JPH0835706A - Air conditioning heat load predicting system - Google Patents

Abstract

PURPOSE:To accurately predict the air conditioning heat load by sequentially correcting a predictive model by using the learning function of a neural network based on the comparison error of the air conditioning heat load predicted value calculated in the previous day based on past observation data and tentative predictive model with the heat load performance value of an appointed day. CONSTITUTION:At the time of starting service, a tentative model forming module is executed, and an air conditioning heat load simulation is conducted for an object building. Then, the output value of a heat load predicting neural network (N.N.) is compared with an air conditioning heat load simulation signal, the N.N. is tuned in response to the error, and an N.N. tentative model for predicting a heat load is formed. An air conditioning load predicted value of next day is calculated from appointed day environmental factor data from a sensor 2 and next day weather factor data from a weather forecast data collector 4, compared with next day air conditioning heat load performance value, the comparison error is corrected, and the corrected value makes use of air conditioning heat load prediction.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a building air-conditioning heat load prediction system, and in particular, it uses a neural network (hereinafter abbreviated as NN) to detect the temperature and humidity of indoor and outdoor air. Environmental factors of the day and predicted temperature of the next day
The present invention relates to an air conditioning heat load prediction system that predicts an air conditioning heat load of the building from a predicted value of a next-day environmental element such as humidity.

[0002]

2. Description of the Related Art FIG. 3 is a schematic diagram showing an outline of a conventional heat storage type air conditioning facility for a building during cooling. FIG.
In, the heat storage tank 80 for storing heat is installed by using a slab of a building or the like. The water in the heat storage tank 80 is 1
By the secondary pump 84, cold water produced using heat source equipment such as the refrigerators 82, 82 is passed through the primary pipe inlet side 86, and the heat storage tank 80 is passed through the primary pipe outlet side 88.
It is stored in the vicinity of the low temperature part 92. After that, the secondary pump 9
0 through the secondary pipe inlet side 94 through the air conditioner 9
6, 96, 96. The water heated via the air conditioner 96 is output to the vicinity of the high temperature part 100 in the heat storage tank 80 via the outlet side of the secondary pipe. As described above, the cold water used for cooling the building has a cycle in which the primary piping system starts from the high temperature section 100 and returns to the low temperature section 92, and the secondary piping system starts from the low temperature section 92 and the high temperature section. In the cycle of returning to 100, all are supplied via the heat storage tank 80, and during heating, the low temperature part and the high temperature part of the heat storage tank are reversed and supplied.
This heat storage type air conditioning system aims to reduce air conditioning running energy by preparing cold (warm) water at night by using late-night electric power, which is cheap, and responding to the air conditioning heat load of the next day. It is often installed in. For this reason, predicting the air-conditioning heat load of the next day has become important in recent years in order to store heat just enough.

[0003]

However, the operation of the heat storage type air conditioning equipment system of this type of conventional example is often dependent on the intuition and experience of the operator who operates the equipment, and the heat storage amount is insufficient for the air conditioning heat load. When it disappears, it is necessary to chase and operate the heat source device, so that heat is often stored more than necessary, and there is a problem that the feature of the heat storage type air conditioning equipment system of reducing air conditioning running energy cannot be fully utilized.

Further, the various proposed prediction methods have a problem that a large amount of actual measurement data is required and the characteristics of an actual building are not taken into consideration, so that the error is large. The present invention has been made in view of such circumstances, and an object thereof is to provide an air conditioning heat load prediction system that appropriately predicts an air conditioning heat load amount for supporting proper operation of a heat storage type air conditioning facility system. And

[0005]

In order to achieve the above object, the present invention provides a sensor unit for detecting environmental element data such as temperature and humidity in a building of a heat storage type air conditioning system, and a collecting unit for collecting weather forecast data. And a processing unit that processes the environmental element data and the weather prediction data to give a heat storage index of the next day to the building and predicts an air conditioning heat load of the building. In the heat load prediction system, the processing unit uses a neural network having a learning function from the environmental element data and the pattern of the air conditioning heat load for the building, and a provisional model creation module that creates a provisional prediction model. , Based on the comparison error between the air-conditioning heat load predicted value and the actual heat load value, the prediction function is sequentially corrected by using the learning function of the neural network. It is characterized with the thermal load forecast module for predicting the load, in that it consists of.

[0006]

According to the present invention, based on the provisional prediction model created by past observation data or simulation data, based on the comparison error between the air conditioning heat load predicted value calculated the day before and the heat load actual value of the day. N. N. There is an effect that a prediction model in which the characteristics of the target building are automatically added is generated by successively correcting the prediction model using the learning function of.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An air conditioning heat load prediction system according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an apparatus that realizes a building air conditioning heat load prediction system according to an embodiment of the present invention. FIG. 2 is a flowchart showing an operating procedure of the air conditioning heat load prediction system according to the embodiment of the present invention.

In FIG. 1, the air-conditioning heat load predicting apparatus includes a sensor unit 2 for measuring environmental elements on the day, a weather forecast data collecting unit 4 for collecting the next-day weather elements, and a signal cable 6 from the sensor unit 2. , 6 and the next day's weather element data transmitted from the weather prediction data collection unit 4 via the signal cable 8 are input to output the air conditioning heat load prediction value of the next day. Main CPU
10 and an air conditioning heat load predicted value display unit 12 that displays the air conditioning heat load predicted value. The sensor unit 2 uses an indoor temperature sensor 14 in the representative room, an indoor humidity sensor 16 in the representative room, an outside air temperature sensor 18, an outside air humidity sensor 20, and a solar radiation sensor 22, respectively.
Measure the indoor humidity of the representative room, the outside air temperature, the outside air humidity, and the amount of solar radiation on the day. In the weather, the collection unit 4 can access the outside air weather prediction database 24 online 25, and automatically calculates the predicted maximum temperature data 21, predicted minimum temperature data 28, predicted maximum humidity data 30, and predicted minimum humidity data 32 of the next day. Collect.

In the main CPU 10, the environmental element data of the day measured by the sensor unit 2 and the weather forecast data collecting unit 4
Before the heat storage operation is started (usually before 10:00 pm), the next day meteorological element data collected in step 1 is taken in as necessary, and is processed with the data regarding the day of the week including the singular day held internally. , Outputs the predicted air conditioning heat load for the next day. This predicted value is displayed as a result signal output from the main CPU 10 on the air conditioning heat load predicted value display unit 12 to assist the equipment operation operator.

Next, an operation procedure of the air conditioning heat load prediction system according to the embodiment of the present invention will be described with reference to FIG. In this embodiment, a case where a building is a new construction or a case where it is difficult to obtain past data even for an existing building will be described. When the operation is started (step 33), the provisional model creation module 35 is implemented, and as a preliminary examination, an air conditioning heat load simulation 36 is performed on the target building using the standard weather database 34. And N. N. According to the comparison error between the output value 40 of 38 and the target building air conditioning heat load simulation signal 42,
The N.V. for heat load prediction is set so that the error of the air-conditioning heat load prediction value becomes a certain value or less. N. 19 is tuned (step 44), and N. N. Create a temporary model. When it is determined in step 46 that the device is put to practical use, the following operation is repeated according to a fixed time loop. If it is determined in step 46 that it will not be put to practical use, the process returns to start step 33, and the entire procedure of the provisional model creation module 35 is repeated.

At the beginning of operation, the N.V. N. Using the provisional model, the same day environmental element data transmitted from the sensor unit 2 shown in FIG. 1,
An air-conditioning heat load prediction value for the next day is calculated from the next-day weather element data transmitted from the weather prediction data collection unit 4 and the day of the week (including a peculiar day) data internally stored (step 5).
0). Furthermore, the air conditioning heat load actual value 54 found on the next day
And the air-conditioning heat load predicted value 52 are compared, the weight is corrected according to the comparison error (step 56), and the N.V. N. To make additional learning and update,
The model is applied in the next time loop 58. Using such a learning function, N. N. , The prediction model is sequentially modified. Therefore, by repeating the time loop 58, the air-conditioning heat load prediction model to which the characteristics of the target building are added is automatically generated.

By the procedure described above, an air conditioning heat load prediction system can be constructed even when sufficient actual measurement data cannot be obtained, and the air conditioning heat load can be predicted in consideration of the characteristics of the actual building. An air conditioning heat load prediction system can be provided. The present invention is not limited to the above-described embodiment, and the next-day weather forecast data to be collected is:
In particular, it is not necessary to obtain it from the external weather forecast database 24 shown in FIG. It may be supplied from a separately prepared weather forecast system into this air conditioning heat load forecast system.

Further, the N.V. for heat load prediction shown in FIG. N. Three
The input items of 8 may include the operating rate of the internal equipment and the fluctuation information data of the personnel according to the operational characteristics of the building. Further, at the time of preliminary examination, the combination of items may be changed so that the error of the predicted air-conditioning heat load value becomes a certain value or less.

[0014]

According to the air conditioning heat load prediction system of the present invention, it is possible to accurately predict the air conditioning heat load in consideration of the actual characteristics of the building. Further, it is possible to reduce variations in the amount of night heat storage due to the operator's intuition and experience when operating the heat storage type air conditioning equipment. Therefore, there is an effect that it is possible to further reduce the air-conditioning running energy, which is the original merit of the heat storage type air-conditioning system.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an apparatus that realizes an air conditioning heat load prediction system for a building of the present invention.

FIG. 2 is a flowchart showing an operation procedure of the air conditioning heat load prediction system according to the present invention.

FIG. 3 is a schematic diagram showing an overview of a conventional heat storage type air conditioning equipment.

[Explanation of symbols]

 2 ... Sensor unit 4 ... Weather forecast data collection unit 10 ... Main CPU 12 ... Air conditioning heat load forecast value display unit

Claims (3)

[Claims] 1. A sensor unit for detecting environmental element data such as temperature and humidity in a building of a heat storage type air conditioning system, a collection unit for collecting weather forecast data, and a heat storage index of the next day for the building. In an air conditioning heat load prediction system including a processing unit that processes the environmental element data and the weather prediction data to give a prediction of the air conditioning heat load of the building, the processing unit is the environmental element data. From the pattern of the air conditioning heat load on the building and the building, using a neural network having a learning function, a provisional model creation module that creates a provisional prediction model, and a comparison error between the air conditioning heat load prediction value and the heat load actual value. A heat load prediction module for predicting the air conditioning heat load by sequentially modifying a prediction model using a learning function of a neural network based on Air conditioning heat load prediction system, characterized and. 2. The air conditioning heat load prediction system according to claim 1, wherein the provisional prediction model is created by an air conditioning heat load simulation for the building. 3. The air conditioning heat load prediction system according to claim 1, wherein the weather prediction data is collected online from a weather information database.