A multifunctional heat pump device and control method thereof

Abstract

The present invention relates to a multifunctional heat pump device and a control method thereof, and relates to the field of heat pump systems. The device includes a multifunctional heat pump and a heat pump control system, and the system includes: a temperature deviation acquisition module, which is used to collect control temperature, collect actual temperature, and calculate the temperature control deviation; a temperature prediction ratio configuration module, which is used to obtain power load information, and configure the temperature adjustment prediction ratio in combination with the temperature control deviation; an adjustment temperature prediction module, which is used to randomly generate heat pump control parameters, and perform temperature control adjustment prediction according to the temperature adjustment prediction ratio to obtain multiple first predicted adjustment temperatures; a fitness calculation module, which is used to calculate the first control accuracy and calculate the first heat pump control fitness; a control optimization module, which is used to continue to optimize and evaluate the heat pump control parameters to obtain the optimal heat pump control parameters. The present invention solves the technical problem that the temperature control of the heat pump system in the prior art is inaccurate and affects the building power load.

Description

Technical Field

[0001] The present invention relates to the technical field of heat pump systems, and in particular to a multifunctional heat pump device and a control method thereof. Background Art

[0002] As an efficient heating and cooling device, heat pump system has been widely used, especially in the field of building energy management. Heat pump transfers heat from low temperature environment to high temperature environment by consuming a small amount of electricity, effectively reducing the cost of building energy use. However, in the actual operation process, there are certain technical bottlenecks in the temperature control of heat pump system, which directly affects its operation stability and efficiency.

[0003] Traditional heat pump control systems rely on temperature sensors and preset control algorithms to adjust the compressor, heat exchanger, and fan according to the set temperature to meet the indoor environment requirements. However, traditional control has the problem of inaccurate temperature control, resulting in temperature overshoot or undershoot, affecting user experience, and frequent adjustment of control to improve temperature control accuracy will cause fluctuations and load impact on the building's power load. Therefore, the heat pump system in the prior art has technical problems of inaccurate temperature control and affecting the building's power load. Summary of the invention

[0004] The present invention aims at the technical problem that the temperature control of the heat pump system in the prior art is inaccurate and affects the building power load, and provides a multifunctional heat pump device and a control method thereof to solve the above technical problem.

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

[0006] In a first aspect, the present invention provides a multifunctional heat pump device, including a multifunctional heat pump and a heat pump control system, wherein the heat pump control system includes: an actual temperature, a temperature control deviation obtained by calculation;

[0007] A temperature prediction ratio configuration module, used to obtain the power load information in the building currently configured with the multifunctional heat pump, and configure the temperature adjustment prediction ratio in combination with the temperature control deviation;

[0008] An adjustment temperature prediction module is used to randomly generate a first heat pump control parameter, perform temperature control adjustment prediction according to the actual temperature according to the temperature adjustment prediction ratio, and obtain multiple first predicted adjustment temperatures;

[0009] a fitness calculation module, configured to calculate a first control accuracy rate according to the plurality of first predicted adjustment temperatures and control temperatures, and calculate a first heat pump control fitness rate in combination with the first heat pump control parameter and the power load caused by the power load information;

[0010] A control optimization module is used to continue to optimize and evaluate the heat pump control parameters based on the first heat pump control adaptability, obtain optimal heat pump control parameters, and control the multifunctional heat pump.

[0011] In a second aspect, the present invention further provides a control method for a multifunctional heat pump device, comprising:

[0012] Collect the control temperature currently set by the user, and collect the actual temperature of the multi-functional heat pump for temperature control, and calculate the temperature control deviation;

[0013] Acquire the power load information in the building currently equipped with the multifunctional heat pump, and configure the temperature adjustment prediction ratio in combination with the temperature control deviation;

[0014] Randomly generate a first heat pump control parameter, perform temperature control adjustment prediction according to the actual temperature according to the temperature adjustment prediction ratio, and obtain multiple first predicted adjustment temperatures;

[0015] Calculate a first control accuracy rate based on the multiple first predicted adjustment temperatures and control temperatures, and calculate a first heat pump control adaptability based on the first heat pump control parameter and the power load caused by the power load information;

[0016] Based on the first heat pump control adaptability, the heat pump control parameters are continuously optimized and evaluated to obtain the optimal heat pump control parameters, and the multifunctional heat pump is controlled.

[0017] The beneficial effects of the present invention are as follows: through the heat pump control system, the technical problems such as temperature control error and large power load influence existing in the traditional heat pump control are solved, and the stability and accuracy of the heat pump system operation are significantly improved. First, the control temperature and actual temperature set by the user are collected in real time, and the temperature control deviation is accurately calculated to provide basic data support for subsequent temperature adjustment, reducing the error accumulation caused by insufficient sensor accuracy. Secondly, combined with the building's power load information and temperature control deviation, the temperature adjustment prediction ratio is dynamically configured to achieve adaptive adjustment of the temperature adjustment strategy according to the power load status, effectively improving the adaptability, accuracy and efficiency of the temperature control adjustment. Thirdly, by randomly generating heat pump control parameters and combining the temperature adjustment prediction ratio to predict the temperature adjustment, multiple candidate predicted adjustment temperatures are obtained, the control accuracy is calculated based on the predicted adjustment temperature and the actual control temperature, and the fitness evaluation is performed in combination with the power load information, the heat pump control parameters are optimized from multiple dimensions to ensure the high efficiency and low energy consumption of the heat pump operation. Finally, the control optimization module iteratively optimizes the heat pump control fitness to obtain the optimal heat pump control parameters, thereby achieving precise control of the multifunctional heat pump, significantly improving the accuracy of temperature control and the stability of the building power system, and avoiding inaccurate temperature control caused by repeated adjustments, which affects the stability of the building power load. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic structural diagram of a multifunctional heat pump device provided by the present invention;

[0019] Figure 2 A schematic flow chart of a control method for a multifunctional heat pump device provided by the present invention.

[0020] In the accompanying drawings, the components represented by the reference numerals are described as follows:

[0021] Temperature deviation acquisition module 01, temperature prediction ratio configuration module 02, temperature adjustment prediction module 03, fitness calculation module 04, control optimization module 05. DETAILED DESCRIPTION

[0022] 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0023] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0024] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0025] Embodiment 1, as Figure 1As shown, an embodiment of the present invention provides a multifunctional heat pump device, including a multifunctional heat pump and a heat pump control system. The multifunctional heat pump is connected to the heat pump control system and performs temperature adjustment under the control of the heat pump control system. For example, the operating parameters of the compressor, heat exchanger and fan are adjusted under the control of the heat pump control system, thereby adjusting the ambient temperature.

[0026] The multifunctional heat pump is a heat pump with multiple functions in the prior art, such as a heat pump that can perform indoor cooling, indoor heating, and hot water, and can monitor and display temperature, and control the cooling and heating temperatures by remote control. The heat pump may also have any existing functions of other heat pumps in the art. In addition, the heat pump also has the control function provided by the heat pump control system in the embodiment of the present application.

[0027] The heat pump control system comprises: a temperature deviation acquisition module 01, a temperature prediction ratio configuration module 02, an adjustment temperature prediction module 03, a fitness calculation module 04 and a control optimization module 05, which are respectively used to execute different steps to perform heat pump temperature regulation control.

[0028] In the embodiment of the present application, the temperature deviation acquisition module 01 is used to collect the control temperature currently set by the user, and collect the actual temperature of the temperature control performed by the multifunctional heat pump, and calculate the temperature control deviation.

[0029] Specifically, in the control process of the multifunctional heat pump, in order to accurately achieve temperature regulation, it is first necessary to collect the control temperature currently set by the user and the actual temperature during the actual operation of the multifunctional heat pump, and calculate the difference between the two to obtain the temperature control deviation, thereby providing basic data support for the subsequent adjustment process.

[0030] Furthermore, the temperature deviation acquisition module 01 is used to:

[0031] Collect the control temperature currently set by the user;

[0032] Collecting the detected temperatures at multiple locations in the building currently equipped with the multifunctional heat pump, and calculating the average to obtain the actual temperature;

[0033] The temperature deviation ratio between the control temperature and the actual temperature is calculated to obtain the temperature control deviation.

[0034] In the embodiment of the present application, it is first necessary to collect the control temperature currently set by the user, which is the temperature set by the user for the heat pump to control the ambient temperature, and is the indoor ambient temperature desired by the user (e.g., the target temperature for heating or cooling). For example, the temperature manually set by the user in the heat pump system or the temperature set by the remote control is, for example, 20°C.

[0035] Furthermore, the temperature of multiple locations in a building is collected and the actual temperature is calculated. The building is a building currently equipped with a multifunctional heat pump for environmental temperature control, such as a residential building. Temperature sensors, such as thermocouples or RTD sensors, are arranged at multiple locations in the building to collect the temperature of multiple locations and obtain multiple monitoring temperatures. The multiple locations are, for example, multiple rooms in the room.

[0036] The average of the detected temperatures at multiple locations is calculated to obtain the actual temperature of the room where the current heat pump is performing ambient temperature control.

[0037] The temperature deviation ratio between the control temperature and the actual temperature is calculated as the temperature control deviation. The temperature deviation ratio refers to the relative difference between the actual temperature and the control temperature, which is used to quantitatively evaluate the degree of deviation between the current ambient temperature and the target temperature. Exemplarily, the difference between the control temperature and the actual temperature is calculated, such as the absolute value of the temperature difference between the control temperature and the actual temperature, and then the ratio of the absolute value of the temperature difference to the control temperature is calculated to obtain the temperature control deviation. Temperature deviation ratio = |actual temperature - control temperature| / control temperature × 100%.

[0038] Through the above steps, the multifunctional heat pump system can perceive user needs and environmental changes in real time, accurately calculate temperature control deviations, and provide data support for the next step of temperature regulation and energy consumption optimization.

[0039] The temperature prediction ratio configuration module 02 is used to obtain the power load information in the building where the multifunctional heat pump is currently configured, and configure the temperature adjustment prediction ratio in combination with the temperature control deviation;

[0040] In the embodiment of the present application, during the operation of the multifunctional heat pump system, in order to maintain the stability of the power system while adjusting the temperature, the power load information in the building is obtained, and the accuracy of the temperature control adjustment prediction is dynamically configured in combination with the temperature control deviation as the temperature adjustment prediction ratio. This process ensures the balance between temperature control efficiency and power stability through load analysis and deviation adjustment mechanism.

[0041] Among them, the greater the deviation of the current temperature control, the more precise the temperature control is needed to improve the user experience, and the larger the temperature adjustment prediction ratio is configured. And if the current building's power load is greater, the greater the impact on the stability of the power system when controlling the heat pump to adjust the temperature, the larger the temperature adjustment prediction ratio needs to be configured to improve the accuracy of temperature control analysis and reduce the impact on the stability of the building's power system.

[0042] In the embodiment of the present application, the temperature prediction ratio configuration module 02 is specifically used for:

[0043] Obtaining the rated power supply power of the building currently equipped with the multifunctional heat pump and the actual power supply load currently required by the building;

[0044] Calculate the load demand coefficient based on the actual power supply load and the rated power supply power;

[0045] Using the load demand factor as a first prediction proportionality factor;

[0046] Calculate and obtain a second prediction proportionality coefficient according to the temperature control deviation, wherein the magnitude of the temperature control deviation is negatively correlated with the magnitude of the second prediction proportionality coefficient;

[0047] The temperature adjustment prediction ratio is calculated based on the first prediction ratio coefficient and the second prediction ratio coefficient.

[0048] In an embodiment of the present application, the rated power supply power and the actual power supply load of the building are obtained in real time through power monitoring. The rated power supply power refers to the designed maximum power supply capacity of the building power grid, which is the upper limit of the power supply of the building. The actual power supply load reflects the real-time total power demand of all electrical equipment in the current building. These data are collected by intelligent power monitoring equipment (such as a multi-function meter or load sensor) and uploaded to the heat pump control system. For example, the rated power supply power is 100kW, and the actual power supply load monitored by the meter is 80kW.

[0049] Furthermore, the load demand coefficient is calculated based on the actual power supply load demand and the rated power supply power. Specifically, the ratio of the actual power supply load demand to the rated power supply power is calculated as the load demand coefficient, such as 80kW / 100kW=0.8.

[0050] The load demand factor quantifies the current electricity load pressure of the building. The closer the value is to 1, the closer the building's electricity demand is to the grid's carrying capacity, and careful adjustment is required to avoid additional load. A smaller value means lower grid pressure, which allows more possible heat pump adjustments.

[0051] The load demand coefficient is used as the first prediction proportional coefficient. The larger the load demand coefficient, the greater the impact on the stability of the power system when controlling the heat pump to adjust the temperature. A larger temperature adjustment prediction ratio needs to be configured to improve the accuracy of the temperature control adjustment prediction, thereby improving the accuracy of the heat pump temperature control and reducing the impact on the building power system.

[0052] Furthermore, a second prediction ratio coefficient is calculated based on the temperature control deviation. The larger the temperature control deviation, the larger the temperature adjustment prediction ratio needs to be configured to improve the accuracy of temperature control analysis. For example, the maximum historical temperature control deviation of controlling the heat pump to adjust the ambient temperature in the historical time is collected, that is, the maximum difference between the control temperature and the ambient temperature in the historical time, that is, the maximum deviation of temperature control, such as 10°C.

[0053] Furthermore, the ratio of the current temperature control deviation to the maximum historical temperature control deviation is calculated as the second prediction proportional coefficient. For example, if the current temperature control deviation is 5° C., the second prediction proportional coefficient is 5° C. / 10° C., which is 0.5.

[0054] According to the first prediction ratio coefficient and the second prediction ratio coefficient, the temperature adjustment prediction ratio of the current temperature adjustment prediction is calculated. For example, the ratio of the first prediction ratio coefficient and the second prediction ratio coefficient is calculated as the temperature adjustment prediction ratio. Optionally, the first prediction ratio coefficient and the second prediction ratio coefficient may be weighted, such as the temperature adjustment prediction ratio = α * first prediction ratio coefficient + β * second prediction ratio coefficient, α and β are weights, for example, 0.6 and 0.4 respectively.

[0055] In this way, based on the current temperature control deviation and the building's power system load, the temperature adjustment prediction ratio is calculated and generated, so that the temperature adjustment strategy can not only meet the user's comfort needs, but also avoid causing additional pressure on the building's power grid, thereby improving the accuracy of the heat pump temperature control adjustment, thereby achieving efficient and stable operation of the heat pump system.

[0056] The temperature adjustment prediction module 03 is used to randomly generate a first heat pump control parameter, adjust the temperature according to the predicted ratio, and perform temperature control adjustment prediction according to the actual temperature to obtain a plurality of first predicted adjustment temperatures.

[0057] In the embodiment of the present application, in the temperature adjustment process of the multifunctional heat pump, in order to achieve precise control, the first heat pump control parameter is randomly generated, combined with the temperature adjustment prediction ratio, and the temperature control adjustment prediction is performed according to the current actual temperature, thereby obtaining multiple first predicted adjustment temperatures. The larger the temperature adjustment prediction ratio, the more the number of the multiple first predicted adjustment temperatures obtained by prediction, and the higher the accuracy of the temperature adjustment control prediction analysis.

[0058] In the embodiment of the present application, the adjustment temperature prediction module 03 is specifically used for:

[0059] Acquire a heat pump control parameter space of the multifunctional heat pump;

[0060] randomly generating a first heat pump control parameter in the heat pump control parameter space;

[0061] Using ensemble learning to train a temperature adjustment prediction channel, the temperature adjustment prediction channel includes a temperature adjustment prediction path array, wherein the number of temperature adjustment prediction paths in the temperature adjustment prediction path array is K, and K is a positive integer;

[0062] Calculate and round up the predicted temperature adjustment ratio and K to obtain J, where J is a positive integer;

[0063] J temperature adjustment prediction paths are randomly selected in the temperature adjustment prediction channel, and temperature control adjustment prediction is performed on the first heat pump control parameter and the actual temperature to obtain J first predicted adjustment temperatures.

[0064] In an embodiment of the present application, in the intelligent temperature regulation process of a multifunctional heat pump, in order to implement an optimized control strategy, multiple first predicted adjustment temperatures are generated by optimizing the heat pump control parameter space, constructing an integrated learning model, and selecting a prediction path to support subsequent regulation optimization.

[0065] First, the heat pump control parameter space is obtained, which includes the combination of heat pump parameters that can be adjusted within the safety range, such as compressor frequency, heat exchanger temperature, fan speed and refrigerant flow. The heat pump control parameter space includes all adjustable heat pump control parameters, and each heat pump control parameter includes a random combination of compressor frequency, heat exchanger temperature, fan speed and refrigerant flow. The compressor frequency ranges from 30 Hz to 90 Hz, the heat exchanger temperature ranges from 5°C to 40°C, the fan speed ranges from 100 RPM to 2000 RPM, and the refrigerant flow ranges from 0.5 L / min to 5 L / min. The compressor frequency, heat exchanger temperature, fan speed and refrigerant flow are randomly combined within the above range to obtain all possible heat pump control parameters in the heat pump control parameter space.

[0066] A first heat pump control parameter is randomly generated in the heat pump control parameter space, for example, a random heat pump control parameter is generated by a random number generation algorithm. For example, the first heat pump control parameter may be a compressor frequency: 50 Hz, a heat exchanger temperature: 25°C, a fan speed: 1500 RPM, and a refrigerant flow rate: 2.5 L / min.

[0067] Further, based on the ensemble learning algorithm, a temperature adjustment prediction channel is trained, and the temperature adjustment prediction channel includes a temperature adjustment prediction path array, wherein the number of temperature adjustment prediction paths in the temperature adjustment prediction path array is K, and K is a positive integer, for example, 20. The K temperature adjustment prediction paths in the temperature adjustment prediction path array are different from each other, for example, the training data is different, and the K temperature adjustment prediction paths can respectively perform temperature control adjustment prediction according to the heat pump control parameters and the current actual temperature, and obtain different predicted adjustment temperatures after the heat pump is controlled for temperature adjustment.

[0068] The larger the temperature adjustment prediction ratio is, the more temperature adjustment prediction paths are selected for temperature adjustment prediction, so as to improve the accuracy of the heat pump temperature adjustment control.

[0069] In the embodiment of the present application, integrated learning is used to train the temperature adjustment prediction channel, which specifically includes the following steps:

[0070] In the heat pump operation log in the historical time, a sample heat pump control parameter set and a sample actual temperature set are collected, and the temperature after control is collected as a sample adjustment temperature set;

[0071] Integrate the sample heat pump control parameter set, the sample actual temperature set and the sample adjusted temperature set, and randomly select K sets of temperature regulation prediction supervision training data with replacement, wherein the amount of data in each set of temperature regulation prediction supervision training data is the same;

[0072] Using the K pieces of temperature adjustment prediction supervision training data, based on machine learning, training to obtain K temperature adjustment prediction paths;

[0073] The K temperature adjustment prediction paths are combined to obtain a temperature adjustment prediction path array as a temperature adjustment prediction channel.

[0074] In the embodiment of the present application, key data is extracted from the historical heat pump operation log, specifically the heat pump control parameters recorded in the previous heat pump control of the building ambient temperature and the actual temperature in the building before control are collected to obtain a sample heat pump control parameter set and a sample actual temperature set. Further, the adjusted temperature reached after adjusting the ambient temperature according to different sample heat pump control parameters at different sample actual temperatures is collected to obtain a sample adjusted temperature set.

[0075] The sample heat pump control parameter set, sample actual temperature set and sample adjusted temperature set constitute the basis of model training and reflect the correlation between heat pump control parameters and ambient temperature changes.

[0076] The sample heat pump control parameter set, the sample actual temperature set and the sample adjusted temperature set are integrated into a complete supervised training sample, and each training data sample contains a sample control parameter, a sample actual temperature and a sample adjusted temperature.

[0077] In order to improve the generalization ability of the temperature regulation prediction path array based on ensemble learning training, a random sampling method with replacement is used to randomly select K pieces of temperature regulation prediction supervision training data from the integrated complete supervision training samples. The amount of data in each piece of temperature regulation prediction supervision training data is the same, for example, the amount of data in each piece of temperature regulation prediction supervision training data is 2 / 3 of the amount of data in the complete supervision training sample. If the complete supervision training sample includes 6000 training data samples, each piece of temperature regulation prediction supervision training data includes 4000 training data samples.

[0078] Among them, random sampling with replacement includes adding the selected data back into the data pool after each sampling, ensuring that the sample may be selected multiple times, improving the reuse of small samples, making the K temperature adjustment prediction paths trained based on K pieces of temperature adjustment prediction supervised training data different from each other, and ensuring the performance of each temperature adjustment prediction path.

[0079] Further, K portions of temperature adjustment prediction supervised training data are used respectively, and K temperature adjustment prediction paths are obtained through training based on machine learning. For example, a feedforward neural network is used to construct an architecture of K temperature adjustment prediction paths, and then K portions of temperature adjustment prediction supervised training data are used to perform supervised training, verification, and testing on the K temperature adjustment prediction paths respectively until the requirements are met.

[0080] Exemplarily, each temperature adjustment prediction path includes an input layer, a hidden layer, and an output layer. The hidden layer design includes 3 layers, each layer includes multiple neurons, and a nonlinear activation function such as a ReLU function is used. During the training process, the sample heat pump control parameters and the actual sample temperature in the training data sample are input into the temperature adjustment prediction path to obtain the adjusted temperature outputted by the sample. The loss is calculated using the mean square error loss function according to the adjusted temperature and the actual sample adjustment temperature. The weight parameters of each neuron in the hidden layer are updated through the Adam optimizer to reduce the loss, for example, to less than 5%. Iterative training is performed in this way until the accuracy of the temperature adjustment prediction path meets the requirements, such as greater than or equal to 95%. After the training is completed, 10% of the validation set and test set can be divided in the temperature adjustment prediction supervision training data for verification and testing. If the accuracy meets the requirements, the training ends. If not, the training tuning continues.

[0081] After the training is completed, K temperature regulation prediction channel paths are generated to form a prediction path array as the temperature regulation prediction channel.

[0082] Since heat pump control has errors, based on ensemble learning training of K temperature adjustment prediction channel paths, the heat pump control temperature adjustment prediction can be performed separately to obtain different adjustment temperatures that may be achieved due to errors under control of the same heat pump control parameters. In this way, the error of heat pump control temperature can be considered to analyze and optimize the heat pump control, improve user experience, and reduce the impact on the building power system.

[0083] By constructing a temperature adjustment prediction path array through the above steps, flexible multi-path parallel prediction is realized, the accuracy and robustness of temperature adjustment are enhanced, and the demand for precise control of heat pumps in complex environments is met. In addition, multiple temperature adjustment prediction paths can be combined with different temperature control deviations and meet the required parameters, and different numbers of temperature adjustment prediction paths can be selected for temperature adjustment prediction, thereby improving the accuracy and efficiency of temperature control adjustment prediction under different conditions.

[0084] Further, calculation is performed based on the temperature adjustment prediction ratio and K and rounded up to obtain J, where J is a positive integer and is less than or equal to K. Exemplarily, J is obtained by multiplying the temperature adjustment prediction ratio by K and rounding up, for example, K is 20 and J is 10.

[0085] J temperature adjustment prediction paths are randomly selected in the temperature adjustment prediction channel, and temperature control adjustment prediction is performed on the first heat pump control parameter and the actual temperature. Specifically, the first heat pump control parameter and the actual temperature are respectively input into the J temperature adjustment prediction paths, and heat pump temperature adjustment control prediction is performed to obtain J first predicted adjustment temperatures after the predicted temperature control adjustment.

[0086] In this way, when the temperature control deviation and the load demand coefficient are large, more temperature adjustment prediction paths are used to perform temperature control adjustment prediction, thereby improving the accuracy of temperature control adjustment prediction, thereby improving user experience, and improving the accuracy of heat pump temperature control optimization, reducing the impact on the power system.

[0087] The fitness calculation module 04 is used to calculate the first control accuracy rate according to the multiple first predicted adjustment temperatures and control temperatures, and calculate the first heat pump control fitness rate in combination with the first heat pump control parameter and the power load caused by the power load information.

[0088] Specifically, the multiple first predicted adjustment temperatures predicted for temperature control adjustment based on the first heat pump control parameter and the actual temperature, i.e., J first predicted adjustment temperatures, reflect multiple adjustment temperatures that may be achieved under the control of the first heat pump control parameter. Based on the multiple first predicted adjustment temperatures and the control temperature, a first control accuracy rate of temperature regulation control under the control of the first heat pump control parameter is calculated.

[0089] And according to the first heat pump control parameter and the power load information, the power load of the building caused by the heat pump temperature adjustment control according to the first heat pump control parameter is analyzed, and the heat pump control adaptability under the heat pump temperature control by the first heat pump control parameter is calculated.

[0090] By considering the temperature control accuracy and power load and calculating the heat pump control adaptability of the heat pump control parameters, the accuracy of the heat pump temperature control can be improved, and the impact on the power system can be analyzed.

[0091] In the embodiment of the present application, the fitness calculation module 04 is specifically used for:

[0092] Calculating temperature deviations between the plurality of first predicted adjusted temperatures and the control temperature to obtain a plurality of first temperature deviations;

[0093] Filter the proportion of the first temperature deviation that is less than or equal to the temperature deviation threshold as the first control accuracy rate;

[0094] Performing power load forecasting according to the first heat pump control parameter and power load information to obtain first forecasted power load information;

[0095] According to the temperature control deviation, the preset temperature weight is corrected and calculated to obtain the temperature weight and the load weight, and the first heat pump control adaptability is calculated and obtained in combination with the first control accuracy and the first predicted power load information, as shown in the following formula:

[0096] ;

[0097] in, is the heat pump control adaptability, TK is the temperature control deviation, is the preset temperature weight, CK is the control accuracy, and DK is the predicted power load information.

[0098] In the embodiment of the present application, the temperature deviations of the multiple first predicted adjustment temperatures and the control temperature are first calculated in the same way as the temperature control deviation, for example, the absolute value of the difference between each first predicted adjustment temperature and the control temperature is calculated, for example, 2°C, 1°C, etc. The first temperature deviation reflects the deviation between the first predicted adjustment temperature that may be achieved under the heat pump temperature adjustment control according to the first heat pump control parameter and the control temperature actually set by the user.

[0099] Furthermore, the proportion of the first temperature deviation that is less than or equal to the temperature deviation threshold is screened, and the temperature deviation threshold is the deviation threshold between the ambient temperature actually controlled by the heat pump and the control temperature that the user expects to set. For example, it is 1°C, and the number of first temperature deviations that are less than or equal to the temperature deviation threshold can be pre-set or set by the user, and then the ratio of this number to the number of multiple first predicted adjustment temperatures (that is, J) is calculated as the first control accuracy. The first control accuracy reflects the probability of reaching the temperature deviation threshold within the control temperature required by the user under the temperature adjustment control of the heat pump according to the first heat pump control parameter, that is, the accuracy of the current temperature control of the heat pump. The more temperature adjustment prediction paths are selected to predict the adjustment temperature (that is, the larger J is), the more are obtained, and the higher the accuracy of the first control accuracy.

[0100] Furthermore, the influence of the heat pump temperature adjustment control according to the first heat pump control parameter on the building power system is predicted, specifically the power load caused by the adjustment control temperature. The power load is predicted according to the first heat pump control parameter and the power load information, and the first predicted power load information of the building power system under the heat pump temperature adjustment control according to the first heat pump control parameter is obtained.

[0101] In the embodiment of the present application, according to the first heat pump control parameter and the power load information, power load forecasting is performed to obtain first predicted power load information, including:

[0102] According to the power operation monitoring log of the building, a sample heat pump control parameter set and a sample power load information set are collected, and the power load information after the heat pump is controlled under different heat pump control parameters and sample power load information is collected, and the sample power load information set is marked;

[0103] The sample heat pump control parameter set, the sample power load information set and the sample power load information set are used as supervised training data, and machine learning is used to train and obtain a power load prediction channel;

[0104] The first heat pump control parameter and power load information are input into the power load prediction channel, and the prediction output is used to obtain the first predicted power load information.

[0105] In the embodiment of the present application, based on the building's power operation monitoring log, such as data monitored by an electric meter, and the operation data of a heat pump, sample heat pump control parameters when the heat pump temperature adjustment control was previously performed and sample power load information before control, such as 80%, are collected. A sample heat pump control parameter set and a sample power load information set are obtained as input data.

[0106] Furthermore, the power load after the heat pump temperature is adjusted and controlled according to different heat pump control parameters under different sample power load information, that is, the power load demand coefficient, such as 85%, is collected to obtain a sample power load information set as output supervision data.

[0107] Then, the sample heat pump control parameter set, the sample power load information set and the sample power load information set are used as supervised training data, and machine learning is used to train and obtain the power load prediction channel. Exemplarily, the training process of the power load prediction channel is the same as the training process of the temperature adjustment prediction path in the above content. The power load prediction channel is constructed based on a feedforward neural network. During the training process, the sample heat pump control parameters and sample power load information in the supervised training data are input into the power load prediction channel to obtain the power load information output by it. According to the power load information and the real sample power load information, the loss is calculated using the mean square error loss function, and the weight parameters of each neuron in the hidden layer are updated through the Adam optimizer to reduce the loss, for example, less than 5%. Iterative training is performed in this way until the accuracy of the power load prediction channel meets the requirements, such as greater than or equal to 90%. After the training is completed, 10% of the validation set and test set can be divided in the supervised training data for verification and testing. If the accuracy meets the requirements, the training ends. If not, the training tuning continues.

[0108] Based on the trained power load prediction channel, the current first heat pump control parameters and power load information are input into the power load prediction channel for prediction processing and output, so as to obtain the first predicted power load information under which the first heat pump control parameters are used to control the heat pump for temperature adjustment, that is, the power load demand coefficient that may be reached in the building after control.

[0109] By predicting the first predicted power load information under the temperature adjustment control of the heat pump through machine learning, the accuracy and efficiency of the power load impact analysis can be improved. The evaluation and optimization of the heat pump control parameters can be carried out in combination with the impact of the power system to improve the accuracy of the heat pump control and the energy-saving effect.

[0110] Furthermore, according to the temperature control deviation, the preset temperature weight is corrected and calculated to obtain the temperature weight, and the load weight is obtained by subtracting the temperature weight from 1. Among them, the temperature control deviation reflects the deviation between the actual temperature of the current environment and the control temperature. The larger the deviation is, the greater the importance of improving the accuracy of temperature control in the current process of controlling the heat pump to adjust the temperature, and the greater the temperature weight. The preset temperature weight is a pre-set temperature weight that reflects the importance of temperature control accuracy in heat pump control, such as 0.5. The load weight reflects the importance of reducing the power load, that is, reducing the impact on the power system, in the process of controlling the heat pump to adjust the temperature. Currently, the heat pump control optimization is carried out from the two dimensions of improving the accuracy of temperature control and reducing the impact on the power system. The sum of the temperature weight and the load weight is 1.

[0111] Specifically, the sum of 1 and the temperature control deviation is used as the temperature weight adjustment coefficient, multiplied by the preset temperature weight to obtain the temperature weight, and 1-temperature weight is the load weight. If the temperature control deviation is 5% and the preset temperature weight is 0.5, then the temperature weight = 1.05*0.5 = 0.525, and the load weight = 0.475.

[0112] Further, according to the temperature weight and the load weight, and in combination with the first control accuracy and the first predicted power load information, the fitness under the heat pump temperature adjustment control according to the first heat pump control parameter is calculated to obtain the first heat pump control fitness, as shown in the following formula:

[0113] ;

[0114] in, is the heat pump control adaptability, TK is the temperature control deviation, is the preset temperature weight, CK is the control accuracy, and DK is the predicted power load information.

[0115] In this way, the first heat pump control parameters are optimized and evaluated from two dimensions: improving temperature control accuracy and reducing the impact on the power system. The first heat pump control adaptability reflects the pros and cons of the first heat pump control parameters for heat pump temperature adjustment control, and uses integrated prediction to adjust the temperature, thereby improving the accuracy and precision of the heat pump control parameter control analysis.

[0116] The control optimization module 05 is used to continue to optimize and evaluate the heat pump control parameters based on the first heat pump control adaptability, obtain the optimal heat pump control parameters, and control the multifunctional heat pump.

[0117] In an embodiment of the present application, based on the first heat pump control fitness, the heat pump control parameters continue to be optimized and evaluated. Specifically, the second heat pump control parameters continue to be randomly generated in the heat pump control parameter space, and the second heat pump control fitness is evaluated and calculated based on the adjustment temperature prediction module 03 and the fitness calculation module 04. The heat pump control parameters are iteratively optimized until convergence. For example, when a preset number of iterative optimizations is reached, such as 100 times, the optimization ends and the heat pump control parameters with the largest heat pump control fitness are output as the optimal heat pump control parameters. The temperature of the multifunctional heat pump is adjusted and controlled to maximize the temperature control accuracy and reduce the impact on the building power system.

[0118] Optionally, an optimization algorithm such as a genetic algorithm may be used to optimize the heat pump control parameters within the heat pump control parameter space, and the heat pump control fitness evaluation calculation of the heat pump control parameter population is performed based on the method in the aforementioned content to obtain the optimal heat pump control parameters.

[0119] A multifunctional heat pump device and a control method thereof provided by an embodiment of the present invention have at least the following technical effects:

[0120] The embodiment of the present application solves the technical problems such as temperature control error and large power load influence in traditional heat pump control through the heat pump control system, and significantly improves the stability and accuracy of the heat pump system operation. First, the control temperature and actual temperature set by the user are collected in real time, and the temperature control deviation is accurately calculated to provide basic data support for subsequent temperature adjustment, reducing the error accumulation caused by insufficient sensor accuracy. Secondly, combined with the building's power load information and temperature control deviation, the temperature adjustment prediction ratio is dynamically configured to achieve adaptive adjustment of the temperature adjustment strategy according to the power load status, effectively improving the adaptability, accuracy and efficiency of the temperature control adjustment. Thirdly, by randomly generating heat pump control parameters and combining the temperature adjustment prediction ratio to predict the temperature adjustment, multiple candidate predicted adjustment temperatures are obtained, the control accuracy is calculated based on the predicted adjustment temperature and the actual control temperature, and the fitness evaluation is performed in combination with the power load information, the heat pump control parameters are optimized from multiple dimensions to ensure the high efficiency and low energy consumption of the heat pump operation. Finally, the control optimization module iteratively optimizes the heat pump control fitness to obtain the optimal heat pump control parameters, thereby achieving precise control of the multifunctional heat pump, significantly improving the accuracy of temperature control and the stability of the building power system, and avoiding inaccurate temperature control caused by repeated adjustments, which affects the stability of the building power load.

[0121] Embodiment 2, as Figure 2 As shown, based on the same concept as the multifunctional heat pump device in the first embodiment, the embodiment of the present invention further provides a control method of the multifunctional heat pump device, including:

[0122] Collect the control temperature currently set by the user, and collect the actual temperature of the multi-functional heat pump for temperature control, and calculate the temperature control deviation;

[0123] Acquire the power load information in the building currently equipped with the multifunctional heat pump, and configure the temperature adjustment prediction ratio in combination with the temperature control deviation;

[0124] Randomly generate a first heat pump control parameter, perform temperature control adjustment prediction according to the actual temperature according to the temperature adjustment prediction ratio, and obtain multiple first predicted adjustment temperatures;

[0125] Calculate a first control accuracy rate based on the multiple first predicted adjustment temperatures and control temperatures, and calculate a first heat pump control adaptability based on the first heat pump control parameter and the power load caused by the power load information;

[0126] Based on the first heat pump control adaptability, the heat pump control parameters are continuously optimized and evaluated to obtain the optimal heat pump control parameters, and the multifunctional heat pump is controlled.

[0127] Among them, the control temperature currently set by the user is collected, and the actual temperature of the temperature control performed by the multifunctional heat pump is collected, and the temperature control deviation is calculated, including:

[0128] Collect the control temperature currently set by the user;

[0129] Collecting the detected temperatures at multiple locations in the building currently equipped with the multifunctional heat pump, and calculating the average to obtain the actual temperature;

[0130] The temperature deviation ratio between the control temperature and the actual temperature is calculated to obtain the temperature control deviation.

[0131] The step of obtaining the power load information in the building currently configured with the multifunctional heat pump and configuring the temperature adjustment prediction ratio in combination with the temperature control deviation includes:

[0132] Obtaining the rated power supply power of the building currently equipped with the multifunctional heat pump and the actual power supply load currently required by the building;

[0133] Calculate the load demand coefficient based on the actual power supply load and the rated power supply power;

[0134] Using the load demand factor as a first prediction proportionality factor;

[0135] Calculating and obtaining a second prediction proportionality coefficient according to the temperature control deviation;

[0136] The temperature adjustment prediction ratio is calculated based on the first prediction ratio coefficient and the second prediction ratio coefficient.

[0137] The first heat pump control parameter is randomly generated, and the temperature control adjustment prediction is performed according to the actual temperature according to the temperature adjustment prediction ratio to obtain multiple first predicted adjustment temperatures, including:

[0138] Acquire a heat pump control parameter space of the multifunctional heat pump;

[0139] randomly generating a first heat pump control parameter in the heat pump control parameter space;

[0140] Using ensemble learning to train a temperature adjustment prediction channel, the temperature adjustment prediction channel includes a temperature adjustment prediction path array, wherein the number of temperature adjustment prediction paths in the temperature adjustment prediction path array is K, and K is a positive integer;

[0141] Calculate and round up the predicted temperature adjustment ratio and K to obtain J, where J is a positive integer;

[0142] J temperature adjustment prediction paths are randomly selected in the temperature adjustment prediction channel, and temperature control adjustment prediction is performed on the first heat pump control parameter and the actual temperature to obtain J first predicted adjustment temperatures.

[0143] The temperature adjustment prediction channel is trained by using integrated learning, and the temperature adjustment prediction channel includes a temperature adjustment prediction path array, including:

[0144] In the heat pump operation log in the historical time, a sample heat pump control parameter set and a sample actual temperature set are collected, and the temperature after control is collected as a sample adjustment temperature set;

[0145] Integrate the sample heat pump control parameter set, the sample actual temperature set and the sample adjusted temperature set, and randomly select K sets of temperature regulation prediction supervision training data with replacement, wherein the amount of data in each set of temperature regulation prediction supervision training data is the same;

[0146] Using the K pieces of temperature adjustment prediction supervision training data, based on machine learning, training to obtain K temperature adjustment prediction paths;

[0147] The K temperature adjustment prediction paths are combined to obtain a temperature adjustment prediction path array as a temperature adjustment prediction channel.

[0148] The first control accuracy is calculated based on the multiple first predicted adjustment temperatures and control temperatures, and the first heat pump control adaptability is calculated based on the power load caused by the first heat pump control parameter and the power load information, including:

[0149] Calculating temperature deviations between the plurality of first predicted adjusted temperatures and the control temperature to obtain a plurality of first temperature deviations;

[0150] Filter the proportion of the first temperature deviation that is less than or equal to the temperature deviation threshold as the first control accuracy rate;

[0151] Performing power load forecasting according to the first heat pump control parameter and power load information to obtain first forecasted power load information;

[0152] According to the temperature control deviation, the preset temperature weight is corrected and calculated to obtain the temperature weight and the load weight, and the first heat pump control adaptability is calculated and obtained in combination with the first control accuracy and the first predicted power load information, as shown in the following formula:

[0153] ;

[0154] in, is the heat pump control adaptability, TK is the temperature control deviation, is the preset temperature weight, CK is the control accuracy, and DK is the predicted power load information.

[0155] According to the first heat pump control parameter and the power load information, power load forecasting is performed to obtain first forecast power load information, including:

[0156] According to the power operation monitoring log of the building, a sample heat pump control parameter set and a sample power load information set are collected, and the power load information after the heat pump is controlled under different heat pump control parameters and sample power load information is collected, and the sample power load information set is marked;

[0157] The sample heat pump control parameter set, the sample power load information set and the sample power load information set are used as supervised training data, and machine learning is used to train and obtain a power load prediction channel;

[0158] The first heat pump control parameter and power load information are input into the power load prediction channel, and the prediction output is used to obtain the first predicted power load information.

[0159] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0160] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0162] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0164] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0165] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A multifunctional heat pump device, characterized in that: It includes a multifunctional heat pump and a heat pump control system, wherein the heat pump control system includes: The temperature deviation acquisition module is used to collect the control temperature currently set by the user, collect the actual temperature of the temperature control performed by the multifunctional heat pump, and calculate the temperature control deviation; The temperature prediction ratio configuration module is used to obtain the power load information in the building currently configured with the multifunctional heat pump, and configure the temperature adjustment prediction ratio in combination with the temperature control deviation, including: Obtaining the rated power supply power of the building currently equipped with the multifunctional heat pump and the actual power supply load currently required by the building; Calculate the load demand coefficient based on the actual power supply load and the rated power supply power; Using the load demand factor as a first prediction proportionality factor; Calculating and obtaining a second prediction proportionality coefficient according to the temperature control deviation; Calculating a temperature adjustment prediction ratio according to the first prediction ratio coefficient and the second prediction ratio coefficient; An adjustment temperature prediction module is used to randomly generate a first heat pump control parameter, perform temperature control adjustment prediction according to the actual temperature according to the temperature adjustment prediction ratio, and obtain multiple first predicted adjustment temperatures; A fitness calculation module, used to calculate a first control accuracy rate according to the multiple first predicted adjustment temperatures and control temperatures, and to calculate a first heat pump control fitness rate in combination with the first heat pump control parameter and the power load caused by the power load information, including: Calculating temperature deviations between the plurality of first predicted adjusted temperatures and the control temperature to obtain a plurality of first temperature deviations; Filter the proportion of the first temperature deviation that is less than or equal to the temperature deviation threshold as the first control accuracy rate; Performing power load forecasting according to the first heat pump control parameter and power load information to obtain first forecasted power load information; According to the temperature control deviation, the preset temperature weight is corrected and calculated to obtain the temperature weight and the load weight, and the first heat pump control adaptability is calculated and obtained in combination with the first control accuracy and the first predicted power load information, as shown in the following formula: ； in, is the heat pump control adaptability, TK is the temperature control deviation, is the preset temperature weight, CK is the control accuracy, and DK is the predicted power load information; A control optimization module is used to continue to optimize and evaluate the heat pump control parameters based on the first heat pump control adaptability, obtain optimal heat pump control parameters, and control the multifunctional heat pump.

2. The multifunctional heat pump device according to claim 1, characterized in that: The temperature deviation acquisition module is used for: Collect the control temperature currently set by the user; Collecting the detected temperatures at multiple locations in the building currently equipped with the multifunctional heat pump, and calculating the average to obtain the actual temperature; The temperature deviation ratio between the control temperature and the actual temperature is calculated to obtain the temperature control deviation.

3. The multifunctional heat pump device according to claim 1, characterized in that: The adjustment temperature prediction module is used for: Acquire a heat pump control parameter space of the multifunctional heat pump; randomly generating a first heat pump control parameter in the heat pump control parameter space; Using ensemble learning to train a temperature adjustment prediction channel, the temperature adjustment prediction channel includes a temperature adjustment prediction path array, wherein the number of temperature adjustment prediction paths in the temperature adjustment prediction path array is K, and K is a positive integer; Calculate and round up according to the temperature adjustment prediction ratio and K to obtain J, where J is a positive integer; J temperature adjustment prediction paths are randomly selected in the temperature adjustment prediction channel, and temperature control adjustment prediction is performed on the first heat pump control parameter and the actual temperature to obtain J first predicted adjustment temperatures.

4. The multifunctional heat pump device according to claim 3, characterized in that: The adjustment temperature prediction module is also used for: In the heat pump operation log in the historical time, a sample heat pump control parameter set and a sample actual temperature set are collected, and the temperature after control is collected as a sample adjustment temperature set; Integrate the sample heat pump control parameter set, the sample actual temperature set and the sample adjusted temperature set, and randomly select K sets of temperature regulation prediction supervision training data with replacement, wherein the amount of data in each set of temperature regulation prediction supervision training data is the same; Using the K pieces of temperature adjustment prediction supervision training data, based on machine learning, training to obtain K temperature adjustment prediction paths; The K temperature adjustment prediction paths are combined to obtain a temperature adjustment prediction path array as a temperature adjustment prediction channel.

5. The multifunctional heat pump device according to claim 1, characterized in that: The fitness calculation module is also used for: According to the power operation monitoring log of the building, a sample heat pump control parameter set and a sample power load information set are collected, and the power load information after the heat pump is controlled under different heat pump control parameters and sample power load information is collected, and the sample power load information set is marked; The sample heat pump control parameter set, the sample power load information set and the sample power load information set are used as supervised training data, and machine learning is used to train and obtain a power load prediction channel; The first heat pump control parameter and power load information are input into the power load prediction channel, and the prediction output is used to obtain the first predicted power load information.

6. A control method for a multifunctional heat pump device, characterized in that: The method is applied to the multifunctional heat pump device according to any one of claims 1 to 5, and the method comprises: Collect the control temperature currently set by the user, and collect the actual temperature of the multi-functional heat pump for temperature control, and calculate the temperature control deviation; Acquire the power load information in the building currently equipped with the multifunctional heat pump, and configure the temperature adjustment prediction ratio in combination with the temperature control deviation; Randomly generate a first heat pump control parameter, perform temperature control adjustment prediction according to the actual temperature according to the temperature adjustment prediction ratio, and obtain multiple first predicted adjustment temperatures; Calculate a first control accuracy rate based on the multiple first predicted adjustment temperatures and control temperatures, and calculate a first heat pump control adaptability based on the first heat pump control parameter and the power load caused by the power load information; Based on the first heat pump control adaptability, the heat pump control parameters are continuously optimized and evaluated to obtain the optimal heat pump control parameters, and the multifunctional heat pump is controlled.

Images