A multi-zone air conditioning control method and system based on intelligent detection

Abstract

The present invention provides a multi-zone air conditioning control method and system based on intelligent detection, which relates to the field of air conditioning technology. The method comprises: obtaining temperature data, set temperature data, wind flux data, working power data, outdoor temperature data, and average number of people in multiple zones at multiple times of the current detection cycle; obtaining first predicted temperature data through a temperature prediction model, and judging whether the temperature prediction model needs to be trained, and if so, training the temperature prediction model; obtaining second predicted temperature data through the temperature prediction model; judging whether the working power data and wind flux data need to be adjusted according to the second predicted temperature data and the set temperature data; and if so, determining the adjusted air conditioning working power data and wind flux data through the trained temperature prediction model. According to the present invention, the accuracy of temperature regulation can be improved, and the possibility that some areas cannot reach the set temperature can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning, and in particular to a multi-zone air conditioning control method and system based on intelligent detection. Background Art

[0002] In the related art, the air conditioner may be provided with a temperature sensor, which can measure the indoor temperature at any time during operation, and reduce the power to maintain the temperature when the temperature reaches the set temperature. When the indoor temperature deviates from the set temperature again, the power can be increased again to make the indoor temperature close to or reach the set temperature. However, the wind flux and power of the air outlet of the air conditioner always operate according to the set data. For example, the air is always delivered according to a certain wind flux. The power of the air conditioner works according to the set power when it is working normally or using reduced power. It may make it difficult to reach the set temperature indoors, or waste electricity. In addition, when the air conditioner adjusts the temperature for multiple areas, the wind flux needs to be set for each area, and the air conditioner works according to the set power, which makes it difficult for all areas to reach the set temperature.

[0003] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0004] The present invention provides a multi-zone air conditioning control method and system based on intelligent detection, which can solve the technical problem that it is difficult to make all zones reach the set temperature in the related technology.

[0005] According to a first aspect of the present invention, there is provided a multi-zone air conditioning control method based on intelligent detection, comprising:

[0006] At multiple moments in the current detection cycle, temperature data of multiple areas are obtained respectively;

[0007] Obtain the set temperature data of multiple areas, as well as the wind flux data of the air conditioner outlet during the current detection period, the working power data of the air conditioner, the outdoor temperature data and the average number of people in the area;

[0008] The temperature prediction model is used to process the wind flux data of the air outlet of the air conditioner in the current detection period, the working power data of the air conditioner, the outdoor temperature data, the average number of people in multiple areas, and the temperature data at the first moment of the current detection period, and obtain the first predicted temperature data of multiple areas at multiple moments of the current detection period respectively;

[0009] determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data;

[0010] If the temperature prediction model needs to be trained, the temperature prediction model is trained according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model;

[0011] The trained temperature prediction model is used to process the wind flux data of the air outlet of the air conditioner in the current detection cycle, the working power data of the air conditioner, the predicted outdoor temperature data of the next detection cycle, the average number of people in multiple areas, and the temperature data at the last moment of the current detection cycle, and obtain the second predicted temperature data of multiple areas at multiple moments in the next detection cycle respectively;

[0012] Determine whether the working power data of the air conditioner and the wind flux data of the air outlets of the air conditioners in each area in the next detection cycle need to be adjusted according to the second predicted temperature data and the set temperature data;

[0013] If adjustment is required, the adjusted air-conditioning working power data and the adjusted wind flux data of the air-conditioning outlets in each area are determined according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in multiple areas, and the trained temperature prediction model;

[0014] In the next detection cycle, the working power data of the air conditioner is set to the adjusted working power data of the air conditioner, and the wind flux data of the air outlets of the air conditioner in each area is set to the adjusted wind flux data.

[0015] According to a second aspect of the present invention, there is provided a multi-zone air conditioning control system based on intelligent detection, comprising:

[0016] The temperature acquisition module is used to obtain temperature data of multiple areas at multiple times in the current detection cycle;

[0017] The first data acquisition module is used to acquire the set temperature data of multiple areas, as well as the wind flux data of the air outlet of the air conditioner in the current detection period, the working power data of the air conditioner, the outdoor temperature data and the average number of people in the area;

[0018] A first prediction module is used to process the wind flux data of the air outlet of the air conditioner, the working power data of the air conditioner, the outdoor temperature data, the average number of people in multiple areas, and the temperature data at the first moment of the current detection cycle through a temperature prediction model, and obtain first predicted temperature data of multiple areas at multiple moments of the current detection cycle respectively;

[0019] A first judgment module, used for determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data;

[0020] a training module, configured to train the temperature prediction model according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model if the temperature prediction model needs to be trained;

[0021] The second prediction module is used to process the wind flux data of the air outlet of the air conditioner in the current detection cycle, the working power data of the air conditioner, the predicted outdoor temperature data of the next detection cycle, the average number of people in multiple areas, and the temperature data at the last moment of the current detection cycle through the trained temperature prediction model, and obtain the second predicted temperature data of multiple areas at multiple moments in the next detection cycle respectively;

[0022] A second judgment module is used to determine whether the working power data of the air conditioner and the wind flux data of the air outlets of the air conditioner in each area in the next detection cycle need to be adjusted according to the second predicted temperature data and the set temperature data;

[0023] An adjustment module, for determining adjusted air conditioning working power data and adjusted wind flux data of the air conditioning outlets of each area according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in multiple areas and the trained temperature prediction model if adjustment is required;

[0024] The setting module is used to set the working power data of the air conditioner to the adjusted working power data of the air conditioner in the next detection cycle, and to set the wind flux data of the air outlet of the air conditioner in each area to the adjusted wind flux data.

[0025] Technical effect: According to the present invention, a temperature prediction model can be trained, and the temperature prediction model can be used to determine whether each area can reach or remain near the set temperature in the next cycle based on the wind flux data of the air-conditioning outlet of each area in the current detection cycle and the working power data of the air-conditioning. If not, the wind flux data of the air-conditioning outlet of some or all areas can be adjusted, and the working power data of the air-conditioning can also be adjusted. When adjusting, the influence of the outdoor temperature data on the indoor temperature adjustment work is referred to improve the accuracy of temperature adjustment, so that intelligent adjustment of each area can be achieved, the possibility of some areas failing to reach the set temperature can be reduced, and electric energy can be saved. When judging whether the temperature prediction model needs to be trained, the total deviation between all the measured temperature data of each area in the current detection cycle and the first predicted temperature data can be solved, and the maximum value thereof can be determined as the maximum temperature error of each area. Therefore, when the maximum temperature error of each area is greater than or equal to the preset error threshold, it is determined that the temperature prediction accuracy of the temperature prediction model for at least one area is poor, and it can be accurately and objectively judged whether the temperature prediction model needs to be trained. When training the temperature prediction model, the total deviation between the measured temperature data corresponding to each area and the first predicted temperature data can be weighted and summed by the wind flux data of each area and the difference between the outdoor temperature and the average temperature of each area, so as to obtain the loss function of the temperature prediction model, so that the loss function can be used not only to represent the total deviation corresponding to each area, but also to represent the difficulty of temperature regulation in each area. After training, the accuracy of the temperature prediction model is improved, as well as the adaptability to factors such as personnel flow and outdoor temperature changes in each area, and the performance of the temperature prediction model is improved. When setting the constraint conditions of the air conditioning control optimization model, the predicted temperature data of the next detection cycle can be represented by the trained temperature prediction model and the parameters such as the pending value of the adjusted wind flux data and the pending value of the adjusted air conditioning working power data, and the constraint conditions are set for the predicted temperature data of the next detection cycle, so that the temperature of each area in the next detection cycle can be effectively regulated, and the regulation effect of the wind flux data and the air conditioning working power data is improved. When determining the objective function of the air-conditioning control optimization model, the overall minimization of the error between the adjusted temperature and the set temperature of each area can be taken as the goal, so that the adjustment effect of the air-conditioning working power and wind flux can be maximized as a whole. By minimizing the adjusted air-conditioning working power data and minimizing the total wind flux, the energy consumption can be minimized while the temperature of the area can be effectively adjusted, thereby improving the temperature adjustment efficiency and saving electricity.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work.

[0028] Figure 1 A schematic diagram of a process flow of a multi-zone air conditioning control method based on intelligent detection according to an embodiment of the present invention is exemplarily shown;

[0029] Figure 2 A block diagram of a multi-zone air-conditioning control system based on intelligent detection according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0032] Figure 1 A flowchart of a multi-zone air conditioning control method based on intelligent detection according to an embodiment of the present invention is exemplarily shown, and the method includes:

[0033] Step S101, acquiring temperature data of multiple areas at multiple moments in the current detection cycle;

[0034] Step S102, obtaining set temperature data of multiple areas, as well as wind flux data of the air outlet of the air conditioner in the current detection period, working power data of the air conditioner, outdoor temperature data and the average number of people in the area;

[0035] Step S103, using a temperature prediction model, processing the wind flux data of the air outlet of the air conditioner, the working power data of the air conditioner, the outdoor temperature data, the average number of people in multiple areas, and the temperature data at the first moment of the current detection cycle in the current detection cycle, and obtaining first predicted temperature data of multiple areas at multiple moments of the current detection cycle respectively;

[0036] Step S104, determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data;

[0037] Step S105, if the temperature prediction model needs to be trained, the temperature prediction model is trained according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model;

[0038] Step S106, using the trained temperature prediction model, the wind flux data of the air outlet of the air conditioner in the current detection cycle, the working power data of the air conditioner, the predicted outdoor temperature data of the next detection cycle, the average number of people in multiple areas, and the temperature data at the last moment of the current detection cycle are processed to obtain second predicted temperature data of multiple areas at multiple moments in the next detection cycle respectively;

[0039] Step S107, determining whether the working power data of the air conditioner and the wind flux data of the air outlets of the air conditioner in each area in the next detection cycle need to be adjusted according to the second predicted temperature data and the set temperature data;

[0040] Step S108, if adjustment is required, the adjusted air conditioning working power data and the adjusted wind flux data of the air conditioning outlets of each area are determined according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in multiple areas and the trained temperature prediction model;

[0041] Step S109, in the next detection cycle, the working power data of the air conditioner is set to the adjusted working power data of the air conditioner, and the wind flux data of the air outlet of the air conditioner in each area is set to the adjusted wind flux data.

[0042] According to the multi-zone air-conditioning control method based on intelligent detection in an embodiment of the present invention, a temperature prediction model can be trained, and the temperature prediction model can be used to determine whether each zone can reach or remain near the set temperature in the next cycle based on the wind flux data of the air-conditioning outlets of each zone in the current detection cycle and the working power data of the air-conditioning. If not, the wind flux data of the air-conditioning outlets of some or all zones can be adjusted, and the working power data of the air-conditioning can also be adjusted. When adjusting, the influence of outdoor temperature data on indoor temperature regulation can be referred to to improve the accuracy of temperature regulation, thereby realizing intelligent regulation of each zone, reducing the possibility that some zones cannot reach the set temperature, and saving electric energy.

[0043] According to one embodiment of the present invention, an air conditioner (e.g., a central air conditioner) can adjust the indoor temperature of a building having multiple areas. For example, a building having multiple areas can be a shopping mall, an office building, a hospital, etc. The present invention does not limit the type of building having multiple areas. Each area can have an air conditioner outlet, which can supply air to the area so that the temperature of the area reaches or approaches the set temperature data.

[0044] According to one embodiment of the present invention, in step S101, at multiple moments in the current detection cycle, temperature data of multiple regions are obtained respectively, that is, temperature data of multiple moments in each region can be obtained. The duration of the detection cycle can be 15 minutes, 20 minutes, 30 minutes, etc., and the time interval between adjacent moments can be 30 seconds, 1 minute, etc. The present invention does not limit the duration of the detection cycle and the time interval between adjacent moments.

[0045] According to one embodiment of the present invention, in step S102, the set temperature data of multiple areas, as well as the wind flux data of the air outlet of the air conditioner in the current detection period, the working power data of the air conditioner and the outdoor temperature data can be obtained, which are used as reference data in the training process of the temperature prediction model. In addition, the number of people in the area at each time can be counted by a living body sensing device (for example, ultrasonic and far-infrared equipment), and the number of people at each time is averaged to determine the average number of people in the area in the current detection period.

[0046] According to one embodiment of the present invention, in step S103, the temperature prediction model may be a deep learning neural network model, and the temperature prediction model may be trained through historical data so that the temperature prediction model can process the wind flux data of the air-conditioning outlet, the working power data of the air-conditioning, the average number of people in the area, the outdoor temperature data, and the temperature data at the first moment of the current detection cycle to obtain the first predicted temperature data at multiple moments in the current detection cycle, that is, based on the wind flux data of the air-conditioning outlet, the working power data of the air-conditioning, the average number of people in the area, and the outdoor temperature data, to predict the temperature change trend of the area, and combine the temperature data at the first moment to predict the temperature data at subsequent moments.

[0047] According to one embodiment of the present invention, in step S104, due to factors such as the flow of people in the area or changes in outdoor temperature, there may be a certain error between the first predicted temperature data output by the temperature prediction model trained with historical data and the actually measured temperature data. Therefore, it can be determined whether the error is large. If the error is large, it is necessary to further train the temperature prediction model through various data in the current detection cycle to make the temperature prediction model more suitable for the current environment of the area. If the error is small, there is no need for training and the temperature prediction model can be used directly for subsequent processing.

[0048] According to an embodiment of the present invention, determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data includes: determining the maximum temperature error E of each area according to formula (1),

[0049]

[0050] Among them, T k,i is the temperature data of the kth area at the i-th moment of the current detection cycle, T 1,k,i is the first predicted temperature data of the kth region at the i-th moment of the current detection cycle, n is the number of moments in the current detection cycle, n k is the number of regions, max is the maximum value function, i≤n, k≤n k , and i, k, n and n k are all positive integers; if the maximum temperature error of each area is greater than or equal to the preset error threshold, it is determined whether the temperature prediction model needs to be trained.

[0051] According to one embodiment of the present invention, in formula (1), |T k,i -T 1,k,i | is the deviation between the measured temperature data of the kth region at the i-th moment of the current detection cycle and the first predicted temperature data, therefore, is the total deviation between all measured temperature data of the kth region in the current detection cycle and the first predicted temperature data, It represents the maximum value of the total deviation of each area.

[0052] According to one embodiment of the present invention, if the maximum temperature error of each area is greater than or equal to a preset error threshold, it means that there is at least one area where the temperature prediction model has poor accuracy in temperature prediction. In this case, it is determined whether the temperature prediction model needs to be trained.

[0053] In this way, the total deviation between all measured temperature data of each area in the current detection cycle and the first predicted temperature data can be solved, and the maximum value therein can be determined as the maximum temperature error of each area. Therefore, when the maximum temperature error of each area is greater than or equal to the preset error threshold, it is determined that the temperature prediction model has poor temperature prediction accuracy for at least one area, and it can be accurately and objectively judged whether the temperature prediction model needs to be trained.

[0054] According to an embodiment of the present invention, in step S105, if the temperature prediction accuracy of the temperature prediction model for at least one region is poor, the temperature prediction model may be trained.

[0055] According to one embodiment of the present invention, if the temperature prediction model needs to be trained, the temperature prediction model is trained according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model, including: determining a loss function of the temperature prediction model according to the first predicted temperature data, the temperature data, the average number of people, the wind flux data, and the outdoor temperature data of each area; and training the temperature prediction model according to the loss function to obtain a trained temperature prediction model.

[0056] According to one embodiment of the present invention, the loss function of the temperature prediction model is determined according to the first predicted temperature data, the temperature data and the wind flux data of each area, and the outdoor temperature data, including: determining the loss function LOSS of the temperature prediction model according to formula (2):

[0057]

[0058] Among them, T k,i is the temperature data of the kth area at the i-th moment of the current detection cycle, T 1,k,i is the first predicted temperature data of the kth region at the i-th moment of the current detection cycle, n is the number of moments in the current detection cycle, n k is the number of regions, T out is the outdoor temperature data, F k is the wind flux data of the kth area in the current detection cycle, N k is the average number of people in the kth area during the current detection cycle, i≤n, k≤n k , and i, k, n and n k All are positive integers.

[0059] According to one embodiment of the present invention, in formula (2), It can represent the total deviation between all measured temperature data of the kth area in the current detection cycle and the first predicted temperature data, Then it can represent the weight of the total deviation corresponding to the kth region, where is the ratio of the wind flux data of the kth region to the total wind flux of all regions. The larger the wind flux data is, the larger the area of ​​the region is, and the more difficult it is to adjust the temperature. Therefore, a higher weight can be set for the region, that is, Set as part of the weight of the region. is the ratio of the average number of people in the kth region to the total number of people in each region. The more people there are in a region, the more heat is generated by human activities, the more difficult it is to regulate the temperature, and the more difficult it is to accurately control the temperature of the region. Therefore, a higher weight can be set for the region, that is, is set as part of the weight of the region. On the other hand, is the average temperature of the kth region in the current detection cycle, is the difference between the outdoor temperature data and the average temperature of the kth region. The larger the difference, the more difficult it is to adjust the temperature of the region, and the more difficult it is to accurately control the temperature of the region. Therefore, a higher weight can be set for the region, that is, Set to the other part of the weight for this region.

[0060] According to an embodiment of the present invention, the total deviations corresponding to each area can be weighted and summed by the above weights to obtain the loss function of the temperature prediction model, which can be used not only to represent the total deviations corresponding to each area, but also to represent the difficulty of temperature adjustment in each area. The loss function can be used for back propagation, and the parameters of the temperature prediction model can be adjusted by the gradient descent method to obtain the trained temperature prediction model, thereby improving the accuracy of the temperature prediction model, as well as the adaptability to factors such as personnel flow in each area and outdoor temperature changes, and improving the performance of the temperature prediction model.

[0061] In this way, the total deviation between the measured temperature data corresponding to each area and the first predicted temperature data can be weighted and summed by the wind flux data of each area, the average number of people, and the difference between the outdoor temperature and the average temperature of each area, so as to obtain the loss function of the temperature prediction model, so that the loss function can be used not only to represent the total deviation corresponding to each area, but also to represent the difficulty of temperature regulation in each area. After training, the accuracy of the temperature prediction model is improved, as well as its adaptability to factors such as personnel flow in each area and outdoor temperature changes, and the performance of the temperature prediction model is improved.

[0062] According to an embodiment of the present invention, in step S106, the temperature data of multiple regions at the last moment of the current detection cycle can be set as the temperature data at the first moment of the next detection cycle, and the predicted outdoor temperature data of the next detection cycle can be determined according to the weather forecast data, and it can be assumed that the average number of people in each region in the next detection cycle is equal to the average number of people in each region in the current detection cycle, that is, in a short period of time, the number of people in each region will not change significantly. Therefore, the wind flux data of the air outlet of the air conditioner in the current detection cycle, the working power data of the air conditioner, the predicted outdoor temperature data of the next detection cycle, the average number of people in multiple regions, and the temperature data of multiple regions at the last moment of the current detection cycle can be input into the trained temperature prediction model, and the second predicted temperature data of multiple regions at multiple moments of the next detection cycle can be obtained, and it can be determined whether it is necessary to adjust the working power data of the air conditioner and the wind flux data of the air outlet of the air conditioner in each region based on the second predicted temperature data and the set temperature data.

[0063] According to an embodiment of the present invention, in step S107, it can be determined whether the working power data of the air conditioner in the next detection cycle and the wind flux data of the air outlet of the air conditioner in each area need to be adjusted according to the second predicted temperature data and the set temperature data. In the example, if the gap between the second predicted temperature data and the set temperature data of a certain area at the first moment of the next detection cycle is smaller than the gap between the second predicted temperature data and the set temperature data at the last moment, that is, after a cycle of adjustment, the gap between the temperature in the area and the set temperature has not only not been reduced, but has increased, then it can be considered that at least one of the wind flux data and the working power data of the air conditioner needs to be adjusted. In another example, if the gap between the second predicted temperature data and the set temperature data of a certain area at the first moment of the next detection cycle is greater than or equal to the gap between the second predicted temperature data and the set temperature data at the last moment, but the gap between the second predicted temperature data and the set temperature data at the last moment is still large, then it can be considered that the temperature adjustment efficiency in the area is low, and at least one of the wind flux data and the working power data of the air conditioner needs to be adjusted.

[0064] According to one embodiment of the present invention, in step S108, if adjustment is required, the adjusted air-conditioning working power data and the adjusted wind flux data of the air-conditioning outlets of each area are determined according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in the multiple areas and the trained temperature prediction model, including: determining the constraints of the air-conditioning control optimization model according to the set temperature data, the average number of people in the multiple areas, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle and the trained temperature prediction model; determining the objective function of the air-conditioning control optimization model according to the trained temperature prediction model; determining the adjusted air-conditioning working power data and the adjusted wind flux data of the air-conditioning outlets of each area according to the constraints and objective function of the air-conditioning control optimization model.

[0065] According to one embodiment of the present invention, the air conditioning control optimization model may be a genetic algorithm model, a nonlinear programming model, etc., and may determine the optimal solution that can achieve the target represented by the objective function to the greatest extent under the constraints of the constraints. That is, under the constraints of the constraints, the adjusted wind flux data of the air conditioning outlets of each area that can achieve the target represented by the objective function to the greatest extent and the optimal solution of the adjusted air conditioning working power data may be calculated.

[0066] According to an embodiment of the present invention, the constraint conditions of the air conditioning control optimization model are determined according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, and the trained temperature prediction model, including: determining the constraint conditions of the air conditioning control optimization model according to formulas (3), (4), (5), (6) and (7),

[0067] f(T set,k ,T out,p ,T k,n ,N k ,P A ,F k,A )=(T k,1,p ,T k,2,p ,…T k,n,p )(3)

[0068]

[0069] |T k,n,p -T set,k |≤α|T k,n -T set,k |(5)

[0070] P A ≤P max (6)

[0071] F k,A ≤F k,max (7)

[0072] Among them, f is the temperature prediction model after training, T set,k is the set temperature data of the kth zone, T out,p is the predicted outdoor temperature data for the next detection cycle, T k,n is the temperature data of the kth region at the last moment of the current detection cycle, P A is the undetermined value of the adjusted air conditioner working power data, F k,A is the pending value of the adjusted wind flux data of the air conditioning outlet in the kth area, N k is the average number of people in the kth area in the current detection cycle, T k,1,p is the predicted temperature data of the kth region at the first moment of the next detection cycle, T k,2,p is the predicted temperature data of the kth region at the second moment of the next detection cycle, T k,n,p is the predicted temperature data of the kth area at the nth moment of the next detection cycle, α is a preset coefficient less than 1, P max is the maximum power of the air conditioner, F k,max is the maximum wind flux of the air-conditioning outlet in the kth area, n is the number of moments in the current detection cycle, i≤n, and i, k and n are all positive integers.

[0073] According to one embodiment of the present invention, in formula (3), the set temperature data of the kth area, the predicted outdoor temperature data of the next detection cycle, the temperature data of the kth area at the last moment of the current detection cycle, the average number of people in the kth area in the current detection cycle (assuming that it is equal to the average number of people in the next detection cycle), the pending value of the adjusted air-conditioning working power data and the pending value of the adjusted wind flux data are input into the trained temperature prediction model, and the predicted temperature data of the kth area at multiple moments in the next detection cycle can be obtained. Changing the pending value of the adjusted air-conditioning working power data and the pending value of the adjusted wind flux data will cause the predicted temperature data to change. Based on the subsequent constraints and objective function, suitable predicted temperature data can be found, thereby finding the optimal solution for the air-conditioning working power data and the adjusted wind flux data.

[0074] According to one embodiment of the present invention, in formula (4), the relationship between the predicted temperature data of the kth region at the first moment of the next detection cycle, the predicted temperature data of the kth region at the nth moment of the next detection cycle, and the set temperature data can be expressed. In the case where the predicted temperature data of the kth region at the first moment of the next detection cycle is higher than or equal to the set temperature data, the predicted temperature data of the kth region at the nth moment of the next detection cycle is lower than or equal to the set temperature data, that is, the set temperature data is within the temperature adjustment range of the kth region in the next detection cycle. On the contrary, in the case where the predicted temperature data of the kth region at the first moment of the next detection cycle is lower than or equal to the set temperature data, the predicted temperature data of the kth region at the nth moment of the next detection cycle is higher than or equal to the set temperature data, that is, the set temperature data is within the temperature adjustment range of the kth region in the next detection cycle. Therefore, based on the constraint conditions represented by formula (4), the temperature in the kth region can be effectively adjusted, that is, the set temperature is reached and maintained near the set temperature.

[0075] According to one embodiment of the present invention, in formula (5), it is indicated that at the end of the next detection cycle, the difference between the predicted temperature data and the set temperature data of the kth region at the nth moment of the next detection cycle is less than or equal to α times the difference between the temperature data at the last moment of the current detection cycle and the predicted temperature data, and α is a preset coefficient less than 1, that is, the difference between the predicted temperature data and the set temperature at the end of the next detection cycle is smaller than the difference between the temperature data and the set temperature at the end of the current detection cycle, thereby enabling the temperature in the kth region to be effectively regulated.

[0076] According to one embodiment of the present invention, formula (6) may indicate that the adjusted air conditioner working power data does not exceed the maximum power of the air conditioner, and formula (7) may indicate that the adjusted wind flux data of each air conditioner outlet does not exceed the maximum wind flux of the air conditioner outlet in the area.

[0077] In this way, the predicted temperature data of the next detection cycle can be represented by the trained temperature prediction model and parameters such as the pending value of the adjusted wind flux data and the pending value of the adjusted air-conditioning working power data, and constraints can be set on the predicted temperature data of the next detection cycle, so that the temperature of each area in the next detection cycle can be effectively adjusted, thereby improving the adjustment effect of the wind flux data and the air-conditioning working power data.

[0078] According to one embodiment of the present invention, determining the objective function of the air conditioning control optimization model according to the trained temperature prediction model includes: determining the objective function of the air conditioning control optimization model according to formulas (8), (9) and (10),

[0079]

[0080] min(P A )(9)

[0081]

[0082] Among them, min is the minimization function, k≤n k , n k Is a positive integer.

[0083] According to one embodiment of the present invention, in formula (8), it can be indicated that the sum of the gaps between the predicted temperature data and the set temperature data of all regions at the end of the next detection cycle is minimized as a goal, that is, the error between the adjusted temperature of each region and the set temperature is minimized as a whole. In formula (9), it can be indicated that the adjusted air conditioning working power data is minimized, so that the power consumption is minimized when the temperature of each region can be effectively adjusted, thereby saving energy. In formula (10), it can be indicated that the sum of the adjusted wind flux data of the air outlet of the air conditioner in each region is minimized, that is, the total wind flux of the next detection cycle is minimized, so that the energy consumption generated during air supply is minimized when the temperature of each region can be effectively adjusted.

[0084] In this way, the goal is to minimize the overall error between the adjusted temperature and the set temperature of each area, so that the adjustment effect of the air conditioner's working power and wind flux can be maximized as a whole. By minimizing the adjusted air conditioner working power data and minimizing the total wind flux, the energy consumption can be minimized while the temperature of the area can be effectively adjusted, thereby improving the temperature adjustment efficiency and saving electricity.

[0085] According to one embodiment of the present invention, after setting the above objective function and constraints, the air-conditioning control optimization model can be solved to obtain the optimal solution of the adjusted air-conditioning working power data as the air-conditioning working power data for the next detection cycle, and to obtain the optimal solution of the adjusted wind flux data of the air-conditioning outlets in each area as the wind flux data of the air-conditioning outlets in each area for the next detection cycle.

[0086] According to one embodiment of the present invention, in step S109, at the beginning of the next detection cycle, the working power data of the air conditioner can be set to the adjusted working power data of the air conditioner, and the wind flux data of the air outlet of the air conditioner in each area can be set to the adjusted wind flux data.

[0087] According to the multi-zone air conditioning control method based on intelligent detection of the embodiment of the present invention, the temperature prediction model can be trained, and the temperature prediction model can be used to determine whether each zone can reach or remain near the set temperature in the next cycle based on the wind flux data of the air conditioning outlet of each zone in the current detection cycle and the working power data of the air conditioning. If not, the wind flux data of the air conditioning outlet of some or all zones can be adjusted, and the working power data of the air conditioning can also be adjusted. When adjusting, the influence of the outdoor temperature data on the indoor temperature adjustment work is referred to improve the accuracy of temperature adjustment, so that the intelligent adjustment of each zone can be realized, the possibility that some zones cannot reach the set temperature is reduced, and electric energy can be saved. When judging whether the temperature prediction model needs to be trained, the total deviation between all the measured temperature data of each zone in the current detection cycle and the first predicted temperature data can be solved, and the maximum value thereof can be determined as the maximum temperature error of each zone, so that when the maximum temperature error of each zone is greater than or equal to the preset error threshold, it is determined that the temperature prediction accuracy of the temperature prediction model for at least one zone is poor, and it can be accurately and objectively judged whether the temperature prediction model needs to be trained. When training the temperature prediction model, the total deviation between the measured temperature data corresponding to each area and the first predicted temperature data can be weighted and summed by the wind flux data of each area and the difference between the outdoor temperature and the average temperature of each area, so as to obtain the loss function of the temperature prediction model, so that the loss function can be used not only to represent the total deviation corresponding to each area, but also to represent the difficulty of temperature regulation in each area. After training, the accuracy of the temperature prediction model is improved, as well as the adaptability to factors such as personnel flow and outdoor temperature changes in each area, and the performance of the temperature prediction model is improved. When setting the constraint conditions of the air conditioning control optimization model, the predicted temperature data of the next detection cycle can be represented by the trained temperature prediction model and the parameters such as the pending value of the adjusted wind flux data and the pending value of the adjusted air conditioning working power data, and the constraint conditions are set for the predicted temperature data of the next detection cycle, so that the temperature of each area in the next detection cycle can be effectively regulated, and the regulation effect of the wind flux data and the air conditioning working power data is improved. When determining the objective function of the air-conditioning control optimization model, the overall minimization of the error between the adjusted temperature and the set temperature of each area can be taken as the goal, so that the adjustment effect of the air-conditioning working power and wind flux can be maximized as a whole. By minimizing the adjusted air-conditioning working power data and minimizing the total wind flux, the energy consumption can be minimized while the temperature of the area can be effectively adjusted, thereby improving the temperature adjustment efficiency and saving electricity.

[0088] Figure 2 A block diagram of a multi-zone air conditioning control system based on intelligent detection according to an embodiment of the present invention is exemplarily shown, and the system includes:

[0089] The temperature acquisition module is used to obtain temperature data of multiple areas at multiple times in the current detection cycle;

[0090] The first data acquisition module is used to acquire the set temperature data of multiple areas, as well as the wind flux data of the air outlet of the air conditioner in the current detection period, the working power data of the air conditioner, the outdoor temperature data and the average number of people in the area;

[0091] A first prediction module is used to process the wind flux data of the air outlet of the air conditioner, the working power data of the air conditioner, the outdoor temperature data, the average number of people in multiple areas, and the temperature data at the first moment of the current detection cycle through a temperature prediction model, and obtain first predicted temperature data of multiple areas at multiple moments of the current detection cycle respectively;

[0092] A first judgment module, used for determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data;

[0093] a training module, configured to train the temperature prediction model according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model if the temperature prediction model needs to be trained;

[0094] The second prediction module is used to process the wind flux data of the air outlet of the air conditioner in the current detection cycle, the working power data of the air conditioner, the predicted outdoor temperature data of the next detection cycle, the average number of people in multiple areas, and the temperature data at the last moment of the current detection cycle through the trained temperature prediction model, and obtain the second predicted temperature data of multiple areas at multiple moments in the next detection cycle respectively;

[0095] A second judgment module is used to determine whether the working power data of the air conditioner and the wind flux data of the air outlets of the air conditioner in each area in the next detection cycle need to be adjusted according to the second predicted temperature data and the set temperature data;

[0096] An adjustment module, for determining adjusted air conditioning working power data and adjusted wind flux data of the air conditioning outlets of each area according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in multiple areas and the trained temperature prediction model if adjustment is required;

[0097] The setting module is used to set the working power data of the air conditioner to the adjusted working power data of the air conditioner in the next detection cycle, and to set the wind flux data of the air outlet of the air conditioner in each area to the adjusted wind flux data.

[0098] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-zone air conditioning control method based on intelligent detection, characterized in that: include: At multiple moments in the current detection cycle, temperature data of multiple areas are obtained respectively; Obtain the set temperature data of multiple areas, as well as the wind flux data of the air conditioner outlet, the working power data of the air conditioner, the outdoor temperature data and the average number of people in the area during the current detection period. The average number of people in the area is obtained by counting the number of people in the area at each time and averaging the number of people at each time to determine the average number of people in the area; Through the temperature prediction model, the wind flux data of the air outlet of the air conditioner, the working power data of the air conditioner, the outdoor temperature data, the average number of people in multiple areas, and the temperature data at the first moment of the current detection cycle are processed to obtain the first predicted temperature data of multiple areas at multiple moments of the current detection cycle respectively. The average number of people in multiple areas represents the average number of people in each of the multiple areas at multiple moments; determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data; If the temperature prediction model needs to be trained, the temperature prediction model is trained according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model; The trained temperature prediction model is used to process the wind flux data of the air outlet of the air conditioner in the current detection cycle, the working power data of the air conditioner, the predicted outdoor temperature data of the next detection cycle, the average number of people in multiple areas, and the temperature data at the last moment of the current detection cycle, and obtain the second predicted temperature data of multiple areas at multiple moments in the next detection cycle respectively; Determine whether the working power data of the air conditioner and the wind flux data of the air outlets of the air conditioners in each area in the next detection cycle need to be adjusted according to the second predicted temperature data and the set temperature data; If adjustment is required, the adjusted air-conditioning working power data and the adjusted wind flux data of the air-conditioning outlets in each area are determined according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in multiple areas, and the trained temperature prediction model; In the next detection cycle, the working power data of the air conditioner is set to the adjusted working power data of the air conditioner, and the wind flux data of the air outlets of the air conditioner in each area is set to the adjusted wind flux data; If the temperature prediction model needs to be trained, the temperature prediction model is trained according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model, including: Determine a loss function of a temperature prediction model according to the first predicted temperature data, the temperature data, the average number of people, the wind flux data, and the outdoor temperature data of each area; According to the loss function, the temperature prediction model is trained to obtain a trained temperature prediction model; Determining a loss function of a temperature prediction model according to the first predicted temperature data, temperature data, average number of people, wind flux data, and outdoor temperature data of each area includes: According to the formula Determine the loss function LOSS of the temperature prediction model, where T k,i is the temperature data of the kth area at the i-th moment of the current detection cycle, T 1,k,i is the first predicted temperature data of the kth region at the i-th moment of the current detection cycle, n is the number of moments in the current detection cycle, n k is the number of regions, T out is the outdoor temperature data, F k is the wind flux data of the kth area in the current detection cycle, N k is the average number of people in the kth area during the current detection cycle, i≤n, k≤n k , and i, k, n and n k All are positive integers.

2. The multi-zone air conditioning control method based on intelligent detection according to claim 1, characterized in that: Determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data includes: According to the formula Determine the maximum temperature error E in each area, where T k,i is the temperature data of the kth area at the i-th moment of the current detection cycle, T 1,k,i is the first predicted temperature data of the kth region at the i-th moment of the current detection cycle, n is the number of moments in the current detection cycle, n k is the number of regions, max is the maximum value function, i≤n, k≤n k , and i, k, n and n k All are positive integers; If the maximum temperature error of each area is greater than or equal to the preset error threshold, it is determined whether the temperature prediction model needs to be trained.

3. The multi-zone air conditioning control method based on intelligent detection according to claim 1, characterized in that: If adjustment is required, the adjusted air conditioning working power data and the adjusted wind flux data of the air conditioning outlets in each area are determined according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in multiple areas, and the trained temperature prediction model, including: Determine the constraint conditions of the air conditioning control optimization model according to the set temperature data, the average number of people in the multiple areas, the predicted outdoor temperature data of the next detection cycle, the temperature data of the multiple areas at the last moment of the current detection cycle, and the trained temperature prediction model; Determining the objective function of the air conditioning control optimization model according to the trained temperature prediction model; According to the constraint conditions and objective function of the air conditioning control optimization model, the adjusted air conditioning working power data and the adjusted wind flux data of the air conditioning outlets in each area are determined.

4. The multi-zone air conditioning control method based on intelligent detection according to claim 3, characterized in that: According to the set temperature data, the average number of people in the multiple areas, the predicted outdoor temperature data of the next detection cycle, the temperature data of the multiple areas at the last moment of the current detection cycle, and the trained temperature prediction model, the constraint conditions of the air conditioning control optimization model are determined, including: According to the formula f(T set,k ,T out,p ,T k,n ,N k ,P A ,F k,A )=(T k,1,p ,T k,2,p ,…T k,n,p ) |T k,n,p -T set,k |≤α|T k,n -T set,k | P A ≤P max F k,A ≤F k,max Determine the constraints of the air conditioning control optimization model, where f is the trained temperature prediction model, T set,k is the set temperature data of the kth zone, T out,p is the predicted outdoor temperature data for the next detection cycle, T k,n is the temperature data of the kth region at the last moment of the current detection cycle, P A is the undetermined value of the adjusted air conditioner working power data, F k,A is the pending value of the adjusted wind flux data of the air conditioning outlet in the kth area, N k is the average number of people in the kth area in the current detection cycle, T k,1,p is the predicted temperature data of the kth region at the first moment of the next detection cycle, T k,2,p is the predicted temperature data of the kth region at the second moment of the next detection cycle, T k,n,p is the predicted temperature data of the kth area at the nth moment of the next detection cycle, α is a preset coefficient less than 1, P max is the maximum power of the air conditioner, F k,max is the maximum wind flux of the air-conditioning outlet in the kth area, n is the number of moments in the current detection cycle, i≤n, and i, k and n are all positive integers.

5. The multi-zone air conditioning control method based on intelligent detection according to claim 4, characterized in that: According to the trained temperature prediction model, determining the objective function of the air conditioning control optimization model includes: According to the formula Determine the objective function of the air conditioning control optimization model, where min is the minimization function, k≤n k , n k Is a positive integer.

6. A multi-zone air conditioning control system based on intelligent detection for executing the method according to any one of claims 1 to 5, characterized in that: include: The temperature acquisition module is used to obtain temperature data of multiple areas at multiple times in the current detection cycle; The first data acquisition module is used to acquire the set temperature data of multiple areas, as well as the wind flux data of the air outlet of the air conditioner in the current detection period, the working power data of the air conditioner, the outdoor temperature data and the average number of people in the area; A first prediction module is used to process the wind flux data of the air outlet of the air conditioner, the working power data of the air conditioner, the outdoor temperature data, the average number of people in multiple areas, and the temperature data at the first moment of the current detection cycle through a temperature prediction model, and obtain first predicted temperature data of multiple areas at multiple moments of the current detection cycle respectively; A first judgment module, used for determining whether the temperature prediction model needs to be trained according to the first predicted temperature data and the temperature data; a training module, configured to train the temperature prediction model according to the first predicted temperature data, the temperature data, the wind flux data, the outdoor temperature data and the average number of people in the area to obtain a trained temperature prediction model if the temperature prediction model needs to be trained; The second prediction module is used to process the wind flux data of the air outlet of the air conditioner in the current detection cycle, the working power data of the air conditioner, the predicted outdoor temperature data of the next detection cycle, the average number of people in multiple areas, and the temperature data at the last moment of the current detection cycle through the trained temperature prediction model, and obtain the second predicted temperature data of multiple areas at multiple moments in the next detection cycle respectively; A second judgment module is used to determine whether the working power data of the air conditioner and the wind flux data of the air outlets of the air conditioner in each area in the next detection cycle need to be adjusted according to the second predicted temperature data and the set temperature data; An adjustment module, for determining adjusted air conditioning working power data and adjusted wind flux data of the air conditioning outlets of each area according to the set temperature data, the predicted outdoor temperature data of the next detection cycle, the temperature data of multiple areas at the last moment of the current detection cycle, the average number of people in multiple areas and the trained temperature prediction model if adjustment is required; The setting module is used to set the working power data of the air conditioner to the adjusted working power data of the air conditioner in the next detection cycle, and to set the wind flux data of the air outlet of the air conditioner in each area to the adjusted wind flux data.

Images