JP2020167891A - Method and system for predicting air conditioning load and method and system for managing air conditioning system energy - Google Patents

Abstract

To provide a method and system for predicting an air conditioning load that can accurately and easily predict an air conditioning load, and a method and system for managing air conditioning system energy.SOLUTION: Based on a plurality of parameters related to an air conditioning load, multiple regression analysis is used to predict the air conditioning load. Some or all parameters selected from outdoor air temperature, date, time and day of the week are used as explanatory variables.SELECTED DRAWING: Figure 9

Description

The present invention relates to a method and system for predicting an air conditioning load, and an energy management method and system for an air conditioning system to which the method and system are applied.

In performing air conditioning, for example, a method of measuring the temperature of a room to be air-conditioned and adjusting the operating status of each device (for example, a heat source machine) according to the measured value is widely adopted. However, in such a method, the fluctuation of the air conditioning load in the room is grasped after the fact, and the operating condition is adjusted in a feedback manner based on the fluctuation. Therefore, the air conditioning fluctuates after the fluctuation of the air conditioning load occurs. There will be a gap between the load and operating conditions until the load is properly addressed. There is a problem that the deviation of the operating condition with respect to such a load condition may bring about an air-conditioned state that is not comfortable for people in the room and is not efficient in terms of energy consumption.

In order to solve such a problem, various techniques for predicting the air conditioning load or its fluctuation have been proposed. For example, Patent Document 1 below describes a technique of reading weather forecast data, power load pattern data, and actual measurement data, and predicting a required amount of heat (air conditioning heat load) by a load prediction program using a neural network. There is.

Japanese Unexamined Patent Publication No. 2014-27784

However, in the technique described in Patent Document 1 above, when creating a neural network, there are many parameters to be determined such as the number of layers and neurons and the type of activation function, and it takes time and effort to create an appropriate load prediction program. In addition to taking time, the calculation load becomes enormous when actually making a prediction. In addition, overfitting is likely to occur in the training of neural network models, and it is necessary for data scientists to go to each site for tuning, which also requires labor and cost. In addition, there is a separate running cost for using the weather forecast API, and there is also the problem that the system cannot be delivered in an environment where the Internet cannot be connected.

In view of such circumstances, the present invention intends to provide an air conditioning load prediction method and system capable of accurately and easily predicting an air conditioning load, and an energy management method and system for an air conditioning system.

The present invention relates to a method for predicting an air conditioning load, which is characterized in that the air conditioning load is predicted by using multiple regression analysis based on a plurality of parameters related to the air conditioning load.

In the method for predicting the air conditioning load of the present invention, some or all parameters selected from the following parameters can be used as explanatory variables.
・ Outside air temperature ・ Date ・ Time ・ Day of the week

In the method for predicting the air conditioning load of the present invention, at least three of the above parameters can be used as explanatory variables, and a third-order or higher interaction term can be included in the regression equation.

In the method for predicting the air conditioning load of the present invention, a part of the interaction term can be deleted by the variable increase / decrease method.

In the method for predicting the air conditioning load of the present invention, when the day of the week is used as the explanatory variable, two dummy variables, holiday and weekday, can be set.

Further, the present invention relates to an air conditioning load prediction system characterized in that the above-mentioned air conditioning load prediction method is feasibly configured.

The present invention also relates to an energy management method for an air conditioning system, which comprises determining an operation schedule of a heat source machine based on a predicted value of the air conditioning load by the above-mentioned air conditioning load prediction method.

Further, the present invention includes a heat source machine and a display unit for displaying the predicted value of the air conditioning load by the above-mentioned method for predicting the air conditioning load, and can determine the operation schedule of the heat source machine based on the predicted value. It is related to the energy management system of the air conditioning system, which is characterized by the fact that it is used.

According to the method and system for predicting the air conditioning load of the present invention and the energy management method and system for the air conditioning system, it is possible to obtain an excellent effect that the air conditioning load can be predicted accurately and easily.

It is a schematic diagram which shows an example of the structure of the air conditioning load prediction system and the energy management system of an air conditioning system to which this invention is applied. It is a flowchart explaining an example of the procedure of the method of predicting the air conditioning load by carrying out this invention. It is a graph which compares the estimated value of the air conditioning load by the regression equation with the measured value. It is a graph which compares the estimated value of the air conditioning load by another regression equation with the measured value. It is a graph which compares the estimated value of the air conditioning load by another regression equation with the measured value. It is a graph which compares the estimated value of the air conditioning load by another regression equation with the measured value. It is a graph which compares the estimated value of the air conditioning load by another regression equation with the measured value. It is a graph which compares the estimated value of the air conditioning load by another regression equation with the measured value. It is a graph which compares the predicted value of the air conditioning load by the regression equation with the measured value. It is a graph explaining the relationship between the date deviation and the prediction accuracy of the air conditioning load when using the past outside air temperature as an explanatory variable.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

The first network 10 is a local network in which the devices constituting the air conditioning system are informationally connected. Here, the heat source machine 1 and the temperature sensor 8 are illustrated as the devices, but in addition to the devices, various devices (for example, a variable air volume unit that supplies conditioned air to the room and the operating status of each device are monitored. A monitoring device, etc.) may be connected to form a network. The second network 16 is, for example, a virtual network constructed on the cloud, and the first network 10 is connected to the second network 16 via the router 14. In this way, various types of information can be exchanged between the devices connected to the first network 10 and the second network 16.

In the second network 16, a data storage unit 17, a model generation unit 18, a load prediction unit 19, and an input / output unit 21 are virtually installed (note that some or all of these units are physical). Of course, it can be installed as standard hardware).

The data storage unit 17 has a function as a server constituting an energy management system, and includes information acquired from devices such as a heat source machine 1 and a temperature sensor 8 connected to the first network 10, and various other types of data. Store information as a database. The model generation unit 18 generates a load prediction model M that predicts the air conditioning load in the room at a certain time point (for example, the present or the current latest future time point) based on the test data stored in the data storage unit 17. To do. The load prediction unit 19 stores the load prediction model M generated by the model generation unit 18, and predicts the air conditioning load in the room at a certain time point based on the parameters described later. The input / output unit 21 writes information acquired from each device connected to the first network 10 to the data storage unit 17, takes out necessary data from the data storage unit 17, and inputs it to the load prediction unit 19. Information is input and output between each part.

The model generation unit 18 has a function of generating a load prediction model M by multiple regression analysis based on test data. The load prediction model M may be generated automatically based on given test data, or may be generated by manual operation input.

The operating status of the equipment constituting the air conditioning system, the log of the data stored in the data storage unit 17, the specifications of the load prediction model M stored in the load prediction unit 19, and the value of the air conditioning load predicted by the load prediction unit 19. , And other various data can be displayed on the information terminal device 22 as a display unit connected to any of the first and second networks 10 and 16 (here, information is displayed on the second network 16). The case where the terminal device 22 is connected is illustrated).

A method of predicting the air conditioning load by such a system will be described with reference to the flowchart of FIG.

First, test data for generating the load prediction model M is created and input to the data storage unit 17 (step S1). The test data is a data set that records various parameters related to the air conditioning load at various time points when the air conditioning system is actually operated, and the actual value of the air conditioning load at that time. The actual value of the air conditioning load can be calculated as the product of "the temperature difference between the air conditioning air supplied indoors and the return air recovered from the room" and "the amount of air conditioning air supplied indoors". .. Further, some of the parameters related to the air conditioning load can be acquired from the heat source machine 1 and the temperature sensor 8.

Various parameters related to the air conditioning load can be used as explanatory variables for the prediction of the air conditioning load, but in the case of this embodiment, a plurality of parameters selected from the following four parameters are used as the explanatory variables. ing.
・ Outside air temperature / time (hour or hour / minute)
・ Date (month, day or month)
・ Sunday (Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, or holiday or weekday)

The outside air temperature affects the inflow and outflow of heat into the room and affects the air conditioning load. The date and time correspond to fluctuations in the outside air temperature depending on the season or time, or fluctuations in the amount of heat generated in the room (for example, it is assumed that there is equipment in the room that operates only in a specific season or time). The days of the week correspond to the schedule of human activities and equipment operation in the room. For example, in an office or the like, there are few people on holidays and the equipment does not operate much, so it is expected that the air conditioning load will be small on the days of the week that correspond to holidays.

From a different point of view, "time" is a parameter that reflects the change in the daily cycle of the air conditioning load, "date" is the parameter that reflects the change in the one-year cycle, and "day of the week" is the parameter that reflects the change in the weekly cycle. .. In addition, it can be said that the "outside air temperature" is a parameter that reflects changes in the annual and daily cycles of the air conditioning load in the estimated value. Of the above parameters, the outside air temperature is set as a qualitative variable, and the date, time, and day of the week are set as dummy variables.
If you want to make a forecast that corresponds to the schedule of the office or the like in more detail, you may add "whether or not it is the air conditioning start day" or "whether or not it is the air conditioning start time" as a dummy variable. For example, on Monday, which is a weekday after a holiday, it can be assumed that the load will be particularly large at the start time of air conditioning due to the influence of heat storage in the skeleton. Further, even on a day in the middle of consecutive weekdays, it is conceivable that the air conditioning load will increase immediately after the start of air conditioning (for example, before and after the start time).

In addition to or in addition to the parameters listed above, it is also possible to use, for example, the following parameters, or other parameters related to the air conditioning load.
・ Measured value of indoor temperature ・ Set value of indoor temperature ・ Temperature difference between indoor and outdoor ・ Actual value of air conditioning load ・ Number of operating heat source machines 1 ・ Supply air temperature of heat source machine 1 (temperature of air conditioning air)
・ Weather, amount of solar radiation, expected temperature

In the test data, the actual value of the air conditioning load at that time is recorded together with each of the above parameters at a plurality of time points. The test data can be created, for example, by operating the air conditioning system by feedback control based on the temperature sensor 8 for a period sufficient for constructing the load prediction model M (for example, about one year), and collecting the above-mentioned parameters during that period.

The model generation unit 18 uses the test data to generate a load prediction model M that predicts the air conditioning load at a target time point by multiple regression based on the above parameters (step S2). The generated load prediction model M is a function that predicts the air conditioning load at a target time point from parameters related to the operating conditions at a certain time point or period in the target air conditioning system. The specific contents of the load prediction model M will be described later.

After the generation of the load prediction model M is completed, the operation of actually predicting the air conditioning load can be started.

The input / output unit 21 acquires the values of the parameters used as explanatory variables in the load prediction model M via the networks 10 and 16 and sequentially records them in the data storage unit 17 (step S3). The load prediction unit 19 reads out the numerical values of the parameters at a certain time point or period from the data storage unit 17, and predicts the air conditioning load at the target time point using the load prediction model M (step S4).

The operation manager can display and refer to the value of the air conditioning load predicted by the load prediction unit 19 on the information terminal device 22. The operation manager determines the operation status of each device related to air conditioning (for example, the start / stop schedule of the heat source unit 1) based on the displayed predicted value, or changes it as necessary. For example, if the predicted value of the air conditioning load at a certain future point is large and it is assumed that the air conditioning load cannot be covered by the heat source machine 1 currently in operation, the start / stop schedule of the heat source is changed to operate the heat source. The number can be increased. Further, in the case of the heat storage type air conditioning system, it is possible to perform the heat storage operation of the heat source machine 1 according to the predicted value of the air conditioning load. Alternatively, the system may be configured so that the determination of the operation schedule based on such predicted values is automatically performed without the intervention of the operation manager. In this way, it is possible to reduce the deviation between the load condition and the operating condition that occur due to the fluctuation of the load, realize a comfortable air-conditioning state, and operate the air-conditioning system with energy saving and cost saving.

Although the case where the heat source unit 1 is turned on and off to deal with the air conditioning load has been described here, the air conditioning load may be dealt with by adjusting the operating conditions of other devices constituting the air conditioning system. For example, it is possible to change the temperature and supply amount of the conditioned air in the air conditioner, or the amount of blown air of the variable air volume unit that sends the conditioned air into the target room according to the prediction result.

In addition, the system configuration and the procedure for predicting the air conditioning load described here are just examples. A load prediction model M by multiple regression analysis is generated from various parameters related to the operation status of the air conditioning system as described above, and the system configuration and prediction are performed as long as the air conditioning load can be predicted from the operation status at an arbitrary time point or period. The procedure can be changed in various ways.

The generation of the load prediction model M, which is a regression equation by multiple regression analysis, will be described. If the objective variable y (t) is expressed by two explanatory variables x ₁ and x ₂ , which are qualitative variables or quantitative variables, respectively, the regression equation is, for example, the following equation 1.

Here, the effect of each explanatory variable x ₁ and x _{2 on} the objective variable y (t) may be affected by the action of the explanatory variables x ₁ and x ₂ (interaction). When considering this interaction, the regression equation is added with an interaction term, for example, as shown in Equation 2 below.

When the number of explanatory variables is 2, the degree of the interaction term is 2 as in the above equation 2. When the number of explanatory variables is three and all the interactions between the explanatory variables are considered, the interaction term having a degree of 2 and the interaction term having a degree of 3 are included in the regression equation. When the number of explanatory variables is four and all the interactions between the explanatory variables are considered, the interaction term having a degree of 2 and the interaction term having a degree of 3 are included in the regression equation.

In the case of this embodiment, as described above, the parameter selected from the four parameters of the outside air temperature, the date, the time, and the day of the week is used for the prediction of the air conditioning load. For regression analysis with these parameters, test data was used to verify the accuracy of estimation or prediction.

As the test data, the outside air temperature measured at each time (0:00 to 23:00) on each day (January 1 to December 31) in 2007 and the actual value of the air conditioning load were used. Then, the air conditioning load at a certain point in 2007 is set as the objective variable, and the "outside air temperature" at that time, "time (0:00, 1:00, 2:00, ..., 23:00)", "month (1)" Month, February, ..., December) ”and“ day of the week (holiday or weekday) ”were used as explanatory variables, and the estimated value of the air conditioning load was calculated by regression analysis.

First, when the interaction was not considered, a regression equation was generated by combining each explanatory variable to obtain an estimated value of the air conditioning load, and the accuracy was evaluated by a coefficient of determination in comparison with the actual value of the air conditioning load at that time. The results are shown in Table 1 below.

In Table 1 above, the circles in each column indicate that the parameters corresponding to the columns are incorporated as explanatory variables. That is, for example, in combination number 1, the air conditioning load is estimated by simple regression using only the outside air temperature as an explanatory variable, and in combination number 8, estimation is performed by multiple regression using time and month as explanatory variables.

As shown in Table 1, it was confirmed that the coefficient of determination tends to increase as the number of explanatory variables increases, but if the interaction is not taken into consideration, the coefficient of determination is 0.6 even if all four parameters are used. Stayed.

Subsequently, a regression equation was created using two or more explanatory variables and incorporating all the interaction terms between the explanatory variables, and the air conditioning load was estimated. The results are shown in Table 2 below.

Considering the interaction, the coefficient of determination was at most 0.6 when the maximum order of the interaction term was 2, but when using three explanatory variables and incorporating the maximum third-order interaction term into the regression equation, , The coefficient of determination has improved significantly. In particular, when "hour", "month", and "weekday or holiday" were used as explanatory variables, a high coefficient of determination of 0.918 was shown (combination number 10). Furthermore, when all four explanatory variables were used and all interactions between the explanatory variables were taken into consideration, the highest coefficient of determination was 0.943 (combination number 11).

The degree of agreement between the estimated values and the measured values will be further described using a graph showing fluctuations in the air conditioning load.

3 to 8 show the actual fluctuation of the air conditioning load during a certain 10 days (July 1 to 11) in 2007 and the regression equation by some combinations shown in Tables 1 and 2 above. It is a graph which compares with the calculated estimated value. First, the actual air conditioning load showed fluctuations depending on the day of the week and time, as shown by the solid lines in FIGS. 3 to 8. That is, the load was large during the daytime on weekdays, and the load was small during the nighttime on holidays and weekdays. On the other hand, in the case of simple regression using only the outside air temperature as an explanatory variable (Table 1, combination number 1), as shown in FIG. 3, the estimated value shown by the broken line became almost flat over 10 days. The outside air temperature does not depend much on the activity schedule of a person, but the amount of heat generated in an office or the like fluctuates greatly depending on the day of the week and the time of day. Regression based only on the outside air temperature cannot reflect such schedule fluctuations well.

Therefore, when multiple regression was performed with time as the explanatory variable in addition to the outside air temperature (Table 1, combination number 5), the estimated value fluctuated in a daily cycle as shown in FIG. 4, and from the results shown in FIG. The estimated value is close to the measured value. Further, considering the interaction between the outside air temperature and the time (Table 2, combination number 1), as shown in FIG. 5, the degree of agreement between the estimated value and the measured value was further increased.

Subsequently, FIGS. 6 and 7 show a case where "holiday or weekday" is further added as an explanatory variable. In FIG. 6, the outside air temperature, time, and whether it is a holiday or a weekday are used as explanatory variables, but the interaction is not considered (Table 1, combination number 12). Since holidays or weekdays are added to the explanatory variables, the estimated value of holidays has decreased as shown by the arrows in the figure, and is approaching the measured value. Here, when the interaction term between each explanatory variable is added to the regression equation (Table 2, combination number 8), as shown in FIG. 7, the estimated value in the daytime on holidays is even lower, which is close to the bottom value, which is the measured value. The degree of agreement with is further increased. If the interaction is not taken into account, the effect of the holiday or weekday explanatory variable remains in the form of simply raising the estimated value of the other explanatory variables on weekdays and lowering it on weekdays, but if the interaction is taken into consideration, for example. This is because the relationship between the explanatory variables can be reflected in the estimated value, such as raising the estimated value during the daytime on weekdays and suppressing it during the daytime on holidays.

Further, FIG. 8 shows a case where all the above four parameters are used as explanatory variables and regression analysis is performed in consideration of all interactions between the explanatory variables (Table 2, combination number 11). Due to the influence of each interaction term, a pattern such as "daytime on weekdays in a hot season" is incorporated into the regression equation, and highly accurate estimation is possible.

Next, the accuracy of the actual prediction was examined. The air conditioning load was predicted based on the 2008 data set using the regression equation generated using the 2007 data set as test data. For the prediction, all four parameters were used as explanatory variables, and a regression equation (Table 2, combination number 11) considering all interactions between the explanatory variables was used.

The result is shown in FIG. The predicted values (broken line) based on the 2007 data set are in good agreement with the measured values in 2008 shown by the solid line, and the air conditioning load can be predicted with sufficient accuracy for practical use.

Furthermore, when using the past outside air temperature as the explanatory variable, the case where the date at the time of predicting the air conditioning load and the date of the outside air temperature used as the explanatory variable deviate from each other was examined. For example, when trying to determine the start / stop schedule of the heat source unit 1 on the day by the previous day, even if the outside air temperature 24 hours before the time when the air conditioning load is to be predicted is referred to, it is not possible to predict all the times on the day. Can not. However, if the air conditioning load at the target time point can be accurately predicted by referring to the outside air temperature 48 hours ago, the start / stop schedule of the day can be determined by the previous day. In addition, if the air conditioning load can be predicted with sufficient accuracy even with reference to the outside air temperature at a time earlier than 2 days, the start / stop schedule can be determined earlier. Therefore, regarding the date that refers to the outside air temperature, we examined how much deviation is acceptable. For the prediction, all four parameters are used as explanatory variables, and a regression equation (Table 2, combination number 11) considering all interactions between the explanatory variables is used to determine the air conditioning load at a specific time point in FY2008 in 2007. Forecasts are based on outside air temperatures on dates before and after the year. The results are shown in the graph of FIG.

In general, if the coefficient of determination is 0.8 or more, it can be considered that the prediction can be made with sufficient accuracy. If the date deviation is within 5 days, the coefficient of determination of 0.8 or more is maintained, but if the deviation is 6 days or more, the coefficient of determination drops significantly and falls below 0.8. From the above, when using the outside air temperature as the explanatory variable, it is considered that the date deviation is practically acceptable as long as it is within 5 days. When the date for referring to the outside air temperature is shifted in this way, for example, by repeatedly using the past outside air temperature 24 hours before the desired time point n times, it is possible to perform an operation of predicting n days ahead.

Further, in Table 2 above, the case where all the interaction terms between the explanatory variables are incorporated into the regression equation has been described when considering the interaction, but it is not always necessary to use the interaction terms based on the combination of all the explanatory variables. , You may delete some interaction terms using the variable increase / decrease method. According to the research by the inventors of the present application, for example, when using four explanatory variables of outside air temperature, time, month, holiday or weekday, "outside air temperature x time x month" and "outside air temperature x time x month x holiday" When the "weekday?" Interaction is deleted, the coefficient of determination of the estimated value (when the air conditioning load in 2007 is predicted from the data in 2007) does not change compared to the case where it is not deleted, but the coefficient of determination is the same, but the predicted value (from the data in 2007) The coefficient of determination (when estimating the air conditioning load in 2008) has improved slightly. As described above, the selection of interaction terms may be effective not only in reducing the amount of calculation but also in improving the prediction accuracy.

In addition, regarding the case where the day of the week is used as the explanatory variable, the case where two types of dummy variables, holiday and weekday, are set has been described above, but it is also possible to set seven types of day, month, Tuesday, Wednesday, Thursday, Friday, and Saturday. However, according to the research by the inventors of the present application, it has been found that the accuracy of prediction is higher when the dummy variable of the day of the week is set to two types, holiday and weekday. If there are seven dummy variables for the day of the week, there is a possibility that overfitting has occurred as a result of too many cases. If the dummy variable for the day of the week is set in two ways, holiday or weekday, the amount of calculation can be further reduced and the prediction accuracy can be improved.

In this way, according to the method and system as in this embodiment, which predicts the air conditioning load by multiple regression analysis using a plurality of parameters related to the air conditioning load as explanatory variables, the air conditioning load is predicted with practically sufficient accuracy. be able to. At this time, in particular, if parameters selected from the outside air temperature, time, date, and day of the week are used, accurate prediction is possible. Furthermore, if at least three or more parameters are used as setting variables and the interaction term of the third order or higher is included in the regression equation in consideration of the interaction between each setting variable, the accuracy of the predicted value is further improved. be able to.

Further, when parameters selected from the outside air temperature, time, date, and day of day are used as explanatory variables, special data (for example, weather forecast data) is not required for creating a regression equation or predicting an air conditioning load. When creating the regression equation, if the outside air temperature is used as the explanatory variable, past test data may be used, and dummy variables may be set for other parameters. Also in the prediction, for example, the current measured value acquired by an outdoor temperature sensor (not shown) may be used. Therefore, it is possible to predict the air conditioning load without acquiring new data or purchasing weather forecast data with the operation of the air conditioning system, which is simple and can reduce the cost.

As described above, in the method for predicting the air conditioning load of the present embodiment, the air conditioning load is predicted by using multiple regression analysis based on a plurality of parameters related to the air conditioning load. In this way, the air conditioning load can be predicted with sufficient accuracy for practical use.

In the method of predicting the air conditioning load of this embodiment, some or all parameters selected from the following parameters are used as explanatory variables. In this way, the air conditioning load can be predicted accurately.
・ Outside air temperature ・ Date ・ Time ・ Day of the week

In the method for predicting the air conditioning load of this embodiment, at least three of the parameters can be used as explanatory variables, and a third-order or higher interaction term can be included in the regression equation. In this way, the accuracy of the predicted value can be further improved. In addition, the air conditioning load can be easily predicted and the cost can be suppressed.

In the method of predicting the air conditioning load of this embodiment, a part of the interaction term can be deleted by the variable increase / decrease method. In this way, the amount of calculation can be reduced and the accuracy can be improved.

In the method of predicting the air conditioning load of the present embodiment, when the day of the week is used as the explanatory variable, two dummy variables, holiday and weekday, can be set. In this way, the prediction accuracy can be improved while further reducing the amount of calculation.

Further, the air conditioning load prediction system of this embodiment is configured to be able to execute the above-mentioned air conditioning load prediction method. In this way, the air conditioning load can be predicted accurately.

Further, in the energy management method of the air conditioning system of the present embodiment, the operation schedule of the heat source unit 1 is determined based on the predicted value of the air conditioning load by the above-mentioned air conditioning load prediction method. In this way, the energy management of the air conditioning system can be suitably performed according to the predicted air conditioning load.

Further, the energy management system of the air conditioning system of the present embodiment includes a heat source machine 1 and a display unit 22 for displaying the predicted value of the air conditioning load by the above-mentioned air conditioning load prediction method, and the heat source machine is based on the predicted value. It is configured so that the operation schedule of 1 can be determined. In this way, the energy management of the air conditioning system can be suitably performed according to the predicted air conditioning load.

Therefore, according to the above embodiment, the air conditioning load can be predicted accurately and easily.

The method and system for predicting the air conditioning load of the present invention, and the energy management method and system for the air conditioning system are not limited to the above-described embodiments, and various changes are made within the scope of the gist of the present invention. Of course it can be added.

1 Heat source machine 22 Display (information terminal device)

Claims (8)

A method for predicting an air conditioning load, which comprises predicting an air conditioning load using multiple regression analysis based on a plurality of parameters related to the air conditioning load. The method for predicting an air conditioning load according to claim 1, wherein some or all of the parameters selected from the following parameters are used as explanatory variables.
・ Outside air temperature ・ Date ・ Time ・ Day of the week The method for predicting an air conditioning load according to claim 1 or 2, wherein at least three of the parameters are used as explanatory variables, and a third-order or higher interaction term is included in the regression equation. The method for predicting an air conditioning load according to claim 3, wherein a part of the interaction term is deleted by the variable increase / decrease method. The method for predicting an air conditioning load according to any one of claims 1 to 4, wherein two dummy variables, a holiday and a weekday, are set when the day of the week is used as an explanatory variable. An air-conditioning load prediction system according to any one of claims 1 to 5, wherein the air-conditioning load prediction method is feasibly configured. An energy management method for an air conditioning system, characterized in that an operation schedule of a heat source unit is determined based on a predicted value of the air conditioning load by the method for predicting the air conditioning load according to any one of claims 1 to 6. With a heat source machine
A display unit for displaying the predicted value of the air conditioning load according to the method for predicting the air conditioning load according to any one of claims 1 to 6 is provided.
An energy management system for an air conditioning system, characterized in that the operation schedule of the heat source machine can be determined based on the predicted value.

Images