Energy-saving control method and system for fresh air system

Abstract

The invention provides an energy-saving control method and system for a fresh air system, and relates to the technical field of fresh air system regulation, and the method comprises the steps: collecting real-time environment data, personnel constitution data, historical fresh air regulation amount, historical environment data and historical personnel constitution data of a target area; a real-time PMV value and a historical PMV value are calculated through a heat balance equation, a BI-LSTM model is constructed, the historical PMV value and the historical fresh air adjusting amount are used for training the model, and the real-time PMV value is input into the trained BI-LSTM model to obtain the predicted fresh air adjusting amount; and the air supply amount of the fresh air system is regulated and controlled according to the predicted fresh air adjusting amount. Energy can be saved to a certain extent, and the comfort degree of indoor personnel is improved.

Description

Technical Field

[0001] The present application relates to the technical field of regulating and controlling fresh air systems, and in particular to an energy-saving control method and system for fresh air systems. Background Art

[0002] In modern buildings, fresh air systems are widely used in various building spaces as crucial equipment for ensuring indoor air quality and comfort. However, current fresh air systems often suffer from regulation lag during operation.

[0003] The Chinese invention patent with application publication number CN117109133A provides a method for controlling an air-conditioning system based on changes in the number of people. The patent designs a simulation model based on parameters, calculates the PMV index based on the simulation results, and then determines whether the absolute value of the PMV index is less than 1. Only when the absolute value of the PMV index is less than 1 will the valve opening be adjusted.

[0004] The above patent only adjusts the valve opening when the absolute value of the PMV index is less than 1. Once the absolute value of the PMV index exceeds 1, it means that some people may have already clearly felt hot or cold, especially for the elderly, children, and people with more sensitive constitutions, this discomfort will be more obvious. If the valve is adjusted only at this time, the person will still have to wait for a period of time in an uncomfortable environment during the adjustment process, which undoubtedly reduces the overall quality of the indoor environment. In addition, while waiting for the PMV index to reach the adjustment threshold, the air conditioning system will continue to operate at inappropriate power. In order to restore the indoor environment to a comfortable state, the air conditioning system needs to consume more energy to adjust, resulting in energy waste. Summary of the Invention

[0005] In order to save energy and improve the comfort of indoor occupants to a certain extent, the present application provides an energy-saving control method and system for a fresh air system.

[0006] In the first aspect, the present application provides an energy-saving control method for a fresh air system, which adopts the following technical solutions: A method for energy-saving control of a fresh air system comprises the following steps: Data collection: Collect real-time environmental data and real-time physical data of personnel in the target area; collect historical fresh air adjustment volume, historical environmental data and historical physical data of personnel in the target area; Prediction: Real-time environmental data and real-time personnel physical data are input into the heat balance equation to calculate the real-time PMV value. Historical environmental data and historical personnel physical data are input into the heat balance equation to calculate the historical PMV value. A BI-LSTM model is constructed and trained using historical PMV values and historical fresh air adjustment values to obtain a trained BI-LSTM model. Real-time PMV values are input into the trained BI-LSTM model to obtain the predicted fresh air adjustment value. Regulation: Regulate the air supply volume of the fresh air system according to the predicted fresh air adjustment amount.

[0007] This application first collects real-time environmental data and real-time personnel physical data, and calculates the real-time PMV value based on this to accurately reflect the current thermal comfort status of personnel in the indoor environment. The PMV value calculated by this application takes into account the physical differences of personnel, such as people of different ages, genders, and health conditions have different feelings about temperature and humidity. The above method can make the regulation more in line with individual needs. Subsequently, this application obtains the fresh air adjustment amount through prediction and adjusts the fresh air system in real time, so that the fresh air system can quickly respond to changes in the indoor environment, try to avoid people from feeling cold or hot due to environmental discomfort, and always maintain a comfortable indoor microclimate. Subsequently, this application uses historical data to train the BI-LSTM model, explore the potential patterns between the data, predict the fresh air adjustment amount in advance, and realize active regulation of the fresh air system before the absolute value of the real-time PMV value exceeds 1. Compared with the traditional fixed air supply volume mode, this application can reduce the situation of lagging adjustment.

[0008] This application can reduce the air supply volume and reduce fan energy consumption when there are few people. On the contrary, when the number of people increases, this application can increase the air supply volume to keep people in a comfort zone while reducing energy waste, effectively improving energy utilization efficiency and reducing the operating cost of the fresh air system. This application comprehensively considers the impact of multiple factors on the indoor environment, covering environmental factors such as temperature and humidity, and incorporating differences in people's physical fitness, thereby constructing a more comprehensive basis for regulation. The BI-LSTM model is trained to learn the complex relationships in historical data in order to accurately predict the fresh air adjustment amount under different circumstances. The predicted fresh air adjustment amount can provide precise control instructions for the fresh air system. Compared with single parameter control, this application can significantly improve the control accuracy.

[0009] Optionally, the data collection step further includes collecting the volume, real-time CO2 concentration, and real-time VOCs concentration of the target area; After performing the step of collecting data, the method further includes: Concentration judgment: judge whether the real-time CO2 concentration and real-time VOCs concentration are both less than the corresponding concentration thresholds. If so, no processing is performed; if not, the step of determining the ventilation volume is executed; Determine ventilation volume: Determine the target ventilation volume based on the volume of the target area, concentration threshold, real-time CO2 concentration, and real-time VOCs concentration, and execute control steps; The control step also includes calculating the sum of the target ventilation volume and the predicted fresh air adjustment volume, recording it as first data, and using the first data instead of the predicted fresh air adjustment volume to control the air supply volume of the fresh air system.

[0010] This application monitors the real-time CO2 concentration and real-time VOCs concentration and compares them with the corresponding concentration thresholds. When the CO2 concentration or VOCs concentration exceeds the corresponding concentration threshold, this application can determine the target ventilation volume based on the volume of the target area, the concentration threshold, the real-time CO2 concentration and the real-time VOCs concentration, and adjust the air supply volume of the fresh air system according to the target ventilation volume, introduce fresh air in time, and expel the indoor polluted air, thereby reducing the pollutant concentration and improving the indoor air quality.

[0011] This application combines the target ventilation volume determined based on the real-time CO2 concentration and the real-time VOCs concentration with the fresh air adjustment volume predicted based on the real-time PMV value, comprehensively considering the two important aspects of indoor thermal comfort and air quality. The above-mentioned multi-factor control method enables the fresh air system to achieve better operating results under different environmental conditions, further enhancing flexibility and adaptability. When the real-time CO2 concentration and the real-time VOCs concentration are both lower than the corresponding concentration thresholds, this application will not take any action, reducing unnecessary fresh air supply when the air quality is good, thereby reducing energy waste. Only when the real-time CO2 concentration or the real-time VOCs concentration exceeds the standard, the fresh air volume will be increased according to the actual situation, making the use of energy more accurate and efficient.

[0012] Optionally, before performing the step of determining the concentration, the method further includes: Calculate the CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount; use the CO2 concentration change to adjust the real-time CO2 concentration to obtain the adjusted CO2 concentration; use the VOCs concentration change to adjust the real-time VOCs concentration to obtain the adjusted VOCs concentration; and perform the concentration judgment step; In the concentration judgment step, the adjusted CO2 concentration is used instead of the real-time CO2 concentration, and the adjusted VOCs concentration is used instead of the real-time VOCs concentration.

[0013] The present application calculates the CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount, and then adjusts the real-time CO2 concentration based on the CO2 concentration change, and adjusts the real-time VOCs concentration based on the VOCs concentration change. By adopting the above scheme, the present application no longer determines whether the fresh air volume needs to be adjusted based solely on the real-time CO2 concentration and real-time VOCs concentration, but can predict in advance the changes in the real-time CO2 concentration and real-time VOCs concentration caused by the fresh air adjustment amount. Traditional methods based on real-time concentration judgment are often based on real-time CO2 concentration and real-time VOCs concentration. However, the predicted fresh air adjustment amount will have the effect of adjusting the real-time CO2 concentration and real-time VOCs concentration after being input into the room. When determining the target ventilation volume, since the adjusted CO2 concentration and VOCs concentration are used for judgment, the present application can more accurately determine the required target ventilation volume based on the effects of real-time CO2 concentration, real-time VOCs concentration and fresh air adjustment amount, which helps to reduce the situation where the ventilation volume is too large or too small, so that the fresh air system can provide just the right fresh air supply according to actual needs, further improving the accuracy of regulation.

[0014] This application no longer simply determines real-time CO2 and VOC concentrations, but instead takes into account the predicted fresh air conditioning effect, making concentration judgments more accurate. By adopting this solution, we can reduce misjudgments caused by not considering the fresh air conditioning effect. For example, when the fresh air volume is about to reduce pollutant concentrations to a reasonable range, we no longer need to increase the fresh air supply unnecessarily, thus reducing energy waste.

[0015] Optionally, the concentration determination step further includes: Calculate the CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount; use the CO2 concentration change and VOCs concentration change to adjust the concentration threshold to obtain a new concentration threshold.

[0016] This application calculates the CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount to adjust the concentration threshold, so that the air quality judgment standard can change in real time according to the adjustment of the fresh air system. After dynamic adjustment, this application can more accurately determine when to increase or maintain the fresh air volume based on the dilution ability of the fresh air adjustment amount on pollutants, so that the indoor pollutant concentration is always within the health standard, effectively improving the air quality. Before adjusting the fresh air system according to the predicted fresh air adjustment amount, this application can lower the concentration threshold according to the concentration change, so that the judgment result in the concentration judgment step is more in line with the actual situation.

[0017] Optionally, when the real-time CO2 concentration and the real-time VOCs concentration are both less than the corresponding new concentration thresholds, the concentration determination step further includes: Determine whether the real-time PMV value is within the range of ±5% within the preset time. If so, control to reduce the opening of the fresh air valve and exhaust valve of the fresh air system, and increase the opening of the return air valve of the fresh air system; if not, no action is taken.

[0018] When the real-time CO2 concentration and the real-time VOCs concentration are both less than the new concentration threshold, it indicates that the current indoor air quality is good. At this time, the present application further determines whether the real-time PMV value is within the range of ±5% within the preset time length. If the conditions are met, the fresh air valve and exhaust valve opening of the fresh air system are controlled to reduce, and the return air valve opening is increased at the same time. By adopting the above scheme, the present application can reduce the energy consumption of the fresh air system by reducing the amount of fresh air introduced from the outside and the amount of indoor air discharged. Because the introduction of outdoor air often requires additional energy to adjust its temperature and humidity to adapt to the indoor environment, reducing the amount introduced can effectively save this part of energy consumption. Increasing the opening of the return air valve means that more indoor air is recycled. Since the temperature and humidity of the indoor air are already close to a comfortable state, recycling it again can reduce the energy required to adjust the temperature and humidity of the new air. That is, the present application can achieve optimal energy allocation while ensuring the comfort of the indoor environment, thereby improving energy utilization efficiency.

[0019] Optionally, after the step of performing the prediction and before the step of performing the regulation, the method further includes: Setting range: Generate multiple control strategies based on the predicted fresh air adjustment amount, including the actual fresh air adjustment amount, the target temperature setting value, and the target PMV value; input the target PMV value into the trained BI-LSTM model to obtain the target fresh air adjustment amount, where the value of the actual fresh air adjustment amount is between the target fresh air adjustment amount and the predicted fresh air adjustment amount; set the value range of the target temperature setting value; Initialization: Set the objective function of the NSGA-II algorithm, which includes an energy-saving function and a comfort function. Perform binary encoding on the actual fresh air adjustment amount and the target temperature setting value in each control strategy, obtain the encoding result of each actual fresh air adjustment amount and each target temperature setting value, concatenate the encoding result of each actual fresh air adjustment amount and the encoding result of each target temperature setting value into a binary string, and use the binary string as a chromosome of the NSGA-II algorithm. Determine the target strategy: Use the NSGA-II algorithm to obtain the target chromosome from multiple chromosomes, and take the control strategy corresponding to the target chromosome as the Pareto optimal solution. Repeat this step multiple times until the number of executions reaches the maximum number of iterations. Then, integrate all Pareto optimal solutions into a Pareto optimal solution set, and select the control strategy with the lowest energy consumption in the Pareto optimal solution set as the target strategy. In the control step, the air supply volume of the fresh air system is regulated according to the target strategy.

[0020] This application uses the NSGA-II algorithm to find the Pareto optimal solution among multiple control strategies, and finally selects the control strategy with the lowest energy consumption as the target strategy. It comprehensively considers the impact of the actual fresh air adjustment amount and the target temperature setting value on energy consumption, and can effectively reduce the energy waste caused by over-adjustment. Under different indoor and outdoor environments and personnel activities, this application can accurately match the fresh air adjustment scheme that can both meet the indoor environment needs and minimize energy consumption. Compared with the traditional fixed mode or simple adjustment method, it can greatly reduce the operating energy consumption of the fresh air system. This application incorporates the target PMV value into the control strategy, and obtains the target fresh air adjustment amount through the trained BI-LSTM model. It can fully consider the thermal comfort of the personnel when adjusting the air supply volume, and then adjust the fresh air adjustment amount and target temperature setting value according to the personnel's demand for thermal comfort on the basis of ensuring good air quality, creating a comfortable and healthy indoor environment and improving the comfort experience of the personnel indoors.

[0021] The introduction of the NSGA-II algorithm enables this application to find a balance between multiple conflicting objectives (i.e., energy saving and comfort). By binary encoding and chromosome selection of the actual fresh air adjustment amount and the target temperature setting value, a refined search and optimization of the control parameters are achieved. Compared with the simple control of a single parameter, the multi-objective optimization method adopted by this application can more accurately adjust the fresh air system according to changes in the indoor and outdoor environment and personnel needs, thereby improving the accuracy and stability of the control. In different seasons, different outdoor climate conditions, and frequent changes in the number of indoor people and activities, this application can quickly adapt to the dynamic changes in the environment by continuously adjusting the control strategy. Multiple strategies are generated using the predicted fresh air adjustment amount, and then the optimal solution is screened through the algorithm, so that the fresh air system can always maintain a good operating state in a complex and changing environment, and maintain the stability and adaptability of the indoor environment.

[0022] Optionally, the initialization step further includes: Determine the control accuracy of the actual fresh air adjustment amount and the target temperature setting value, determine the number of binary coding bits based on the control accuracy, and binary code the actual fresh air adjustment amount and the target temperature setting value in each control strategy according to the number of binary coding bits; The calculation model of the binary code bit number is as follows: ; in, The number of bits of binary code for the actual fresh air adjustment value; The number of bits of binary encoding for the target temperature setting value; It is the maximum value of the actual fresh air adjustment amount; The maximum value of the target temperature setting; It is the minimum value of the actual fresh air adjustment amount; The minimum value of the target temperature setting; The control accuracy of the actual fresh air adjustment amount; It is the control accuracy of the target temperature setting value.

[0023] By determining the control accuracy and calculating the number of binary coding bits based on it, the present application can binary encode the actual fresh air control amount and the target temperature setting value with appropriate accuracy, so that the present application can more accurately adjust the fresh air system according to actual needs, try to avoid the control error caused by insufficient accuracy, and improve the accuracy of indoor environment control. The present application can adjust the control accuracy according to actual conditions, thereby enhancing the adaptability and flexibility of the fresh air system.

[0024] The appropriate number of binary coding bits helps the NSGA-II algorithm search and optimize more efficiently. Too few coding bits will result in a search space that is too small and the optimal solution cannot be found; while too many coding bits will increase the algorithm's computational complexity and search time. This application calculates the number of binary coding bits based on the control accuracy. Under the premise of ensuring search accuracy, it can reasonably control the size of the search space, improve the algorithm's search efficiency, find the Pareto optimal solution more quickly, and thus adjust the fresh air system's operating strategy in a timely manner.

[0025] Optionally, the step of determining the target strategy further includes: The target chromosome is segmented according to the binary coding bit number of the actual fresh air adjustment amount and the binary coding bit number of the target temperature setting value to obtain the coding block of the actual fresh air adjustment amount and the coding block of the target temperature setting value, and the coding block of the actual fresh air adjustment amount and the coding block of the target temperature setting value are decoded respectively to obtain the decoded actual fresh air adjustment amount and the decoded target temperature setting value; Determine whether the decoded actual fresh air adjustment amount and the decoded target temperature setting value are consistent with expectations. If so, no processing is performed; if not, an alarm signal is issued.

[0026] This application obtains the decoded real fresh air adjustment amount and the decoded target temperature setting value by segmenting and decoding the target chromosome, so that the optimization results obtained by the NSGA-II algorithm can be converted into parameters that can actually be used to regulate the fresh air system. By judging whether these parameters meet expectations, this application can promptly discover abnormal situations that may occur during the algorithm optimization process. If the decoded parameters do not meet expectations, it means that there is a problem with the control strategy derived by the algorithm. At this time, the target strategy cannot be applied to actual control, thereby reducing the situation where the fresh air system operates abnormally due to incorrect parameters and ensuring the reliability of the fresh air system operation.

[0027] In a second aspect, the present application provides an energy-saving control system for a fresh air system, which adopts the following technical solutions: An energy-saving control system for a fresh air system, comprising: a processor and a memory, The memory stores program code; When the processor calls the program code in the memory, the steps of the method described in the first aspect are executed.

[0028] In summary, this application includes at least one of the following beneficial technical effects: 1. This application comprehensively considers the impact of multiple factors on the indoor environment, including environmental factors such as temperature and humidity, as well as differences in individual physical conditions, to build a more comprehensive basis for control. The BI-LSTM model is trained to learn complex relationships within historical data to accurately predict the amount of fresh air adjustment required under different circumstances, providing precise control instructions for the fresh air system. This significantly improves control accuracy compared to single-parameter control.

[0029] 2. This application takes into account the influence of fresh air adjustment volume when judging the concentration of pollutants, which helps to reduce the situation where the ventilation volume is too large or too small, so that the fresh air system can provide just the right fresh air supply according to actual needs, reducing energy waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flow chart of Example 1 of the present application; Figure 2 This is a flow chart of Example 2 of the present application; Figure 3 This is a flowchart of Example 4 of the present application. DETAILED DESCRIPTION

[0031] The following combination Figures 1 to 3 This application is described in further detail.

[0032] Example 1: This example discloses an energy-saving control method for a fresh air system. Figure 1 The method includes: S11 data collection, S12 prediction and S13 regulation. First, real-time environmental data, personnel physical data, and historical fresh air adjustment amount, historical environmental data and historical personnel physical data of the target area are collected; then, the real-time PMV value and the historical PMV value are calculated respectively through the heat balance equation, a BI-LSTM model is constructed, and the model is trained with the historical PMV value and the historical fresh air adjustment amount. Then, the real-time PMV value is input into the trained BI-LSTM model to obtain the predicted fresh air adjustment amount; finally, the air supply volume of the fresh air system is regulated according to the predicted fresh air adjustment amount. This embodiment includes the following steps: S11 data collection, collects real-time environmental data and real-time physical data of personnel in the target area; collects historical fresh air adjustment volume, historical environmental data and historical physical data of personnel in the target area.

[0033] Real-time environmental data includes: temperature, humidity, wind speed and radiant temperature.

[0034] Personnel physical data includes: metabolic rate, clothing thermal resistance, and personnel distribution and number. The personnel physical data is collected in the following manner: Metabolic Rate: Estimates a person's metabolic rate by measuring their activity level.

[0035] Clothing thermal resistance: Estimate the thermal resistance of clothing based on the person's wearing conditions.

[0036] Personnel distribution and quantity: Statistics on the number and distribution of personnel in the target area.

[0037] Historical environmental data is the same type as real-time environmental data, but the data covers a longer time range and is collected over weeks or months.

[0038] Historical personnel physical data is of the same type as real-time personnel physical data, and its collection time is the same as that of historical environmental data. In principle, the collection timestamps of historical personnel physical data, historical environmental data, and historical fresh air adjustment amount should be similar or the same.

[0039] In other embodiments, if the collection timestamps of historical personnel physical data, historical environmental data, and historical fresh air adjustment amounts are different, the DTW algorithm is used for time alignment processing. In the S12 prediction, the historical personnel physical data, historical environmental data, and historical fresh air adjustment amounts processed by the DTW algorithm are used to train the BI-LSTM model.

[0040] S12 prediction: Input the real-time collected environmental data and personnel physical data into the heat balance equation to calculate the real-time PMV value. Input the historical environmental data and personnel physical data into the heat balance equation to calculate the historical PMV value.

[0041] The PMV value is based on Fanger's heat balance equation and comprehensively considers the impact of environmental parameters (temperature, humidity, radiant temperature) and human parameters (metabolic rate, clothing thermal resistance) on human thermal comfort. The calculation method of the PMV value is very mature and will not be further described in this embodiment. The core concept of the PMV value is to calculate the heat exchange balance between the human body and the environment, and obtain a value between -3 (cold) and +3 (hot), with 0 indicating a thermal neutral state.

[0042] Build a BI-LSTM model. The BI-LSTM model is a variant of a recurrent neural network (RNN) that can simultaneously learn forward and backward information of the data and has a strong ability to capture long-term dependencies in the data. The bidirectional structure of the BI-LSTM model allows the BI-LSTM model to consider both past and future information simultaneously (future information is not available during prediction but can be used during training), thereby improving prediction accuracy.

[0043] The training process of the BI-LSTM model is as follows: historical PMV values are used as training data, historical fresh air adjustment values are used as training labels, training sets and validation sets are divided, the weights and biases of the BI-LSTM model are initialized, the backpropagation algorithm and optimizer (such as Adam) are used to adjust the model parameters, the error between the predicted value and the true value (such as the mean square error (MSE)) is minimized, and the model performance is evaluated on the validation set to reduce overfitting.

[0044] The real-time calculated PMV value is input into the trained BI-LSTM model to output the predicted fresh air adjustment amount.

[0045] S13 controls the fresh air adjustment amount, sending the predicted fresh air adjustment amount directly to the controller of the fresh air system, and accordingly controls the speed of the blower or the opening of the fresh air valve in the fresh air system to change the air supply volume of the fresh air system.

[0046] This embodiment first obtains the real-time environmental parameters and real-time physical data of the target area, and simultaneously collects the historical fresh air adjustment amount, historical environmental data and historical physical data of the target area; then, based on the heat balance equation, the real-time PMV value and the historical PMV value are calculated to quantify thermal comfort, the historical PMV and the fresh air adjustment amount are used to train the BI-LSTM model, and the fresh air adjustment amount is predicted by the trained BI-LSTM model; finally, this application converts the predicted fresh air adjustment amount into a fresh air system control instruction, dynamically adjusts the air supply volume of the fresh air system, and forms a closed-loop feedback in combination with the real-time monitoring data. This embodiment can predict the fresh air adjustment amount based on the real-time PMV value and the BI-LSTM model after the fresh air system is turned on and the real-time PMV value is initially adjusted to a comfortable state, so as to adjust the indoor environment in advance before the absolute value of the real-time PMV value exceeds 1, thereby achieving energy-saving optimization while ensuring the thermal comfort of the personnel.

[0047] Example 2: Reference Figure 2 The difference between this embodiment and embodiment 1 is that the S11 data collection also includes: collecting the volume of the target area, real-time CO2 concentration and real-time VOCs (volatile organic compounds) concentration.

[0048] After executing S11 data collection, the method further includes: S21 concentration judgment, performs dual threshold verification on the real-time CO2 concentration and the real-time VOCs concentration, that is, verifies whether the real-time CO2 concentration and the real-time VOCs concentration meet the conditions that the real-time CO2 concentration is less than the CO2 concentration threshold and the real-time VOCs concentration is less than the VOCs concentration threshold. If so, it is determined that the indoor air quality meets the standard and the current state of the fresh air system is maintained; otherwise, S22 is immediately executed to determine the ventilation volume.

[0049] S22 determines the ventilation volume. This step determines the target ventilation volume based on the volume of the target area, the concentration threshold, the real-time CO2 concentration, and the real-time VOCs concentration. This allows the target ventilation volume to dilute the concentration of pollutants (i.e., the real-time CO2 concentration and the real-time VOCs concentration) while reducing energy waste caused by excessive ventilation. S13 control is then executed.

[0050] The target ventilation volume is calculated as follows: Target ventilation volume = |real-time CO2 concentration - CO2 concentration threshold| × volume of target area + |real-time VOCs concentration - VOCs concentration threshold| × volume of target area.

[0051] S13 controls the ventilation volume by summing the target ventilation volume with the predicted fresh air adjustment volume to obtain a first value. This first value is used instead of the predicted fresh air adjustment volume to generate control instructions for the fresh air system. This step, by superimposing the rigid ventilation volume driven by pollutants and the flexible adjustment volume driven by thermal comfort, balances air quality assurance with energy efficiency optimization, forming a multi-objective coordinated control solution.

[0052] In other embodiments, before performing S21 concentration determination, the method further includes: S20 concentration adjustment, calculate the CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount, the calculation method is as follows: CO2 concentration change = real-time CO2 amount ÷ target area volume; Real-time CO2 amount = original CO2 amount - exhaust CO2 amount + fresh air CO2 amount; Original CO2 amount = CO2 concentration before fresh air input × target area volume; Amount of CO2 discharged = CO2 concentration before fresh air is input × amount of indoor air discharged; Fresh air CO2 amount = predicted fresh air adjustment amount × 0.00042; The amount of indoor air exhausted = the exhaust speed of the fresh air system × the exhaust time.

[0053] The exhaust time is equal to the time the fresh air system takes to input the supply air into the room.

[0054] The calculation method of the VOCs concentration change is the same as the calculation method of the CO2 concentration change, and will not be repeated in this embodiment.

[0055] The real-time CO2 concentration is adjusted using the CO2 concentration change to obtain an adjusted CO2 concentration, which is equal to the difference between the real-time CO2 concentration and the CO2 concentration change.

[0056] The real-time VOCs concentration is adjusted using the VOCs concentration change to obtain an adjusted VOCs concentration, which is equal to the difference between the real-time VOCs concentration and the VOCs concentration change.

[0057] In the S21 concentration judgment, the adjusted CO2 concentration is used instead of the real-time CO2 concentration, and the adjusted VOCs concentration is used instead of the real-time VOCs concentration, that is, it is judged whether the adjusted CO2 concentration and the adjusted VOCs concentration are both less than the corresponding concentration thresholds.

[0058] After executing S12 prediction, this embodiment will obtain the fresh air adjustment amount input into the room (i.e., the predicted fresh air adjustment amount). The predicted fresh air adjustment amount will have a certain dilution effect on the real-time CO2 concentration and the real-time VOCs concentration. When making concentration judgments on the real-time CO2 concentration and the real-time VOCs concentration, this application takes into account the impact of the predicted fresh air adjustment amount on the real-time CO2 concentration and the real-time VOCs concentration, thereby reducing the target ventilation volume. Therefore, this embodiment can reduce the energy consumption of the fresh air system with a multi-objective collaborative strategy while ensuring indoor air quality.

[0059] Example 3: This example differs from Example 2 in that S21 concentration determination further includes: The CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount are calculated. The calculation method of the CO2 concentration change and VOCs concentration change in this embodiment is the same as that in Example 2 and will not be repeated here.

[0060] The concentration threshold is adjusted using the CO2 concentration change and the VOCs concentration change to obtain a new concentration threshold. Each time this step is performed, the original concentration threshold, i.e., the original CO2 concentration threshold and the original VOCs concentration threshold, are adjusted. The adjustment method is as follows: New CO2 concentration threshold = original CO2 concentration threshold - CO2 concentration change; New VOCs concentration threshold = original VOCs concentration threshold - VOCs concentration change; In the S21 concentration judgment, determine whether the real-time CO2 concentration and the real-time VOCs concentration are both less than the corresponding new concentration threshold. If so, further determine whether the real-time PMV value is within the range of ±5% within the preset time length. If so, control to reduce the opening of the fresh air valve and exhaust valve of the fresh air system, and increase the opening of the return air valve of the fresh air system; if not, no processing is performed.

[0061] This embodiment adjusts the original concentration threshold value through the predicted fresh air adjustment amount, so that the concentration judgment step can judge the real-time CO2 concentration and the real-time VOCs concentration according to the actual environment. When the real-time CO2 concentration and the real-time VOCs concentration are both lower than the corresponding new concentration threshold value, it is also judged whether the PMV value continues to be within the range of ±5%. If so, the opening of different valves is adjusted. This embodiment can lower the concentration threshold value according to the concentration change amount, so that the judgment result in the concentration judgment step is more in line with the actual situation.

[0062] Example 4: Reference Figure 3 The difference between this embodiment and embodiment 1 is that after executing S12 prediction and before executing S13 control, the following steps are further included: S41 sets a range and generates multiple control strategies based on the predicted fresh air adjustment amount, wherein the control strategies include: a real fresh air adjustment amount, a target temperature setting value, and a target PMV value.

[0063] The actual fresh air adjustment amount refers to the air volume control value actually acting on the fresh air system, that is, the actual fresh air adjustment amount input into the room. The value range of the actual fresh air adjustment amount is between the target fresh air adjustment amount and the predicted fresh air adjustment amount.

[0064] The target temperature setting value refers to the target temperature control value of the air-conditioning system linked to the fresh air system. The value range of the target temperature setting value is set to [Tmin, Tmax] (such as Tmin=24 degrees Celsius, Tmax=30 degrees Celsius). In other embodiments, the value range of the target temperature setting value can be set according to needs.

[0065] The target PMV value refers to the thermal comfort control target. The target PMV value range is set to [-0.5, 0.5] (i.e., the comfort range). By adjusting the target PMV value, the coordination between the fresh air system and the air-conditioning system can be indirectly affected.

[0066] The target PMV value is input into the trained BI-LSTM model to derive the target fresh air adjustment amount required to maintain the target PMV value.

[0067] S42 is initialized. This step uses the NSGA-II algorithm (non-dominated sorting genetic algorithm) to solve the multi-objective optimization problem and sets the objective function of the NSGA-II algorithm. The objective function includes an energy-saving function and a comfort function. The calculation models of the energy-saving function and the comfort function are as follows: Energy saving function The calculation model is as follows: ; in, is the weight of the fresh air system energy consumption; Energy consumption of the fresh air system; is the weight of energy consumption of air conditioning system; Energy consumption of the air conditioning system.

[0068] Comfort function The calculation model is as follows: ; in, is the weight of the PMV change; the value of the PMV change is equal to the target PMV value With real-time PMV value The absolute value of the difference; is the weight of the temperature change; the temperature change is equal to the target temperature setting value With real-time temperature value The absolute value of the difference.

[0069] In other embodiments, the energy-saving function and the comfort function may be set according to needs.

[0070] The control accuracy of the actual fresh air adjustment amount and the target temperature setting value is determined, the number of binary coding bits is determined based on the control accuracy, and the actual fresh air adjustment amount and the target temperature setting value in each control strategy are binary-coded according to the number of binary coding bits to obtain the coding result of each actual fresh air adjustment amount and the coding result of each target temperature setting value. The coding result of the actual fresh air adjustment amount and the coding result of the target temperature setting value are spliced into a binary string, and the binary string is used as a chromosome of the NSGA-II algorithm.

[0071] The calculation model of the binary code bit number is as follows: ; in, The number of bits of binary code for the actual fresh air adjustment value; The number of bits of binary encoding for the target temperature setting value; It is the maximum value of the actual fresh air adjustment amount; The maximum value of the target temperature setting; It is the minimum value of the actual fresh air adjustment amount; The minimum value of the target temperature setting; The control accuracy of the actual fresh air adjustment amount; It is the control accuracy of the target temperature setting value.

[0072] S43 determines the target strategy and uses the NSGA-II algorithm to obtain the target chromosome from multiple chromosomes. The control strategy corresponding to the target chromosome is used as the Pareto optimal solution. This step is repeated multiple times until the number of executions reaches the maximum number of iterations. All Pareto optimal solutions are integrated into a Pareto optimal solution set. The control strategy with the lowest energy consumption is selected as the target strategy in the Pareto optimal solution set. The process is as follows: For each chromosome, the value of its objective function is calculated and recorded as the target value, which is equal to the sum of the function value of the energy-saving function and the function value of the comfort function.

[0073] Chromosomes are sorted according to the target value and the non-dominated level of each chromosome is determined.

[0074] In each non-dominated level, the crowdedness of the chromosome is calculated.

[0075] Select one or more chromosomes from the non-dominated level 1 as Pareto optimal solutions.

[0076] Repeat the above steps (which may include crossover and mutation to generate new chromosomes) until the maximum number of iterations is reached. Each iteration may generate a new Pareto optimal solution. Therefore, this embodiment integrates the Pareto optimal solutions selected from all iterations into a set and removes duplicate solutions from the set to obtain a Pareto optimal solution set.

[0077] From the Pareto optimal solution set, the control strategy with the minimum energy consumption is selected as the target strategy.

[0078] In S13 control, the air supply volume of the fresh air system is regulated according to the target strategy.

[0079] In other embodiments, S43 determining the target strategy further includes: The target chromosome is segmented according to the binary coding bit number of the actual fresh air adjustment amount and the binary coding bit number of the target temperature setting value to obtain the coding block of the actual fresh air adjustment amount and the coding block of the target temperature setting value.

[0080] The decoding method corresponding to the binary code is adopted to decode the coding block of the actual fresh air adjustment amount and the coding block of the target temperature setting value respectively to obtain the decoded actual fresh air adjustment amount and the decoded target temperature setting value.

[0081] It is determined whether the decoded actual fresh air adjustment amount and the decoded target temperature setting value are in line with expectations. If so, no processing is performed; if not, an alarm signal is issued.

[0082] This embodiment uses the NSGA-II algorithm to balance the energy saving and comfort of the fresh air system, reducing the risk of thermal comfort deterioration caused by traditional single-objective control (such as minimizing energy consumption only).

[0083] Embodiment 5: This embodiment provides an energy-saving control system for a fresh air system, including: a processor and a memory, The memory stores program code; The processor executes the steps of the method when calling the program code in the memory.

[0084] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. An energy-saving control method for a fresh air system, characterized in that: include: Data collection: Collect real-time environmental data and real-time physical data of personnel in the target area; collect historical fresh air adjustment volume, historical environmental data and historical physical data of personnel in the target area; Prediction: Real-time environmental data and real-time personnel physical data are input into the heat balance equation to calculate the real-time PMV value. Historical environmental data and historical personnel physical data are input into the heat balance equation to calculate the historical PMV value. A BI-LSTM model is constructed and trained using historical PMV values and historical fresh air adjustment amounts to obtain a trained BI-LSTM model. The real-time PMV value is input into the trained BI-LSTM model to obtain the predicted fresh air adjustment amount; Regulation: Regulate the air supply volume of the fresh air system according to the predicted fresh air adjustment amount.

2. The energy-saving control method for the fresh air system according to claim 1, characterized in that: The data collection step also includes collecting the volume, real-time CO2 concentration and real-time VOCs concentration of the target area; After performing the step of data collection, the method further comprises: Concentration judgment: judge whether the real-time CO2 concentration and real-time VOCs concentration are both less than the corresponding concentration thresholds. If so, no processing is performed; if not, the step of determining the ventilation volume is executed; Determine ventilation volume: Determine the target ventilation volume based on the volume of the target area, concentration threshold, real-time CO2 concentration, and real-time VOCs concentration, and execute control steps; The control step also includes calculating the sum of the target ventilation volume and the predicted fresh air adjustment volume, recording it as first data, and using the first data instead of the predicted fresh air adjustment volume to control the air supply volume of the fresh air system.

3. The energy-saving control method for the fresh air system according to claim 2, characterized in that: Before performing the step of determining the concentration, the method further includes: Calculate the CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount; use the CO2 concentration change to adjust the real-time CO2 concentration to obtain the adjusted CO2 concentration; use the VOCs concentration change to adjust the real-time VOCs concentration to obtain the adjusted VOCs concentration; and perform the concentration judgment step; In the concentration judgment step, the adjusted CO2 concentration is used instead of the real-time CO2 concentration, and the adjusted VOCs concentration is used instead of the real-time VOCs concentration.

4. The energy-saving control method for the fresh air system according to claim 2, characterized in that: The step of determining the concentration further comprises: Calculate the CO2 concentration change and VOCs concentration change corresponding to the predicted fresh air adjustment amount; use the CO2 concentration change and VOCs concentration change to adjust the concentration threshold to obtain a new concentration threshold.

5. The energy-saving control method for the fresh air system according to claim 4, characterized in that: When the real-time CO2 concentration and the real-time VOCs concentration are both less than the corresponding new concentration thresholds, the concentration determination step further includes: Determine whether the real-time PMV value is within the range of ±5% within the preset time. If so, control to reduce the opening of the fresh air valve and exhaust valve of the fresh air system, and increase the opening of the return air valve of the fresh air system; if not, no action is taken.

6. The energy-saving control method for a fresh air system according to any one of claims 1 to 5, characterized in that: After the prediction step and before the control step, the following steps are also included: Setting range: Generate multiple control strategies based on the predicted fresh air adjustment amount, including the actual fresh air adjustment amount, the target temperature setting value, and the target PMV value; input the target PMV value into the trained BI-LSTM model to obtain the target fresh air adjustment amount, where the value of the actual fresh air adjustment amount is between the target fresh air adjustment amount and the predicted fresh air adjustment amount; set the value range of the target temperature setting value; Initialization: Set the objective function of the NSGA-II algorithm, which includes an energy-saving function and a comfort function. Perform binary encoding on the actual fresh air adjustment amount and the target temperature setting value in each control strategy, obtain the encoding result of each actual fresh air adjustment amount and each target temperature setting value, concatenate the encoding result of each actual fresh air adjustment amount and the encoding result of each target temperature setting value into a binary string, and use the binary string as a chromosome of the NSGA-II algorithm. Determine the target strategy: Use the NSGA-II algorithm to obtain the target chromosome from multiple chromosomes, and take the control strategy corresponding to the target chromosome as the Pareto optimal solution. Repeat this step multiple times until the number of executions reaches the maximum number of iterations. Then, integrate all Pareto optimal solutions into a Pareto optimal solution set, and select the control strategy with the lowest energy consumption in the Pareto optimal solution set as the target strategy. In the control step, the air supply volume of the fresh air system is regulated according to the target strategy.

7. The energy-saving control method for the fresh air system according to claim 6, characterized in that: The initialization step also includes: Determine the control accuracy of the actual fresh air adjustment amount and the target temperature setting value, determine the number of binary coding bits based on the control accuracy, and binary code the actual fresh air adjustment amount and the target temperature setting value in each control strategy according to the number of binary coding bits; The calculation model of the binary code bit number is as follows: ； in, The number of bits of binary code for the actual fresh air adjustment value; The number of bits of binary encoding for the target temperature setting value; It is the maximum value of the actual fresh air adjustment amount; The maximum value of the target temperature setting; It is the minimum value of the actual fresh air adjustment amount; The minimum value of the target temperature setting; The control accuracy of the actual fresh air adjustment amount; It is the control accuracy of the target temperature setting value.

8. The energy-saving control method for the fresh air system according to claim 7, characterized in that: The step of determining the target strategy also includes: The target chromosome is segmented according to the binary coding bit number of the actual fresh air adjustment amount and the binary coding bit number of the target temperature setting value to obtain the coding block of the actual fresh air adjustment amount and the coding block of the target temperature setting value, and the coding block of the actual fresh air adjustment amount and the coding block of the target temperature setting value are decoded respectively to obtain the decoded actual fresh air adjustment amount and the decoded target temperature setting value; Determine whether the decoded actual fresh air adjustment amount and the decoded target temperature setting value are consistent with expectations. If so, no processing is performed; if not, an alarm signal is issued.

9. An energy-saving control system for a fresh air system, characterized in that: include: processor and memory, The memory stores program code; When the processor calls the program code in the memory, the steps of the method according to any one of claims 1 to 8 are executed.

Images