Photovoltaic hot air stratification collaborative control method for improving building temperature uniformity

Abstract

The application discloses a photovoltaic hot air layering cooperative control method for improving building temperature uniformity, and relates to the technical field of data processing. The method comprises the following steps: collecting indoor temperatures of each floor of a building, temperatures of each layer section of a photovoltaic ventilation cavity and an outdoor environment temperature in real time, and calculating vertical temperature non-uniformity. According to the non-uniformity and the indoor-outdoor temperature difference, a control mode is adaptively selected. Based on the selected mode, temperature deviations of each floor and the cavity temperature, a baffle target opening degree adjustment instruction is calculated layer by layer through a baffle opening degree fuzzy decision model, and the instruction is sent to a baffle actuator to complete adjustment, and the whole process is repeated based on updated temperature data in the next control cycle to form a closed-loop feedback control. The application realizes layered refinement and adaptive intelligent regulation and control of the photovoltaic hot air system, effectively improves the vertical temperature uniformity of the building, and improves the comprehensive energy efficiency of the system.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a photovoltaic hot air stratified collaborative control method for improving building temperature uniformity. Background Technology

[0002] With increasingly stringent building energy efficiency requirements and the large-scale application of renewable energy, building-integrated photovoltaics (BIPV) technology is gaining increasing attention. In multi-story or high-rise buildings, the ventilation cavities formed between photovoltaic panels and walls can effectively utilize the hot air generated by solar energy to provide heat to the building interior or achieve natural ventilation. However, due to the thermal pressure effect within the building, hot air naturally rises in the vertical cavities, often leading to excessive heat accumulation in the upper part of the building. This results in excessively high temperatures in the top-floor spaces, while the lower-floor spaces suffer from insufficient heating, creating a vertical temperature gradient. This not only reduces indoor thermal comfort but also affects the effective utilization of thermal energy from the photovoltaic system, becoming a key issue restricting the overall energy efficiency improvement of photovoltaic hot air systems.

[0003] Currently, existing technical solutions for building temperature control, such as integrated ventilation control or zoned adjustment based on simple temperature thresholds, mostly focus on on / off or proportional adjustment of the overall building ventilation volume or a single heat source. They lack targeted analysis and dynamic response to the characteristics of vertical temperature distribution, making it difficult to effectively identify and suppress vertical temperature unevenness caused by thermal pressure effects. The control process is also characterized by lag and crudeness. Furthermore, existing methods typically do not comprehensively analyze and coordinate decisions regarding outdoor ambient temperature, indoor temperatures on each floor, and the temperature inside ventilation cavities. They lack intelligent optimization mechanisms that can balance local temperature comfort with overall temperature uniformity, making it difficult to achieve adaptive, layered, and refined control of photovoltaic thermal air systems. Therefore, they are significantly insufficient in improving the vertical temperature uniformity of buildings and enhancing the overall energy efficiency of the system. Summary of the Invention

[0004] This invention addresses the technical problems of existing control methods being crude, lagging, and lacking the ability to collaboratively optimize vertical thermal distribution. It provides a photovoltaic thermal air stratified collaborative control method for improving building temperature uniformity.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] This invention provides a photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity, comprising: The system acquires the indoor temperature of each floor of the target building, the cavity temperature of each independent section of the ventilation cavity, and the outdoor ambient temperature in real time. Based on the indoor temperature of each floor, the system calculates the vertical temperature non-uniformity of the target building. Based on the vertical temperature non-uniformity and the indoor-outdoor temperature difference, a target control mode is determined from a variety of preset control modes. The selection of the target control mode is based on the comparison between the vertical temperature non-uniformity and a preset threshold. Based on the target control mode, the deviation between the indoor temperature of each floor and the set target temperature, and the cavity temperature of each floor segment, combined with the layered fine-tuning strategy, the target opening adjustment command of the baffle corresponding to each independent floor segment cavity is independently calculated and output through the pre-constructed baffle opening fuzzy decision model. The target opening adjustment command is sent to the actuator of the corresponding baffle to control each baffle to adjust to the target opening. In the next control cycle, the real-time acquisition, mode determination, command calculation and issuance steps are repeated based on the updated temperature data to form a closed-loop feedback control.

[0007] The beneficial effects of this invention are: This invention first constructs an evaluation index that comprehensively reflects the vertical heat distribution of a building by integrating multi-source temperature sensing data from indoor, outdoor, and cavity sources, laying a data foundation for accurately identifying temperature unevenness issues. Second, based on the dynamic comparison of this index with preset thresholds, it achieves intelligent switching between various control modes, such as strong temperature uniformity and weak regulation, enabling the control strategy to adapt to different thermal environment conditions. Third, it innovatively combines fuzzy decision-making and particle swarm optimization algorithms to establish an intelligent model capable of independently calculating the optimal baffle opening layer by layer. This model effectively coordinates the balance between local temperature comfort and overall temperature uniformity through optimized individual cost coefficients. Finally, by forming a closed-loop feedback control architecture, it achieves continuous, automatic, and refined regulation of the photovoltaic thermal air system, thereby improving the vertical temperature uniformity of the building and simultaneously enhancing the overall energy utilization efficiency of the photovoltaic system. Attached Figure Description

[0008] Figure 1 This is a schematic flowchart of the photovoltaic hot air stratified collaborative control method for improving building temperature uniformity provided by the present invention. Figure 2 This is a flowchart illustrating the process of calculating and outputting the target opening adjustment command of the baffle based on the fuzzy decision model of baffle opening provided by the present invention. Detailed Implementation

[0009] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0010] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0011] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0012] Example 1, as Figure 1 As shown, this embodiment of the invention provides a photovoltaic-thermal air stratified collaborative control method for improving building temperature uniformity, including: S10: Real-time acquisition of indoor temperature on each floor of the target building, cavity temperature of each independent section in the ventilation cavity, and outdoor ambient temperature; calculation of vertical temperature non-uniformity of the target building based on the indoor temperature of each floor. First, indoor temperature data is continuously collected from each floor of the target building using temperature sensors deployed indoors. Simultaneously, temperature data is collected from each independent segment of the vertical ventilation cavity formed between the photovoltaic panels on the exterior wall and the interior wall, using temperature sensors deployed within this cavity. Furthermore, outdoor ambient temperature data is simultaneously collected using temperature sensors deployed in the outdoor environment. Ultimately, the indoor temperatures of each floor of the target building, the temperatures of each independent segment of the ventilation cavity, and the outdoor ambient temperature are obtained, achieving comprehensive and synchronous monitoring of key temperature parameters related to the building's thermal environment. The target building refers to a building that applies this photovoltaic-thermal-air layered collaborative control method and has a corresponding photovoltaic ventilation cavity physical structure, such as a multi-story office building, residential building, or public building using a building-integrated photovoltaic (BIPV) design.

[0013] Furthermore, based on the real-time indoor temperatures of each floor, the vertical temperature non-uniformity of the target building is calculated. This vertical temperature non-uniformity is a comprehensive quantitative index, whose value comprehensively reflects the dispersion and unevenness of the indoor temperature distribution of each floor in the vertical spatial dimension of the target building. It characterizes the level of temperature difference between upper and lower floors caused by factors such as thermal pressure effect, uneven distribution of solar radiation, or improper ventilation control. This vertical temperature non-uniformity is the core basis for assessing the vertical uniformity of the building's thermal environment and for triggering and guiding subsequent layered collaborative control decisions.

[0014] Specifically, the calculation steps for the vertical temperature non-uniformity include: Calculate the standard deviation of indoor temperature on each floor as a basic non-uniformity index to characterize the overall dispersion. Calculate the absolute value of the indoor-outdoor temperature difference based on the outdoor ambient temperature and the average indoor temperature of the building; The cavity temperature of each independent section within the ventilation cavity is obtained, and the absolute value of the difference between the cavity temperature and the corresponding floor temperature is calculated to obtain the cavity-indoor temperature difference set for each floor. Based on the aforementioned basic non-uniformity index, the absolute value of the indoor-outdoor temperature difference and the average value of the cavity-indoor temperature difference set of each floor are introduced as correction factors and weighted fusion is performed to obtain a comprehensive vertical temperature non-uniformity.

[0015] First, based on the real-time indoor temperatures of each floor, the standard deviation is calculated. This standard deviation directly characterizes the dispersion of the indoor temperature of each floor relative to its average value, reflecting the uniformity of the indoor temperature distribution in the vertical direction of the target building. This standard deviation is used as a preliminary, basic non-uniformity index based on the single dimension of indoor temperature to characterize the overall dispersion.

[0016] Secondly, to further improve the accuracy and environmental adaptability of the vertical temperature non-uniformity assessment, two types of correction factors are introduced. First, the absolute value of the difference between the outdoor ambient temperature and the average indoor temperature is calculated to obtain the absolute value of the indoor-outdoor temperature difference. This absolute value reflects the driving potential difference of the external environment on the overall thermal state of the building and serves as a correction factor for the external environment's influence. Second, for each independent section of the ventilation cavity, its temperature data is acquired and compared with the indoor temperature data of the corresponding floor. The absolute value of the difference between the two is calculated. After traversing all floors, a set containing the corresponding differences for each floor is obtained, i.e., the cavity-indoor temperature difference set for each floor. The average value of this cavity-indoor temperature difference set is calculated. This average value reflects the average heat exchange potential and state between the ventilation cavity and the building's indoor space along the vertical direction and serves as a correction factor for internal heat transfer.

[0017] Finally, a comprehensive weighted fusion of indicators is implemented. Specifically, the basic non-uniformity index calculated above is used as the core benchmark value, and the absolute value of the indoor-outdoor temperature difference and the average value of the cavity-indoor temperature difference set are used as correction factors. These are then fused using preset weighting coefficients. This weighted fusion process quantitatively superimposes the influence of external environmental factors and internal heat transfer states onto the basic temperature dispersion, thereby generating a comprehensive vertical temperature non-uniformity. This vertical temperature non-uniformity overcomes the limitations of single-dimensional evaluation and can more comprehensively and accurately characterize the overall unevenness of the vertical temperature distribution of a building under specific indoor and outdoor thermal environments.

[0018] Specifically, vertical temperature non-uniformity In this formula, This represents the standard deviation calculated from the indoor temperature of each floor, i.e., the base non-uniformity. It represents the absolute value of the difference between the outdoor ambient temperature and the average indoor temperature of the building, i.e., the absolute value of the indoor-outdoor temperature difference; This represents the average absolute value of the difference between the temperature of the ventilation cavity section corresponding to each floor and the indoor temperature of that floor, i.e., the average cavity-indoor temperature difference. The coefficients 0.6, 0.2, and 0.2 in the formula are weighting coefficients for the corresponding calculation items, and are only examples. Specific values ​​can be calibrated and optimized according to the specific building type, such as residential or office buildings, and the adjustment range is usually within ±0.1. This vertical temperature non-uniformity not only considers the dispersion of indoor temperature but also integrates the relationship between indoor and outdoor thermal environments and the thermal interaction state between the cavity and the indoor environment. Therefore, it can more comprehensively and accurately quantify the overall temperature non-uniformity in the vertical direction of the building, thus providing a reliable and multi-dimensional evaluation basis for subsequent control decisions.

[0019] S20: Based on the vertical temperature non-uniformity and the indoor-outdoor temperature difference, determine the currently applicable target control mode from a variety of preset control modes, wherein the selection of the target control mode is based on the comparison relationship between the vertical temperature non-uniformity and a preset threshold. Among them, the multiple control modes include at least a strong temperature uniformity mode and a weak adjustment mode. The strong temperature uniformity mode is activated when the vertical temperature non-uniformity is higher than a first preset threshold for active intervention. The weak adjustment mode is activated when the vertical temperature non-uniformity is lower than a second preset threshold for maintenance fine-tuning.

[0020] Specifically, based on the vertical temperature non-uniformity and the indoor-outdoor temperature difference, the target control mode is determined from a variety of preset control modes. Its core lies in intelligently selecting the most suitable control strategy based on the real-time building thermal state assessment results.

[0021] First, the calculated vertical temperature non-uniformity is compared with preset thresholds. Specifically, a first preset threshold and a second preset threshold are set. The first preset threshold is a critical value used to determine whether the building's vertical temperature distribution has become significantly non-uniform, representing the threshold for initiating strong intervention and control. The second preset threshold is a critical value used to determine whether the building's vertical temperature distribution is in a good and uniform state, representing the threshold for entering maintenance fine-tuning. The setting of the first and second preset thresholds refers to the general thermal comfort standard ISO 7730 to ensure that the control target meets the scientific requirements of human thermal comfort. The value of the first preset threshold is greater than the second preset threshold. For example, in a typical office building scenario, the first preset threshold can be set to 1.5℃ and the second preset threshold to 0.5℃.

[0022] If the current vertical temperature non-uniformity is greater than or equal to the first preset threshold, it indicates that the building's vertical temperature distribution has significantly deviated from a uniform state, and at this time, the strong temperature uniformity mode is activated. This strong temperature uniformity mode aims to actively and forcefully intervene in the building's thermal environment, with the core objective of rapidly reducing vertical temperature differences. If the current vertical temperature non-uniformity is less than or equal to the second preset threshold, it indicates that the building's vertical temperature distribution is in a relatively good uniform state, and at this time, the weak adjustment mode is activated. This weak adjustment mode focuses on maintaining fine-tuning, with the goal of optimizing the local thermal comfort of each floor while maintaining overall uniformity.

[0023] Furthermore, when the vertical temperature non-uniformity falls between the first and second preset thresholds, it indicates that the temperature distribution is in a critical transition state, requiring the introduction of a second layer of decision logic. At this point, it is further determined whether the absolute value of the indoor-outdoor temperature difference exceeds a preset temperature difference threshold. The preset temperature difference threshold refers to the temperature difference limit used to assess the significant driving effect of the outdoor environment on the building's indoor thermal environment. It represents the critical condition requiring attention to the influence of external climatic factors. This temperature difference threshold is comprehensively set based on local climate characteristics, the thermal performance of the building envelope, and indoor thermal comfort goals. For example, in transitional seasons or temperate climate regions, this temperature difference threshold can be set to 5°C.

[0024] Specifically, this assessment aims to evaluate the potential impact of the external environment on the building's thermal state. If the absolute value of the indoor-outdoor temperature difference exceeds this threshold, it indicates strong external thermal disturbance, which may exacerbate the trend of uneven internal temperature. Therefore, a strong temperature uniformity mode should still be applied for preventative intervention. If the absolute value of the indoor-outdoor temperature difference does not exceed this threshold, it indicates that the external environment is relatively stable and poses little threat to internal uniformity. In this case, a weak regulation mode can be applied for mild control.

[0025] Through the above two-level logical judgment, the applicable target control mode can be determined, thereby providing a clear strategic guide for subsequent tiered regulation.

[0026] S30: Based on the target control mode, the deviation between the indoor temperature of each floor and the set target temperature, and the cavity temperature of each floor segment, combined with the layered fine-tuning strategy, the target opening adjustment command of the baffle corresponding to each independent floor segment cavity is independently calculated and output through the pre-constructed baffle opening fuzzy decision model. Specifically, this step utilizes the established target control mode, the deviation between the real-time indoor temperature of each floor and its set target temperature, and the cavity temperature of each floor segment. Combined with a stratified fine-tuning strategy aimed at coordinating local and overall thermal comfort, the target opening adjustment command for the corresponding ventilation cavity segment baffle is independently calculated and generated for each floor through a pre-built fuzzy decision model for baffle opening.

[0027] Among them, the fuzzy decision model for damper opening is an intelligent decision-making algorithm based on fuzzy control and historical data training. This model establishes a nonlinear mapping relationship from the real-time thermal environment state of the building to the optimal opening adjustment of ventilation dampers on each floor by learning the historical optimal control strategy. Specifically, this mapping relationship transforms multi-dimensional input variables such as the target control mode, floor indoor temperature deviation, and floor cavity temperature into specific and quantified opening adjustment commands for each independent floor damper. This enables refined, adaptive, and tiered control that balances overall temperature uniformity and local thermal comfort in complex and dynamically changing building thermal environments.

[0028] Specifically, the construction steps of the fuzzy decision model for the baffle opening include: Collect historical temperature datasets of the target building under typical seasons and operating conditions, and label the ideal baffle opening state of each floor at each time to obtain a training sample set. The historical temperature data includes at least the indoor temperature of each floor, the cavity temperature of the corresponding floor section, and the outdoor ambient temperature. The ideal baffle opening state is determined by the historical optimal control record. Based on fuzzy control theory, the structure of the fuzzy decision model for the baffle opening degree is constructed. Using the training sample set, the membership function parameters and fuzzy rule weights in the baffle opening fuzzy decision model are learned and optimized until the model's control accuracy on the validation set converges, thus completing the model construction.

[0029] First, training data is collected and labeled. Historical temperature datasets of the target building are collected under typical seasons, such as summer and winter, and various typical operating conditions. This historical temperature dataset must include at least the indoor temperature of each floor, the cavity temperature of the corresponding floor section, and the outdoor ambient temperature. Simultaneously, based on the historical operating records of the target building, the optimal damper opening degree for each floor at each historical moment is determined through analysis, and this opening degree is labeled as the ideal damper opening degree at that moment. This results in a training sample set consisting of historical environmental data and corresponding ideal control actions.

[0030] Secondly, based on fuzzy control theory, a fuzzy decision-making model for the baffle opening is constructed. Fuzzy control theory is an intelligent control method for handling imprecision, uncertainty, and nonlinear problems. Its core lies in mimicking the experience and reasoning methods of human experts, transforming precise numerical inputs into fuzzy linguistic variables such as high, medium, and low, and then performing approximate reasoning through a fuzzy rule base based on if-then rules. Finally, the fuzzy conclusions derived from the reasoning are transformed back into precise numerical outputs. This process does not rely on a precise mathematical model of the controlled object, but rather utilizes a knowledge base and experience for decision-making, making it particularly suitable for complex systems such as building thermal environments that exhibit strong nonlinearity, large time lag, and are difficult to model precisely.

[0031] Specifically, the input variables of the fuzzy decision model for baffle opening are the target control mode, the deviation between the indoor temperature of each floor and the set target temperature, and the cavity temperature of the corresponding floor segment; the output variable is the target opening adjustment command for the corresponding baffle. For example, the fuzzy decision model for baffle opening can consist of three functional modules: a fuzzification interface, a fuzzy inference engine, and a defuzzification interface. The fuzzification interface is responsible for converting the precise input values, including the target control mode, the deviation between the indoor temperature of each floor and the set target temperature, and the cavity temperature of the corresponding floor segment, into the membership degrees of the corresponding fuzzy set.

[0032] The fuzzy decision model for damper opening requires defining the universe of discourse and fuzzy set for each input and output variable. For example, the universe of discourse for indoor temperature deviation can be set to -5℃ to +5℃, and divided into five fuzzy subsets (negative large, negative small, zero, positive small, and positive large) using a triangular membership function. Each subset is described by a membership function to quantify the degree to which any precise temperature deviation value belongs to a certain fuzzy concept. The target control mode, as a categorical variable, is directly mapped to the switching signal that activates different fuzzy rule subsets. The core of the fuzzy decision model for damper opening is a rule base containing multiple fuzzy rules. Each rule is in the form that if the input conditions satisfy certain fuzzy combinations, then the output should be a certain fuzzy opening adjustment amount. This initial rule base can be initially established based on the experience of domain experts or the analysis and summarization of historical optimal control records. The defuzzification interface uses methods such as the centroid method to transform the fuzzy inference results into precise output instructions. Through three steps—fuzzification, rule-based reasoning, and defuzzification—this fuzzy decision model for baffle opening can ultimately convert precise real-time measurement data into precise baffle opening adjustment commands.

[0033] Specifically, the preliminarily constructed fuzzy decision-making model is trained using the aforementioned training sample set. For example, during model training, the training sample set can be divided into a training subset and a validation subset according to a preset ratio, such as a 7:3 ratio. The training process mainly focuses on automatically learning and optimizing the membership function parameters and the weights of each fuzzy rule in the baffle opening fuzzy decision-making model. The mean squared error between the output command and the ideal opening state labeled in the samples is used as the loss function. A gradient descent algorithm, combined with an adaptive learning rate optimizer, is used to iteratively optimize the adjustable parameters in the baffle opening fuzzy decision-making model. By backpropagating the error signal, continuous adjustments are made to continuously improve the consistency between the baffle opening adjustment command output by the model and the ideal opening state labeled in the training subset. The training process is monitored and evaluated using an independent validation subset, with the evaluation metric being the adjustment accuracy, i.e., the degree of matching between the opening command output by the model on the validation subset and the ideal opening state.

[0034] Specifically, through iterative optimization, the consistency between the baffle opening adjustment command output by the fuzzy decision model and the ideal opening state labeled in the training samples is continuously improved. When the control accuracy index of the fuzzy decision model on the validation dataset reaches stability and convergence, the training is considered complete, and the fuzzy decision model is deemed to have reliable decision-making ability. The convergence condition is set based on the performance change trend of the validation set and a preset threshold; for example, the improvement in the control accuracy of the validation set is less than 5‰ in 10 consecutive iterations.

[0035] Specifically, based on the target control mode, the deviation between the indoor temperature of each floor and the set target temperature, and the cavity temperature of each floor segment, combined with a layered fine-tuning strategy, and through a pre-constructed fuzzy decision model for baffle opening, the target opening adjustment command for the baffle corresponding to each independent floor segment cavity is independently calculated and output layer by layer, including: For each floor, the primary optimization objective for the current fuzzy decision is determined based on the target control mode. The deviation between the current floor's indoor temperature and the set target temperature, the cavity temperature of the corresponding floor section, and the vertical temperature non-uniformity are used as inputs to the pre-constructed fuzzy decision model for baffle opening. In the decision-making process of the baffle opening fuzzy decision model, an individual cost coefficient is set for the baffle opening adjustment of each floor. The individual cost coefficient is used to adjust the weight ratio between the local target of temperature control of this floor and the global target of temperature uniformity of the whole building. The particle swarm optimization algorithm is used to iteratively optimize the individual cost coefficients by using the weighted sum of the vertical temperature non-uniformity of the building as a whole and the temperature deviation of each floor under long-term operation as the fitness function, so as to obtain a set of individual cost coefficients that optimize the overall control performance. The optimized individual cost coefficient is used to configure the fuzzy decision model for the baffle opening. Fuzzy reasoning is performed independently for each floor to calculate and output the corresponding baffle target opening adjustment command.

[0036] First, for each floor, based on the target control mode, determine the primary optimization objective for the current fuzzy decision, including: If the target control mode is a strong uniform temperature mode, then minimizing the vertical temperature non-uniformity is the primary optimization objective. If the target control mode is a weak adjustment mode, then minimizing the deviation between the current floor's indoor temperature and the set target temperature is the primary optimization objective.

[0037] Specifically, when making independent decisions on the opening degree of the baffle for each floor, the first step is to clarify the core orientation of this fuzzy decision based on the established target control mode, namely the primary optimization target. This step aims to transform the macro control strategy into specific optimization criteria that guide local regulation of each floor.

[0038] If the current target control mode is a strong uniform temperature mode, it indicates that the uneven vertical temperature distribution of the building is a significant problem, and the primary task of control is to quickly improve the overall uniformity. Therefore, for the fuzzy decision-making of each floor, the primary optimization objective is set to minimize the overall vertical temperature non-uniformity of the building. This means that when calculating the baffle opening for this floor, the fuzzy decision-making model will prioritize the contribution of the floor's control action to reducing the temperature difference across the entire building.

[0039] If the current target control mode is a weak adjustment mode, it indicates that the overall temperature distribution of the building is already in a good and uniform state. The core of control then shifts to maintaining this state and optimizing local comfort. Therefore, for the fuzzy decision-making of each floor, the primary optimization objective is to minimize the deviation between the current floor's indoor temperature and the set target temperature. This means that when calculating the baffle opening for this floor, the fuzzy decision model will prioritize how to make the indoor temperature of this floor approach the set target value more accurately and quickly, in order to achieve optimal local thermal comfort.

[0040] Through the differentiated target setting based on the target control mode, the fuzzy decision model can dynamically adjust its decision focus, thereby achieving an adaptive balance between overall uniformity and local comfort. This ensures that the control strategy can both cope with significant uniformity imbalance and make fine-tuning adjustments to comfort.

[0041] Furthermore, the deviation between the current floor's indoor temperature and the set target temperature, the cavity temperature of the corresponding floor segment, and the vertical temperature non-uniformity are used as inputs to the pre-constructed fuzzy decision model for baffle opening. The set target temperature refers to the desired temperature value pre-set for the indoor space of this floor. This target temperature value is set according to human thermal comfort standards, seasonal patterns, and specific functional requirements; for example, it can be set to 20℃ under winter heating conditions and 26℃ under summer cooling conditions.

[0042] Specifically, the first type of input is the deviation between the current floor's indoor temperature and the set target temperature. This deviation directly quantifies the immediate comfort deviation of the local thermal environment on that floor. The second type of input is the temperature of the cavity in the independent segment corresponding to that floor. This cavity temperature reflects the immediate thermal state of the photovoltaic ventilation cavity at that floor height and is a key driving parameter affecting the inflow of hot air into that floor. The third type of input is the overall vertical temperature non-uniformity of the building calculated above. This vertical temperature non-uniformity characterizes the macroscopic uniformity level of temperature distribution in the vertical dimension of the building. Among these, the deviation between the current floor's indoor temperature and the set target temperature provides a local control demand signal for the baffle opening fuzzy decision model, the segment cavity temperature provides information on available heat sources or heat dissipation conditions, and the vertical temperature non-uniformity provides the baffle opening fuzzy decision model with global constraints and optimization target information that need to be considered.

[0043] In the decision-making process of the fuzzy decision model for baffle opening, an individual cost coefficient needs to be set for the baffle opening adjustment of each floor. This individual cost coefficient is an adjustable weight parameter, and its core function is to quantify and balance the relative importance between the two potential effects involved in the temperature control of this floor: one effect is the direct adjustment effect of the control action on the indoor temperature of this floor itself, that is, the contribution to achieving the local comfort goal; the other effect is the indirect effect of the control action on the overall vertical temperature non-uniformity of the building by changing the hot air distribution, that is, the contribution to achieving the global uniformity goal. Specifically, the value range of the individual cost coefficient is usually set between 0 and 1. When the individual cost coefficient is small, it indicates that the decision-making process tends to prioritize the local temperature control needs of this floor, that is, to focus on quickly reducing the deviation between the indoor temperature and the set target. When the individual cost coefficient is large, it indicates that the decision-making process tends to prioritize the improvement of the overall temperature uniformity of the building, that is, to focus on reducing the vertical temperature non-uniformity through the control action of this floor, even if it may not be possible to optimally meet the local comfort goal of this floor temporarily.

[0044] By independently configuring and optimizing the individual cost coefficient for each floor, the fuzzy decision model for baffle opening can achieve refined and differentiated adjustment of the weight ratio between local and global objectives within a unified decision framework.

[0045] Furthermore, a particle swarm optimization algorithm is adopted, using the weighted sum of the overall vertical temperature non-uniformity of the building and the temperature deviation of each floor under long-term operation as the fitness function to iteratively optimize the individual cost coefficients, so as to obtain a set of individual cost coefficients that optimize the overall control performance.

[0046] Specifically, a particle swarm optimization algorithm is used, with the weighted sum of the overall vertical temperature non-uniformity of the building and the temperature deviation of each floor under long-term operation as the fitness function, to iteratively optimize the individual cost coefficients, thereby obtaining a set of individual cost coefficients that optimize the overall control performance, including: The individual cost coefficient of each floor is regarded as an optimization variable, and the individual cost coefficients of all floors are used to form a position vector of a particle in multidimensional space. Initialize a particle swarm, where each particle represents a set of candidate values ​​for the individual cost coefficient; In each iteration of the particle swarm optimization algorithm, for each individual cost coefficient value scheme represented by each particle, the corresponding baffle opening fuzzy decision model is configured, and the long-term comprehensive control performance corresponding to each individual cost coefficient value scheme is evaluated based on the historical operation data of the target building. The evaluation is achieved by calculating the fitness function, which is the weighted sum of the vertical temperature non-uniformity of the building as a whole and the temperature deviation of each floor under long-term operation. Based on the evaluation results of the fitness function, update the individual historical best position of each particle and the global historical best position of the entire particle swarm. Based on the individual historical best position, the global historical best position, and the particle's motion pattern, the position vector of each particle is iteratively updated; Repeat the evaluation and update steps until the preset iteration termination condition is met. At this point, the set of individual cost coefficient values ​​corresponding to the global historical best position of the particle swarm is the optimal solution obtained by optimization.

[0047] The optimization process first requires formalizing the optimization problem. Specifically, the individual cost coefficient associated with each floor is defined as an independent optimization variable. The individual cost coefficients of all floors are then combined in floor order to form a multidimensional vector. In particle swarm optimization, this multidimensional vector, composed of all optimization variables, is defined as the position of a particle in the multidimensional search space.

[0048] Secondly, a particle swarm is initialized. For each particle in the swarm, an initial value is randomly assigned to each dimension of its position vector, i.e., each individual cost coefficient, within a preset feasible range of 0-1. Therefore, for each particle in the swarm, its complete position vector represents a specific and complete set of numerical assignments for the individual cost coefficient across all floors, i.e., a set of candidate value schemes available for subsequent evaluation.

[0049] Furthermore, in each iteration of the particle swarm optimization algorithm, an evaluation operation is performed for each particle in the swarm. The core of this evaluation lies in calculating a fitness function to assess its corresponding long-term comprehensive control performance. Specifically, the fitness function is defined as the weighted sum of the average value of the overall vertical temperature non-uniformity of the building and the average value of the deviation of the indoor temperature of each floor from its set target temperature during the simulated long-term operation period, according to preset weights. The value of this fitness function directly quantifies the comprehensive performance of the baffle opening fuzzy decision model in long-term control under the configuration of this set of individual cost coefficients. The lower the fitness value, the better the comprehensive performance of this set of individual cost coefficients in balancing overall uniformity and local comfort.

[0050] Specifically, in each iteration of the particle swarm optimization algorithm, for each individual cost coefficient value scheme represented by each particle, a corresponding fuzzy decision model for the baffle opening is configured, and based on the historical operating data of the target building, the long-term comprehensive control performance corresponding to each individual cost coefficient value scheme is evaluated, including: Extract multiple consecutive runtime data segments from the historical operation dataset of the target building to serve as a test dataset for evaluation; For each particle, a set of individual cost coefficient value schemes is configured one by one into the baffle opening fuzzy decision model; On each test dataset, the baffle opening fuzzy decision model under the corresponding configuration is used to simulate and control the control time step by step, and record the vertical temperature non-uniformity sequence and the temperature deviation sequence of each floor generated during the control process. Based on the vertical temperature non-uniformity sequence and the temperature deviation sequence of each floor, the fitness function value corresponding to the current segment test dataset is calculated. The average value of the fitness function calculated from each data segment is taken to obtain the overall evaluation result of the current group's individual cost coefficient value scheme, which is used as the fitness value of the current particle.

[0051] First, multiple consecutive runtime data segments are extracted from the historical operational dataset of the target building to form a test dataset for performance evaluation. The selection of these consecutive runtime data segments should cover different typical seasons and diverse weather conditions to ensure the comprehensiveness and robustness of the evaluation results.

[0052] Secondly, for the particle to be evaluated, the specific set of individual cost coefficient numerical schemes it represents are fully configured into the baffle opening fuzzy decision model, thereby forming a baffle opening fuzzy decision model instance with specific parameter settings.

[0053] Then, using this configured fuzzy decision model instance for baffle opening, simulations are performed on each continuous runtime segment of the test dataset. Specifically, the simulation process takes the time sequence in the test dataset as input, and the fuzzy decision model for baffle opening calculates and outputs baffle opening adjustment commands time by time, thereby simulating the complete control process. During this process, two key indicator sequences generated at each simulation moment need to be continuously recorded: one is the vertical temperature non-uniformity sequence of the entire building, and the other is the temperature deviation sequence between the indoor temperature of each floor and the set target temperature.

[0054] For each segment of test data and simulation results, based on the recorded vertical temperature non-uniformity sequence and the temperature deviation sequence of each floor, the fitness function value corresponding to that segment of data is calculated according to a preset weighting ratio. Specifically, this fitness function value is the weighted sum of the overall vertical temperature non-uniformity of the building and the temperature deviation of each floor under long-term operation. Its value directly reflects the comprehensive control performance of the current individual cost coefficient configuration scheme under the operating conditions of that segment. The weighting coefficients corresponding to vertical temperature non-uniformity and temperature deviation of each floor in the weighted calculation are set according to the priority requirements of the control strategy for overall uniformity and local comfort. For example, the weight corresponding to overall uniformity can be set to 0.7, and the weight corresponding to local comfort can be set to 0.3.

[0055] Finally, the arithmetic mean of the multiple fitness function values ​​calculated across all test data segments is taken. This average value represents the overall performance evaluation result of the current individual cost coefficient value scheme and is assigned as the fitness value of the current particle in this iteration. This fitness value will serve as the core basis for driving the subsequent updates of the individual historical best position and the global historical best position in the particle swarm optimization algorithm, as well as the iterative adjustment of the position vectors of each particle.

[0056] Furthermore, based on the evaluation results of the fitness function, the individual best-in-class position remembered by each particle, as well as the global best-in-class position shared by the entire particle swarm, are updated. Specifically, for each particle, its fitness value calculated in the current iteration is compared with the fitness value corresponding to its recorded individual best-in-class position. If the current fitness value is better, its individual best-in-class position is updated using the particle's current position vector. Simultaneously, among all the individual best-in-class positions of all particles, the position with the best fitness value is selected and compared with the global best-in-class position recorded by the current particle swarm. If this position is better than the current global best-in-class position, the global best-in-class position is updated using this position.

[0057] Furthermore, based on each particle's individual historical best position, the global historical best position shared by the particle swarm, and the motion rules defined in the particle swarm optimization algorithm, the position vector of each particle is iteratively updated. Specifically, the motion rules are implemented through velocity update formulas and position update formulas.

[0058] The velocity update formula determines the particle's movement direction and step size in each iteration, comprehensively considering three components: the inertial component that maintains the particle's velocity from the previous generation, the cognitive component that drives the particle towards its individual historical best position, and the social component that guides the particle towards the global historical best position. Each component is assigned an adjustable weight coefficient, controlling the degree of influence of historical velocity, individual experience, and group experience on the current velocity, respectively. The velocity vector of the next generation of particles can be obtained through this formula. Specifically, the particle's movement direction is determined by the direction of this velocity vector, with the overall trend driving the particle towards a better region indicated by its individual historical best position and the global historical best position of the particle swarm. The step size is determined by the magnitude of this velocity vector, and the basic scale of the step size is adjusted by preset weight coefficients for each component; for example, the inertial weight coefficient can be set to 0.8, and the individual cognitive coefficient and social learning coefficient can both be set to 1.0.

[0059] The position update formula corrects the particle's current position based on the updated velocity vector. Specifically, it adds the particle's current position vector to the calculated new velocity vector to determine the particle's position in the next iteration. This new position corresponds to a set of updated individual cost coefficient values.

[0060] Therefore, the update process is essentially a process of dynamically adjusting the search direction and movement distance of a particle based on its own historical optimal experience and the global optimal experience shared by the particle swarm, thereby exploring and generating new, potentially better combinations of individual cost coefficient values ​​in the solution space, and then assigning these new combinations of individual cost coefficient values ​​to the particle.

[0061] Furthermore, the evaluation and update steps are repeated. Specifically, the steps of evaluating particle performance, updating historical best positions, and updating particle position vectors are repeated as a loop. This iterative process continues until a preset iteration termination condition is met. The iteration termination condition is set based on the computational resources and convergence requirements of the optimization process, such as reaching a maximum of 1000 iterations, or the fitness value improvement corresponding to the global historical best position being less than 1‰ in 20 consecutive iterations. When the iteration terminates, the specific set of individual cost coefficient values ​​corresponding to the currently recorded global historical best position of the particle swarm are determined as the optimal solution obtained through the optimization search. This set of individual cost coefficient values ​​enables the baffle opening fuzzy decision model to achieve the comprehensive optimal control performance by minimizing the weighted sum of the overall vertical temperature non-uniformity of the building and the temperature deviation of each floor during long-term operation.

[0062] Finally, using the optimal solution obtained after optimization, namely the fuzzy decision model for configuring the baffle opening based on the individual cost coefficient, fuzzy inference is performed independently for each floor to calculate and output the corresponding baffle target opening adjustment command.

[0063] Specifically, such as Figure 2 As shown, the optimized individual cost coefficient is used to configure the fuzzy decision model for the baffle opening. Fuzzy inference is performed independently for each floor, and the corresponding baffle target opening adjustment command is calculated and output, including: The optimized set of individual cost coefficients are used as corresponding weight parameters and configured into the decision logic units corresponding to each floor in the fuzzy decision model of the baffle opening. For each floor currently awaiting decision, the corresponding subset of fuzzy rules in the fuzzy decision model for the baffle opening is activated according to the target control mode. Using the deviation between the current floor's indoor temperature and the set target temperature, the cavity temperature of the corresponding floor segment, and the vertical temperature non-uniformity as real-time inputs, and combining the individual cost coefficient configured for the current floor, fuzzy inference is performed within the activated fuzzy rule subset, wherein the output result of the fuzzy inference is the target opening adjustment amount of the baffle corresponding to the current floor. Based on the target opening adjustment amount, a corresponding control command is generated and output as the target opening adjustment command for the baffle of the current floor.

[0064] Specifically, the optimized set of individual cost coefficients is first used as key weight parameters to configure the decision logic units corresponding to each floor in the fuzzy decision model for baffle opening. This configuration process essentially solidifies the optimized individual cost coefficients, which can balance local and global objectives, into the decision logic of the fuzzy decision model for baffle opening.

[0065] Secondly, for each floor currently requiring a decision, based on the determined target control mode, the corresponding preset subset of fuzzy rules in the baffle opening fuzzy decision model is activated. Specifically, if the current target control mode is a strong uniform temperature mode, the preset subset of fuzzy rules in the model with minimizing the current building's vertical temperature non-uniformity as the primary optimization objective is activated; if the current target control mode is a weak regulation mode, the preset subset of fuzzy rules in the model with minimizing the deviation between the current floor's indoor temperature and the set target temperature as the primary optimization objective is activated. This step ensures that the reasoning direction of the baffle opening fuzzy decision model is consistent with the macro-control strategy.

[0066] Furthermore, using the real-time deviation between the current floor's indoor temperature and the set target temperature, the cavity temperature of the corresponding floor segment, and the current vertical temperature non-uniformity of the building as inputs, and combining this with the individual cost coefficient configured for that floor, fuzzy inference is performed within the activated subset of fuzzy rules. This inference process comprehensively considers the local thermal state, available heat source conditions, and overall uniformity requirements, and its output is the target opening adjustment amount of the baffle corresponding to the current floor. This target opening adjustment amount is a specific value used to indicate the percentage by which the baffle opening needs to be changed from its current level, such as increasing by 15% or decreasing by 8%.

[0067] Finally, based on the calculated target opening adjustment amount, a clear and formatted control command is generated, such as increasing by 15% or decreasing by 8%. This control command is output as the target opening adjustment command for the current floor's baffle, and can be directly sent to the actuator of the corresponding baffle to drive it to complete precise physical adjustment.

[0068] S40: The target opening adjustment command is sent to the actuator of the corresponding baffle to control each baffle to adjust to the target opening. In the next control cycle, the real-time acquisition, mode determination, command calculation and issuance steps are repeated based on the updated temperature data to form a closed-loop feedback control.

[0069] The control cycle is the same as the sampling cycle for real-time acquisition of indoor temperatures on each floor of the target building, cavity temperatures in each independent section of the ventilation cavity, and outdoor ambient temperature.

[0070] Specifically, the baffle target opening adjustment instructions for all floors calculated above are sent to the baffle actuators of the corresponding ventilation cavity sections through the control network. The baffle actuators drive the physical baffles to move according to the received instructions, and precisely adjust their opening to the target opening required by the instructions, so as to change the ventilation resistance and hot air flow of each cavity section, thereby realizing the fine adjustment of the amount of hot air flowing into each floor.

[0071] This control process constitutes a continuous closed-loop control flow. After completing the baffle adjustment for the current control cycle, the flow enters the next control cycle. At the start of the new cycle, the steps of real-time acquisition and updating of the indoor temperature of each floor, the cavity temperature of each floor section, and the outdoor ambient temperature are repeated. Based on this latest temperature data, the steps of calculating the vertical temperature non-uniformity, determining the target control mode, and then generating a new round of baffle target opening adjustment commands through the baffle opening fuzzy decision model and issuing them for execution are repeated.

[0072] The length of the control cycle is consistent with the temperature data sampling cycle. This setting means that the entire process, from data acquisition, status assessment, intelligent decision-making to physical execution, is completed synchronously at fixed time intervals. Through a recurring cycle of monitoring, decision-making, execution, and re-monitoring, changes in the building's thermal environment can be continuously sensed, and control strategies and damper openings can be dynamically adjusted to respond promptly and compensate for temperature distribution fluctuations caused by external weather changes and internal thermal disturbances. Ultimately, this closed-loop feedback control mechanism can ensure that the temperature on each floor of the building remains stable within a uniform and comfortable target range over the long term, thereby achieving adaptive and intelligent management of the building's thermal environment.

[0073] In summary, the embodiments of this application have at least the following technical effects: Compared to existing technologies, this application first constructs a comprehensive evaluation index for vertical temperature non-uniformity, which can more accurately reflect the overall temperature distribution in the vertical direction of a building, providing a reliable basis for intelligent control. Secondly, it designs a two-layer decision logic based on this index and the indoor-outdoor temperature difference, realizing automatic switching between strong intervention and weak fine-tuning control modes, enabling the control strategy to adapt to different thermal environment conditions. Thirdly, it creatively employs a combination of fuzzy control and particle swarm optimization to establish an intelligent decision model that can independently calculate the optimal baffle opening layer by layer. This model, through optimized weight parameters, can effectively balance the control objectives between local comfort and overall uniformity.

[0074] Finally, this application forms a complete closed-loop feedback control process. Through continuous data collection, decision-making and execution, it can achieve dynamic, precise and adaptive management of the building thermal environment, thereby improving the building temperature uniformity and the overall energy efficiency of the photovoltaic hot air system.

[0075] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0076] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0077] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A photovoltaic-thermal air stratified synergistic control method for improving building temperature uniformity, characterized in that, The method includes: The system acquires the indoor temperature of each floor of the target building, the cavity temperature of each independent section of the ventilation cavity, and the outdoor ambient temperature in real time. Based on the indoor temperature of each floor, the system calculates the vertical temperature non-uniformity of the target building. Based on the vertical temperature non-uniformity and the indoor-outdoor temperature difference, a target control mode is determined from a variety of preset control modes. The selection of the target control mode is based on the comparison between the vertical temperature non-uniformity and a preset threshold. Based on the target control mode, the deviation between the indoor temperature of each floor and the set target temperature, and the cavity temperature of each floor segment, combined with the layered fine-tuning strategy, the target opening adjustment command of the baffle corresponding to each independent floor segment cavity is independently calculated and output through the pre-constructed baffle opening fuzzy decision model. The target opening adjustment command is sent to the actuator of the corresponding baffle to control each baffle to adjust to the target opening. In the next control cycle, the real-time acquisition, mode determination, command calculation and issuance steps are repeated based on the updated temperature data to form a closed-loop feedback control.

2. The photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity according to claim 1, characterized in that, The calculation steps for the vertical temperature non-uniformity include: Calculate the standard deviation of indoor temperature on each floor as a basic non-uniformity index to characterize the overall dispersion. Calculate the absolute value of the indoor-outdoor temperature difference based on the outdoor ambient temperature and the average indoor temperature of the building; The cavity temperature of each independent section within the ventilation cavity is obtained, and the absolute value of the difference between the cavity temperature and the corresponding floor temperature is calculated to obtain the cavity-indoor temperature difference set for each floor. Based on the aforementioned basic non-uniformity index, the absolute value of the indoor-outdoor temperature difference and the average value of the cavity-indoor temperature difference set of each floor are introduced as correction factors and weighted fusion is performed to obtain a comprehensive vertical temperature non-uniformity.

3. The photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity according to claim 1, characterized in that, The multiple control modes include at least a strong temperature uniformity mode and a weak adjustment mode. The strong temperature uniformity mode is activated when the vertical temperature non-uniformity is higher than a first preset threshold for active intervention. The weak adjustment mode is activated when the vertical temperature non-uniformity is lower than a second preset threshold for maintenance fine-tuning.

4. The photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity according to claim 1, characterized in that, The steps for constructing the fuzzy decision model for the baffle opening include: Collect historical temperature datasets of the target building under typical seasons and operating conditions, and label the ideal baffle opening state of each floor at each time to obtain a training sample set. The historical temperature data includes at least the indoor temperature of each floor, the cavity temperature of the corresponding floor section, and the outdoor ambient temperature. The ideal baffle opening state is determined by the historical optimal control record. Based on fuzzy control theory, the structure of the fuzzy decision model for the baffle opening degree is constructed. Using the training sample set, the membership function parameters and fuzzy rule weights in the baffle opening fuzzy decision model are learned and optimized until the model's control accuracy on the validation set converges, thus completing the model construction.

5. The photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity according to claim 1, characterized in that, Based on the target control mode, the deviation between the indoor temperature of each floor and the set target temperature, and the cavity temperature of each floor segment, combined with a layered fine-tuning strategy, and through a pre-constructed fuzzy decision model for baffle opening, the target opening adjustment command for the baffle corresponding to each independent floor segment cavity is independently calculated and output layer by layer, including: For each floor, the primary optimization objective for the current fuzzy decision is determined based on the target control mode. The deviation between the current floor's indoor temperature and the set target temperature, the cavity temperature of the corresponding floor section, and the vertical temperature non-uniformity are used as inputs to the pre-constructed fuzzy decision model for baffle opening. In the decision-making process of the baffle opening fuzzy decision model, an individual cost coefficient is set for the baffle opening adjustment of each floor. The individual cost coefficient is used to adjust the weight ratio between the local target of temperature control of this floor and the global target of temperature uniformity of the whole building. The particle swarm optimization algorithm is used to iteratively optimize the individual cost coefficients by using the weighted sum of the vertical temperature non-uniformity of the building as a whole and the temperature deviation of each floor under long-term operation as the fitness function, so as to obtain a set of individual cost coefficients that optimize the overall control performance. The optimized individual cost coefficient is used to configure the fuzzy decision model for the baffle opening. Fuzzy reasoning is performed independently for each floor to calculate and output the corresponding baffle target opening adjustment command.

6. The photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity according to claim 5, characterized in that, For each floor, based on the target control mode, determine the primary optimization objective for the current fuzzy decision, including: If the target control mode is a strong uniform temperature mode, then minimizing the vertical temperature non-uniformity is the primary optimization objective. If the target control mode is a weak adjustment mode, then minimizing the deviation between the current floor's indoor temperature and the set target temperature is the primary optimization objective.

7. The photovoltaic hot air stratified coordinated control method for improving building temperature uniformity according to claim 5, characterized in that, The particle swarm optimization algorithm is used, with the weighted sum of the overall vertical temperature non-uniformity of the building and the temperature deviation of each floor under long-term operation as the fitness function, to iteratively optimize the individual cost coefficients, thereby obtaining a set of individual cost coefficients that optimize the overall control performance, including: The individual cost coefficient of each floor is regarded as an optimization variable, and the individual cost coefficients of all floors are used to form a position vector of a particle in multidimensional space. Initialize a particle swarm, where each particle represents a set of candidate values ​​for the individual cost coefficient; In each iteration of the particle swarm optimization algorithm, for each individual cost coefficient value scheme represented by each particle, the corresponding baffle opening fuzzy decision model is configured, and the long-term comprehensive control performance corresponding to each individual cost coefficient value scheme is evaluated based on the historical operation data of the target building. The evaluation is achieved by calculating the fitness function, which is the weighted sum of the vertical temperature non-uniformity of the building as a whole and the temperature deviation of each floor under long-term operation. Based on the evaluation results of the fitness function, update the individual historical best position of each particle and the global historical best position of the entire particle swarm. Based on the individual historical best position, the global historical best position, and the particle's motion pattern, the position vector of each particle is iteratively updated; Repeat the evaluation and update steps until the preset iteration termination condition is met. At this point, the set of individual cost coefficient values ​​corresponding to the global historical best position of the particle swarm is the optimal solution obtained by optimization.

8. The photovoltaic hot air stratified coordinated control method for improving building temperature uniformity according to claim 7, characterized in that, In each iteration of the particle swarm optimization algorithm, for each individual cost coefficient value scheme represented by each particle, a corresponding fuzzy decision model for the baffle opening is configured, and based on the historical operating data of the target building, the long-term comprehensive control performance corresponding to each individual cost coefficient value scheme is evaluated, including: Extract multiple consecutive runtime data segments from the historical operation dataset of the target building to serve as a test dataset for evaluation; For each particle, a set of individual cost coefficient value schemes is configured one by one into the baffle opening fuzzy decision model; On each test dataset, the baffle opening fuzzy decision model under the corresponding configuration is used to simulate and control the control time step by step, and record the vertical temperature non-uniformity sequence and the temperature deviation sequence of each floor generated during the control process. Based on the vertical temperature non-uniformity sequence and the temperature deviation sequence of each floor, the fitness function value corresponding to the current segment test dataset is calculated. The average value of the fitness function calculated from each data segment is taken to obtain the overall evaluation result of the current group's individual cost coefficient value scheme, which is used as the fitness value of the current particle.

9. The photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity according to claim 5, characterized in that, The optimized individual cost coefficient is used to configure the fuzzy decision model for the baffle opening. Fuzzy inference is performed independently for each floor, and the corresponding baffle target opening adjustment command is calculated and output, including: The optimized set of individual cost coefficients are used as corresponding weight parameters and respectively configured into the decision logic units corresponding to each floor in the fuzzy decision model of the baffle opening. For each floor currently awaiting decision, the corresponding subset of fuzzy rules in the fuzzy decision model for the baffle opening is activated according to the target control mode. Using the deviation between the current floor's indoor temperature and the set target temperature, the cavity temperature of the corresponding floor segment, and the vertical temperature non-uniformity as real-time inputs, and combining the individual cost coefficient configured for the current floor, fuzzy inference is performed within the activated fuzzy rule subset, wherein the output result of the fuzzy inference is the target opening adjustment amount of the baffle corresponding to the current floor. Based on the target opening adjustment amount, a corresponding control command is generated and output as the target opening adjustment command for the baffle of the current floor.

10. The photovoltaic-thermal air stratified coordinated control method for improving building temperature uniformity according to claim 1, characterized in that, The control cycle is the same as the sampling cycle for real-time acquisition of indoor temperature on each floor of the target building, cavity temperature in each independent section of the ventilation cavity, and outdoor ambient temperature.

Images