a coefficient with the dimension of time per temperature, for example, it can be taken as 0.12 s / ℃. This value is a curing parameter that balances gain and stability. Substituting it into the calculation yields... It is used to perform expansion operations on the boundary points of the basic distribution range.

[0069] Conversely, if the system detects For 1.5℃ for This indicates that the system is converging; the two signs are opposite, and at this time the system calculates... This value is then compared with a preset lower protection threshold in the memory, and the maximum value is taken. In one embodiment, the lower threshold... For example, it can take the value 0.5. Because... System output adjustment coefficient It is used to perform shrinkage operations on boundary points.

[0070] If the system detects or ,but and When the product of the two factors is 0, the system determines that the temperature deviation is at the target point or the trend of change is at a critical static state, and the adjustment coefficient is adjusted accordingly. Set the value to 1, keeping the basic boundary coordinates obtained in the previous stage unchanged.

[0071] After obtaining the adjustment coefficient Then, the system uses this coefficient to scale the boundary coordinates of each fuzzy subset relative to its core fixed-point coordinates using a difference product. Taking the PM subset as an example again, the core fixed-point coordinates are 2℃, and the left boundary point calculated in the previous step is 1.274℃. The system calculates the relative offset from the boundary to the core point, i.e. If the current operating condition is one of diverging signs, substitute the adjustment coefficient. The new offset difference is then corrected to After mapping back to the global coordinate system, the left boundary point becomes 1.2723℃, and the right boundary point is expanded outward through similar logic.

[0072] This scaling mechanism, which is tied to the real-time temperature change rate, expands the membership function boundary outward when the current temperature of the heated space deviates from the set value, smoothing the output gradient to enhance the stability of equipment operation; conversely, it tightens the boundary inward when the current temperature of the heated space approaches the target, improving the sensitivity of the control system and thus ensuring temperature tracking performance. Without this adaptive boundary adjustment mechanism based on the temperature deviation change rate, and instead using a fixed membership function boundary, when the temperature deviation expands rapidly, the fuzzy subset coverage cannot expand in time, resulting in an excessively small controller output airflow gradient and slow response, potentially causing the indoor temperature to continuously deviate from the set value. Conversely, when the temperature deviation converges rapidly, an excessively wide fuzzy subset boundary will cause the controller to maintain a high airflow output even when approaching the target temperature, leading to temperature overshoot and energy waste.

[0073] S3: Control instructions for generating matching rule weights and upper bound of discourse.

[0074] In an optional embodiment, the thermal response characteristic parameters of the heating space and the current temperature deviation are used as inputs. A two-dimensional rule weight mapping table is queried, and the target weight coefficient of the corresponding fuzzy rule is selected by combining the temperature deviation change rate. The upper limit of the universe of discourse of the output air volume fuzzy subset is determined according to the outdoor ambient temperature and the heat source temperature, so as to form different upper limit limits of air volume under low temperature heating, conventional heating and insufficient heat source temperature conditions. Fuzzy inference and defuzzification calculation are performed using the membership function, the target weight coefficient and the upper limit of the universe of discourse to generate air volume control instructions.

[0075] In the actual execution process, the system first presets multiple fuzzy subsets of the output air volume. These fuzzy subsets include low air volume, low-medium air volume, medium air volume, medium-high air volume, high air volume, and maximum air volume. The core points of each fuzzy subset of the output air volume are determined according to the upper limit of the current universe of discourse. The preset proportion distribution. Specifically, the preset proportions corresponding to the above six fuzzy subsets can be set to 0, 0.2, 0.4, 0.6, 0.8 and 1 respectively, that is, the x-coordinates of their core points are located at 0, 0.2, 0.4, 0.6, 0.8 and 1 respectively. 0.4 , , and Furthermore, the system pre-defines a basic fuzzy rule table with the current temperature deviation fuzzy level and the temperature deviation change rate fuzzy level as antecedents, and the output air volume fuzzy level as a consequent. The basic fuzzy rule table is pre-defined according to the following principles: when the current temperature deviation is positive and the absolute value is larger, the output air volume level is higher; when the current temperature deviation is negative or close to zero, the output air volume level is reduced; when the current temperature deviation and the temperature deviation change rate have the same sign and the absolute value of the deviation increases, the corresponding output air volume level is increased; when the two have opposite signs and the deviation converges, the corresponding output air volume level is reduced or maintained. A two-dimensional rule weight mapping table is used to adjust the activation intensity of corresponding rules based on the basic fuzzy rule table, thereby achieving differentiated rule weight configuration according to the thermal response characteristics of different heating spaces.

[0076] Subsequently, the system constructs parameters in computer memory, using discrete levels of the current absolute value of temperature deviation as row indices, to represent the thermal response characteristics of the heated space. The discrete interval is a two-dimensional regular weighted mapping table matrix with column indices. The system calls a preset two-dimensional array lookup table addressing program, matches the row coordinates according to the absolute value of the current temperature deviation, and determines the parameters. Match column coordinates, read the rule weight vector stored in the corresponding cell, and select the target weight coefficient based on the fuzzy level with the highest membership degree of the current temperature deviation change rate, as the activation weight of the currently activated fuzzy rule.

[0077] To determine the upper limit of the universe of discourse for the fuzzy subset of output airflow, the system compares the outdoor ambient temperature with a pre-stored preset low-temperature threshold. When the outdoor ambient temperature is lower than the preset low-temperature threshold and the heat source temperature meets the heating conditions, the system sets the upper limit of the universe of discourse for the fuzzy subset of output airflow to the rated maximum airflow value. When the outdoor ambient temperature is not lower than the preset low-temperature threshold and the heat source temperature meets the heating conditions, the system sets the upper limit of the universe of discourse to the rated maximum airflow value multiplied by a preset derating factor. Simultaneously, the system compares the heat source temperature with a preset anti-cold airflow threshold. If the heat source temperature is insufficient, the upper limit of the universe of discourse is overwritten with a preset low-level airflow value, and the fuzzy set of output airflow is initialized within this universe of discourse.

[0078] Based on this, the system, using the aforementioned assigned weights and adjusted membership functions, employs the Mamdani fuzzy inference algorithm to truncate the output of each rule using a minimum-value approach, obtaining the original derived membership degree of each rule. Subsequently, the system multiplies the original derived membership degree by the corresponding target weight coefficient and then performs upper limit truncation, taking the smaller value between the weighted result and 1 as the weighted rule membership degree, ensuring that the membership degree participating in aggregation is no greater than 1. Next, the system aggregates the outputs of multiple rules using a maximum-value approach, obtaining a fuzzy subset of the aggregated output airflow.

[0079] Finally, the system calls the centroid method defuzzification calculation program embedded in the controller to perform defuzzification operations on the aggregated output air volume fuzzy subset, outputting an air volume control command value representing the specific air volume. Subsequently, based on the fan performance curve or linear mapping relationship, the system converts the air volume control command value into a PWM duty cycle command, and transmits the PWM duty cycle command to the general input / output pins or motor drive interface of the heating fan controller, outputting a pulse width modulation signal with the corresponding duty cycle to drive the brushless DC motor, AC variable frequency motor, or EC fan of the heating fan to adjust its speed, thus completing the air volume control. At the same time, the system feeds back the current air volume command to the main controller of the heating equipment for coordinated adjustment of the hot water valve opening, electric heating power, or heat pump compressor power, achieving closed-loop matching between the fan air volume and the heating pattern.

[0080] In an optional embodiment, the specific process of querying the two-dimensional rule weight mapping table and combining it with the temperature deviation change rate to select the target weight coefficient of the corresponding fuzzy rule includes: the system matching the corresponding row coordinates according to the absolute value of the current temperature deviation, and matching the corresponding column coordinates according to the thermal response characteristic parameters of the heating space; querying the two-dimensional rule weight mapping table to obtain the corresponding rule weight vector; and extracting the corresponding coefficient from the rule weight vector as the target weight coefficient according to the fuzzy level with the largest membership degree of the temperature deviation change rate.

[0081] More specifically, the system pre-codes a two-dimensional rule-weighted mapping matrix table in the ROM space of the main control chip. Preferably, the row coordinates of the mapping table are divided into 5 deviation levels, and the column coordinates are divided into 10 thermal response characteristic parameter intervals. Specifically, the row coordinates are divided into 5 level intervals based on the absolute value of the current temperature deviation: that is... , , , , The unit is °C. The column coordinates are based on historical thermal response characteristic parameters. The time duration is divided into 10 assessment intervals, with the first 9 intervals being as follows: , , , , , , , , The 10th interval is set as The unit is s, used for coverage. In extremely slow response conditions, this prevents the inability to look up tables after parameters exceed their limits.

[0082] The matrix table contains 50 grid nodes, each storing a regular weight vector corresponding to the fuzzy level of the temperature deviation change rate. The weight coefficients in these regular weight vectors range from 0.1 to 3. For example, in a grid condition where the temperature deviation is large and the thermal response parameters indicate difficulty in heating the space, some weight coefficients in the corresponding regular weight vector can be configured to be close to 2.85 to enhance the output response under the corresponding fuzzy level of change rate. The specific values ​​of the weight coefficients can be determined through experimental calibration by conducting comprehensive tests in a standard thermal response testing environment on combinations of different deviation levels and thermal response characteristic parameters, using the heating rate, steady-state error, and overshoot as optimization objectives.

[0083] To clearly and completely disclose the internal data structure of this mapping table, this embodiment presents the representative weight vector distribution of the four boundary vertex nodes and the center node of the 5-row × 10-column matrix. The data of the remaining grid nodes can be smoothly generated using the bilinear interpolation algorithm through these five reference nodes: the node located in the first row and first column represents the condition of extremely small deviation and extremely fast response, and its corresponding weight vector is [0.5, 0.65, 0.8, 10, 1.1, 1.2, 1.3]; the node located in the first row and tenth column represents the condition of extremely small deviation but extremely slow response, and its corresponding weight vector is [0.8, 0.95, 1.1, 1]. [1.25, 1.4, 1.55, 1.7]; The node in the fifth row and first column represents the condition with extremely large deviation but extremely fast response, and its corresponding weight vector is [1.5, 1.7, 1.9, 20, 2.2, 2.4, 2.6]; The node in the fifth row and tenth column represents the condition with extremely large deviation and extremely slow response, and its corresponding weight vector is [20, 2.15, 2.3, 2.5, 2.7, 2.85, 3]; The node in the third row and fifth column represents the center node of the normal centering condition, and its corresponding weight vector is [1.1, 1.25, 1.35, 1.45, 1.6, 1.75, 1.9].

[0084] As can be seen from the above benchmark data, the data distribution of the mapping table follows the following calibration principle: the larger the absolute value of the current temperature deviation (i.e., the larger the row coordinate), and the more difficult the space is to heat (i.e., the thermal response parameter P, the smaller the column coordinate), the more non-linearly the overall weight coefficient increases towards 3; conversely, it decreases towards 0.1. Using this distribution pattern and interpolation method, those skilled in the art can construct a complete two-dimensional regular weight mapping table containing 50 nodes.

[0085] In the real-time control cycle of the equipment, assuming the calculated absolute value of the current temperature deviation is 1.2℃, and the thermal response parameters... The time was 249 seconds. The lookup table algorithm matched 1.2℃ to the third level of the row coordinate axis, falling into... The interval; matching 249s to the 5th level category on the column axis, falling into... The system uses row and column pointers to interleave the address space. The matrix memory block reads the rule weight vector stored at the node in the 3rd row and 5th column, assuming this vector is... Each element corresponds to one of seven fuzzy levels, ranging from negative to positive, for the rate of change of temperature deviation. The system selects the corresponding target weight coefficient from this vector based on the fuzzy level with the highest membership degree for the current rate of change of temperature deviation.

[0086] In the subsequent Mamdani fuzzy inference step, when the system activates a specific fuzzy rule, it multiplies the initial membership degree of the rule's original output by the target weight coefficient selected based on the rule weight vector. If the product exceeds 1, the system truncates the result to 1; if the product does not exceed 1, the original weighted result is retained. By utilizing this two-dimensional mapping adjustment matrix, the control logic solves the problem that conventional constant rule tables cannot adequately consider the thermal response characteristics of different heating spaces, achieving differentiated heating adjustment under different room temperature conditions and different insulation characteristics. If this two-dimensional weight mapping mechanism based on thermal response characteristic parameters and temperature deviation is not introduced, and fixed rule weights are used instead, in heating spaces with high thermal inertia, fixed weights will lead to overly conservative airflow adjustment and slow temperature rise, while in spaces with low thermal inertia, fixed weights may lead to overly aggressive airflow adjustment, causing frequent start-stop cycles and temperature fluctuations.

[0087] In an optional embodiment, the specific process of determining the upper limit of the universe of discourse of the fuzzy subset of output air volume based on the outdoor ambient temperature and the heat source temperature includes: if the heat source temperature is lower than the preset anti-cold air threshold, the system sets the upper limit of the universe of discourse to a preset low-level air volume value; if the heat source temperature is not lower than the preset anti-cold air threshold and the outdoor ambient temperature is lower than the preset low-temperature threshold, the system sets the upper limit of the universe of discourse to the rated maximum air volume of the device; if the heat source temperature is not lower than the preset anti-cold air threshold and the outdoor ambient temperature is not lower than the preset low-temperature threshold, the system sets the upper limit of the universe of discourse to a limit value of the rated maximum air volume of the device multiplied by a preset derating factor.

[0088] More specifically, the system acquires real-time outdoor ambient temperature data through NTC thermistor sensors located on the outdoor side, near the air inlet, or on the shaded side of the building's exterior wall, or through a networked temperature control module, at a set refresh rate. The comparator routine within the control logic compares the current outdoor ambient temperature value with a preset low-temperature threshold. Preferably, the preset low-temperature threshold can be, for example, a value of... The preset anti-cold air threshold can be set to, for example, 40℃, and the preset derating factor can be set to, for example, 0.8. The selection of the above parameters is based on the following: -15℃ corresponds to the typical design field temperature boundary of my country's severe cold climate zone; 40℃ is the safe heat exchange temperature limit for water coils / heaters set to prevent excessive draft sensation, based on the average human body surface temperature combined with air outlet attenuation and heat convection effect; the derating factor of 0.8 is set based on the statistical ratio of the steady-state heat transfer load of the building envelope under normal operating conditions to the load under extreme operating conditions, ensuring heating capacity while avoiding the high-frequency noise zone of the fan.

[0089] in, This is the rated maximum air volume of the current heating exhaust fan, which can be set according to the parameters on the equipment nameplate. In examples of wall-mounted, cabinet-type, or duct-type heating exhaust fans, It can be set according to the equipment model. , or The corresponding limits are respectively .

[0090] When the system determines the outdoor ambient temperature collected and transmitted, for example... Below that When the temperature of the heat source is not lower than the preset cold air prevention threshold, it indicates that the external ambient temperature of the heated space is low and the heat source has the conditions to provide heating. Under this condition, the system sets the global upper limit parameter of the wind quantum set domain output by the fuzzy inference engine to the rated maximum air volume parameter of the heating air outlet fan. This applies to wall-mounted or duct-type heating air outlet fans. For example, the upper limit of the domain of discourse is set to 1. This mechanism ensures that, under extremely cold conditions, the system can fully release its heating capacity to compensate for additional heat loss from the building envelope, thus maintaining the indoor temperature at the set level.

[0091] Conversely, if the heat source temperature is not lower than the preset anti-cold air threshold, and the returned outdoor ambient temperature is... Meeting no less than If the condition is met, it indicates that the external heat loss is within the system's normal compensation range. The system calls the derating function in the background to adjust the current equipment's rated maximum airflow. Multiply by the preset reduction factor of 0.8 to obtain the limit value. And set the upper bound of the universe of discourse for the fuzzy subset of the airflow output at this limit value. For example, the limit value is This derating constraint effectively prevents the fan from operating at ineffective high airflow under normal weather conditions when full-load heating is not required, thereby reducing fan power consumption and operating noise.

[0092] Furthermore, if the system detects that the heat source temperature is below the preset anti-cold air threshold of 40°C, it indicates that the heat source has not yet reached a state where it can provide heat. In this case, if a higher airflow is used, it will result in direct cold air blowing, severely impacting user comfort. Under this condition, the system will forcibly overwrite the upper bound of the universe of discourse of the output airflow fuzzy subset with a preset low-level airflow value, causing the fan to operate only at the lowest level or remain in standby preheating mode until the heat source temperature rises above the preset anti-cold air threshold. The feedforward constraint relationship between outdoor ambient temperature, heat source temperature, and the upper bound of the airflow universe of discourse is illustrated below. Figure 3 As shown. This feedforward compensation logic limits ineffective high airflow output to reduce energy consumption under normal outdoor temperatures, fully releases heating capacity to ensure comfort under low-temperature conditions, and prevents cold air from blowing directly when the heat source is insufficient, thus achieving adaptive limitation of the airflow domain under various operating conditions.

[0093] In an optional embodiment, the specific process of generating airflow control commands by performing fuzzy inference and defuzzification calculations using membership functions, target weight coefficients, and upper limits of the universe of discourse includes: during fuzzy inference, the system multiplies the target weight coefficients with the derived membership degrees of the corresponding fuzzy rules, and performs upper limit truncation by taking the smaller value between the multiplication result and 1 to obtain the weighted rule membership degrees; aggregates each fuzzy rule using the maximum value finding method to obtain an aggregated fuzzy subset of output airflow; performs defuzzification calculations on the aggregated fuzzy subset of output airflow based on the centroid method, generates discrete feature reference values ​​between 0 and the upper limit of the universe of discourse, calculates the sum of the products of all discrete feature reference values ​​and their corresponding membership degrees in the aggregated fuzzy subset of output airflow, and divides by the sum of all membership degrees to generate airflow control commands.

[0094] More specifically, after completing the aforementioned fuzzy rule derivation and decomposition, the system generates a comprehensive fuzzy set surface in memory to represent the probability distribution of the target airflow, and then calls the centroid method to perform inverse fuzzy decomposition calculations. The system then sets the airflow operating range... Mapped to a coordinate system, and the universe of discourse divided equally into... Generate intervals A discrete feature reference value. The discrete feature reference value ranges from 0 to the upper limit of the current universe of discourse. The average step value between, where This is the upper limit of the current output air volume domain, determined based on the outdoor ambient temperature and the heat source temperature. If the heat source temperature meets the heating requirements and the current operating condition is low temperature... Then the grid step resolution is If the heat source temperature meets the heating requirements and the current operating condition is not low temperature, Then the grid step resolution is Therefore, the defuzzification calculation is always completed within the currently constrained output domain, avoiding obtaining an air volume output that exceeds the limit.

[0095] At this resolution, the algorithm sequentially generated 61 discrete feature reference values. ,index From 0 to 60. The system extracts each discrete reference point sequentially. The membership value of the ordinate corresponding to the projection onto the comprehensive fuzzy set surface The data is then fed into an accumulator for weighted summation to calculate the airflow control command value.

[0096] Specifically, the air volume control command value satisfies the following relationship:

[0097] in, Indicates the air volume control command value; This indicates a cumulative summation operation; Indicates the first One discrete feature reference value; Indicates the first The membership value of the vertical axis corresponding to each discrete feature reference value in the aggregated output air volume fuzzy subset.

[0098] because belong Therefore, the output It must not exceed the upper limit of the current domain of discourse. If the sum of all memberships is 0, the controller can maintain the valid airflow command from the previous cycle or output a preset safe airflow to avoid division by zero anomalies.

[0099] If the processor calculates the molecule after traversing all reference nodes... The value is 750, and the sum of all membership degrees in the denominator is... If the value is 2.5, then the derived and output airflow control command value is... .Should This is the airflow control command value, and the unit of the control command value is... .

[0100] In an optional embodiment, after generating the air volume control command, the method further includes: the system converting the air volume control command into a PWM duty cycle command based on a preset mapping relationship; and sending the PWM duty cycle command to the hardware communication module to control the operating frequency of the blower motor.

[0101] More specifically, in obtaining, for example Following this airflow control command, the underlying controller interface routine converts it into underlying duty cycle modulation data based on the mapping relationship between airflow and motor speed or PWM duty cycle on the fan performance curve. Assuming the equipment is known... Full load air volume corresponds to 100% full duty cycle, then this After the control quantity is converted by the duty cycle mapping function, the output target is a PWM duty cycle of 50%.

[0102] Subsequently, the timer output inside the main control MCU chip is locked at a base frequency of 20kHz, with a pulse width corresponding to a 50% duty cycle, producing a PWM digital speed control signal. The system sends this PWM signal to the heating fan motor drive board, which controls the motor drive switching devices to adjust the motor speed and impeller airflow, ensuring that the airflow from the heating equipment impeller is controlled in real-time. At the operating point. In one implementation, the hardware communication module can be a general-purpose input / output pin of the heating fan controller or a motor drive interface. The fan motor can be a brushless DC motor, an AC inverter motor, or an EC fan. Furthermore, the system sends the target airflow load to the main controller of the heating equipment to achieve closed-loop matching between the fan airflow and the hot water valve opening, electric heating power, or heat pump compressor power, thereby ensuring coordinated operation of the entire heating system between the heat source and the terminal side. If the airflow control command is not converted into a PWM duty cycle and sent to the motor drive via the hardware communication module, but only remains at the level of airflow value calculation, the optimized airflow command derived from fuzzy reasoning cannot actually drive the fan to execute, resulting in a disconnect between the control strategy and the physical execution layer, and the entire adaptive airflow adjustment link cannot form an effective closed loop.

[0103] Figure 2 This is a schematic diagram of the membership function distribution of the current temperature deviation fuzzy subsets. The dashed triangles in the background represent the initial baseline membership functions; the thick solid broken lines represent the membership function distribution calculated and expanded using the thermal response characteristic parameters of the heating space. The image shows that the base distribution range covered by the thick solid lines is significantly wider than that of the dashed lines. This proves that the algorithm successfully achieved nonlinear widening of the universe of discourse based on the thermal inertia differences of the heating space without destroying the initial values ​​of the core fixed-point coordinates of each fuzzy subset. Observing the edge regions where the negative and positive large fuzzy subsets are located, it can be found that the solid lines remain horizontal straight lines in the direction away from zero, meaning that the membership degree is constant at one within this interval. As the deviation moves towards zero, the membership degree exhibits a piecewise slope change. This corresponds to the mathematical characteristic described in the specific implementation method of dynamically compensating for the interval radius increment using the natural logarithm analytical function.

[0104] Figure 3This diagram illustrates the feedforward constraint on the upper limit of the universe of discourse for the fuzzy subset of output airflow. The filled area in the background represents the effective airflow regulation universe of discourse; the horizontally thick solid line represents the upper limit of the universe of discourse dynamically locked according to environmental conditions; the vertical dashed line represents the switching boundary of the preset low-temperature threshold of -15℃. The image shows that the effective airflow regulation universe of discourse is not constant at the rated maximum airflow level, but rather undergoes a step-like numerical switch as the outdoor ambient temperature crosses -15℃. This proves that the control logic successfully truncates the output upper limit according to external meteorological conditions and does not unduly reduce the heating capacity under severe cold conditions. Observing the normal temperature range to the right of the -15℃ threshold, it can be seen that the upper limit of the universe of discourse is suppressed within... At this point; as the ambient temperature drops to the left and crosses the threshold, the upper limit rapidly jumps to... This corresponds to the logical feature described in the specific implementation that adaptively limits the power consumption and noise of the fan under normal heating conditions based on a preset derating factor.

[0105] This invention also discloses a heating fan air volume regulation system, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a heating fan air volume regulation method according to the present invention.

[0106] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0107] It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept, and these all fall within the scope of protection of this invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for adjusting the air volume of a heating exhaust fan, characterized in that, include: S1: Obtain the indoor set temperature, current temperature, outdoor ambient temperature and heat source temperature, and calculate the current temperature deviation and the rate of change of temperature deviation; The thermal response characteristic parameters of the heating space are calculated based on historical temperature log data. The thermal response characteristic parameters of the heating space are the weighted average time during which the current temperature deviation drops from the first positive deviation threshold to the second positive deviation threshold in the historical heating cycle. S2: Input the thermal response characteristic parameters of the heating space into a preset nonlinear function to calculate the basic distribution interval of the fuzzy subset of temperature deviation. Calculate the adjustment coefficient based on the rate of change of temperature deviation to scale the boundary points of the basic distribution interval. Dynamically construct the membership function of the current temperature deviation and call the preset membership function of the rate of change of temperature deviation. When the current temperature deviation and the rate of change of temperature deviation have the same sign, the adjustment coefficient is used to expand the boundary points; when they have opposite signs, the adjustment coefficient is used to shrink the boundary points. S3: Use the thermal response characteristic parameters of the heating space and the current temperature deviation as inputs. Query the two-dimensional rule weight mapping table and combine it with the rate of change of temperature deviation to select the target weight coefficient of the corresponding fuzzy rule. Determine the upper limit of the universe of discourse of the fuzzy subset of output air volume based on the outdoor ambient temperature and the heat source temperature to form different upper limits of air volume under low-temperature heating, conventional heating, and insufficient heat source temperature conditions. Use the membership function, target weight coefficient, and upper limit of discourse to perform fuzzy inference and defuzzification calculations to generate air volume control instructions.

2. The method for adjusting the air volume of a heating exhaust fan according to claim 1, characterized in that, The calculation of the thermal response characteristic parameters of the heating space based on historical temperature log data includes: extracting historical temperature log data for the most recent preset number of heating cycles; obtaining the first timestamp when the current temperature deviation reaches the first positive deviation threshold and the second timestamp when it drops to the second positive deviation threshold in each heating cycle; using the difference between the second timestamp and the first timestamp as the time consumed for a single heating cycle; wherein the first positive deviation threshold is greater than the second positive deviation threshold; and multiplying the time consumed for a single heating cycle in time order by the corresponding attenuation weight and summing them to obtain the thermal response characteristic parameters of the heating space.

3. The method for adjusting the air volume of a heating exhaust fan according to claim 2, characterized in that, The step of multiplying the single heating time of each heating cycle by the corresponding attenuation weight and summing them according to the time sequence to obtain the heating space thermal response characteristic parameter further includes: when there is a cycle in the most recent preset number of heating cycles that does not fully cross the first positive deviation threshold and the second positive deviation threshold, the corresponding cycle is removed as an invalid cycle; when the number of valid cycles is less than the preset number, the attenuation weight corresponding to the valid cycle is normalized, and the weighted average of the existing valid cycles is calculated as the heating space thermal response characteristic parameter; if there is no valid cycle, the heating space thermal response characteristic parameter is set to a preset reference duration.

4. The method for adjusting the air volume of a heating exhaust fan according to claim 1, characterized in that, The step of inputting the thermal response characteristic parameters of the heating space into a preset nonlinear function to calculate the basic distribution interval of the fuzzy subset of temperature deviation includes: obtaining the initial value of the core fixed point coordinates of the built-in basic universe of discourse; calculating the natural logarithm of the ratio of the thermal response characteristic parameters of the heating space to the preset reference time by a preset expansion coefficient, and adding a preset compensation offset constant to obtain the calculated value of the interval radius; taking the maximum value between the calculated value of the interval radius and the preset positive lower limit radius value as the final interval radius increment; adding and subtracting the final interval radius increment from the initial value of the core fixed point coordinates of the basic universe of discourse to obtain the boundary coordinates of each fuzzy subset as the basic distribution interval.

5. The method for adjusting the air volume of a heating exhaust fan according to claim 1, characterized in that, The step of calculating the adjustment coefficient based on the temperature deviation change rate to scale and adjust the boundary points of the basic distribution interval includes: when the product of the current temperature deviation and the temperature deviation change rate is greater than 0, using the sum of the product of the first preset coefficient and the absolute value of the temperature deviation change rate plus 1 as the adjustment coefficient to expand the boundary points of the basic distribution interval; when the product of the current temperature deviation and the temperature deviation change rate is less than 0, comparing the difference between 1 and the product of the first preset coefficient and the absolute value of the temperature deviation change rate with a lower threshold, taking the maximum value as the adjustment coefficient to shrink the boundary points of the basic distribution interval; when the product of the current temperature deviation and the temperature deviation change rate is equal to 0, setting the adjustment coefficient to 1; and using the adjustment coefficient to perform product scaling processing on the difference between the boundary coordinates of the basic distribution interval and the initial values ​​of the corresponding core fixed point coordinates.

6. The method for adjusting the air volume of a heating exhaust fan according to claim 1, characterized in that, The process of querying the two-dimensional rule weight mapping table and selecting the target weight coefficient of the corresponding fuzzy rule in conjunction with the temperature deviation change rate includes: matching the corresponding row coordinates based on the absolute value of the current temperature deviation, and matching the corresponding column coordinates based on the thermal response characteristic parameters of the heating space; querying the two-dimensional rule weight mapping table to obtain the corresponding rule weight vector; and extracting the corresponding coefficient from the rule weight vector as the target weight coefficient based on the fuzzy level with the largest membership degree of the temperature deviation change rate.

7. The method for adjusting the air volume of a heating exhaust fan according to claim 1, characterized in that, The step of determining the upper limit of the universe of discourse for the fuzzy subset of output air volume based on the outdoor ambient temperature and the heat source temperature includes: if the heat source temperature is lower than a preset anti-cold air threshold, then setting the upper limit of the universe of discourse to a preset low-level air volume value; if the heat source temperature is not lower than the preset anti-cold air threshold and the outdoor ambient temperature is lower than a preset low-temperature threshold, then setting the upper limit of the universe of discourse to the rated maximum air volume of the device; if the heat source temperature is not lower than the preset anti-cold air threshold and the outdoor ambient temperature is not lower than the preset low-temperature threshold, then setting the upper limit of the universe of discourse to a limit value of the rated maximum air volume of the device multiplied by a preset derating factor.

8. The method for adjusting the air volume of a heating exhaust fan according to claim 1, characterized in that, The step of generating airflow control instructions by performing fuzzy inference and defuzzification calculations using membership functions, target weight coefficients, and upper limits of the universe of discourse includes: during the fuzzy inference process, multiplying the target weight coefficients by the derived membership degrees of the corresponding fuzzy rules, and taking the smaller value between the multiplication result and 1 to perform upper limit truncation processing to obtain weighted rule membership degrees; aggregating each fuzzy rule using the maximum value finding method to obtain an aggregated output airflow fuzzy subset; performing defuzzification calculations on the aggregated output airflow fuzzy subset based on the centroid method, generating discrete feature reference values ​​between 0 and the upper limit of the universe of discourse, calculating the sum of the products of all discrete feature reference values ​​and their corresponding membership degrees in the aggregated output airflow fuzzy subset, and dividing by the sum of all membership degrees to generate the airflow control instructions.

9. A method for adjusting the air volume of a heating exhaust fan according to claim 8, characterized in that, After generating the air volume control command, the method further includes: converting the air volume control command into a PWM duty cycle command based on a preset mapping relationship; and sending the PWM duty cycle command to a hardware communication module to control the operating frequency of the blower motor.

10. A heating exhaust fan air volume regulating system, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement a method for regulating the air volume of a heating exhaust fan according to any one of claims 1-9.

Images