Patent Application
20250137672
The present disclosure relates to an integrated control system and method for a heat pump to optimize the energy of a solar power generation building. More specifically, the present disclosure relates to an integrated control system and method for a heat pump to optimize the energy of a solar power generation building that searches for an optimal control variable of the heat pump, using each dynamic behavior prediction value, derived using multiple dynamic behavior prediction models.
Conventional heat pump control searches the control strategy of a heat hump based on local state information. For example, an indoor air temperature measured at the current time point is utilized to search whether the indoor air temperature is within a target temperature range, and the heat pump is stopped or intensified (e.g., increasing the airflow rate, changing the set temperature) according to pre-defined rules and conditions to adjust the temperature.
However, since the dynamic behavior of the heat pump is affected by an interaction with the thermal behavior of other systems within the building (e.g., photovoltaic power generation system, control target room), a control strategy considering only this local state information may not achieve optimal energy saving effects.
Furthermore, there is a problem where conventional building heating and cooling equipment control systems do not comprehensively manage the consumption system (e.g., heat pump) with the power generation system (e.g., photovoltaic power generation system), resulting in a failure to consider the energy offset effect between the two systems.
Therefore, there is a problem where conventional control can deteriorate the utilization efficiency of renewable energy and may make it difficult to achieve optimal energy efficiency from the perspective of the entire building.
A purpose of the present disclosure is to provide an integrated control system and method for a heat pump to optimize the energy of a solar power generation building, that uses real-time data measured from at least any one of the heat pump of the solar power generation building, an external environment, a room and a photovoltaic power generation system, where interactions occur in a dynamic behavior of the heat pump.
Another purpose of the present disclosure is to provide an integrated control system and method for a heat pump to optimize the energy of the solar power generation building, that minimizes the number of sensors for real-time data measurement, by using weather data collected from an external weather system and/or data derived from stored formulas, together.
Another purpose of the present disclosure is to provide an integrated control system and method for a heat pump to optimize the energy of the solar power generation building, that derives each dynamic behavior prediction value using multiple dynamic behavior prediction models, at least some of which are interconnected.
Another purpose of the present disclosure is to provide an integrated control system and method for a heat pump to optimize the energy of the solar power generation building, that searches for an optimal control variable of the heat pump using collected real-time data and each dynamic behavior prediction value derived using multiple dynamic behavior prediction models.
The above and other purposes of the present disclosure may all be achieved by the integrated control system and method for a heat pump to optimize the energy of a solar power generation building according to the present disclosure.
The integrated control system for a heat pump to optimize the energy of a solar power generation building according to one embodiment of the present disclosure includes a computer including a processor and a memory within which a code for execution by the processor is stored.
The computer is configured to collect real-time data measured from at least any one of the heat pump of the solar power generation building, an external environment, a room and a photovoltaic power generation system, where interactions occur in a dynamic behavior of the heat pump.
Furthermore, the computer is configured to input the real-time data of before pre-set time into multiple dynamic behavior prediction models, to derive an indoor air temperature prediction value, heat pump electricity consumption prediction value, and photovoltaic power generation prediction value, of after pre-set time.
Furthermore, the computer is configured to search for an optimal control variable of the heat pump using the derived indoor air temperature prediction value, heat pump electricity consumption prediction value, and photovoltaic power generation prediction value.
The real-time data may include at least one of a set temperature of the heat pump and/or airflow rate of the heat pump, which are control variables; and an outdoor air temperature, solar radiation, solar panel incident angle, solar panel surface temperature, photovoltaic electricity generation, indoor air temperature, heat pump on/off, heat pump supply air temperature, heat pump return air temperature, and heat pump electricity consumption, which are state variables.
The multiple dynamic behavior prediction models may consist of an indoor air temperature prediction model, a heat pump electricity consumption prediction model, and a photovoltaic electricity generation prediction model, stored in the memory, and the heat pump electricity consumption prediction model may derive the heat pump electricity consumption prediction value, using the indoor air temperature prediction value predicted by the indoor air temperature prediction model.
The indoor air temperature prediction model is a model that derives the indoor air temperature prediction value of after pre-set time according to a control variable setting using the outdoor air temperature, solar radiation, indoor air temperature, and heat pump supply air temperature, of before pre-set time.
The heat pump electricity consumption prediction model is a model that derives the heat pump electricity consumption prediction value of after pre-set time according to the control variable setting using the outdoor air temperature, solar radiation, indoor air temperature, heat pump on/off, heat pump supply air and return air temperatures, and heat pump electricity consumption, of before pre-set time, and the indoor air temperature prediction value predicted by the indoor air temperature prediction model.
The photovoltaic electricity generation prediction model is a model that derives a photovoltaic electricity generation prediction value of after pre-set time using the outdoor air temperature, solar radiation, solar panel incident angle, solar panel surface temperature, and photovoltaic electricity generation, of before pre-set time.
The computer may be configured to derive a net energy consumption calculated from a difference between the heat pump electricity consumption prediction value and a photovoltaic electricity generation prediction value, and to search for the optimal control variable using the derived net energy consumption.
Furthermore, the computer may be configured to search for the optimal control variable that minimizes a total sum of the net energy consumption of after pre-set time while maintaining an average value of the indoor air temperature prediction values of after pre-set time within the predetermined range, and to give greater weight to the indoor air temperature prediction value and/or the net energy consumption, closer to the current time.
The computer may be configured to use data collected from an external weather system and/or data derived using a formula stored in the memory, instead of at least some of the real-time data measured from at least any one of the external environment, room, and photovoltaic power generation system.
The weather data may include at least one of an outdoor air temperature prediction value, rainfall prediction value, rainfall condition prediction value, and cloud condition prediction value.
The computer may be configured to use the heat pump return air temperature as the indoor air temperature, and to derive a solar radiation prediction value using the stored formula.
The multiple dynamic behavior prediction models may consist of a heat pump return air temperature prediction model, heat pump supply air temperature prediction model, heat pump electricity consumption prediction model, and photovoltaic electricity generation prediction model, stored in the memory.
The heat pump supply air temperature prediction model may derive a heat pump supply air temperature prediction value using a heat pump return air temperature prediction value predicted by the heat pump return air temperature prediction model, and the heat pump electricity consumption prediction model may derive the heat pump electricity consumption prediction value using the heat pump supply air temperature prediction value predicted by the heat pump supply air temperature prediction model.
The heat pump return air temperature prediction model is a model that derives the heat pump return air temperature prediction value of after pre-set time according to a control variable setting using the heat pump return air temperature, heat pump supply air temperature, and heat pump set temperature, of before pre-set time, and the outdoor air temperature prediction value of after pre-set time.
The heat pump supply air temperature prediction model is a model that derives the heat pump supply air temperature prediction value of after pre-set time according to a control variable setting using the heat pump return air temperature and heat pump supply air temperature, of before pre-set time, and the outdoor air temperature prediction value and heat pump return air temperature prediction value, of after pre-set time.
The heat pump electricity consumption prediction model is a model that derives a heat pump electricity consumption prediction value of after pre-set time according to a control variable setting using a heat pump electricity consumption of before pre-set time, and the heat pump return air temperature prediction value and the heat pump supply air temperature prediction value, of after pre-set time.
The photovoltaic electricity generation prediction model is a model that derives a photovoltaic electricity generation prediction value of after pre-set time using a photovoltaic electricity generation of before pre-set time, and a solar radiation prediction value, outdoor air temperature prediction value, rainfall prediction value, rainfall condition prediction value, and cloud condition prediction value, of after pre-set time.
The computer may be configured to derive a net energy consumption calculated from a difference between the heat pump electricity consumption prediction value and the photovoltaic electricity generation prediction value, and to search for the optimal control variable using the derived net energy consumption.
Furthermore, the computer may be configured to search for the optimal control variable that minimizes a total sum of the net energy consumption of after pre-set time while maintaining an average value of the heat pump return air temperature prediction values of after pre-set time within the predetermined range, and to give greater weight to the heat pump return air temperature prediction value and/or the net energy consumption closer to the current time.
The present disclosure collects real-time data measured from at least any one of the heat pump of the solar power generation building, external environment, room, and photovoltaic power generation system, allowing it to perform a control strategy that comprehensively considers the correlated dynamic behavior of multiple systems within the building, instead of treating the heat pump as an independent control target.
Furthermore, the present disclosure has the effect of minimizing the sensors for real-time data measurement, by using weather data collected from the external weather system and/or data derived from stored formulas.
Furthermore, the present disclosure has the effect of predicting the correlated dynamic behavior of the heat pump, control target room, and photovoltaic energy generation system, using the multiple dynamic behavior prediction models (simulation), at least some of which are interconnected, and implementing integrated optimal control for the heat pump using the prediction results of the model.
Through this, the present disclosure has the effect of being sufficiently utilized in optimal control that comprehensively manages not only the photovoltaic power generation system and the cooling system of the heat pump but also various heating systems, cooling systems, and renewable energy generation systems, applied to buildings.
In the following description, only the portions necessary for understanding the integrated control system and method for a heat pump to optimize the energy of a solar power generation building according to the embodiments of the present disclosure will be described, and descriptions of other portions may be omitted so as not to obscure the gist of the present disclosure.
Furthermore, the terms and words used in this specification and claims described hereinbelow should not be interpreted in their conventional or dictionary meanings, but should be interpreted in meanings and concepts that best express the present disclosure and conform to its technical spirit.
Throughout this specification, when a certain portion is described as “including (comprising)” a certain component, this means that it may include additional components unless specifically stated otherwise; it does not exclude other components. Furthermore, terms such as “ . . . part,”, “ . . . er”, and “ . . . module,” etc., as used in this specification, refer to a unit that processes at least one function or operation, which may be implemented by hardware, software, or a combination of hardware and software.
In various embodiments, the same reference numerals will be used for components that have the same configuration, with representative explanations provided for one embodiment. Other embodiments will be described with respect to configurations that differ from those of the representative embodiment.
Referring to
More specifically, the real-time data collection part 110 may collect real-time data regarding control variables according to at least any one of a set temperature of the heat pump and airflow rate of the heat pump; and state variables according to at least any one of an outdoor air temperature, solar radiation, photovoltaic panel incident angle, photovoltaic panel surface temperature, photovoltaic electricity generation, indoor air temperature, heat pump on/off, heat pump supply air temperature, heat pump return air temperature, and heat pump electricity consumption, using sensors installed in the solar power generation building, from at least any one of the heat pump, external environment, room, and photovoltaic power generation system, as shown in Table 1 below.
Furthermore, the real-time data collection part 110 may collect data from an external weather system that provides weather data.
More specifically, by collecting data such as that shown in Table 2 below, from the external weather system, the real-time data collection part 110 may replace some or all of the data that would otherwise need to be collected from the sensors installed in the external environment, room, and photovoltaic power generation system of the solar power generation building.
Furthermore, the real-time data collection part 110 may also derive data in real-time using formulas pre-stored in a memory. For example, the real-time data collection part may input information such as a location (latitude, longitude), date and time of the solar power generation building into the pre-stored formula to derive solar radiation prediction values.
The dynamic behavior prediction part 120 used artificial neural network technology, which corresponds to a data-driven model instead of physical formulas. An artificial neural network model is one that learns patterns between variables through input variable data and output variable data. The basic form of the artificial neural network model is ‘perceptron,’ which receives multiple input values, multiplies them by weights, and then sums them up, and generates outputs through an activation function. Multiple perceptrons are gathered to constitute one layer, and output values of each layer are transmitted as input values of a next layer, creating a model consisting of multiple layers, known as a multilayer perceptron model. The artificial neural network model may be created using only minimal information, and due to automation of the model training process, it has the advantage of significant data availability and field applicability.
Thus, the dynamic behavior prediction part 120 derives each dynamic behavior prediction value using multiple dynamic behavior prediction models, which are artificial neural network models. The dynamic behavior prediction model is a model that learned a pattern between the variables using the data before a reference time as input variable data and the data after the reference time as output variable data. Here, the data before the reference time may be, for example, data measured at 1-minute intervals for 20 minutes prior to the reference time, or data measured at 10-minute intervals for 30 minutes before the reference time, and the data measured at the reference time may also be included. Additionally, the data after the reference time may include, for example, data measured at 1-minute intervals for 20 minutes after the reference time, or data measured at 10-minute intervals for 30 minutes after the reference time. The durations such as 20 minutes or 30 minutes, and data measurement intervals like 1 minute or 10 minutes may be appropriately adjusted by those skilled in the art. Furthermore, the time intervals and/or total durations of the data before the reference time and after the reference time may differ.
Moreover, the training of the dynamic behavior prediction model for predicting multiple dynamic behaviors is a process that increases prediction accuracy through the updating of weights of the multiple dynamic behavior prediction models. An optimization algorithm may be used to update the weights of the artificial neural network so that an error between the output value () of the artificial neural network and the corresponding actual value (
) is minimized. For example, in the model training process, the root mean square error as defined by mathematical formula 1 may be used as a reference error.
In a system according to one embodiment of the present disclosure, the dynamic behavior prediction part 120 may consist of an indoor air temperature prediction model (ANN #1), a heat pump electricity consumption prediction model (ANN #2) and a photovoltaic electricity generation prediction model (ANN #3), and may be trained as shown in
Additionally, the heat pump electricity consumption prediction model (ANN #2) derives a heat pump electricity consumption (EHP) prediction value of after pre-set time based on the current time according to the control variable settings, using the outdoor air temperature (OAT), solar radiation (Qsolar), indoor air temperature (IAT), heat pump on/off status (xHP), heat pump supply air temperature (SAT), return air temperature (RAT), and heat pump electricity consumption (EHP), of before pre-set time, and the indoor air temperature (IAT), of after pre-set time predicted in the indoor air temperature prediction model.
Furthermore, the photovoltaic electricity generation prediction model (ANN #3) derives a photovoltaic electricity generation (EPV) prediction value of after pre-set time based on the current time, using the outdoor air temperature (OAT), solar radiation (Qsolar), solar panel incident angle (θpv), solar panel surface temperature (Tpv), and photovoltaic electricity generation (EPV), of before pre-set time.
In this way, the dynamic behavior prediction part 120 may take into account different dynamic behaviors and interactions of the photovoltaic power generation system, room and heat pump system through the structural connection of multiple dynamic behavior prediction models. For example, the indoor air temperature predicted for the next 30 minutes by the indoor air temperature prediction model may be used as an input value for a model predicting the heat pump electricity consumption, explaining the relationship between indoor air temperature changes and cooling energy. Additionally, by considering the prediction period of 30 minutes, not only immediate changes but also thermal responses that appear slowly due to time lag can be taken into account.
Furthermore, as shown in
Furthermore, the multiple dynamic behavior prediction models of the dynamic behavior prediction part 120 may consist of a trained heat pump return air temperature prediction model (ANN #4), heat pump supply air temperature prediction model (ANN #5), heat pump electricity consumption prediction model (ANN #6), and photovoltaic electricity generation prediction model (ANN #7), as shown in
At this time, the heat pump return air temperature prediction model (ANN #4) derives a heat pump return air temperature (RAT) prediction value of after pre-set time based on the current time according to the control variable (e.g. heat pump set temperature (SET)) settings using a heat pump return air temperature (RAT), heat pump supply air temperature (SAT), and heat pump set temperature (SET), of before pre-set time, and an outdoor air temperature (OAT) prediction value of after pre-set time.
In addition, the heat pump supply air temperature prediction model (ANN #5) derives a heat pump supply air temperature (SAT) prediction value of after pre-set time based on the current time according to the control variable (e.g. heat pump set temperature (SET)) settings, using the heat pump return air temperature (RAT) and heat pump supply air temperature (SAT), of before pre-set time, and the outdoor air temperature (OAT) prediction value and heat pump return air temperature (RAT) prediction value, of after pre-set time.
Moreover, the heat pump electricity consumption prediction model (ANN #6) derives a heat pump electricity consumption (Ehp) prediction value of after pre-set time based on the current time according to the control variable (e.g. heat pump set temperature (SET)) settings using the heat pump electricity consumption (Ehp) of before pre-set time, and the heat pump return air temperature (RAT) prediction value and heat pump supply air temperature (SAT) prediction value, of after pre-set time.
In addition, the photovoltaic electricity generation prediction model (ANN #7) derives a photovoltaic electricity generation (Epv) prediction value of after pre-set time, based on the current time using the photovoltaic electricity generation (Epv) of before pre-set time, and the solar radiation (Qsolar) prediction value, outdoor air temperature (OAT) prediction value, rainfall (RN) prediction value, rainfall condition (RC) prediction value, and cloud condition (CC) prediction value, of after pre-set time.
Thus, the dynamic behavior prediction part 120 may take into account different dynamic behaviors and interactions of the photovoltaic power generation system and heat pump system through the structural connection of multiple dynamic behavior prediction models.
Additionally, the dynamic behavior prediction part 120 may use data (outdoor air temperature (OAT) prediction value, rainfall (RN) prediction value, rainfall condition (RC) prediction value, cloud condition (CC) prediction value) collected from the external weather system and data (solar radiation (Qsolar) prediction value) derived using stored formulas, thereby minimizing the number of sensors that need to be installed in the solar power generation building compared to when constructing the dynamic behavior prediction part and reducing the time and cost for optimal control as shown in
Next, the optimal control variable search part 130 searches for optimal control variables of the heat pump using the collected real-time data and each dynamic behavior prediction value derived from the multiple dynamic behavior prediction models.
The optimal control variable search part 130 derives an expected total energy consumption based on the difference between the heat pump electricity consumption prediction value, derived from the heat pump electricity consumption prediction model, and the photovoltaic electricity generation prediction value, derived from the photovoltaic electricity generation prediction model, among the multiple dynamic behavior prediction models, and searches for the optimal control variable based on the derived expected total energy consumption.
In more detail, the optimal control variable search part 130 calculates a predicted cost for all possible combinations of the control variables and searches for the control variable that minimizes the cost using an exhaustive search method.
As shown in
The net energy consumption is calculated as the difference between the heat pump electricity consumption prediction value from the heat pump electricity consumption prediction model and the photovoltaic electricity generation prediction value from the photovoltaic electricity generation prediction model. In this case, if no surplus photovoltaic electricity generation storage system exists, all negative values are considered as ‘0’.
Thus, the optimal control variable search part 130 may induce the offsetting effect between the energy consumption of the heat pump and the photovoltaic power generation by using the difference value between the heat pump electricity consumption prediction value and the photovoltaic electricity generation prediction value as COST, and may induce energy optimization from the perspective of the entire building rather than just the energy consumption of the heat pump itself.
In addition, the optimal control variable search part 130 derives a control strategy that can minimize the total energy consumption of the building on the condition of maintaining the indoor air temperature within the target temperature range. Specifically, the control strategy may be derived under the condition that the average value of the indoor air temperature prediction values, or the average value of the return air temperature prediction values of the heat pump that can replace the indoor air temperature prediction values are maintained within the target temperature range.
Also, as shown in
Next, the heat pump control part 140 controls the set temperature and/or airflow rate of the heat pump, based on the searched optimal control variable.
The integrated control system of heat pump to optimize energy of the solar power generation building according to the present disclosure, which has been described so far, may be implemented by a computer that includes a processor and a memory within which instructions for execution by the processor are stored.
At this time, the computer may function as the aforementioned data collection part, dynamic behavior prediction part, optimal control variable search part, and heat pump control part.
Specifically, the computer of the integrated control system of the heat pump to optimize the energy of the solar power generation building according to one embodiment of the present disclosure is configured to collect real-time data measured from at least any one of the heat pump of the solar power generation building, external environment, room and photovoltaic power generation system where interactions occur in the dynamic behavior of the heat pump, for the integrated control of the heat pump to optimize the energy of the solar power generation building. Here, the collected data may be the type of data displayed in Table 1, for example.
Then, the computer is configured to predict the dynamic behavior for the next 30 minutes, for example, at 10-minute intervals, using the dynamic behavior prediction model developed in the method as shown in
The multiple dynamic behavior prediction models may consist of a photovoltaic electricity generation prediction model, indoor air temperature prediction model, and heat pump electricity consumption prediction model, that are based on artificial neural network models, and may be stored in the computer's memory.
The computer derives the indoor air temperature prediction value for the future 30 minutes according to the control variable settings using the indoor prediction model (ANN #1) and the outdoor air temperature, solar radiation, indoor air temperature, and heat pump supply air temperature for the past 30 minutes.
Additionally, the computer derives the heat pump electricity consumption prediction value for the future 30 minutes according to the control variable settings, using the heat pump electricity consumption prediction model (ANN #2), the outdoor air temperature, solar radiation, indoor air temperature, heat pump on/off, heat pump supply air and return air temperature, and heat pump electricity consumption, for the past 30 minutes, and the indoor air temperature for the future 30 minutes predicted in the indoor air temperature prediction model.
Additionally, the computer derives the photovoltaic electricity generation prediction value for the future 30 minutes, using the photovoltaic electricity generation prediction model (ANN #3), the outdoor air temperature, solar radiation, solar panel incident angle, solar panel surface temperature, and photovoltaic electricity generation, for the past 30 minutes.
Furthermore, the computer is configured to derive a control strategy that can minimize the building's net energy consumption while maintaining the indoor air temperature within the target temperature range, using the prediction values of the photovoltaic electricity generation prediction model, indoor air temperature prediction model, and heat pump electricity consumption prediction model.
Specifically, as shown in
Moreover, the computer is configured to transmit the derived control strategy to the heat pump system, allowing the heat pump, for instance, to be controlled to be updated to the optimal set temperature and airflow rate every 10 minutes.
Furthermore, the computer of the integrated control system of the heat pump to optimize the energy of the solar power generation building according to another embodiment of the present disclosure is configured to collect real-time data measured from the heat pump of the solar power generation building and the photovoltaic power generation system, for the integrated control of the heat pump to optimize the energy of the solar power generation building. Furthermore, the computer is configured to collect data from the external weather system or derive necessary data using the formulas stored in the memory. In this case, the data collected from the external weather system may be of the type displayed in Table 2, for example.
Then, the computer is configured to predict the dynamic behavior for the next 10 minutes, for example at 1-minute intervals, using the dynamic behavior prediction model developed in the method as shown in
The multiple dynamic behavior prediction models consist of the heat pump return air temperature prediction model, heat pump supply air temperature prediction model, heat pump electricity consumption prediction model, and photovoltaic electricity generation prediction model, that are based on artificial neural network models, which may be stored in the computer's memory.
The computer derives the heat pump return air temperature (RAT) prediction value for the future 10 minutes according to the heat pump set temperature (SET) setting which is the control variable, using the heat pump return air temperature prediction model (ANN #4), the heat pump return air temperature (RAT), heat pump supply air temperature (SAT), and heat pump set temperature (SET), for the past 10 minutes, and the outdoor air temperature (OAT) prediction value of the future 20 minutes.
Additionally, the computer derives the heat pump supply air temperature (SAT) prediction value for the future 10 minutes according to the heat pump set temperature (SET) settings, which is the control variable, using the heat pump supply air temperature prediction model (ANN #5), the heat pump return air temperature (RAT) and heat pump supply air temperature (SAT), for the past 10 minutes, and the outdoor air temperature (OAT) prediction value and the heat pump return air temperature (RAT) prediction value, for the future 10 minutes.
Furthermore, the computer derives the heat pump electricity consumption (Ehp) prediction value for the future 10 minutes according to the heat pump set temperature (SET) settings, which is the control variable, using the heat pump electricity consumption prediction model (ANN #6), the heat pump electricity consumption (Ehp) for the past 10 minutes, and the heat pump return air temperature (RAT) prediction value and heat pump supply air temperature (SAT) prediction value, for the future 10 minutes.
Furthermore, the computer derives the photovoltaic electricity generation (Epv) prediction value for the future 10 minutes using the photovoltaic electricity generation prediction model (ANN #7), the photovoltaic electricity generation (Epv) for the past 10 minutes, and solar radiation (Qsolar) prediction value, outdoor air temperature (OAT) prediction value, rainfall (RN) prediction value, rainfall condition (RC) prediction value, and cloud condition (CC) prediction value, for the future 20 minutes.
In addition, the computer is configured to derive the control strategy that can minimize the building's net energy consumption while maintaining the indoor air temperature within the target temperature range, using the prediction values of the heat pump return air temperature prediction model, heat pump electricity consumption prediction model, and photovoltaic electricity generation prediction model.
The integrated control system for the heat pump to optimize the energy of the solar power generation building according to another embodiment of the present disclosure does not measure or predict the indoor air temperature; but instead, uses the prediction value of the heat pump return air temperature, returning indoors to the heat pump, as the indoor air temperature prediction value.
Specifically, as shown in
Additionally, the computer is configured to transmit the derived control strategy to the heat pump system, allowing the heat pump to be controlled to be updated to the optimal set temperature every minute.
Referring to
More specifically, the integrated control method for the heat pump to optimize the energy of the solar power generation building according to the present disclosure may be performed by the computer of the integrated control system for the heat pump according to the present disclosure, and thus redundant explanations will be omitted hereinafter.
First, the integrated control system 100 collects real-time data measured from at least any one of the heat pump of the solar power generation building, external environment, room, and photovoltaic power generation system where interactions occur in the dynamic behavior of the heat pump (S100).
In step S100, as shown in
In addition, in step S100, as shown in
The real-time data collected and/or derived in step S100 may be collected and/or derived at pre-set time intervals (for example, at 1-minute intervals, 10-minute intervals, etc.).
Next, the integrated control system 100 inputs data of before pre-set time based on the current time, of the collected real-time data, as input variable data of the multiple dynamic behavior prediction models, to derive the dynamic behavior prediction value of after pre-set time (S200).
In step S200, the integrated control system 100 may derive the indoor air temperature prediction value, heat pump electricity consumption prediction value, and photovoltaic power generation prediction value, as the dynamic behavior prediction values, using the learned indoor air temperature prediction model, heat pump electricity consumption prediction model, and photovoltaic electricity generation prediction model, as the multiple dynamic behavior prediction models, as shown in
Furthermore, the integrated control system 100 may derive the heat pump return air temperature prediction value, heat pump supply air temperature prediction value, heat pump electricity consumption prediction value, and photovoltaic power generation prediction value, as the dynamic behavior prediction values, using the learned heat pump return air temperature prediction model, heat pump supply air temperature prediction model, heat pump electricity consumption prediction model, and photovoltaic electricity generation prediction model, as the multiple dynamic behavior prediction models, as shown in
Then, the integrated control system 100 searches for the optimal control variables of the heat pump using the derived dynamic behavior prediction values (S300). Here, the optimal control variables may be the set temperature and/or the airflow rate of the heat pump to optimize the solar power generation building from an energy perspective.
In step S300, the integrated control system 100 searches for the optimal control variables from the net energy consumption derived from the difference between the heat pump electricity consumption prediction value derived from the heat pump electricity consumption prediction model, and the photovoltaic electricity generation prediction value derived from the photovoltaic electricity generation prediction model, among the multiple dynamic behavior prediction models (refer to
More specifically, the system searches for the optimal control variables that minimize the total sum of the net energy consumption of after pre-set time. In this case, the average value of the indoor air temperature prediction values of after pre-set time being maintained within a pre-defined range may set as the condition. Here, if there are no indoor air temperature prediction values among the dynamic behavior prediction values, the heat pump return air temperature prediction value may be used as the indoor air temperature prediction value.
Additionally, of the prediction values, since the prediction value that is closer to the current time is more reliable, the system may be configured to give greater weight to the indoor air temperature prediction value and/or the net energy consumption, that are closer to the current time, when calculating the average value of the indoor air temperature prediction values and/or the total sum of the net energy consumption. Then, the integrated control system 100 controls the set temperature and/or airflow rate of the heat pump based on the searched optimal control variables (S400).
If the integrated control system can directly control the heat pump, in step S400, the integrated control system controls the set temperature and/or airflow rate of the heat pump based on the searched optimal control variables.
If the integrated control system cannot directly control the heat pump, in step S400, the integrated control system provides the searched optimal control variables to a system that controls the heat pump.
The functional operations and embodiments related to the present subject matter, as described in this specification, can be implemented in digital electronic circuits, computer software, firmware, or hardware, or a combination of one or more of these, including the structures disclosed in this specification and their structural equivalents.
The embodiments of the subject matter described in this specification may be implemented, at least in part, as one or more modules in a computer program product, that is, as computer program instructions encoded on a tangible program medium for execution by, or to control the operation of, a data processing device. The tangible program medium may be a propagated signal or a computer-readable medium. A propagated signal is an artificially generated signal, such as an electrically, optically, or electromagnetically generated signal, created to encode information for transmission to a suitable receiver device for execution by a computer. A computer-readable medium may be a machine-readable storage device, a machine-readable storage substrate, a memory device, a combination of materials that affect a machine-readable propagated signal, or a combination of one or more of these.
A computer program (also known as a program, software, software application, script, or code) may be written in any form of programming language, whether compiled or interpreted, including declarative or procedural languages. It may be deployed in any form, including as a standalone program, a module, a component, a subroutine, or another unit suitable for use in a computing environment.
A computer program does not necessarily correspond to a file in a file device. A program may be stored in a single file provided to the requested program, or in multiple interacting files (e.g., one or more files that store parts modules, subprograms, or codes), or as part of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document).
A computer program may be deployed to be executed on a single computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.
Additionally, the logical flows and structural block diagrams described in this patent document are intended to illustrate corresponding actions and/or specific methods supported by the corresponding functions and steps of the disclosed structural means. They can also be used to define corresponding software structures, algorithms, and their equivalents.
The processes and logical flows described in this specification may be performed by one or more programmable processors executing one or more computer programs to carry out functions by operating on input data and generating output.
A processor suitable for executing a computer program includes, for example, any one or more of a general-purpose and special-purpose microprocessor and any form of digital computer. Generally, a processor will receive instructions and data from read-only memory and/or random access memory.
The essential components of a computer are one or more memory devices for storing instructions and data, and a processor for executing the instructions. Additionally, a computer is generally configured to receive data from one or more mass storage devices, such as magnetic, magnetic optical disks, or optical disks, or to transfer data to such devices, or to perform both operations. However, the computer is not required to have such devices.
The description provided herein presents the best mode of the present disclosure, and offers examples for explaining the present disclosure and enabling those skilled in the art to make and use it. This specification is not intended to limit the present disclosure to the specific terms presented.
Thus, while the present disclosure has been described in detail with reference to the aforementioned examples, those skilled in the art may make modifications, changes, and variations to these examples without departing from the scope of the present disclosure. In summary, it is not necessary for all functional blocks illustrated in the drawings to be included separately or for all sequences shown in the drawings to be followed in the exact order presented in order to achieve the intended effects of the present disclosure. Even if they do not, they may still fall within the technical scope of the present disclosure as defined in the claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10-2023-0146553 | Oct 2023 | KR | national |
