Indoor ventilation airflow intelligent design method oriented to multiple requirements

Abstract

The invention discloses a multi-demand-oriented indoor ventilation airflow intelligent design method, which comprises the following steps of S1, establishing a ventilation parameter database according to changes of temperature, relative humidity, gas pollutant concentration and wind speed under different design parameters; s2, establishing a ventilation evaluation mechanism according to the quantitative relation between the design parameters and the performance indexes; s3, a BP neural network is utilized to construct a mapping relation between design parameters and performance indexes, and a proxy model is constructed; s4, performing multi-objective optimization on the agent model, and obtaining an optimization result; and S5, evaluating an optimization result according to a design requirement, and determining an optimal design scheme. The demand-oriented optimal design scheme can be determined according to the specific ventilation demand.

Description

Technical Field

[0001] The present invention relates to the technical field of ventilation system control and design, and in particular to an intelligent design method for indoor ventilation airflow oriented to multiple requirements. Background Art

[0002] Residents have diverse expectations for their indoor environments, placing stricter demands on mechanical ventilation design strategies, encompassing both the energy efficiency of the ventilation system itself and the comfort and cleanliness it provides. Addressing these diverse ventilation needs, it is necessary to integrate all relevant indicators into a numerical framework, clarify the relationships between ventilation design parameters and performance, and identify potential interdependencies to achieve optimal ventilation system design that meets these diverse requirements.

[0003] Existing studies generally have one or more of the following limitations: (1) There is a lack of multi-index evaluation of a single ventilation mode considering multiple control parameter combinations; (2) There are limited studies exploring the mapping relationship between various performance indicators and design factors of ventilation systems; (3) There is a lack of research on the multi-objective optimization design and decision-making methods for indoor mechanical ventilation, especially the measurement under the multi-index framework. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent design method for indoor ventilation airflow oriented to multiple needs, so as to solve the technical problems pointed out in the above background technology.

[0005] The present invention is implemented by adopting the following technical solution: a multi-demand oriented indoor ventilation airflow intelligent design method, comprising the following steps: S1: Establish a ventilation parameter database based on the changes in temperature, relative humidity, gas pollutant concentration and wind speed under different design parameters; S2: Establish a ventilation evaluation mechanism based on the quantitative relationship between design parameters and performance indicators; S3: Using BP neural network, the mapping relationship between design parameters and performance indicators is constructed to build the agent model; S4: Perform multi-objective optimization on the surrogate model and find the optimization result; S5: Evaluate the optimization results according to the design requirements and determine the optimal design solution.

[0006] Furthermore, step S1 specifically includes the following steps: S11: Obtain the changes in temperature, relative humidity, gas pollutant concentration, and wind speed under different design parameters, and establish a sample database based on the specified ventilation mode; S12: Determine the design parameters that need to be optimized, such as ventilation volume, air inlet opening area, fresh air temperature, etc. Perform Latin hypercube sampling on the design parameters that need to be optimized within a specified range, and perform CFD calculations on the sampled samples to obtain the corresponding temperature, relative humidity, gas pollutant concentration and wind speed under the corresponding parameters to form a ventilation parameter database.

[0007] Furthermore, step S2 specifically includes the following steps: S21: Establish a numerical evaluation framework for ventilation performance based on the analytic hierarchy process, using environmental indicators in the ventilation parameter database as the input layer, and inputting environmental indicators, wherein the environmental indicators include temperature, relative humidity, gas pollutant concentration, and wind speed; S22: Convert environmental indicators into satisfaction levels to form a secondary indicator evaluation layer; S23: Calculate the weights of the secondary indicators according to the entropy weight method to form the environmental comfort level; calculate the energy efficiency and space cleanliness at the same time to form the main indicators; subjectively weight the environmental comfort level, energy efficiency and space cleanliness to form the target layer.

[0008] Furthermore, step S3 is specifically as follows: establishing an agent model using the prediction function of the back propagation artificial neural network (BPANN), with the input parameters being the design parameters and the output results being the environmental comfort, energy efficiency and space cleanliness, thereby forming an agent model of environmental comfort, energy efficiency and space cleanliness.

[0009] Furthermore, step S4 is specifically as follows: based on the agent model, a non-dominated sorting genetic algorithm (NSGA-II) is used to optimize the three objectives of energy efficiency, comfort and cleaning ability to obtain a Pareto solution set.

[0010] Furthermore, step S5 is specifically as follows: according to the design requirements, TOPSIS is used to identify the ideal solution of the optimization results, which is the final optimization design decision.

[0011] The beneficial effects of the present invention are: 1. The flexibility, efficiency and versatility of computational fluid dynamics can be utilized to establish a representative ventilation system CFD database.

[0012] 2. The established multi-index numerical evaluation framework for ventilation systems based on the analytic hierarchy process can directly link multiple design parameters with multiple performance indicators.

[0013] 3. The Latin hypercube sampling technique is used to ensure that only a small number of samples are needed to represent the entire sample space, reducing research costs.

[0014] 4. Using genetic algorithm to optimize BP neural network can avoid the local optimal situation of neural network model and solve the problem that it is difficult to obtain the design objective function due to strong nonlinear relationship such as ventilation control.

[0015] 5. The NSGA-II multi-objective optimization algorithm is combined with the BP neural network agent model to perform multi-objective optimization for multiple ventilation indicators.

[0016] 6. The TOPSIS strategy can be used to concentrate the optimal parteo solutions obtained in NSGA-II and determine the optimal design solution based on specific needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0018] Figure 1 This is the framework diagram for the numerical evaluation of ventilation performance; Figure 2 This is the schematic diagram of the proxy model; Figure 3 It is a comparison between the predicted values ​​and actual values ​​of the proxy model on three indicators: environmental comfort, energy efficiency and cleaning ability; Figure 4 The sample distribution and Pareto solution of the three indicators of environmental comfort, energy efficiency and cleaning ability are as follows: Figure 5 It is the optimal design solution in the Pareto solution set. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0020] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0021] The following embodiments of the present invention are described in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0022] A multi-demand oriented indoor ventilation airflow intelligent design method includes the following steps: The Latin hypercube sampling method was used to sample the design parameters to be optimized and determine the sample library. CFD modeling and calculation were then performed on each sample to obtain the corresponding data on temperature, relative humidity, gas pollutant concentration (CO2), and wind speed, and to construct a ventilation parameter database.

[0023] Establish as Figure 1 The numerical evaluation framework of ventilation performance is shown.

[0024] The environmental indicators obtained by CFD calculation are used as the input layer. The four environmental indicators shown in the input layer are converted into corresponding satisfaction scores through equations 1 and 2.

[0025] (1) (2) Where, For indicators The percentage of samples that fall outside the design value range; Indicates the number of samples where the indicator exceeds the design value range; is the total number of samples; Provide satisfaction ratings for each indicator.

[0026] Calculate the environmental comfort according to formula 3: (3) Where, For environmental comfort; is the weight of each indicator.

[0027] Calculate energy efficiency and space cleanliness according to formulas 4 to 7 respectively: (4) (5) (6) (7) Where, is the energy utilization coefficient; is the exhaust air temperature; is the supply air temperature; is the occupied zone temperature; for energy efficiency; is the gaseous pollutant removal efficiency; is the pollutant concentration in the exhaust air; is the concentration of pollutants in the supply air; is the pollutant concentration in the occupied area; For the cleanliness of space.

[0028] The final score is calculated according to formula 8: (8) Where, Design satisfaction for ventilation; 、 and are the weights of energy efficiency, environmental comfort and space cleanliness respectively.

[0029] The prediction function of the back-propagation artificial neural network (BPANN) is used to build the agent model, see Figure 2 The input parameters include ventilation volume, air inlet opening area, and fresh air temperature, and the output results are energy efficiency, comfort, and cleanliness. The sample set consists of samples obtained using Latin hypercube sampling and is divided into a training set and a test set in a ratio of 7:3. The training set is used to train the BPANN agent model, while the test set is used to evaluate the model's accuracy and generalization ability. The hidden layer H can be determined by trial and error: (9) Where H is the number of hidden layer nodes; is the number of nodes in the input layer; is the number of nodes in the output layer; An integer between 1 and 10.

[0030] A genetic algorithm (GA) is used to determine weights and biases to avoid the problem of network training falling into local optimality. First, a set of weights and biases are randomly selected, and the individual with the highest fitness is selected for inheritance by calculating the fitness value. This process is repeated until the error requirement is met. The definition of the fitness value F is shown in formula (10). Figure 3 It can be seen that the predicted values ​​are consistent with the actual values, which shows that the proxy model is accurate.

[0031] (10) Where F is the fitness value; N is the number of samples; is the predicted value; is the actual value.

[0032] The non-dominated sorting genetic algorithm (NSGA-II) is used to optimize the three objectives of energy efficiency, comfort, and cleaning ability. The objective function in this embodiment is expressed by the GA-BPANN agent model as follows: (11) Where y(x) is the vector function composed of the objective function; x is the vector composed of the design parameters.

[0033] The optimization results are as follows Figure 4 shown. Figure 4 (a) is the distribution of all sample points in terms of energy efficiency, comfort and cleaning ability, and the Pareto solution set obtained by multi-objective optimization based on NSGA-II is as follows: Figure 4 (b) shown.

[0034] Furthermore, the TOPSIS method is used to identify the ideal solution from the Pareto solutions obtained by NSGA-II multi-objective optimization. The TOPSIS method involves constructing a performance matrix ,in , represents the number of individuals in the Pareto solution set, , represents the number of objective functions in the set. In addition, by The normalized value and weight matrix Combined, a new performance matrix can be created The weights are calculated using the entropy weight method.

[0035] The principle of TOPSIS method is shown in equations (12) to (14): (12) Where PIS and NIS are the best ideal solution and the worst ideal solution, respectively.

[0036] (13) Where, and are the Euclidean distances of the Pareto solution from the optimal ideal solution and the worst ideal solution, respectively.

[0037] (14) Where, is the relative closeness of the Pareto solution, which is a number between 0 and 1. , represents the optimal solution.

[0038] The ideal solution obtained by TOPSIS is as follows Figure 5 As shown in the figure, if the emphasis is on environmental comfort, the optimal design of air exchange rate, air inlet area and fresh air temperature is 13.3ACH, 0.072m 2 If the emphasis is on energy efficiency, the optimal design of air exchange rate, air inlet area and fresh air temperature is 11.5ACH, 0.098m 2 If the emphasis is on cleaning performance, the optimal design of air exchange rate, air inlet area and fresh air temperature is 10.48ACH, 0.099m 2 and 20.62℃.

[0039] For the sake of simplicity, the aforementioned embodiments are described as a series of actions. However, those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are preferred embodiments, and the actions involved are not necessarily required by this application.

[0040] The above embodiments describe the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Without departing from the spirit and scope of the present invention, modifications and variations made by those skilled in the art without departing from the spirit and scope of the present invention should be within the scope of protection of the appended claims.

Claims

1. A multi-demand oriented indoor ventilation airflow intelligent design method, characterized in that: The steps include: S1: Establish a ventilation parameter database based on the changes in temperature, relative humidity, gas pollutant concentration and wind speed under different design parameters; S2: Establish a ventilation evaluation mechanism based on the quantitative relationship between design parameters and performance indicators; S3: Using BP neural network, the mapping relationship between design parameters and performance indicators is constructed to build the agent model; S4: Perform multi-objective optimization on the agent model and obtain the optimization results; S5: Evaluate the optimization results according to the design requirements and determine the optimal design solution.

2. The multi-demand oriented indoor ventilation airflow intelligent design method according to claim 1, characterized in that: Step S1 specifically includes the following steps: S11: Obtain the changes in temperature, relative humidity, gas pollutant concentration, and wind speed under different design parameters, and establish a sample database based on the specified ventilation mode; S12: Determine the design parameters that need to be optimized, perform Latin hypercube sampling on the design parameters that need to be optimized within a specified range, and perform CFD calculations on the sampled samples to obtain the corresponding temperature, relative humidity, gas pollutant concentration and wind speed under the corresponding parameters to form a ventilation parameter database.

3. The method for intelligent design of indoor ventilation airflow for multiple requirements according to claim 1, characterized in that: Step S2 specifically includes the following steps: S21: Input environmental indicators, including temperature, relative humidity, gas pollutant concentration, and wind speed; S22: Convert environmental indicators into satisfaction levels to form a secondary indicator evaluation layer; S23: Calculate the weights of the secondary indicators according to the entropy weight method to form the environmental comfort level; calculate the energy efficiency and space cleanliness at the same time to form the main indicators; subjectively weight the environmental comfort level, energy efficiency and space cleanliness to form the target layer.

4. The method for intelligent design of indoor ventilation airflow for multiple requirements according to claim 1, characterized in that: Step S3 is specifically as follows: establishing an agent model using the prediction function of the back-propagation artificial neural network, with the input parameters being the design parameters and the output results being the environmental comfort, energy efficiency and space cleanliness, thereby forming an agent model of environmental comfort, energy efficiency and space cleanliness.

5. The method for intelligent design of indoor ventilation airflow for multiple requirements according to claim 4, characterized in that: Step S4 is specifically as follows: based on the agent model, a non-dominated sorting genetic algorithm is used to optimize the three objectives of energy efficiency, comfort and cleaning ability to obtain a Pareto solution set.

6. The method for intelligent design of indoor ventilation airflow for multiple requirements according to claim 1, characterized in that: Step S5 specifically includes: according to the design requirements, using TOPSIS to identify the ideal solution of the optimization results, which is the final optimization design decision.

Images