JP2019066135A - Air-conditioning control system - Google Patents

Abstract

To provide an air-conditioning control system capable of performing appropriate air-conditioning control considering an installation situation and operating situation of a machine.SOLUTION: An air-conditioning control system 1 includes: a condition designation section 110 for designating a condition in an environment; a state amount detection section 140 for detecting state amount indicating the environment; a characteristic amount preparation section 210 for preparing characteristic amount characterizing the environment from the state amount; an inference calculation section 220 for inferring a control method for an air conditioner in the environment; an air-conditioning control section 400 for controlling the air conditioner on the basis of the inferred control method; a learning model generation section 500 for generating and updating a learning model through mechanical learning using the characteristic amount; and a learning model storage section 300 for storing the learning model in association with a combination of environment conditions.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an air conditioning control system, and more particularly to an air conditioning control system that controls air conditioning while switching a learning model according to the installation status and operation status of machines in a factory.

  The factory is provided with a plurality of machines (for example, machine tools), and in precision machining, an air conditioner is provided to control the environmental conditions in the factory because the temperature in the factory influences the machining accuracy. The production at the factory is instructed by the production plan, and the operation status and processing conditions of multiple machines are controlled so that the products instructed by the production plan are produced at the indicated delivery date, and the operation status of the air conditioner To control the temperature conditions (temperature and uniformity) in the factory required for processing.

  For example, in Patent Documents 1 and 2, a plurality of air conditioners are centrally controlled to set the temperature in the factory to a predetermined temperature as a technology for controlling the air conditioning to satisfy the temperature conditions in the factory required for processing. Techniques for keeping are disclosed.

Japanese Patent Application Publication No. 06-307703 JP, 2015-206519, A

  However, the temperature distribution in the factory differs depending on the installation status of machines (machine tools, robots, etc.) as heat sources installed in the factory and the operating status of each machine. Therefore, in order to realize the temperature conditions in the factory required for processing, it is necessary to control the air conditioner while taking into consideration the installation status and operation status of each machine. It is difficult to carry out appropriate air conditioning control.

  Although it is conceivable to introduce a machine learning device to control air conditioning, in order to create a general-purpose machine learning device (general purpose learning model) that can cope with various situations as described above. This requires a lot of state information to be detected in various situations, and also requires many parameters including situational data, which may cause known problems such as over-learning.

  SUMMARY OF THE INVENTION An object of the present invention is to provide an air conditioning control system capable of performing a wide range of appropriate air conditioning control in consideration of the installation state and the operation state of a machine.

  The air conditioning control system of the present invention solves the above-mentioned problems by providing a mechanism for switching a learning model used to determine the control operation of the air conditioning to be used according to the installation status and operation status of the machine in the factory. The air conditioning control system according to the present invention has a plurality of learning models, selects a learning model according to the installation status and operation status of machines in a factory, and detects state quantities detected from within the factory for the selected learning model. Machine learning is performed based on the above, and the learning model created in this way is used according to the installation status and operation status of the machine in the factory to control the air conditioner.

  Then, one aspect of the present invention is an air conditioning control system for controlling an air conditioner in an environment in which at least one machine is installed, comprising: a condition specifying unit for specifying a condition in the environment; The air conditioner is controlled based on a state quantity detection unit that detects an amount, an inference calculation unit that infers a control method of the air conditioner in the environment from the state amount, and a control method inferred by the inference calculation unit. An air conditioning control unit, a learning model generation unit that generates or updates a learning model by machine learning using the state quantity, and at least one learning model generated by the learning model generation unit is specified by the condition specification unit. A learning model storage unit for storing in association with a combination of the above, and the inference calculation unit is based on the condition in the environment specified by the condition specification unit. An air conditioning control system that calculates a control method of the air conditioner in an environment managed by the air conditioning control system 1 by selectively using at least one learning model among learning models stored in the learning model storage unit. It is.

  Another aspect of the present invention is a method of specifying a condition for controlling an air conditioner in an environment where at least one machine is installed, detecting a state quantity indicating the environment, and determining the condition in the environment from the state quantity. An air conditioning control that executes a step of inferring a control method of an air conditioner, a step of controlling the air conditioner based on the control method, and a step of generating or updating a learning model by machine learning using the state quantity A method, wherein the inferring step is based on the condition in the environment specified in the step of designating the condition out of at least one learning model previously associated with a combination of conditions in the environment. Select the learning model to be used and calculate the control method of the air conditioner in the environment using the selected learning model It is the air conditioning control method.

  Another aspect of the present invention is a learning model set in which each of a plurality of learning models is associated with a combination of conditions for controlling an air conditioner in an environment in which at least one machine is installed, the plurality of learning models Each is a learning model generated or updated based on a state quantity indicating the environment performed under conditions in the environment, and among the plurality of learning models, based on conditions set in the environment One learning model is selected, and the selected learning model is a learning model set used in a process of inferring a control method of the air conditioner in the environment.

  According to the present invention, machine learning can be performed on a learning model selected according to the installation status and operation status of machines in a factory, based on the amount of state detected in each status. It is possible to perform efficient machine learning while preventing it, and to control the air conditioner using the learning model selected according to the installation situation and operation situation of the machines in the factory. Accuracy is improved.

It is a rough functional block diagram of an air-conditioning control system by a 1st embodiment. It is a figure which illustrates the model of the environment which air-conditioning control system 1 manages. It is a rough functional block diagram of an air-conditioning control system by a 2nd embodiment. It is a schematic functional block diagram of an air-conditioning control system by a 3rd embodiment. It is a rough functional block diagram of an air-conditioning control system by a 4th embodiment. It is a schematic functional block diagram of the air-conditioning control system by a 5th embodiment. It is a schematic functional block diagram which shows the modification of the air-conditioning control system by 5th Embodiment. It is a schematic functional block diagram of the air-conditioning control system by a 6th embodiment. It is a schematic flowchart of the process performed on an air-conditioning control system. It is a schematic flowchart of the process performed on an air-conditioning control system.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a schematic functional block diagram of the air conditioning control system 1 according to the first embodiment. In each functional block shown in FIG. 1, a processor such as a CPU or GPU provided in a computer such as a numerical control device, a cell computer, a host computer, or a cloud server controls the operation of each part of the device according to each system program. Is realized by

  The air conditioning control system 1 of the present embodiment includes an environment management unit 100 that manages at least an environment (a room in which the air conditioner 130 is installed, etc.) to be observed and inferred, and an inference processing unit 200 that infers the environment state. And a learning model storage unit 300 that stores and manages a plurality of learning models. The air conditioning control system 1 according to the present embodiment further includes the air conditioning control unit 400 that controls air conditioning based on the result that the inference processing unit 200 infers the state of the environment, and creation and update of a learning model stored in the learning model storage unit 300. The learning model generation unit 500 is provided.

  The environment management unit 100 according to the present embodiment specifies the condition of control of the air conditioner 130 in an environment in which the air conditioner 130 controlled by the air conditioning control system 1 is installed (environment managed by the air conditioning control system 1) and the state of the environment. Acquisition of the state quantity indicating. The environment management unit 100 can be mounted on, for example, a central management device of an air conditioner or a numerical control device in an environment in which an environment managed by the air conditioning control system 1 is installed. The environment management unit 100 is connected to the wired / wireless network 150 and can exchange data with the machine 120 installed in the environment managed by the air conditioning control system 1 and the air conditioner 130 controlled by the air conditioning control system 1 Is configured.

  The condition designation unit 110 included in the environment management unit 100 is a functional unit that designates conditions (the installation status of the machine 120 and the air conditioner 130, the operation status of the machine 120, etc.) in the environment managed by the air conditioning control system 1.

  For example, as shown in FIG. 2, the installation status of the machine 120 and the air conditioner 130 divides the environment managed by the air conditioning control system 1 into a plurality of areas, and how the machine 120 as a heat source is installed in each area It is expressed by For example, in the example shown in FIG. 2, the installation condition of the machine 120 and the air conditioner 130 is as follows: the machine 120 as a heat source is installed in the areas 2, 7, 11 and 12; ing. The environment management unit 100 may acquire the installation status of the machine 120 and the air conditioner 130 from, for example, a setting made by an operator via an input / output device (not shown).

  The operating status of the machine 120 is acquired as the operating status of each machine 120 operating in the environment managed by the air conditioning control system 1. The operating status of the machine 120 can be classified into, for example, the calorific value according to the operation of the machine 120 such as stop, low operation (low temperature), medium operation (medium temperature), high operation (high temperature). The environmental management unit 100 is based on the operation status of the machine 120 (the internal temperature of the machine, the operation of the main shaft, the feed shaft, the load status, etc.) acquired from the machines 120 via the network 150. It may be determined.

  The condition designation unit 110 designates (outputs) the condition in the environment managed by the air conditioning control system 1 acquired in this manner to the learning model storage unit 300 and the learning model generation unit 500. The condition designation unit 110 has a role of informing each unit of the air conditioning control system 1 of the current condition of air conditioner control of the environment management unit 100 as a condition for selecting a learning model.

  The state quantity detection unit 140 included in the environment management unit 100 is a functional unit that detects the state of the environment managed by the air conditioning control system 1 as a state quantity. Examples of the state quantity of the environment managed by the air conditioning control system 1 include, for example, the ambient temperature and the internal temperature of each machine 120, the operating state, and the temperature of each area of the environment managed by the air conditioning control system 1. The state quantity detection unit 140 detects, as state quantities, detection values detected by temperature sensors provided in the machine 120 and the air conditioner 130 and temperature sensors installed in each area of the environment managed by the air conditioning control system 1. . The state quantities detected by the state quantity detection unit 140 are output to the inference processing unit 200 and the learning model generation unit 500.

  The inference processing unit 200 of the present embodiment observes the state of the environment managed by the air conditioning control system 1 acquired from the environment management unit 100, and infers the environment managed by the air conditioning control system 1 based on the observation result. . The inference processing unit 200 can be mounted on, for example, an air conditioning control device, a cell computer, a host computer, a cloud server, a machine learning device, or the like.

  The feature amount creating unit 210 included in the inference processing unit 200 is a functional unit that creates a feature amount indicating a feature of the environment managed by the air conditioning control system 1 based on the state amount detected by the state amount detecting unit 140. The feature quantity indicating the feature of the environment managed by the air conditioning control system 1 created by the feature quantity creating unit 210 is information that is useful as a material for the air conditioning control determination by the air conditioning control unit 400. Further, the feature quantities indicating the features of the environment managed by the air conditioning control system 1 created by the feature quantity creating unit 210 are input data when the inference calculating unit 220 described later performs inference using a learning model. The feature quantity indicating the feature of the environment managed by the air conditioning control system 1 created by the feature quantity creation unit 210 is, for example, a predetermined peripheral temperature of each machine 120 detected by the state quantity detection unit 140 for a predetermined period in the past. It may be sampled at a sampling cycle, for example, it may be a peak value within a predetermined period in the past of the operation status of each machine 120 detected by the state quantity detection unit 140, for example, the state quantity detection unit 140 Signal processing such as integrating and converting the temperature of each detected area to the time domain frequency domain, normalizing the amplitude or power density, adapting it to a transfer function, or reducing the dimension to a specific time or frequency width It may be a combination of The feature amount creating unit 210 performs preprocessing to normalize the state amount detected by the state amount detecting unit 140 so that the inference calculation unit 220 can handle it.

  The inference calculation unit 220 included in the inference processing unit 200 includes the learning model selected from the learning model storage unit 300 based on the conditions in the environment managed by the current air conditioning control system 1, and the feature amount created by the feature amount creation unit 210. And the functional means for inferring the control method of the air conditioner 130 in the environment managed by the air conditioning control system 1. The inference calculation unit 220 is realized by applying a learning model stored in the learning model storage unit 300 to a platform capable of executing inference processing by machine learning. The inference calculation unit 220 may be for performing inference processing using, for example, a multi-layered neural network, and uses a known learning algorithm as machine learning such as a Bayesian network, a support vector machine, or a mixed Gaussian model. It may be for performing inference processing. The inference calculation unit 220 may be for performing inference processing using a learning algorithm such as supervised learning, unsupervised learning, reinforcement learning, and the like. In addition, the inference calculation unit 220 may be able to execute inference processing based on a plurality of types of learning algorithms. The inference calculation unit 220 configures a machine learning device based on a learning model selected from the learning model storage unit 300 for machine learning, and uses the feature amount created by the feature amount creating unit 210 as input data of the machine learning device. The control method of the air conditioner 130 in the environment managed by the air conditioning control system 1 is inferred by executing the inference process. The control method of the air conditioner 130 as a result of inference by the inference calculation unit 220 is, for example, the set temperature and the wind direction (including the waving operation and the like) of each air conditioner 130 installed in the environment managed by the air conditioning control system 1 Etc.

  The learning model storage unit 300 according to the present embodiment is a functional unit capable of storing a plurality of learning models associated with a combination of conditions in the environment managed by the air conditioning control system 1 designated by the condition designation unit 110. . The learning model storage unit 300 can be mounted on, for example, the environment management unit 100, a cell computer, a host computer, a cloud server, a database server, or the like.

  The learning model storage unit 300 includes a plurality of learning models 1 associated with a combination of conditions (such as the installation status and operation status of machines in a factory) in the environment managed by the air conditioning control system 1 designated by the condition designation unit 110. , 2, ..., N are stored. The combination of the conditions (the installation status and operation status of machines in the factory, etc.) in the environment managed by the air conditioning control system 1 mentioned here is the combination of the values, the range of values, the enumeration of the values that each condition can take. For example, when the environment managed by the air conditioning control system 1 is modeled as shown in FIG. 2, for example, a matrix (None, machine (high temperature), , None, machine (medium temperature), none, none, none, machine (medium temperature, machine (stop)) can be used as one of the combinations of conditions in the environment controlled by the air conditioning control system 1.

  The learning model stored in the learning model storage unit 300 is stored as information capable of constructing one learning model adapted to the inference processing in the inference calculation unit 220. For example, when the learning model stored in the learning model storage unit 300 is a learning model using a learning algorithm of a multi-layer neural network, the number of neurons (perceptrons) in each layer, weight parameters between neurons (perceptrons) in each layer, etc. In the case of a learning model using a Bayesian network learning algorithm, it can be stored as a transition probability between nodes constituting the Bayesian network and the like. Each of the learning models stored in the learning model storage unit 300 may be learning models using the same learning algorithm, or may be learning models using different learning algorithms. The learning model may be any learning algorithm that can be used for inference processing.

  The learning model storage unit 300 may store one learning model in association with a combination of conditions in an environment managed by one air conditioning control system 1, and in an environment managed by one air conditioning control system 1. A learning model using two or more different learning algorithms may be associated with a combination of conditions and stored. The learning model storage unit 300 may associate and store learning models using different learning algorithms for each combination of conditions in the environment managed by the plurality of air conditioning control systems 1 in which the range of the combination is superimposed. At this time, the learning model storage unit 300 further determines, for example, the use conditions such as the required processing capacity and the type of learning algorithm for the learning model corresponding to the combination of the conditions in the environment managed by the air conditioning control system 1. It becomes possible to select the learning model according to the inference calculation part 220 which differs in the inference processing which can be performed, or processing capacity with respect to the combination of the conditions in the environment which the air-conditioning control system 1 manages.

  When the learning model storage unit 300 receives an external request for reading / writing a learning model including a combination of conditions in the environment managed by the air conditioning control system 1, the learning model storage unit 300 associates it with the combination of conditions in the environment managed by the air conditioning control system 1. Read / write the stored learning model. At this time, the reading / writing request of the learning model may include information on inference processing and processing capacity that can be executed by the inference calculation unit 220, and in such a case, the learning model storage unit 300 performs air conditioning. Reading / writing is performed on a combination of conditions in the environment managed by the control system 1 and a learning model associated with the inference processing that can be executed by the inference calculation unit 220 and the processing capacity. The learning model storage unit 300 reads the learning model associated with (the combination of) the conditions based on the condition designated by the condition designation unit 110 in response to the read / write request of the learning model from the outside. / A function may be provided to enable writing. By providing such a function, it is not necessary to provide the inference calculation unit 220 or the learning model generation unit 500 with a function of requesting a learning model based on the condition designated by the condition designation unit 110.

  The learning model storage unit 300 encrypts and stores the learning model generated by the learning model generation unit 500, and decodes the encrypted learning model when the learning model is read by the inference calculation unit 220. It is good.

  The air conditioning control unit 400 operates the air conditioner 130 installed in the environment managed by the air conditioning control system 1 based on the control method of the air conditioner 130 in the environment managed by the air conditioning control system 1 inferred by the inference processing unit 200. Function means for controlling The air conditioning control unit 400 controls each air conditioner 130 by transmitting a control command to each air conditioner 130 via the network 150, for example. The air conditioning control unit 400 may control each air conditioner 130 via a communication channel (infrared, other wireless means, etc.) different from the network 150.

  The learning model generation unit 500 is a condition amount in the environment managed by the air conditioning control system 1 designated by the condition designation unit 110, and a feature amount indicating a feature of the environment managed by the air conditioning control system 1 created by the feature amount creation unit 210. , And is a functional unit that generates or updates (machine learning) a learning model stored in the learning model storage unit 300. The learning model generation unit 500 selects a learning model to be generated or updated based on the conditions in the environment managed by the air conditioning control system 1 designated by the condition designation unit 110, and the feature amount for the selected learning model. Machine learning is performed using feature quantities that indicate the features of the environment managed by the air conditioning control system 1 created by the creation unit 210. The timing at which the learning model generation unit 500 performs learning is performed, for example, when the operator manually changes the setting of each air conditioner 130. In this case, the learning model generation unit 500 controls the environment managed by the air conditioning control system 1 created by the feature amount creating unit 210 with respect to the learning model selected based on the conditions in the environment managed by the air conditioning control system 1. A learning parameter is generated or updated (machine learning) using, as a state variable, a feature amount indicating the feature of the above and a set temperature, a wind direction, and the like of each air conditioner 130 as label data.

  When the learning model generation unit 500 does not store the learning model associated with (the combination of) the conditions in the environment managed by the air conditioning control system 1 designated by the condition designation unit 110 in the learning model storage unit 300, A learning model associated with the condition (combination of) the condition in the environment managed by the air conditioning control system 1 designated by the condition designation unit 110 is newly generated as a learning model associated with the condition (combination thereof). If it is stored in 300, the learning model is updated by performing machine learning on the learning model. When a plurality of learning models associated with (a combination of) the conditions in the environment managed by the air conditioning control system 1 designated by the condition designation unit 110 are stored in the learning model generation unit 500, Machine learning may be performed on each learning model, and machine learning may be performed on only some learning models based on learning processing and processing capabilities that can be executed by the learning model generation unit 500. You may do so.

  The learning model generation unit 500 may change a learning model stored in the learning model storage unit 300 to generate a new learning model. As an example of modification of the learning model by the learning model generation unit 500, for example, generation of a distillation model is exemplified. The distillation model is a learned model obtained by performing learning from 1 in another machine learner using an output obtained for an input to a machine learner incorporating a learned model. The learning model generation unit 500 can store and use the distillation model obtained through such a process (referred to as a distillation process) in the learning model storage unit 300 as a new learning model. In general, the distillation model is smaller in size than the original learned model, and can be as accurate as the original learned model, so it is more suitable for distribution to other computers via a network or the like. As another example of modification of the learning model by the learning model generation unit 500, integration of the learning model is illustrated. When the learning model generation unit 500 has two or more learning models having similar structures stored in association with (a combination of) conditions in the environment managed by the air conditioning control system 1, for example, the value of each weight parameter is In the case where it is within the predetermined predetermined threshold, (the combination of) the conditions in the environment managed by the air conditioning control system 1 associated with these learning models are integrated and then the structure is similar 2 in association with this. One of the learning models described above may be stored.

  FIG. 3 is a schematic functional block diagram of the air conditioning control system 1 according to the second embodiment. In the air conditioning control system 1 of the present embodiment, each functional block is mounted on one air conditioning control device 2 (built on a central management device of an air conditioner, a numerical control device, etc.). By configuring in this manner, the air conditioning control system 1 according to the present embodiment is managed by the air conditioning control system 1 using a different learning model according to the installation status and operation status of the machine 120 in the environment managed by the air conditioning control system 1. The control method of the air conditioner 130 in the environment to be set is inferred, and the air conditioner 130 in the environment managed by the air conditioning control system 1 is controlled. Further, each air conditioning control device 2 can generate / update each learning model according to the conditions in the environment managed by the air conditioning control system 1.

  FIG. 4 is a schematic functional block diagram of the air conditioning control system 1 according to the third embodiment. In the air conditioning control system 1 of the present embodiment, the environment management unit 100, the inference processing unit 200, and the air conditioning control unit 400 are mounted on the air conditioning control device 2, and the learning model storage unit 300 and the learning model generating unit 500 are provided. It is mounted on a machine learning device 3 connected to the air conditioning control device 2 via a standard interface or network. The machine learning device 3 may be implemented on a cell computer, a host computer, a cloud server, or a database server. By configuring in this way, the inference process using the learned model, which is a relatively light process, is executed on the air conditioning control device 2, and the process of creating / updating the learning model, which is a relatively heavy process, is performed. Since it can be performed on the learning device 3, the operation of the air conditioning control system 1 can be performed without interfering with the original operation of the air conditioning control device 2.

  FIG. 5 is a schematic functional block diagram of the air conditioning control system 1 according to the fourth embodiment. In the air conditioning control system 1 of the present embodiment, the environment management unit 100 is mounted on the air conditioning control device 2, and the inference calculation unit 220, the learning model storage unit 300, and the learning model generation unit 500 are standard with the air conditioning control device 2. It is mounted on a machine learning device 3 connected via an interface or a network. In addition, an air conditioning control unit 400 is separately prepared. In the air conditioning control system 1 according to the present embodiment, the state quantities detected by the state quantity detection unit 140 can be used as they are for the inference process by the inference calculation unit 220 or the learning model generation / update process by the learning model generation unit 500. The configuration of the feature quantity creation unit 210 is omitted on the assumption that the data is data. By configuring in this manner, the inference process using the learned model and the process of generating / updating the learning model can be executed on the machine learning device 3, so the original operation of the air conditioning control device 2 can be realized. The operation of the air conditioning control system 1 can be performed without interference.

  FIG. 6 is a schematic functional block diagram of the air conditioning control system 1 according to the fifth embodiment. In the air conditioning control system 1 of the present embodiment, each functional block is mounted on one air conditioning control device 2. In the air conditioning control system 1 of the present embodiment, the learning model storage unit 300 already stores a plurality of learned learning models associated with combinations of conditions in the environment managed by the air conditioning control system 1. The configuration of the learning model generation unit 500 is omitted on the assumption that no model is generated / updated. By configuring in this manner, the air conditioning control system 1 according to the present embodiment uses, for example, different learning models according to the installation status and operation status of the machine 120 in the environment managed by the air conditioning control system 1. The control method of the air conditioner 130 in the environment to be managed is inferred, and the air conditioner 130 in the environment managed by the air conditioning control system 1 is controlled. In addition, since the self-directed learning model is not updated, it can be adopted, for example, as the configuration of the air conditioning control device 2 shipped to the customer.

  FIG. 7 is a schematic functional block diagram showing a modification of the air conditioning control system 1 according to the fifth embodiment. The air conditioning control system 1 of the present embodiment is an example in which the learning model storage unit 300 is mounted on the external storage 4 connected to the air conditioning control device 2 in the fifth embodiment. In this modification, storing a learning model with a large capacity in the external storage 4 makes it possible to use many learning models and also allows reading of the learning model without going through a network etc. It is effective when you need real-time capability.

  FIG. 8 is a schematic functional block diagram of the air conditioning control system 1 according to the sixth embodiment. In the air conditioning control system 1 of the present embodiment, the environment management unit 100 is mounted on the air conditioning control device 2, and the inference calculation unit 220 and the learning model storage unit 300 with the air conditioning control device 2 via a standard interface or network. It is mounted on the machine learning device 3 connected. The machine learning device 3 may be implemented on a cell computer, a host computer, a cloud server, or a database server. In the air conditioning control system 1 of the present embodiment, the learning model storage unit 300 already stores a plurality of learned learning models associated with combinations of conditions in the environment managed by the air conditioning control system 1. The configuration of the learning model generation unit 500 is omitted on the assumption that no model is generated / updated. Further, in the air conditioning control system 1 of the present embodiment, assuming that the amount of state detected by the state amount detection unit 140 is data that can be directly used for inference processing by the inference calculation unit 220, the feature amount generation unit 210 Is omitted. By configuring in this manner, the air conditioning control system 1 according to the present embodiment uses, for example, different learning models according to the installation status and operation status of the machine 120 in the environment managed by the air conditioning control system 1. The control method of the air conditioner 130 in the environment to be managed is inferred, and the air conditioner 130 in the environment managed by the air conditioning control system 1 is controlled. In addition, since the self-directed learning model is not updated, it can be adopted, for example, as the configuration of the air conditioning control device 2 shipped to the customer.

FIG. 9 is a schematic flowchart of the process executed by the air conditioning control system 1 of the present invention. The flowchart shown in FIG. 9 exemplifies the flow of processing when the learning model in the air conditioning control system 1 is not updated (the fifth and sixth embodiments).
[Step SA01] The condition designation unit 110 designates conditions in the environment managed by the air conditioning control system 1.
[Step SA02] The state quantity detection unit 140 detects the state of the environment managed by the air conditioning control system 1 as a state quantity.

[Step SA03] The feature quantity creation unit 210 creates a feature quantity indicating the feature of the environment managed by the air conditioning control system 1 based on the state quantity detected in step SA02.
[Step SA04] The inference calculation unit 220 selects and reads from the learning model storage unit 300 a learning model corresponding to the condition in the environment managed by the air conditioning control system 1 designated in step SA01 for use in inference. .
[Step SA05] The inference calculation unit 220 infers the control method of the air conditioner 130 in the environment managed by the air conditioning control system 1 based on the learning model read in step SA04 and the feature amount created in step SA03.
[Step SA06] The air conditioning control unit 400 controls the air conditioning based on the air conditioning control method inferred in step SA05.

FIG. 10 is a schematic flowchart of the process executed by the air conditioning control system 1 of the present invention. The flowchart shown in FIG. 10 exemplifies the flow of processing when generation / update of a learning model in the air conditioning control system 1 is performed (first to fourth embodiments).
[Step SB01] The condition designation unit 110 designates conditions in the environment managed by the air conditioning control system 1.
[Step SB02] The state quantity detection unit 140 detects the state of the environment managed by the air conditioning control system 1 as a state quantity.

[Step SB03] The feature quantity creation unit 210 creates a feature quantity indicating the feature of the environment managed by the air conditioning control system 1 based on the state quantity detected in step SB02.
[Step SB04] The inference calculation unit 220 selects and reads from the learning model storage unit 300 a learning model corresponding to the condition in the environment managed by the air conditioning control system 1 designated in step SB01 for use in inference. .

[Step SB05] The learning model generation unit 500 determines whether the learning model storage unit 300 has generated a learned learning model corresponding to the conditions in the environment managed by the air conditioning control system 1 designated in step SB01. judge. If the learned model has been generated, the process proceeds to step SB07. If the learned model is not generated, the process proceeds to step SB06.
[Step SB06] The learning model generation unit 500 generates and updates a learning model corresponding to the condition in the environment managed by the air conditioning control system 1 designated in step SB01 based on the feature quantity created in step SB03. Then, the process proceeds to step SB01.

[Step SB07] The inference calculation unit 220 infers the control method of the air conditioner 130 in the environment managed by the air conditioning control system 1 based on the learning model read out in step SB04 and the feature amount created in step SB03.
[Step SB08] The air conditioning control unit 400 controls the air conditioning based on the air conditioning control method inferred in step SB05.

  As mentioned above, although embodiment of this invention was described, this invention can be implemented in various aspects by adding an appropriate change, without being limited only to the example of embodiment mentioned above.

Reference Signs List 1 air conditioning control system 2 air conditioning control device 3 machine learning device 4 external storage 100 environment management unit 110 condition designation unit 120 machine 130 air conditioner 140 state quantity detection unit 150 network 200 inference processing unit 210 feature amount generation unit 220 inference calculation unit 300 learning Model storage unit 400 Air conditioning control unit 500 Learning model generation unit

Claims (12)

An air conditioning control system for controlling an air conditioner in an environment in which at least one machine is installed, comprising:
A condition specification unit that specifies a condition in the environment;
A state amount detection unit that detects a state amount indicating the state of the environment;
An inference calculation unit for inferring a control method of the air conditioner in the environment from the state quantity;
An air conditioning control unit that controls the air conditioner based on the control method inferred by the inference calculation unit;
A learning model generation unit that generates or updates a learning model by machine learning using the state quantities;
A learning model storage unit that stores at least one learning model generated by the learning model generation unit in association with a combination of conditions specified by the condition specification unit;
Equipped with
The inference calculation unit selectively uses at least one learning model among the learning models stored in the learning model storage unit based on the condition in the environment designated by the condition designation unit. Calculate the control method of the air conditioner in the environment controlled by 1;
Air conditioning control system. The apparatus further comprises a feature amount creation unit that creates a feature amount that characterizes the environment from the state amounts detected by the state amount detection unit,
The inference calculation unit infers a control method of the air conditioner in the environment from the feature amount,
The learning model generation unit generates or updates a learning model by machine learning using the feature amount.
The air conditioning control system according to claim 1. The learning model generation unit generates a new learning model by modifying the existing learning model stored in the learning model storage unit.
The air conditioning control system according to claim 1 or 2. The learning model storage unit encrypts and stores the learning model generated by the learning model generation unit, and decodes the encrypted learning model when the learning model is read by the inference calculation unit.
The air conditioning control system according to any one of claims 1 to 3. An air conditioning control system for controlling an air conditioner in an environment in which at least one machine is installed, comprising:
A condition specification unit that specifies a condition in the environment;
A state amount detection unit that detects a state amount indicating the environment;
An inference calculation unit for inferring a control method of the air conditioner in the environment from the state quantity;
An air conditioning control unit that controls the air conditioner based on the control method inferred by the inference calculation unit;
A learning model storage unit storing at least one learning model previously associated with a combination of conditions in the environment;
Equipped with
The inference calculation unit selectively uses at least one learning model among the learning models stored in the learning model storage unit based on the condition in the environment designated by the condition designation unit. An air conditioning control system that calculates the control method of the air conditioner. The apparatus further comprises a feature amount creation unit that creates a feature amount that characterizes the environment from the state amount.
The inference calculation unit infers a control method of the air conditioner in an environment managed by the air conditioning control system 1 from the feature amount.
The air conditioning control system according to claim 5. A condition designation unit according to any one of claims 1 to 6, a state quantity detection unit,
Air conditioning control device with. Specifying conditions for controlling the air conditioner in an environment in which at least one machine is installed;
Detecting a state quantity indicative of the environment;
Inferring a control method of the air conditioner in the environment from the state quantity;
Controlling the air conditioner based on the control method;
Generating or updating a learning model by machine learning using the state quantities;
An air conditioning control method for performing
The inferring step uses a learning model using a condition in the environment specified in the step of designating the condition among at least one learning model pre-associated with a combination of conditions in the environment. Calculating the control method of the air conditioner in the environment using the selected and selected learning model;
Air conditioning control method. Further executing a step of creating a feature that characterizes the environment from the state quantity;
The inferring step infers a control method of the air conditioner in the environment from the feature amount;
The step of generating or updating the learning model generates or updates a learning model by machine learning using the feature amount.
The air conditioning control method according to claim 8. Specifying conditions for controlling the air conditioner in an environment in which at least one machine is installed;
Detecting a state quantity indicative of the environment;
Inferring a control method of the air conditioner in the environment from the state quantity;
Controlling the air conditioner based on the control method;
An air conditioning control method for performing
The step of inferring selects a learning model to be used based on the condition in the environment designated in the step of designating the condition out of at least one learning model previously associated with the combination of the condition in the environment. Calculating the control method of the air conditioner in the environment using the selected learning model,
Air conditioning control method. Further executing a step of creating a feature that characterizes the environment from the state quantity;
The inferring step infers a control method of the air conditioner in the environment from the feature amount.
The air conditioning control method according to claim 10. A learning model set comprising each of a plurality of learning models associated with a combination of conditions for controlling an air conditioner in an environment in which at least one machine is installed,
Each of the plurality of learning models is a learning model generated or updated based on a state quantity indicating the environment performed under conditions in the environment,
Among the plurality of learning models, one learning model is selected based on the conditions set in the environment, and the selected learning model is used for a process of inferring a control method of the air conditioner in the environment.
Learning model set.

Images