Air conditioning unit parameter tuning method, device, equipment and medium

Abstract

The invention provides an air conditioning unit parameter tuning method and device, equipment and a medium, and belongs to the technical field of air conditioners. Taking the energy consumption of the air conditioning unit as an optimization target; the method comprises the following steps: initializing a bee population, presetting a search strategy and a stop criterion, and starting an optimization process; configuring a ParaSeer model, converting the energy consumption data of the air conditioning unit into a ParaSeer model processing form by a pattern converter based on the ParaSeer model, and constructing multi-dimensional time sequence data; using a predictor of a ParaSeer model to predict the energy consumption of the air conditioning unit in a future preset time period based on the multi-dimensional time sequence data; and the air conditioning unit is controlled based on the predicted energy consumption parameters. And an artificial bee colony algorithm is used for carrying out parameter search, and the indexes of the air conditioning unit are predicted through a ParaSeer time sequence model, so that the green and energy-saving requirements are met.

Description

Technical Field

[0001] The present invention belongs to the technical field of air conditioners, and in particular relates to a method, device, equipment and medium for optimizing parameters of an air conditioner unit. Background Art

[0002] Air conditioning equipment has become an indispensable part of daily life and work. While providing a comfortable environment, air conditioning equipment is also accompanied by huge energy consumption. Especially in large buildings or industrial sites, the energy consumption of air conditioning units often accounts for a large proportion of the total energy consumption. Therefore, how to effectively reduce the energy consumption of air conditioning units and improve their operating efficiency has become an urgent problem to be solved in the current field of energy conservation and emission reduction.

[0003] Most of the existing air conditioning unit parameter tuning methods are based on traditional trial and error or empirical methods. Although these methods are simple and easy to use, they often fail to find the optimal parameter combination, resulting in high energy consumption and low efficiency during the operation of the air conditioning unit. At the same time, since the operation of the air conditioning unit is affected by many factors, such as indoor and outdoor temperature, humidity, and human flow, changes in these factors will cause changes in the energy consumption of the air conditioning unit. Therefore, how to dynamically adjust the air conditioning unit parameters based on relevant data to achieve energy consumption optimization and effectively control the air conditioning unit is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides a method for optimizing parameters of an air-conditioning unit. The method uses a heuristic swarm intelligence algorithm to perform parameter search, and predicts air-conditioning unit indicators through a ParaSeer time series model to meet green energy-saving requirements.

[0005] The method includes: retrieving parameters of the air conditioning unit; The energy consumption of the air-conditioning unit is taken as the optimization target, and the parameters affecting the energy consumption are taken as the decision variables; the optimization process begins by initializing the bee population, and presetting the search strategy and stop criteria; In the optimization process, individual bees exchange information and collaborate with each other to continuously update their positions and search directions in order to find a parameter combination that achieves the preset energy consumption. Configure the ParaSeer model, and use the ParaSeer model-based graph converter to convert the energy consumption data of the air-conditioning unit into a form that can be processed by the ParaSeer model, and construct multi-dimensional time series data; The predictor of the ParaSeer model is used to predict the energy consumption of the air-conditioning unit in the future preset time period based on the multi-dimensional time series data; The air conditioning unit is controlled based on the predicted energy consumption parameters.

[0006] It should be further explained that the air-conditioning unit parameters retrieved include but are not limited to: chilled water inlet and outlet water temperatures of the chiller, start and stop of the chiller, electric meter reading of the chiller, inlet and outlet water temperatures of the cooling tower, electric meter reading of the cooling tower, start and stop of the cooling tower fan, cooling tower fan frequency, chilled water pump operating frequency, start and stop of the chilled water pump, electric meter reading of the chilled water pump, operating frequency of the cooling water pump, start and stop of the cooling water pump, electric meter reading of the cooling water pump, and inlet and outlet water flow of the main water pipe.

[0007] It should be further explained that the method also defines air conditioning unit parameter thresholds, which include: air conditioning unit host water outlet temperature threshold, cooling tower water outlet temperature threshold, cooling tower fan frequency threshold and chilled water pump frequency threshold.

[0008] It is further necessary to explain that the optimal solution in the parameters of the air conditioning unit is found by simulating the communication and cooperation between bees; In the artificial bee colony algorithm, bees are divided into three roles: scout bees, collecting bees, and observer bees; The scout bees randomly search for new air-conditioning unit parameters, the collecting bees search near the current air-conditioning unit parameters, and the observation bees select the better solution for search based on the air-conditioning unit parameters provided by the collecting bees.

[0009] It should be further explained that, in the method, the predictor adopts a variety of machine learning algorithms to learn the law of energy consumption changes by training historical air-conditioning unit parameters and predict future energy consumption.

[0010] It should be further explained that, in the method, the predicted energy consumption parameter is compared with the set energy consumption threshold, and if the predicted energy consumption exceeds the threshold, the control logic is triggered to adjust the air conditioning unit; Adjustment measures may include, but are not limited to: adjusting temperature set points, changing wind speed, and optimizing refrigerant flow.

[0011] It should be further explained that the method also includes: using the mean square error as the loss function to calculate the loss of the energy consumption parameters of the air-conditioning unit in the future preset time period, and the loss function is

[0012] Among them, N, S and d represent the number of samples, the length of the time series and the number of monitoring indicators respectively; when When missing, is 0, otherwise it is 1; The method also defines a projection module as a converter from latent representation to monitoring indicator time series; The projection module randomly sets part of the mask to zero and calculates the loss for the masked part based on the loss function.

[0013] The present application also provides an air conditioning unit parameter tuning device, the device comprising: a parameter optimization module and a parameter prediction module; The parameter optimization module is used to retrieve the parameters of the air-conditioning unit; the energy consumption of the air-conditioning unit is taken as the optimization target, and the parameters affecting the energy consumption are taken as the decision variables; the optimization process begins by initializing the bee population, and presetting the search strategy and stop criteria; during the optimization process, individual bees continuously update their positions and search directions through information exchange and collaboration to find a parameter combination that achieves the preset energy consumption; The parameter prediction module is used to configure the ParaSeer model. The ParaSeer model-based graph converter converts the energy consumption data of the air-conditioning unit into the form processed by the ParaSeer model and constructs multi-dimensional time series data. The predictor of the ParaSeer model is used to predict the energy consumption of the air-conditioning unit in the future preset time period based on the multi-dimensional time series data; The air conditioning unit is controlled based on the predicted energy consumption parameters.

[0014] According to another embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the air conditioning unit parameter tuning method when executing the program.

[0015] According to another embodiment of the present application, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the air conditioning unit parameter tuning method are implemented.

[0016] It can be seen from the above technical solutions that the present invention has the following advantages: The air conditioning unit parameter tuning method of the present application adopts an artificial bee colony algorithm to perform an optimization operation within a preset time period for the energy consumption of the air conditioning unit parameters. The algorithm simulates the foraging behavior of bees in nature, and finds the optimal parameter combination through information exchange and collaboration between bees, thereby minimizing the energy consumption of the air conditioning unit.

[0017] The air conditioning unit parameter tuning method of the present application predicts energy consumption by configuring the ParaSeer model, and by configuring the graphic converter of the ParaSeer model, it is possible to model the energy consumption information of the air conditioning unit based on the time series, and then use the predictor of the model to accurately predict the energy consumption of the air conditioning unit in the future preset time period. According to the predicted energy consumption parameters output by the ParaSeer model, the present invention can realize intelligent control of the air conditioning unit. By comparing the predicted parameters with the actual operating parameters, the system can automatically adjust the working state of the air conditioning unit to ensure that it operates in the best energy efficiency state, thereby achieving the purpose of energy saving and emission reduction.

[0018] The present invention realizes the automation and intelligence of air conditioning unit parameter tuning by combining artificial bee colony algorithm and ParaSeer model. The device can automatically identify and adjust parameters, improving the tuning efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the description will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 It is a flow chart of the method for optimizing parameters of air-conditioning units; Figure 2 This is the ParaSeer model architecture diagram; Figure 3 To build a dependency graph between model indicators; Figure 4 Schematic diagram of Transformer-based time series prediction; Figure 5 This is a schematic diagram of a parameter tuning device for an air conditioning unit; Figure 6 Schematic diagram of an electronic device. DETAILED DESCRIPTION

[0021] The air conditioning unit parameter tuning method of the present application uses a heuristic swarm intelligence algorithm based on the ParaSeer time series prediction model to search for the optimal parameter combination of the central air conditioning unit, and then uses ParaSeer to predict the impact of parameter changes on the energy consumption of the air conditioning unit, and uses it as an evaluation basis to find the most energy-efficient central air conditioning unit parameter combination in a certain environment. The present invention embeds causal dependencies in the time series prediction model, uses the physical modeling formula of the air conditioning unit to correct the model prediction results, integrates the advantages of the control scheme based on the mechanism model and the control scheme based on the data drive, and realizes an efficient parameter tuning mechanism.

[0022] The air-conditioning unit parameter tuning method involved in the present application is to combine the artificial bee colony algorithm with the ParaSeer model to intelligently tune the air-conditioning unit parameters.

[0023] This application uses an artificial bee colony algorithm to perform parameter optimization, and performs optimization operations within a preset time period for the energy consumption of air conditioning unit parameters. The algorithm simulates the foraging behavior of bees in nature, and through information exchange and collaboration between bees, finds the optimal parameter combination, thereby minimizing the energy consumption of the air conditioning unit.

[0024] This application introduces the ParaSeer model, which has powerful data processing and prediction capabilities. By configuring the ParaSeer model's graph converter, the present invention can model the air-conditioning unit energy consumption information based on the time series, and then use the model's predictor to accurately predict the air-conditioning unit energy consumption in the future preset time period.

[0025] Based on the predicted energy consumption parameters output by the ParaSeer model, the present invention can realize intelligent control of the air-conditioning unit. By comparing the predicted parameters with the actual operating parameters, the system can automatically adjust the working state of the air-conditioning unit to ensure that it operates in the best energy efficiency state, thereby achieving the purpose of energy saving and emission reduction.

[0026] The specific process of the air conditioning unit parameter tuning method involved in the present application is described in detail below. For the purpose of explanation rather than limitation, specific details such as specific system structures and technologies are proposed to facilitate a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details.

[0027] The phrases such as "one embodiment" or "some embodiments" described in the present application mean that the specific features, structures or characteristics described in the embodiment are included in one or more embodiments of the present application. Therefore, the phrases such as "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments" etc. that appear in different places in the present application do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways.

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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.

[0029] See also Figure 1 The figure is a flow chart of a method for optimizing parameters of an air conditioning unit in a specific embodiment, the method comprising: S101: Retrieve air conditioning unit parameters.

[0030] In some embodiments, the retrieved air conditioning unit parameters include but are not limited to: chiller chilled water inlet and outlet water temperature, chiller start and stop, chiller electric meter reading, cooling tower cooling water inlet and outlet water temperature, cooling tower electric meter reading, cooling tower fan start and stop, cooling tower fan frequency, chilled water pump operating frequency, chilled water pump start and stop, chilled water pump electric meter reading, cooling water pump operating frequency, cooling water pump start and stop, cooling water pump electric meter reading, main water pipe inlet and outlet water flow. Environmental parameters are also included: external temperature and humidity.

[0031] The adjustable parameters include main unit outlet water temperature, main unit frequency, cooling tower outlet water temperature, and water pump operating frequency.

[0032] This embodiment also defines air conditioning unit parameter thresholds, which include: air conditioning unit host outlet water temperature threshold, cooling tower outlet water temperature threshold, cooling tower fan frequency threshold, and chilled water pump frequency threshold.

[0033] Table 1

[0034] It should be noted that in the process of parameter optimization, in order to ensure the safe and stable operation of the air-conditioning system, the optimization parameters are limited to the feasible range. To avoid the risk of freezing of the evaporator and pipes due to overcooling of the chilled water, the lower limit of the outlet water temperature of the chiller is about 5°C. To avoid excessive temperature and difficulty in effectively removing indoor heat, the upper limit of the outlet water temperature of the host is about 12-15°C. To ensure the safe operation of the motors of the water pump and cooling tower, the operating frequency of the water pump and fan is limited to 30-50Hz.

[0035] S102: The energy consumption of the air-conditioning unit is taken as the optimization target, and the parameters affecting the energy consumption are taken as decision variables; the optimization process is started by initializing the bee population, and presetting the search strategy and stop criteria.

[0036] In this embodiment, the energy consumption of the air conditioning unit is taken as the optimization target to achieve the lowest energy consumption of the air conditioning unit under certain environmental parameters. This implementation combines the historical indicator trend and the indicator change trend brought by the historical parameter change to predict the impact of the current parameter change on the energy consumption indicator.

[0037] Optionally, in this embodiment, the collected meter reading index is used as the energy consumption of the air-conditioning unit, and a model is used to predict the change of the meter reading after the parameter changes.

[0038] This embodiment uses an artificial bee colony algorithm to optimize energy consumption. The artificial bee colony algorithm is a swarm intelligence optimization algorithm that simulates the foraging behavior of bees. The artificial bee colony algorithm simulates the communication and cooperation between bees to find the optimal solution in the solution space.

[0039] In the algorithm, bees are divided into three roles: scout bees, collector bees, and observer bees. Scout bees randomly search for new solutions, collector bees search near the current solution, and observer bees select better solutions for search based on the information provided by collector bees. The artificial bee colony algorithm can effectively search in the solution space and gradually approach the global optimal solution.

[0040] S103: During the optimization process, individual bees continuously update their positions and search directions through information exchange and collaboration to find a parameter combination that achieves a preset energy consumption.

[0041] For the artificial bee colony algorithm provided in this embodiment for parameter optimization, the population size is first determined. The number of leading bees and follower bees can be set equal, and the number of scout bees changes dynamically according to the algorithm process.

[0042] Set the problem dimension D, that is, the number of air conditioning unit parameters to be optimized. Randomly generate initial solutions xi (i∈1, 2, ..., GN), each solution represents a possible parameter combination, corresponding to a bee individual.

[0043] Initialize the maximum number of iterations UL as one of the stopping criteria. The bee collector performs local search near the current solution to generate new candidate solutions for the energy consumption of the air-conditioning unit.

[0044] The follower bee selects a nectar source for search based on the energy consumption of the air-conditioning unit provided by the leader bee, using a probabilistic selection mechanism such as roulette. The scout bee randomly searches for new nectar sources in the solution space to avoid falling into a local optimum.

[0045] For the initial optimization process of this embodiment, the fitness value of each solution is calculated, that is, the energy consumption of the air conditioning unit under the corresponding parameter combination. The leading bee and the following bee update their positions and search directions according to the search strategy. If a solution is not improved within a period of time, the corresponding leading bee is transformed into a scout bee. The above process is repeated until the maximum number of iterations is reached or other stopping criteria are met.

[0046] During the optimization process, individual bees exchange information and collaborate in the following ways: after the leading bee finds a new candidate solution, it shares its fitness value and information with the follower bees, and the follower bees select nectar sources for search based on this information.

[0047] When a solution has not been improved for a long time, the corresponding leader bee will transform into a scout bee and randomly search for new nectar sources in the solution space to increase the diversity of the population.

[0048] In each iteration, the currently found optimal solution is recorded and the global optimal solution is continuously updated.

[0049] Through continuous information exchange and collaboration, individual bees continue to explore and search in the solution space, and eventually converge to one or more air conditioning unit parameter combinations that achieve the preset energy consumption target. The air conditioning unit parameter combination is the optimal solution or approximate optimal solution found by the algorithm.

[0050] This embodiment can conduct a comprehensive search in the solution space through the division of labor and cooperation among the leading bees, the following bees and the scout bees. The artificial bee colony algorithm adopts a greedy strategy to select a better solution and can quickly converge to the global optimal solution or the approximate optimal solution. Moreover, the artificial bee colony algorithm does not need to understand the special information of the problem, but only needs to compare the pros and cons of the problem.

[0051] S104: Configure the ParaSeer model, and convert the energy consumption data of the air-conditioning unit into a form processed by the ParaSeer model based on the graph converter of the ParaSeer model, and construct multi-dimensional time series data.

[0052] S105: Using the predictor of the ParaSeer model to predict the energy consumption of the air-conditioning unit in a future preset time period based on the multi-dimensional time series data.

[0053] S106: Control the air conditioning unit based on the predicted energy consumption parameter.

[0054] In some specific embodiments, Figure 2 The ParaSeer model contains several transformers for monitoring indicator groups. Figure 3 The graphical model in ParaSeer is organized into a graphical transformer (GT). GT provides ParaSeer with the basic ability of time series prediction.

[0055] This embodiment also models the energy consumption of the air conditioning system and configures it as a PEP predictor (Parameter Effect Predictor, PEP) to adjust the prediction of the graph transformer for the direct impact index. The graph transformer will further propagate the effect of parameter changes to other variables in the graph model.

[0056] The ParaSeer model in this example uses Transformer to perform time series prediction. The architecture is as follows: Figure 4 As shown in Figure 2, the encoder module processes historical data and generates hidden representations. On the other hand, the decoder module connects two types of data: one is the recent historical data, and the other is the preliminary estimate of the data of interest (e.g., the average of historical data) or the adopted condition (e.g., future timestamp).

[0057] For one implementation of this embodiment, the air conditioning unit parameters are retrieved, and an encoder and decoder are constructed based on the Transformer architecture. The air conditioning unit parameters are converted into continuous vector representations. An encoder module is defined, and the Transformer self-attention mechanism is used to process the historical air conditioning unit parameters to generate hidden representations.

[0058] This embodiment further defines a decoder that connects historical air conditioning unit parameters and air conditioning unit energy consumption within a preset time period, hidden representations, and a preliminary estimate or condition of the energy consumption of the air conditioning unit of interest to generate a prediction or estimate.

[0059] Define the loss function and define a suitable loss function according to the energy consumption analysis requirements of the air-conditioning unit. Specifically, the mean square error (MSE) is used as the loss function. In the case of missing data, fill the missing points with zeros and exclude the contribution of these points from the loss. The loss function is as follows:

[0060] In fact, N, S and d represent the number of samples, the length of the time series and the number of monitoring indicators respectively. When missing, is 0 if yes, otherwise it is 1.

[0061] In this embodiment, multi-task learning can also be established to improve the generalization benefits of the ParaSeer model. Specifically, the projection module can be used as a converter from the hidden representation to the monitoring indicator time series. The projection module is applied on the encoder output to obtain the reconstructed ,..., .

[0062] This embodiment is inspired by the success of masked autoencoders in the language and vision fields, and randomly ,..., The mask is zeroed and the loss function is calculated for the masked part. Since the projection module is shared between the prediction and reconstruction tasks, the output of the encoder module in the reconstruction task should be the same as the output of the decoder module in the prediction task for the same time series.

[0063] This embodiment can calculate the MSE between two hidden representations to express this consideration. In this embodiment, the losses related to reconstruction are all calculated based on the second half of the time series, so that the sum of the three losses can be used as the final loss function.

[0064] Furthermore, the preprocessed air conditioner unit energy consumption pairs are input into the ParaSeer model, the loss is calculated by the loss function, and the ParaSeer model parameters are updated by back propagation. The training process is repeated until the ParaSeer model converges or the preset number of training rounds is reached.

[0065] Of course, in this embodiment, evaluation indicators suitable for the task can be selected, such as accuracy, F1 score, BLEU score, etc. The model is tuned according to the evaluation results. The performance of the model is evaluated on the test set to ensure that the model can meet the actual needs. The model is then deployed in the actual application environment for real-time prediction and decision-making.

[0066] On the basis of the above embodiments, in order to further improve the reliability of the air conditioning unit parameter tuning method provided in the above embodiments, the following is another implementation method. In one embodiment, when the air conditioning unit parameter tuning method is used, relevant operating parameters are retrieved from the control system of the air conditioning unit.

[0067] After obtaining the parameters of the air-conditioning unit, the artificial bee colony algorithm is used to perform an optimization operation on the energy consumption within a preset time period.

[0068] In the specific implementation, the energy consumption of the air-conditioning unit is taken as the optimization target, and the parameters that affect energy consumption (such as temperature setting value, wind speed, refrigerant flow rate, etc.) are taken as decision variables. The optimization process begins by initializing the bee population and setting appropriate search strategies and stop criteria. During the optimization process, individual bees continuously update their positions and search directions through information exchange and collaboration to find the parameter combination that minimizes energy consumption.

[0069] After completing parameter optimization, this embodiment configures the ParaSeer model and performs time series modeling. The ParaSeer model includes a graph converter and a predictor.

[0070] The graph converter is used to convert the energy consumption data of the air-conditioning unit into a form suitable for model processing. In specific implementation, according to the characteristics of the energy consumption data (such as periodicity, trend, etc.), select appropriate graph transformation methods (such as Fourier transform, wavelet transform, etc.) to convert the original energy consumption data into time series data. At the same time, it can also be combined with other relevant parameters (such as indoor and outdoor temperature, humidity, etc.) to construct multi-dimensional time series data to more comprehensively reflect the operating status of the air-conditioning unit.

[0071] Based on the graphical transformation, the predictor of the ParaSeer model is used to predict the energy consumption of the air-conditioning unit in the future preset time period. The predictor can use a variety of machine learning algorithms (such as neural networks, support vector machines, etc.) to learn the law of energy consumption changes by training historical data, and predict future energy consumption based on this.

[0072] This embodiment controls the air conditioning unit in real time according to the predicted energy consumption parameters output by the ParaSeer model. In specific implementation, the predicted energy consumption parameters are compared with the set energy consumption threshold. If the predicted energy consumption exceeds the threshold, the control logic is triggered to adjust the air conditioning unit.

[0073] Adjustments to the air conditioning unit may include, but are not limited to: adjusting the temperature setting, changing the wind speed, optimizing the refrigerant flow, etc. By adjusting the operating parameters of the air conditioning unit in real time, energy consumption can be effectively reduced while ensuring comfort, achieving the goal of energy conservation and emission reduction.

[0074] This embodiment adopts the artificial bee colony algorithm to perform parameter optimization, configures the ParaSeer model for time series modeling, and controls the air-conditioning unit based on predicted energy consumption parameters, thereby realizing the intelligent tuning of the air-conditioning unit parameters and the energy consumption prediction and control functions, improving the operating efficiency of the air-conditioning system, and reducing energy consumption costs.

[0075] Furthermore, as a refinement and extension of the specific implementation of the above embodiment, another method for optimizing parameters of an air-conditioning unit is provided below. In this method, an artificial bee colony algorithm is also used to perform an optimization operation on energy consumption within a preset time period.

[0076] For the ParaSeer model and time series modeling, this embodiment constructs a time series data set based on the parameters and energy consumption data of the air conditioning unit. The graphical converter based on the ParaSeer model configures the time series and converts the original data into a format that the model can process.

[0077] This embodiment uses the deep learning network of the ParaSeer model to model time series data and learn the characteristics and rules in the data. During the modeling process, the prediction accuracy and generalization ability of the model can be improved by adjusting the model's hyperparameters, optimizing the algorithm, and other means.

[0078] After completing the modeling of the ParaSeer model in this embodiment, the present invention uses the model to predict the energy consumption of the air-conditioning unit in a preset time period in the future. In specific implementation, the future time series is input into the trained ParaSeer model, and the model will output the corresponding predicted energy consumption parameters. The predicted energy consumption parameters include energy consumption values ​​at various time points in the future, energy consumption change trends, etc. By analyzing and comparing the predicted energy consumption parameters, energy consumption anomalies or potential problems can be discovered in a timely manner, and corresponding measures can be taken to adjust and optimize.

[0079] This embodiment can also control the air conditioning unit according to the predicted energy consumption parameter. In specific implementation, the predicted energy consumption parameter is compared with a set energy consumption threshold, and if the predicted energy consumption exceeds the threshold, the control logic is triggered to adjust the air conditioning unit.

[0080] The following is an embodiment of an air-conditioning unit parameter tuning device provided in an embodiment of the present disclosure. The device and the air-conditioning unit parameter tuning method of the above-mentioned embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the air-conditioning unit parameter tuning device, reference can be made to the embodiment of the above-mentioned air-conditioning unit parameter tuning method.

[0081] like Figure 5 As shown, the device includes: a parameter optimization module and a parameter prediction module.

[0082] The parameter optimization module is used to retrieve the parameters of the air-conditioning unit; the energy consumption of the air-conditioning unit is taken as the optimization target, and the parameters affecting the energy consumption are taken as decision variables; the optimization process begins by initializing the bee population and presetting the search strategy and stopping criteria; during the optimization process, individual bees continuously update their positions and search directions through information exchange and collaboration to find a parameter combination that achieves the preset energy consumption.

[0083] The parameter prediction module is used to configure the ParaSeer model. The graph transformer based on the ParaSeer model converts the energy consumption data of the air-conditioning unit into the form processed by the ParaSeer model and constructs multi-dimensional time series data.

[0084] The predictor of the ParaSeer model is used to predict the energy consumption of the air-conditioning unit in a preset time period in the future based on multi-dimensional time series data; and the air-conditioning unit is controlled based on the predicted energy consumption parameters.

[0085] The artificial bee colony algorithm of the air-conditioning unit parameter tuning device provided in the embodiment of the present disclosure performs parameter optimization, and performs optimization operations within a preset time period for the energy consumption in the air-conditioning unit parameters. The algorithm simulates the foraging behavior of bees in nature, and finds the optimal parameter combination through information exchange and collaboration between bees, thereby minimizing the energy consumption of the air-conditioning unit. Combined with the use of ParaSeer to predict the impact of parameter changes on the energy consumption of the air-conditioning unit, and using it as an evaluation basis, the most energy-saving air-conditioning unit parameter combination under different environments of the air-conditioning unit is found. The present disclosure embeds causal dependencies in the time series prediction model, uses physical modeling of the air-conditioning unit, and corrects the model prediction results, thereby realizing an efficient parameter tuning mechanism.

[0086] like Figure 6 As shown, the present application also provides an electronic device, including a display module 103, a memory 102, a processor 101, and a computer program stored in the memory and executable on the processor 101, wherein the processor 101 implements the steps of the air conditioning unit parameter tuning method when executing the program.

[0087] In embodiments of the present invention, electronic devices include, but are not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0088] In the embodiment of the present application, the processor 101 can be implemented by using at least one of an application-specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, and an electronic unit designed to perform the functions described herein. In some cases, such an implementation can be implemented in a controller. For software implementation, implementations such as processes or functions can be implemented with separate software modules that allow execution of at least one function or operation. The software code can be implemented by a software application (or program) written in any appropriate programming language, and the software code can be stored in a memory and executed by a controller.

[0089] The display module 103 is used to display the information input by the user, the process information of the air conditioning unit parameter tuning method, and the air conditioning unit parameter tuning result information. The display module 103 may include a display panel, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc.

[0090] The memory 102 may be used to store software programs and various data. The memory 102 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0091] The present application also provides a storage medium having a computer program stored thereon, wherein the computer program implements the air-conditioning unit parameter tuning method when executed by a processor, and the specific steps of storage include: retrieving the air-conditioning unit parameters; taking the energy consumption of the air-conditioning unit as the optimization target, and taking the parameters affecting the energy consumption as the decision variables; starting the optimization process by initializing the bee population, and presetting the search strategy and stopping criteria; during the optimization process, individual bees continuously update their own positions and search directions through information exchange and collaboration to find a parameter combination that achieves the preset energy consumption; configuring the ParaSeer model, and using a graphic converter based on the ParaSeer model to convert the energy consumption data of the air-conditioning unit into a form processed by the ParaSeer model, and constructing multi-dimensional time series data; using the predictor of the ParaSeer model to predict the energy consumption of the air-conditioning unit in a preset time period in the future based on the multi-dimensional time series data.

[0092] The storage medium provided by the present application can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0093] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing parameters of an air conditioning unit, characterized in that: include: Retrieve air conditioning unit parameters; The energy consumption of the air-conditioning unit is taken as the optimization target, and the parameters affecting the energy consumption are taken as the decision variables; The optimization process begins by initializing the bee population and presetting the search strategy and stopping criteria; In the optimization process, individual bees exchange information and collaborate with each other to continuously update their positions and search directions in order to find a parameter combination that achieves the preset energy consumption. Configure the ParaSeer model, and use the ParaSeer model-based graph converter to convert the energy consumption data of the air-conditioning unit into a form that can be processed by the ParaSeer model, and construct multi-dimensional time series data; The predictor of the ParaSeer model is used to predict the energy consumption of the air-conditioning unit in the future preset time period based on the multi-dimensional time series data; The air conditioning unit is controlled based on the predicted energy consumption parameters.

2. The air conditioning unit parameter tuning method according to claim 1, characterized in that: The retrieved air-conditioning unit parameters include but are not limited to: chilled water inlet and outlet water temperature of the chiller, start and stop of the chiller, electric meter reading of the chiller, cooling water inlet and outlet water temperature of the cooling tower, electric meter reading of the cooling tower, start and stop of the cooling tower fan, cooling tower fan frequency, chilled water pump operating frequency, chilled water pump start and stop, chilled water pump electric meter reading, cooling water pump operating frequency, cooling water pump start and stop, cooling water pump electric meter reading, and main water pipe inlet and outlet water flow.

3. The air conditioning unit parameter tuning method according to claim 1, characterized in that: The method also defines air conditioning unit parameter thresholds, which include: air conditioning unit host water outlet temperature threshold, cooling tower water outlet temperature threshold, cooling tower fan frequency threshold, and chilled water pump frequency threshold.

4. The air conditioning unit parameter tuning method according to claim 1, characterized in that: In the method, Find the optimal solution in air conditioning unit parameters by simulating the communication and cooperation between bees; In the artificial bee colony algorithm, bees are divided into three roles: scout bees, collecting bees, and observer bees; The scout bees randomly search for new air-conditioning unit parameters, the collecting bees search near the current air-conditioning unit parameters, and the observation bees select the better solution for search based on the air-conditioning unit parameters provided by the collecting bees.

5. The air conditioning unit parameter tuning method according to claim 1, characterized in that: In the method, the predictor adopts a variety of machine learning algorithms to learn the law of energy consumption changes by training historical air-conditioning unit parameters and predict future energy consumption.

6. The method for optimizing parameters of an air conditioning unit according to claim 1, characterized in that: In the method, the predicted energy consumption parameter is compared with the set energy consumption threshold. If the predicted energy consumption exceeds the threshold, the control logic is triggered to adjust the air conditioning unit; Adjustment measures may include, but are not limited to: adjusting temperature set points, changing wind speed, and optimizing refrigerant flow.

7. The air conditioning unit parameter tuning method according to claim 1, characterized in that: The method also includes: using the mean square error as a loss function to calculate the loss of the energy consumption parameters of the air conditioning unit in a future preset time period, and the loss function is Among them, N, S and d represent the number of samples, the length of the time series and the number of monitoring indicators respectively; when When missing, is 0, otherwise it is 1; The method also defines a projection module as a converter from latent representation to monitoring indicator time series; The projection module randomly sets part of the mask to zero and calculates the loss for the masked part based on the loss function.

8. An air conditioning unit parameter tuning device, characterized in that: The device is used to implement the air conditioning unit parameter tuning method according to any one of claims 1 to 7; The device comprises: a parameter optimization module and a parameter prediction module; The parameter optimization module is used to retrieve the parameters of the air-conditioning unit; the energy consumption of the air-conditioning unit is taken as the optimization target, and the parameters affecting the energy consumption are taken as the decision variables; the optimization process begins by initializing the bee population, and presetting the search strategy and stop criteria; during the optimization process, individual bees continuously update their positions and search directions through information exchange and collaboration to find a parameter combination that achieves the preset energy consumption; The parameter prediction module is used to configure the ParaSeer model. The ParaSeer model-based graph converter converts the energy consumption data of the air-conditioning unit into the form processed by the ParaSeer model and constructs multi-dimensional time series data. The predictor of the ParaSeer model is used to predict the energy consumption of the air-conditioning unit in the future preset time period based on the multi-dimensional time series data; The air conditioning unit is controlled based on the predicted energy consumption parameters.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the air conditioning unit parameter tuning method according to any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the air conditioning unit parameter tuning method according to any one of claims 1 to 7 are implemented.

Images