A linear heater intelligent control method and system

Abstract

The application discloses a kind of linear warmer intelligent control method and system, it is related to warmer technical field, the method is divided into several equidistant control sections by linear warmer along bathroom floor length direction, each control section is set with integrated temperature acquisition module, hot inertia detection module and the collector of heating response rate record module, the real-time temperature T of control section, unit time temperature rising rate ΔT and humidity Wc and other thermal load data can be synchronously acquired;Combining bathroom floor dry density Ps and specific heat capacity Cs parameter, the derivation method of bathroom floor thermal inertia coefficient K based on thermal inertia theory construction, can quantitatively characterize the thermal response delay characteristics of different bathroom floor area.Simultaneously, central control server is based on setting time window to carry out data cleaning, exception elimination and interpolation completion processing and normalization and unit standardization processing, constructs standardized data set, provides accurate and general data basis for subsequent thermal load grade evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of warmers, in particular to a linear warmer intelligent control method and system. BACKGROUND

[0002] In the existing home bathroom environment, electric towel racks, wall-mounted warm air heaters or whole-house hot water circulation heating systems are widely used, which can alleviate the cold problem to a certain extent when bathing in winter, but their temperature response speed is slow, the regional control ability is poor, especially in the humid ground structure, there are problems of heat lag diffusion, local overheating or slow heating for a long time. This kind of system mostly uses fixed power continuous heating mode, which cannot dynamically sense the heat load state of different ground areas, so it often leads to high heating energy consumption and uneven comfort, especially in the shower area with slow brick temperature sensing or the cold wind interference area near the door, users often face the uncomfortable experience of "cold feet after showering".

[0003] At present, some high-end bathrooms use embedded electric heating floor as a heating scheme, which has certain timing heating and overall temperature control functions, but it mainly uses unified heating control logic and does not model or adjust differently for different areas. In actual use, due to the differences in heat capacity distribution caused by the structure of the floor tiles, the sunken design and the drainage slope, it is easy to cause overheating in some areas and slow heating in some areas. Especially in humid environments, thermal inertia is intensified, and the efficiency of ground heating is significantly reduced. The existing system lacks the ability to feed forward control of this heat diffusion delay, and cannot realize real personalized and accurate temperature control.

[0004] The fundamental reason for the above-mentioned poor heating balance lies in the fact that in most families or intelligent residential systems, the bathroom heating equipment usually does not have self-adaptive parameter adjustment capability during operation. The parameters of the traditional heating system are mostly set by artificial pre-setting or single temperature sensor feedback, and cannot be dynamically adjusted in real time according to the temperature distribution of each area of the ground, the heating response rate or the humidity change. Lack of quantitative evaluation methods for temperature control effect, such as temperature balance score, heat load level feedback mechanism, etc., makes it difficult for the system to be intelligently optimized and energy-saving controlled in long-term operation. With the improvement of users' requirements for bathroom comfort, it is particularly necessary to develop a new type of intelligent bathroom floor heating system that can "heat on demand, dynamically compensate, fine control and continuously adapt". SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a linear warmer intelligent control method and system, which solves the problems mentioned in the background art.

[0006] To achieve the above purpose, the present application is realized by the following technical scheme:

[0007] S1, divide the linear heater into multiple control sections, set collectors at the positions of each control section, obtain the thermal load data of each control section, and transmit the thermal load data to the central control server for preprocessing to obtain a standardized data set;

[0008] S2, construct a thermal load calculation model based on the standardized data set, calculate the thermal load grade value E of each control section, and perform multi-level heating strategy control according to the thermal load grade value E;

[0009] S3, based on the multi-level heating strategy control result, trigger the heat diffusion lagging feedforward regulation mechanism, calculate the pre-compensation heating extension time Δt, and adjust the heating strategy in time to complete the heating control update;

[0010] S4, after each control period ends, calculate the temperature dispersion score Qs of all control sections, compare it with the set balance threshold interval, and based on the comparison result, optimize and reconstruct the parameters in the thermal load calculation model.

[0011] Preferably, the S1 includes S11;

[0012] S11, divide the linear heater along the length direction of the bathroom floor heating floor into several equidistant control sections, each control section as an independent logical heating area, and set collectors in each control section, wherein the collector of each control section is set on the bottom of the bathroom floor and the side edge structure under the corresponding control section center area, and real-time thermal load data is collected;

[0013] The collector includes a temperature collection module, a bathtub thermal inertia detection module, and a heating response rate recording module;

[0014] The thermal load data includes the real-time temperature Tn of the nth control section area n , the temperature rise rate ΔTn per unit time of the nth control section area n , and the humidity Wc of the nth control section area n .

[0015] Preferably, the S1 further includes S12;

[0016] S12, send the thermal load data collected by the collector in real time to the central control server through the wireless communication module, preprocess the thermal load data in the central control server, and preprocess the thermal load data based on the set time window to perform data cleaning, remove outliers, and interpolate complete operation;

[0017] For each segment of humidity Wc in the preprocessed thermal load data, combine the pre-set dry density Ps of the bathroom floor and the specific heat capacity Cs, and use the thermal inertia calculation formula to obtain the thermal inertia coefficient K for evaluating the thermal response lag degree of all segments of the bathroom floor;

[0018] and the real-time temperature T of the nth control section region in the heat load data n and the temperature rise rate per unit time AT of the nth control section region n , combined with the thermal inertia coefficient K of each control section, dimension normalization processing and unit standardization processing are performed to form a standardized data set in a unified format.

[0019] Preferably, S2 includes S21;

[0020] S21, based on the standardized data set, a heat load calculation model is constructed to calculate the heat load level value E of each control section to measure the heat load state of each control section at present;

[0021] The heat load level value E is calculated by the following heat load calculation model;

[0022] ;

[0023] In the formula, E n represents the heat load level value of the nth control section, Ttarget represents the preset target temperature, which is a dimensionless value initially set, K n represents the thermal inertia coefficient of the nth control section, represents the temperature rise rate response factor, represents the thermal inertia compensation factor, represents a small constant to prevent division by zero, and the value is 0.01.

[0024] Preferably, S2 further includes S22;

[0025] S22, by fitting regression of different temperature gradients, thermal response rates and final thermal equilibrium times in the historical heating period, a heating load threshold model is constructed to automatically generate a first judgment threshold F1, a second judgment threshold F2 and a third judgment threshold F3;

[0026] In the heat load level division of the heat load level value E of each control section according to the first judgment threshold F1, the second judgment threshold F2 and the third judgment threshold F3, a multi-level heating strategy control is simultaneously executed; the specific division content is as follows:

[0027] When the heat load level value E of the nth control section n ≥ the first judgment threshold F1, it is determined as heating level 3, and the heating strategy with a time proportion of 90% in the set control period is executed;

[0028] When the second judgment threshold F2 ≤ the heat load level value E of the nth control section n < the first judgment threshold F1, it is determined as heating level 2, and the heating strategy with a time proportion of 60% in the set control period is executed;

[0029] When the third judgment threshold F is less than the thermal load level value E of the nth control section n When the second judgment threshold F2, it is determined that the heating level is 1, and the heating strategy is executed for 30% of the time in the set control period;

[0030] When the thermal load level value E of the nth control section n When the third judgment threshold F, it is determined that the heating level is 0, and no heating is performed in the current period;

[0031] When there are two consecutive heating control periods in which the heating level remains unchanged, and the unit time temperature rise rate ΔT of the nth control section area n Less than 0.2, it is automatically identified as thermal response lag.

[0032] Preferably, the S3 includes S31;

[0033] S31, when identified as thermal response lag, trigger the thermal diffusion lag feedforward control mechanism, the thermal diffusion lag feedforward control mechanism is calculated by outputting the pre-compensation heating extension time Δt based on the unit time temperature rise rate ΔT of the nth control section area of the current control section standard data set n Combined with the thermal inertia coefficient K of the nth control section area n , heating compensation is carried out for the control section of thermal response lag;

[0034] The pre-compensation heating extension time Δt is calculated by the following algorithm formula;

[0035] ;

[0036] In the formula, △t n Indicates the pre-compensation heating extension time of the nth control section, Indicates the minimum temperature rise rate threshold, dimensionless, Kavg represents the average thermal inertia value of all control ends, Indicates the feedforward response adjustment factor.

[0037] Preferably, the S3 further includes S32;

[0038] S32, the pre-compensation heating extension time Δt of the nth control section calculated according to the response lag section n Is applied to the heating time ratio in the current control period according to the following strategy, which extends the heating strategy in time and updates the heating control:

[0039] If the current control section thermal load level is level 1, based on the original set heating time ratio of 30%, extend 30%+ the pre-compensation heating extension time Δt of the nth control section n / Complete heating control cycle time Tp;

[0040] If the current control section thermal load level is level 2, on the basis of the original set heating time ratio of 60%, the heating time is extended by 60%+ the pre-compensation heating extension time Δt of the nth control section n / complete heating control cycle time Tp;

[0041] If the current control section thermal load level is level 3, the heating time ratio of 90% is kept unchanged to prevent the floor from being overheated.

[0042] Preferably, the S4 comprises S41;

[0043] S41, after the heating control update is completed, the updated real-time temperature T' of each control section is reacquired, the updated real-time temperature T' of all control sections is subjected to a normalized difference value calculation with the preset target temperature Ttarget, the temperature dispersion score Qs is calculated based on the total number of control sections N, and the deviation conditions of all control sections are summarized and evaluated;

[0044] The temperature dispersion score Qs is calculated by the following algorithm formula:

[0045] ;

[0046] In the formula, N represents the total number of control sections, represents the updated real-time temperature of the nth control section.

[0047] Preferably, the S4 further comprises S42;

[0048] S42, based on the actual bathroom floor temperature control, the preset target temperature Ttarget is set as the balance threshold interval of ±5%, the balance threshold interval is compared with the temperature dispersion score Qs obtained in real time, the balance of the temperature control effect is judged, and based on the comparison result, the parameters in the thermal load calculation model are adjusted and reconstructed; the specific comparison content is as follows:

[0049] When the temperature dispersion score Qs belongs to the balance threshold interval, it indicates that the temperature control running state is qualified, and the current parameters are kept;

[0050] When the temperature dispersion score Qs does not belong to the balance threshold interval, it indicates that the temperature control running state is unqualified, and at this time the parameter adaptive adjustment process of the thermal load calculation model is executed, and after adjustment, the heating control update is re-performed;

[0051] The parameter adaptive adjustment process of the thermal load calculation model is constructed by using the original thermal load calculation model form to construct a target function based on the historical standardized data set of M control cycles, and the linear regression is adopted to adjust the temperature rise rate response factor with thermal inertia compensation factor Perform parameter re-estimation, output optimal combination.

[0052] A linear heater intelligent control system, comprising a control section acquisition module, a thermal load grade division module, a multi-stage heating control module and a temperature discrete analysis module.

[0053] The control section acquisition module divides the linear heater into multiple control sections, sets an acquisition device at each control section position, obtains thermal load data of each control section, and transmits the thermal load data to a central control server for preprocessing to obtain a standardized data set.

[0054] The thermal load grade division module constructs a thermal load calculation model based on the standardized data set, calculates the thermal load grade value E of each control section, and executes multi-stage heating strategy control according to the thermal load grade value E.

[0055] The multi-stage heating control module triggers the thermal diffusion lagging feedforward regulation mechanism based on the multi-stage heating strategy control result, calculates the pre-compensation heating extension time Δt, and adjusts the heating strategy in time to complete heating control update.

[0056] The temperature discrete analysis module calculates the temperature discrete score Qs of all control sections at the end of each control period, compares it with the set balance threshold interval, and optimizes and reconstructs the parameters in the thermal load calculation model based on the comparison result.

[0057] The present application provides a linear heater intelligent control method and system.

[0058] (1) The method divides the linear heater along the length direction of the bathroom floor heating floor into several equidistant control sections, sets an acquisition device integrated with a temperature acquisition module, a bathtub thermal inertia detection module and a heating response rate recording module in each control section, can synchronously obtain real-time temperature T, unit time temperature rise rate ΔT and humidity Wc and other thermal load data of the control section; combined with the dry density Ps and specific heat capacity Cs parameters of the bathroom floor, the deducing method of the bathroom floor thermal inertia coefficient K based on the thermal inertia theory can quantitatively represent the thermal response delay characteristics of different bathroom floor areas. At the same time, the central control server performs data cleaning, abnormality elimination and interpolation completion processing based on the set time window, and executes unified dimension normalization and unit standardization processing on multi-dimensional data, constructs a standardized data set, and provides accurate and universal data basis for subsequent thermal load grade evaluation. Compared with the traditional single-parameter thermal control system, the adaptability and data precision in complex agricultural environment are significantly improved.

[0059] (2) The method fuses the current temperature difference, the temperature rise rate per unit time ΔT and the thermal inertia coefficient K to construct a calculation model of the thermal load grade value E coupled with three factors; in combination with the automatically generated thermal load grade boundary threshold, the thermal load grade value E is graded and divided, three pulse control strategies are matched, i.e. grade 1, 30% heating proportion, grade 2, 60% heating proportion, and grade 3, 90% heating proportion, to realize flexible adjustment; in the case that the grade is unchanged for two consecutive heating periods and ΔT n <0.2, the thermal response lag is automatically identified, and a thermal diffusion lag feedforward control mechanism based on the temperature rise rate and the thermal inertia is further triggered to output a pre-compensation heating extension time Δt n , which is applied to the current heating proportion in the form of “gain within the grade” to improve the response gentleness and system stability, avoid overheating and oscillation problems, and is particularly suitable for greenhouse environments with differences in bathroom floor humidity gradient and structural thermal inertia.

[0060] (3) The method sets the updated real-time temperature T' at the end of the control period, calculates the temperature dispersion score Qs in combination with the preset target temperature Ttarget, takes the temperature dispersion score Qs as a quantitative index for measuring the overall temperature control uniformity of the system, and sets the temperature dispersion score Qs to be in the qualified interval, otherwise it is in the unqualified interval. When the score is in the unqualified interval, the parameter re-estimation process of the temperature rise rate response factor and the thermal inertia compensation factor in the thermal load calculation model is automatically triggered; the parameter re-estimation is based on the historical standardized data set in M control periods, constructs a linear regression path with the minimum temperature control error as the objective function, extracts the optimal parameter combination, and realizes adaptive updating of the model. This mechanism can dynamically respond to the thermal control mismatch problems caused by the actual bathroom floor state, thermal disturbance and environmental changes, construct a data-driven feedback optimization loop, and significantly enhance the overall temperature control robustness and self-learning ability of the system. BRIEF DESCRIPTION OF DRAWINGS

[0061] Figure 1 It is a step schematic diagram of the intelligent control method of the linear heater.

[0062] Figure 2 It is a module block diagram of the intelligent control system of the linear heater.

[0063] Figure 3 It is a schematic diagram of the segmented collection structure of the linear heater. DETAILED DESCRIPTION

[0064] With reference to the accompanying drawings: clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application. Embodiments

[0065] Please refer to Figure 1 and Figure 3 The present application provides an intelligent control method for linear heater. To achieve the above purpose, the present application is implemented by the following technical solutions: comprising the following steps:

[0066] S1, divide the linear heater into multiple control sections, set collectors at each control section position, obtain the thermal load data of each control section, and transmit the thermal load data to the central control server for preprocessing to obtain a standardized data set;

[0067] S2, construct a thermal load calculation model based on the standardized data set, calculate the thermal load grade value E of each control section, and execute multi-level heating strategy control according to the thermal load grade value E;

[0068] S3, based on the multi-level heating strategy control result, trigger the heat diffusion lagging feedforward regulation mechanism, calculate the pre-compensation heating extension time Δt, and adjust the heating strategy time extension, complete the heating control update;

[0069] S4, after each control cycle ends, calculate the temperature dispersion score Qs of all control sections, and compare it with the set balance threshold interval, and based on the comparison result, optimize and reconstruct the parameters in the thermal load calculation model.

[0070] In this embodiment, the method significantly improves the response efficiency and temperature uniformity of the heating system by constructing an intelligent control process integrating data acquisition, thermal load modeling, feedforward compensation control, and closed-loop adaptive optimization. First, in step S1, the linear heater is divided into multiple control sections, and an integrated collector is set in each section to achieve accurate acquisition of multi-modal thermal load data such as real-time temperature T, temperature rise rate per unit time ΔT, and humidity Wc, avoiding the problem of miscontrol caused by single temperature judgment, especially suitable for complex environments with structural heat dissipation differences or bathroom floor humidity gradients. Subsequently, in step S2, a thermal load calculation model based on a standardized data set is constructed, considering indicators such as thermal inertia coefficient K and response rate ΔT, to calculate the thermal load grade value E and execute multi-level heating strategy control, achieving differentiated heating for different control sections and effectively improving overall energy efficiency. Further, in step S3, to solve the response lag problem caused by thermal diffusion delay, a thermal diffusion lag feedforward control mechanism based on insufficient unit time temperature rise rate ΔT is introduced, a pre-compensation heating extension time Δt is calculated using a self-defined empirical compensation algorithm, and the pre-compensation heating extension time Δt is applied to the original heating proportion in proportion to achieve continuous gain adjustment within the grade. This method avoids the dramatic temperature fluctuations caused by heating grade jumps, improving system stability. Finally, in step S4, the overall level of temperature deviation for each control section is evaluated by temperature dispersion score Qs, and an equilibrium threshold interval is set to determine the control effect. If the target is not met, the parameters of the thermal load model are automatically adapted and reconstructed to improve long-term adaptability and avoid control performance degradation due to environmental changes. Through the above implementation process, the invention not only improves the system's ability to perceive different regional thermal load differences but also enhances its ability to predict and compensate for heating delays, constructing a complete "detection-control-evaluation-optimization" closed-loop temperature control link to achieve high uniformity, high stability, and low energy consumption in heating control under different crops, different greenhouse structures, and different climate conditions.

[0071] Embodiment 2

[0072] See Figure 1 and Figure 3 , specifically: S1 includes S11;

[0073] S11, the linear heater is evenly divided into several equidistant control sections along the length direction of the bathroom floor heating floor, each control section is set as an independent logical heating area, and a collector is set in each control section, wherein the collector of each control section is set on the bottom of the bathroom floor and the side edge structure of the corresponding control section center area to collect thermal load data in real time;

[0074] The collector includes a temperature acquisition module, a bathtub thermal inertia detection module, and a heating response rate recording module.

[0075] the real-time temperature T of the nth control section area n , the temperature rise rate per unit time ΔT of the nth control section area n , and the moisture Wc of the nth control section area n ;

[0076] The degree acquisition module includes a thermistor for periodically collecting the thermal changes between the surface and the bottom of the bathroom floor, and obtaining the real-time temperature T of each control section in real time.

[0077] The bath tub thermal inertia detection module is based on the moisture Wc of the nth control section area n ;

[0078] The heating response rate recording module is used to record the temperature rise amplitude and time difference of the control section in the last heating period, and calculate the temperature rise rate per unit time ΔT of each control section.

[0079] S1 also includes S12;

[0080] S12, the heat load data collected by the collector in real time is sent to the central control server through the wireless communication module, and in the central control server, the heat load data is pretreated, and the pretreatment is based on the set time window to perform data cleaning, remove outliers and interpolation completion operation;

[0081] For each segment of moisture Wc in the pretreated heat load data, the thermal inertia calculation formula is used to obtain the thermal inertia coefficient K for evaluating the thermal response delay degree of all segments of the bathroom floor, based on the preset dry density Ps of the bathroom floor and the specific heat capacity Cs.

[0082] Wherein, the thermal conductivity of the thermal inertia calculation formula is replaced by the moisture Wc, the main reason is that there is a positive correlation between the thermal conductivity of the bathroom floor and the moisture Wc, the main reason is that the thermal conductivity of water is high and the bathroom floor is filled with water pores.

[0083] And the real-time temperature T of the nth control section area n and the temperature rise rate per unit time ΔT of the nth control section area n in the heat load data, combined with the thermal inertia coefficient K of each control section, are subjected to dimension normalization processing and unit standardization processing to form a standardized data set in a unified format.

[0084] In this embodiment, the method divides the linear heater into multiple equidistant control sections along the length direction of the bathroom floor heating, and sets temperature collection modules, bathtub thermal inertia detection modules and heating response rate recording modules in each control section, so as to realize accurate monitoring of the thermal behavior in each logical area, and avoid the problems of overheating or insufficient heating in the area caused by the single-point temperature feedback of the traditional system. For example, if a certain area is close to the greenhouse vent, the heat loss is large, and only the edge probe sampling will underestimate the overall heat load. By setting the center and edge multi-point collection structure, the problem can be effectively avoided, and the representativeness of the heat load evaluation is enhanced. After data collection, the real-time data is uploaded to the central control server by using the wireless communication module, and the data is cleaned and interpolated based on the sliding time window, so as to ensure the continuity and effectiveness of the data, and effectively eliminate the interference of the collection jitter, transmission packet loss and other problems on the calculation model. In addition, the moisture Wc and the dry density Ps of the bathroom floor and the specific heat capacity Cs are jointly introduced into the thermal inertia formula to calculate the thermal inertia coefficient K, which can replace the traditional thermal conductivity detection hardware, reduce the system complexity, and because the thermal conductivity of water is high and the pore filling effect is strong, the thermal conductivity of the bathroom floor is positively correlated with the moisture Wc, which has a good physical replacement basis. Finally, the real-time temperature T, the temperature rise rate per unit time AT and the thermal inertia coefficient K of the control section are normalized and unit standardized, the data scale and dimension are unified, and the standardized data set is formed, which is convenient for subsequent heat load modeling and control decision, improves the algorithm adaptability and stability, and ensures the universality and expansibility under different environments and crop types.

[0085] Embodiment 3

[0086] Please refer to Figure 1 , specifically: S2 includes S21;

[0087] S21, based on the standardized data set, a heat load calculation model is constructed, and the heat load grade value E of each control section is calculated to measure the heat load state of each control section;

[0088] The heat load grade value E is calculated by the following heat load calculation model;

[0089] ;

[0090] In the formula, E n represents the heat load grade value of the nth control section, Ttarget represents the preset target temperature, which is the initial setting of the dimensionless value, K n represents the thermal inertia coefficient of the nth control section, represents the temperature rise rate response factor, represents the thermal inertia compensation factor, represents a small constant to prevent division by zero, and the value is 0.01.

[0091] In the construction process of the thermal load calculation model, the heating rate response factor is set and the thermal inertia compensation factor , the heating rate response factor is used to adjust the weight of the heating rate ΔT, and the thermal inertia compensation factor is used to adjust the weight of the thermal inertia coefficient K, and the heating rate response factor is used to adjust the weight of the thermal inertia compensation factor, and the heating rate response factor is used to adjust the weight of the thermal inertia coefficient K, and the heating rate response factor is in the range of 0.1 to 1.0, and the thermal inertia compensation factor

[0092] is in the range of 0.05 to 0.6, and can be set by historical data fitting, structural thermal resistance measurement or heating system feedback self-adaptation;

[0093] The thermal load calculation model is constructed based on the basic theory of heat conduction and the dynamic response characteristic estimation method of heat transfer;

[0094] represents the temperature difference term, represents the deviation of the current temperature from the target temperature, and is the most intuitive heating demand index;

[0095] represents the response rate correction term, and its reciprocal represents the time required for unit temperature rise, and the larger it is, the more difficult it is to heat and the poorer the heat transfer capacity, so as to prevent the unit time heating rate ΔT n of the nth control section from being zero, which causes a division by zero error, and a small constant ε such as 0.01 is introduced for numerical protection;

[0096] represents the thermal inertia term, which represents the response delay degree of the control section to temperature change, and the larger it is, the stronger the heating inertia, and the higher the energy consumption and the longer the time required to maintain heating, and the thermal inertia compensation factor is used to adjust the weight of the thermal inertia coefficient Kn of the nth control section in the evaluation of the thermal load level value E;

[0097] dimensional consistency analysis, wherein the three factors in the standardized data set are dimensionless and unit standardized, wherein the preset target temperature Ttarget is the initial input, and is a dimensionless value adapted to the thermal load calculation model, so that the thermal load level value E is a dimensionless numerical value index for thermal load level division, and has relative and step characteristics, and therefore belongs to numerical control factor in the control algorithm, and is not a physical output quantity.

[0098] S2 also includes S22;

[0099] S22, by fitting regression of different temperature gradients, thermal response rate and final thermal equilibrium time in the history heating period, a heating load threshold model is constructed to automatically generate the first judgment threshold F1, the second judgment threshold F2 and the third judgment threshold F3, which correspond to the division boundaries of the control section reaching high heat load, medium heat load and low heat load respectively; the above heating load threshold model can be constructed by using polynomial regression or logistic regression, and the regression output corresponds to the comprehensive heat load grade response of the control section, and the different grade boundary values are extracted by clustering or quantile analysis method;

[0100] The heat load grade value E of each control section is divided into heat load grade according to the first judgment threshold F1, the second judgment threshold F2 and the third judgment threshold F3, and the multi-level heating strategy control is simultaneously executed; the specific division content is as follows:

[0101] When the heat load grade value E of the nth control section n ≥ the first judgment threshold F1, it is determined as heating grade 3, and the heating strategy with 90% time proportion in the set control period is executed;

[0102] When the second judgment threshold F2≤ the heat load grade value E of the nth control section n < the first judgment threshold F1, it is determined as heating grade 2, and the heating strategy with 60% time proportion in the set control period is executed;

[0103] When the third judgment threshold F≤ the heat load grade value E of the nth control section n < the second judgment threshold F2, it is determined as heating grade 1, and the heating strategy with 30% time proportion in the set control period is executed;

[0104] When the heat load grade value E of the nth control section n < the third judgment threshold F, it is determined as heating grade 0, and no heating is executed in the current period;

[0105] When the heating grade remains unchanged in the continuous two heating control periods, and the unit time temperature rise rate ΔT of the nth control section area n is less than 0.2, it is automatically identified as thermal response lag;

[0106] The time proportion heating control here refers to the pulse control of the control section heater in the fixed heating control period, for example, 60 seconds, for example, the heating strategy of grade 3 corresponds to the heating time of the heater in the period not less than 90%, that is, more than 54 seconds, to realize higher heat input, and this way realizes the dynamic response of segmented temperature control by controlling the heating on-off ratio.

[0107] In this embodiment, the heat load calculation model based on the standardized data set outputs the heat load level value E for each control section, so that the heating decision can consider multiple factors such as real-time temperature difference, temperature rise rate per unit time ΔT, and thermal inertia coefficient K, and especially, the thermal inertia is used as a modeling factor to effectively identify the heat transfer delay area and avoid misjudgment caused by the current temperature level, thereby improving the robustness of heat response identification. The heat load calculation model introduces the temperature rise rate response factor and the thermal inertia compensation factor as adjustment weights, which can be flexibly and adaptively set according to the historical operation, thereby realizing dynamic modeling capability. For example, under the condition of sudden change of water content on the bathroom floor after taking a bath, the thermal inertia increases dramatically. If the thermal inertia coefficient K parameter is not considered, the control section will be misjudged as a "sufficient temperature state" with slow temperature rise, and the thermal inertia term can effectively avoid such risks. After the heat load level value E is calculated, the multi-level heating threshold model is constructed by fitting regression in S22, the control section state is automatically divided into four levels, and the precise control is performed in the form of "heating time ratio", thereby avoiding the energy waste or temperature overshoot caused by the traditional step switching. Taking a 60-second control cycle as an example, the power-on time of level 3 is limited to 54 seconds 90% upper limit, which ensures the temperature control strength and reserves 10% non-power-on time for observing the key parameters such as temperature rise rate per unit time ΔT, thereby maintaining the system regulation window. In particular, "no change for two consecutive cycles + temperature rise rate per unit time ΔT < 0.2" is set as the criterion for heat response lag, which helps to identify the temperature rise stagnation section and trigger the compensation mechanism in time to prevent control inertia. Through the implementation of the strategy, the intelligence, discrimination accuracy and energy efficiency of the heating control are improved, the dynamic heat load change in the complex bathroom floor environment is adapted, and the regulation accuracy and temperature control stability of the system are enhanced.

[0108] Embodiment 4

[0109] Please refer to Figure 1 , in particular: S3 includes S31;

[0110] S31, when the heat response lag is identified, a heat diffusion lag feedforward regulation mechanism is triggered, which calculates and outputs a pre-compensation heating extension time Δt based on the temperature rise rate per unit time ΔT of the nth control section area of the current control section standard data set n , in combination with the thermal inertia coefficient K of the nth control section area n , to implement heating compensation for the control section with heat response lag;

[0111] The pre-compensation heating extension time Δt is calculated and output by the following algorithm formula;

[0112] ;

[0113] In the formula, Δt nPrecompensation heating extension time of the nth control section, Minimum heating rate threshold, dimensionless, Kavg represents the average thermal inertia value of all control sections, Feedforward response adjustment factor, used to adjust the preheating compensation amplitude required for the response delay of insufficient heating rate, used to quantify the influence of the current section heating lag on the compensation time in the control system, with a value of 10 seconds representing the compensation time in the maximum lag state, v represents the structure thermal inertia adjustment factor, used to adjust the weight of the structural response delay influence of the thermal inertia coefficient K, with a value of 5 seconds representing the compensation time when the thermal inertia coefficient K is equal to the average thermal inertia value Kavg of the control section, used to quantify the adjustment ability of different bath floors or environmental thermal inertia on the heating advance in the control system;

[0114] This formula is not directly derived from a classical physical formula, but is an empirical control formula constructed for feedforward compensation heating adjustment of the response lag control section in the control system, combining the control compensation needs of thermal diffusion lag phenomenon, with the following derivation basis:

[0115] First term Response rate compensation term, which is based on the ratio between the heating rate ΔT and the minimum heating rate threshold , represents the deviation between the actual heating rate of this section and the minimum reasonable heating threshold; if ΔT n approaches 0, there is no significant heating, and this term approaches , extending the time to the maximum; if ΔT n ≥ , this term approaches 0, indicating that no compensation is needed;

[0116] Second term Thermal diffusion lag term, derived from thermal inertia theory, used to measure the delay characteristics of different control sections in response to heat; the thermal inertia coefficient K n of the nth control section area is larger, indicating that the section responds slowly to heat conduction and requires longer compensation time; the average thermal inertia value Kavg of all control sections is used to normalize the characteristics of each section to prevent absolute values from affecting the stability of the adjustment;

[0117] Dimension consistency explanation:

[0118] Δt n : unit in seconds (s); ΔT n , : dimensionless; K n , Kavg: thermal inertia, unitless; , V: unit in seconds, used to ensure the dimension consistency of the overall Δt n output;

[0119] Therefore, both weighting terms are dimensionless coefficients x unit factors = seconds, and the final result Δt n The physical dimension is consistent, and the unit is unified as "seconds", which is logically reasonable.

[0120] S3 also includes S32;

[0121] S32, the pre-compensation heating extension time Δt n of the nth control section calculated according to the response lag section is applied to the heating time proportion in the current control period according to the following strategy, which extends the heating strategy in time and adjusts the heating control update:

[0122] If the current control section heat load level is level 1, based on the original set heating time proportion of 30%, the heating time proportion is extended by 30% + the pre-compensation heating extension time Δt n of the nth control section / the complete heating control period time Tp;

[0123] If the current control section heat load level is level 2, based on the original set heating time proportion of 60%, the heating time proportion is extended by 60% + the pre-compensation heating extension time Δt n of the nth control section / the complete heating control period time Tp;

[0124] If the current control section heat load level is level 3, the heating time proportion of 90% is kept unchanged to prevent the ground from being overheated and scalded;

[0125] Wherein, the pre-compensation heating extension time Δt n of the nth control section / the complete heating control period time Tp represents the relative time extension ratio, which is directly applied to the current heating proportion, and limits the final heating proportion to not more than 100%;

[0126] For example: the complete heating control period time Tp is 60s, the original heating level 1 is 30%, and if the pre-compensation heating extension time Δt n of the nth control section = 6 seconds, then the heating becomes (30% + 6 / 60) = 40%;

[0127] Instead of adjusting it greatly through "level jump", it is better to use the pre-compensation heating extension time Δtn / complete heating control period time Tp of the nth control section in a "continuous gain" way to realize a gentle transition.

[0128] It can significantly reduce the risk of system shock and improve stability, especially suitable for greenhouse scenarios with slow thermal response.

[0129] Level 3 is set to 90% upper limit, which is considered from two aspects: to prevent overheating risk: to avoid high temperature caused by scald or heating band damage; to ensure the remaining control freedom: to leave out the non-heating time for system to detect temperature rise rate ΔT n If it is 100% power, the observation window will be lost.

[0130] In this embodiment, the overheat diffusion lag feedforward control mechanism starts the feedforward control strategy in real time, which is based on the physical characteristics of insufficient thermal response to estimate the required heating compensation in advance. Especially considering the significant influence of bathroom floor thermal inertia coefficient Kn on heat conduction delay, if no compensation is added, it is easy to appear "false steady state" misjudgment, that is, the temperature is long-term stagnation at the critical point and does not adjust the power. By introducing ΔTn and Kn as the basis for feedforward calculation, the system can compensate for the time proportion in the existing response period, and improve the adjustment sensitivity. In S32, the pre-compensation heating time Δt n To compensate for the index, the time proportion of the current heating level is accurately extended, rather than taking the extensive "level +1" control. For example, the original heating time of level 1 is 30%, and when Δtn=6s, Tp=60s, it is adjusted to 40%. This not only maintains the stability of the level, but also smoothly increases the output, avoiding system shock caused by frequent jumps. In addition, level 3 is forced to cap at 90% power time, which is to reserve 10% observation window for subsequent discrimination of key parameters such as ΔTn and Kn, to avoid control blindness. The above mechanism is particularly suitable for bathroom radiant heating temperature control scenes. In the case of high structural thermal resistance, large heat capacity and slow response, the control lag time can be greatly reduced, and the continuity and energy saving of the whole temperature control process can be improved.

[0131] Embodiment 6

[0132] Please refer to Figure 1 Specifically, S4 includes S41;

[0133] S41, after the heating control is updated, the updated real-time temperature T' of each control section is reacquired, the normalized difference value calculation is performed between the updated real-time temperature T' of all control sections and the preset target temperature Ttarget, based on the total number of control sections N, the temperature dispersion score Qs is calculated and output, and the deviation of all control sections is summarized and evaluated;

[0134] The temperature dispersion score Qs is calculated and output by the following algorithm formula;

[0135] ;

[0136] In the formula, N represents the total number of control sections, T'n represents the updated real-time temperature of the nth control section.

[0137] S4 further comprises S42;

[0138] S42, based on the actual bathroom floor temperature control, based on the preset target temperature Ttarget±5%, set as the uniformity threshold interval, compare the uniformity threshold interval with the real-time acquired temperature discrete score Qs with the uniformity threshold interval, judge the uniformity of the temperature control effect, and based on the comparison result, optimize and reconstruct the parameters in the heat load calculation model; The specific comparison content is as follows:

[0139] When the temperature discrete score Qs belongs to the uniformity threshold interval, it means that the temperature control running state is qualified, and the current parameters are kept;

[0140] When the temperature discrete score Qs does not belong to the uniformity threshold interval, it means that the temperature control running state is unqualified, at this time the parameter adaptive adjustment process of the heat load calculation model is executed, and after adjustment, heating control is updated again;

[0141] The parameter adaptive adjustment process of the heat load calculation model constructs a target function by using the original heat load calculation model form based on the historical standardized data set of M control cycles, and adopts linear regression to re-estimate the temperature rising rate response factor And the heat inertia compensation factor Output the optimal combination, realize the adaptive optimization of the heat load level calculation logic, and improve the overall uniformity and response efficiency of the system temperature control.

[0142] In this embodiment, the method calculates the normalized deviation of the temperature of each control section relative to the target temperature Ttarget through the temperature discrete score Qs, and evaluates the overall thermal uniformity performance of the system. This processing method can directly reflect the consistency of the current heating strategy in each region, and avoid covering up the overall problem due to the lagging or overheating of a certain region. For example, if a control section far from the heat source is always below the target temperature by more than 5°C, but the temperature rise in other regions is normal, the traditional mean analysis will ignore this anomaly, but the temperature discrete score Qs can truly reflect the distribution of the overall control error. S42 sets the uniformity threshold interval based on ±5% target temperature as the judgment standard of temperature control precision, to avoid excessive convergence leading to control fluctuation. When the temperature discrete score Qs is not up to standard, the re-estimation logic of the temperature rising rate response factor and the heat inertia compensation factor in the heat load calculation model is triggered, a set of more optimal parameter combination is output through linear regression method, so as to dynamically optimize the response ability of the system. This adaptive mechanism is particularly suitable for seasonal changes or sudden changes of bathroom floor water content, and can significantly improve the environmental adaptation ability and robustness of the temperature control model.

[0143] Embodiment 6

[0144] Referring to Figure 1 and Figure 2 The application discloses a linear heater intelligent control system, which comprises a control section acquisition module, a heat load grade division module, a multi-stage heating control module and a temperature dispersion analysis module.

[0145] The control section acquisition module divides the linear heater into multiple control sections, sets an acquisition device at the position of each control section, acquires the heat load data of each control section, and transmits the heat load data to a central control server for preprocessing to obtain a standardized data set.

[0146] The heat load grade division module constructs a heat load calculation model based on the standardized data set, calculates the heat load grade value E of each control section, and executes multi-stage heating strategy control according to the heat load grade value E.

[0147] The multi-stage heating control module triggers a heat diffusion lagging feedforward regulation mechanism based on the control result of the multi-stage heating strategy, calculates the pre-compensation heating extension time Δt, and adjusts the heating strategy in time to complete heating control update.

[0148] The temperature dispersion analysis module calculates the temperature dispersion score Qs of all control sections after each control cycle, compares the temperature dispersion score Qs with a set balance threshold interval, and optimizes and reconstructs the parameters in the heat load calculation model based on the comparison result.

[0149] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application.

Claims

1. A method for intelligent control of a linear heater, characterized in that: Includes the following steps: S1. Divide the linear heater into multiple control segments, set up a data collector at each control segment location to acquire the heat load data of each control segment, and transmit the heat load data to the central control server for preprocessing to obtain a standardized dataset. S2. Construct a heat load calculation model based on a standardized dataset, calculate the heat load level value E for each control segment, and execute multi-level heating strategy control according to the heat load level value E. S2 includes S21; S21. Construct a heat load calculation model based on a standardized dataset, calculate the heat load level value E of each control segment, and measure the current heat load status of each control segment. The heat load level value E is calculated using the following heat load calculation model; In the formula, E n This represents the heat load level value of the nth control segment, Ttarget represents the preset target temperature, which is the initially set dimensionless value, K n denoted by ρ, the thermal inertia coefficient of the nth control segment; ρ represents the heating rate response factor; σ represents the thermal inertia compensation factor; and ε represents a small constant to prevent division by zero, with a value of 0.

01. S3. Based on the control results of the multi-level heating strategy, a triggering heat diffusion lag feedforward regulation mechanism is implemented to calculate the pre-compensation heating extension time Δt, and the heating strategy is adjusted by time extension to complete the heating control update. S4. After each control cycle ends, calculate the temperature discrete score Qs of all control segments and compare it with the set equilibrium threshold range. Based on the comparison results, optimize and reconstruct the parameters in the heat load calculation model. S4 includes S41; S41. After the heating control update is completed, the updated real-time temperature T' of each control segment is reacquired. The normalized difference between the updated real-time temperature T' of all control segments and the preset target temperature Ttarget is calculated. Based on the total number of control segments N, the temperature discrete score Qs is calculated and output. The deviation of all control segments is summarized and evaluated. The temperature discrete score Qs is calculated and output using the following algorithm formula; In the formula, N represents the total number of control segments, and T n ' indicates the updated real-time temperature of the nth control segment.

2. The intelligent control method for a linear heater according to claim 1, characterized in that: S1 includes S11; S11. The linear heater is evenly divided into several equidistant control segments along the length of the bathroom floor heating system. Each control segment is set as an independent logical heating area. At the same time, a data collector is set in each control segment. The data collector of each control segment is set on the bottom and side edge structure of the bathroom floor below the center area of ​​the corresponding control segment to collect heat load data in real time. The data acquisition device includes a temperature acquisition module, a humidity recognition module, and a heating response rate recording module; The heat load data includes the real-time temperature T of the nth control segment region. n The unit time temperature rise rate ΔT of the nth control segment region n and the humidity Wc of the nth control segment area n .

3. The intelligent control method for a linear heater according to claim 2, characterized in that: S1 further includes S12; S12. The heat load data collected in real time by the collector is sent to the central control server through the wireless communication module. In the central control server, the heat load data is preprocessed. The preprocessing is based on a set time window and performs data cleaning, outlier removal and interpolation completion operations. For each segment of humidity Wc in the pre-processed heat load data, combined with the preset bathroom floor dry density Ps and specific heat capacity Cs, the thermal inertia coefficient K, which is used to evaluate the degree of hysteresis of the bathroom floor thermal response in all segments, is obtained using the thermal inertia calculation formula. And the real-time temperature T of the nth control segment region in the heat load data. n and the unit time temperature rise rate ΔT of the nth control segment region n By combining the thermal inertia coefficient K of each control segment, dimensional normalization and unit standardization are performed to form a standardized dataset with a unified format.

4. The intelligent control method for a linear heater according to claim 1, characterized in that: S2 further includes S22; S22. By fitting and regressing different temperature gradients, thermal response rates and final thermal equilibrium times in historical heating cycles, a heating load threshold model is constructed, and the first judgment threshold F1, the second judgment threshold F2 and the third judgment threshold F3 are automatically generated. The heat load level value E of each control segment is divided into heat load levels according to the first judgment threshold F1, the second judgment threshold F2, and the third judgment threshold F3, and a multi-level heating strategy control is executed simultaneously; the specific division is as follows: When the heat load level value E of the nth control segment n When the value is greater than or equal to the first judgment threshold F1, it is determined to be heating level 3, and a heating strategy of 90% time percentage within the set control cycle is executed. When the second judgment threshold F2 ≤ the heat load level value E of the nth control segment n When the first judgment threshold F1 is reached, it is determined to be heating level 2, and a heating strategy with a time percentage of 60% within the set control cycle is executed. When the third judgment threshold F ≤ the heat load level value E of the nth control segment n When the second judgment threshold F2 is reached, it is determined to be heating level 1, and a heating strategy with a time percentage of 30% within the set control cycle is executed. When the heat load level value E of the nth control segment n When the value is less than the third judgment threshold F, the heating level is determined to be 0, and heating is not performed in the current cycle; When the heating level remains constant for two consecutive heating control cycles, and the unit time temperature rise rate ΔT in the nth control segment region... n If the value is less than 0.2, it will be automatically identified as thermal response hysteresis.

5. The intelligent control method for a linear heater according to claim 4, characterized in that: S3 includes S31; S31. When thermal response hysteresis is identified, a thermal diffusion hysteresis feedforward control mechanism is triggered. The thermal diffusion hysteresis feedforward control mechanism uses the unit time temperature rise rate ΔT of the nth control segment region based on the current control segment standard dataset. n Combined with the thermal inertia coefficient K of the nth control segment region n The pre-compensation heating extension time Δt is calculated and output, and heating compensation is implemented for the control section with thermal response lag. The pre-compensation heating extension time Δt is calculated and output using the following algorithm formula; In the formula, △t n δ represents the pre-compensation heating extension time of the nth control segment, δ represents the minimum heating rate threshold (dimensionless), Kavg represents the average thermal inertia value of all control terminals, and μ represents the feedforward response adjustment factor.

6. The intelligent control method for a linear heater according to claim 5, characterized in that: S3 further includes S32; S32. The pre-compensation heating extension time Δt of the nth control segment calculated based on the response lag segment. n The heating strategy is adjusted by extending the heating time percentage within the current control cycle according to the following strategy, thus completing the heating control update: If the current heat load level of the control segment is Level 1, then based on the original heating time percentage of 30%, the heating time will be extended by 30% + the pre-compensation heating extension time Δt of the nth control segment. n / Complete heating control cycle time Tp; If the current heat load level of the control segment is Level 2, the heating time percentage will be extended by 60% + the pre-compensation heating extension time Δt of the nth control segment, based on the original setting of 60% heating time. n / Complete heating control cycle time Tp; If the current control section heat load level is level 3, then the heating time percentage remains unchanged at 90%.

7. The intelligent control method for a linear heater according to claim 1, characterized in that: S4 also includes S42; S42. Based on actual bathroom floor temperature control, a balance threshold range is set at ±5% of the preset target temperature Ttarget. This balance threshold range is compared with the real-time temperature discrete score Qs to determine the balance of the temperature control effect. Based on the comparison results, the parameters in the heat load calculation model are optimized and reconstructed. The specific comparison content is as follows: When the temperature dispersion score Qs is within the balance threshold range, it indicates that the temperature control operation is qualified and the current parameters should be maintained. When the temperature discrete score Qs is not within the equilibrium threshold range, it indicates that the temperature control operation is not qualified. At this time, the parameter adaptive adjustment process of the heat load calculation model is executed, and the heating control is updated again after adjustment. The parameter adaptive adjustment process of the heat load calculation model constructs an objective function based on a historical standardized dataset of M control cycles using the original heat load calculation model. With the goal of minimizing temperature control error, linear regression is used to re-estimate the parameters of the heating rate response factor ρ and the thermal inertia compensation factor σ, and outputs the optimal combination.

8. A linear heater intelligent control system, applied to the linear heater intelligent control method according to any one of claims 1-7, characterized in that: It includes a control section acquisition module, a heat load level classification module, a multi-level heating control module, and a temperature discrete analysis module; The control segment acquisition module divides the linear heater into multiple control segments, sets an acquisition device at the location of each control segment, acquires the heat load data of each control segment, and transmits the heat load data to the central control server for preprocessing to obtain a standardized dataset. The heat load level classification module constructs a heat load calculation model based on a standardized dataset, calculates the heat load level value E for each control segment, and executes multi-level heating strategy control based on the heat load level value E. The multi-level heating control module calculates the pre-compensation heating extension time Δt by triggering a heat diffusion lag feedforward regulation mechanism based on the control results of the multi-level heating strategy, and adjusts the heating strategy by time extension to complete the heating control update. The temperature discrete analysis module calculates the temperature discrete score Qs of all control segments after each control cycle ends, compares it with the set equilibrium threshold range, and optimizes and reconstructs the parameters in the heat load calculation model based on the comparison results.

Images