Evaluation device, evaluation method, recording medium recording evaluation program, control device, and recording medium recording control program

Abstract

Provided is an evaluation device including: an environment data acquisition unit that acquires environment data indicating an operation environment of a device; an achievement data acquisition unit that acquires achievement data indicating an operation achievement of the device; an estimation unit that estimates, based on the environment data and the achievement data of a learning target period, an operation achievement based on operation of the learning target period in an operation environment of an evaluation target period; an evaluation unit that calculates an index that evaluates, relative to an estimated value of the estimated operation achievement, an actual measured value of the operation achievement of the evaluation target period; and an output unit that outputs the index.

Description

Technical Field

[0001] The present invention relates to an evaluation device, an evaluation method, a recording medium recording an evaluation program, a control device, and a recording medium recording a control program. Background Art

[0002] Patent Document 1 describes "providing an energy-saving effect amount calculation method and apparatus capable of improving the reliability of the energy-saving effect amount."

[0003] Patent Document 1: Japanese Patent No. 4426243 Summary of the Invention

[0004] In the first aspect of the present invention, an evaluation device is provided. The evaluation device may include an environment data acquisition unit that acquires environment data representing the operating environment of the equipment. The evaluation device may include a result data acquisition unit that acquires result data representing the operating results of the equipment. The evaluation device may include an estimation unit that estimates the operating results based on the operation during the learning object period under the operating environment during the evaluation object period based on the environment data and the result data during the learning object period. The evaluation device may include an evaluation unit that calculates an index that evaluates the actual measurement value of the operating result during the evaluation object period relative to the estimated value of the estimated operating result. The evaluation device may include an output unit that outputs the index.

[0005] The above-mentioned estimation unit can estimate the operation results based on the operation during the above-mentioned learning object period under the operating environment during the above-mentioned evaluation object period based on the output of the learning model that has been machine-learned in a manner that uses the above-mentioned environmental data and the above-mentioned result data during the above-mentioned learning object period as learning data and outputs the operation results corresponding to the operating environment.

[0006] The evaluation device may further include a learning unit that generates the learning model.

[0007] The evaluation device may further include a learning model storage unit configured to store the learning model in association with each of the learning target periods.

[0008] The learning unit may generate a learning model corresponding to the designated learning period using the environment data and the achievement data of the designated learning period as the learning data when no learning model corresponding to the designated learning period is stored.

[0009] The output unit may output the index for each of the plurality of evaluation target periods.

[0010] The output unit may output the index for each of a plurality of periods obtained by dividing the one evaluation target period.

[0011] The environmental data acquisition unit may acquire data indicating an external air condition as the environmental data.

[0012] The environmental data acquisition unit may acquire, as the environmental data, data indicating an operating state of a facility where the equipment is installed.

[0013] The performance data acquisition unit may acquire, as the performance data, data indicating at least one of a fuel usage amount of the equipment and a power consumption amount of the equipment.

[0014] In a second aspect of the present invention, an evaluation method is provided. The evaluation method may include the following steps: obtaining environmental data representing the operating environment of the equipment. The evaluation method may include the following steps: obtaining result data representing the operating results of the equipment. The evaluation method may include the following steps: estimating the operating results based on the operation during the learning object period under the operating environment during the evaluation object period based on the environmental data and the result data during the learning object period. The evaluation method may include the following steps: calculating an index that relatively evaluates the actual measured value of the operating result during the evaluation object period relative to the estimated value of the estimated operating result. The evaluation method may include the following steps: outputting the index.

[0015] In a third aspect of the present invention, a recording medium having an evaluation program recorded thereon is provided. The evaluation program can be executed by a computer. The evaluation program can cause the computer to function as an environmental data acquisition unit that acquires environmental data representing the operating environment of the device. The evaluation program can cause the computer to function as a result data acquisition unit that acquires result data representing the operating results of the device. The evaluation program can cause the computer to function as an estimation unit that estimates the operating results based on the operation during the learning object period under the operating environment during the evaluation object period based on the environmental data during the learning object period and the result data. The evaluation program can cause the computer to function as an evaluation unit that calculates an index that relatively evaluates the actual measured value of the operating result during the evaluation object period relative to the estimated value of the estimated operating result. The evaluation program can cause the computer to function as an output unit that outputs the index.

[0016] In a fourth aspect of the present invention, a control device is provided. The control device may include an environment data acquisition unit that acquires environment data representing the operating environment of the device. The control device may include a result data acquisition unit that acquires result data representing the operating results of the device. The control device may include an estimation unit that estimates the operating results based on the operation during the learning object period under the operating environment during the evaluation object period based on the environment data during the learning object period and the result data. The control device may include an evaluation unit that calculates an index that relatively evaluates the actual measurement value of the operating results during the evaluation object period relative to the estimated value of the estimated operating results. The control device may include an output unit that outputs the index. The control device may include a control unit that controls the device based on the index.

[0017] The control unit may control the device based on an output of a control model that has been machine-learned to output an operation amount to be applied to the device, using the indicator.

[0018] The evaluation unit may calculate the index by dividing the estimated value of the operating result by the actual measured value of the operating result. The control unit may generate the control model by performing reinforcement learning so that an operation amount having a higher reward that at least partially includes the index is output as a more recommended operation amount.

[0019] In a fifth aspect of the present invention, a recording medium having a control program recorded thereon is provided. The control program can be executed by a computer. The control program can cause the computer to function as an environmental data acquisition unit that acquires environmental data representing the operating environment of the device. The control program can cause the computer to function as a performance data acquisition unit that acquires performance data representing the operating performance of the device. The control program can cause the computer to function as an estimation unit that estimates the operating performance of the device during the learning period under the operating environment of the evaluation period based on the environmental data and performance data during the learning period. The control program can cause the computer to function as an evaluation unit that calculates an index that evaluates the actual measured value of the operating performance during the evaluation period relative to the estimated value of the estimated operating performance. The control program can cause the computer to function as an output unit that outputs the index. The control program can cause the computer to function as a control unit that controls the device based on the index.

[0020] The above summary of the invention does not list all the essential features of the present invention. In addition, sub-components of the above-mentioned feature groups can also constitute the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 An example of a block diagram of the evaluation device 10 according to the present embodiment is shown together with a facility 20 in which equipment 200 to be evaluated is installed.

[0022] Figure 2 An example of an evaluation flow of the evaluation device 10 according to this embodiment is shown.

[0023] Figure 3 An output example of the evaluation result of the evaluation device 10 according to the present embodiment is shown.

[0024] Figure 4 Another output example of the evaluation result of the evaluation device 10 according to the present embodiment is shown.

[0025] Figure 5 An example of a block diagram of the control device 500 according to the present embodiment is shown.

[0026] Figure 6 This shows an example of a computer 9900 that can embody all or part of the various aspects of the present invention. DETAILED DESCRIPTION

[0027] The present invention will be described below by way of embodiments of the invention, but the following embodiments do not limit the invention as defined in the claims. Furthermore, not all combinations of features described in the embodiments are essential for the solution to the problem of the invention.

[0028] Figure 1 An example of a block diagram of the evaluation device 10 involved in the present embodiment is shown together with a facility 20 in which a device 200 serving as an evaluation object is installed. The evaluation device 10 involved in the present embodiment estimates the results based on the previous operation when attempting to replace the previous environment with the same environment as the period of the evaluation object. Moreover, the evaluation device 10 performs a relative evaluation of the results (performance) actually measured during the period of the evaluation object with respect to the estimated results based on the previous operation. As an example, the evaluation device 10 involved in the present embodiment evaluates the energy usage of the device 200 as an outcome, thereby outputting the energy saving (abbreviated as "energy saving") effect of the device 200 as an evaluation result. In addition, the "energy" mentioned here includes energy resources that form the basis of energy in addition to the energy itself.

[0029] Facility 20 is a building established for a specific purpose or use. This facility 20 is equipped with various instruments and other equipment to achieve various functions. Examples of facility 20 include factories, office buildings, commercial buildings, hospitals, schools, shops, and hotels. As an example, the case where facility 20 is a workshop will be described. Examples of such workshops include, in addition to chemical and biological industrial plants, workshops responsible for managing and controlling wellheads and their surrounding areas at gas and oil fields, workshops responsible for managing and controlling hydropower, thermal power, and nuclear power generation, workshops responsible for managing and controlling environmental resource-based power generation such as solar and wind power, workshops responsible for managing and controlling water supply and drainage systems, dams, and semiconductor manufacturing plants that perform various semiconductor manufacturing processes. As shown in the figure, facility 20 includes, for example, n (where n is an integer greater than or equal to 1) pieces of equipment 200, including devices 200a to 200n (collectively referred to as "equipment 200"). The evaluation device 10 according to this embodiment can evaluate these n pieces of equipment 200 located in this facility 20.

[0030] The equipment 200 is an instrument (group) set as an evaluation object by the evaluation device 10 involved in this embodiment. The equipment 200 can be, for example, air-conditioning equipment, water supply and drainage equipment, and charging and power supply equipment. First, as an example, the case where the equipment 200 is an air-conditioning equipment installed in a workshop is described. That is, the evaluation device 10 involved in this embodiment sets such air-conditioning equipment installed in the workshop as the evaluation object. However, it is not limited to this. The evaluation device 10 involved in this embodiment can set various equipment 200 installed in various facilities 20 as evaluation objects. In particular, the evaluation device 10 involved in this embodiment can set instruments (groups) in various facilities 20 whose operating results have a strong correlation with the operating environment as evaluation objects.

[0031] The evaluation device 10 can be a computer such as a PC (personal computer), a tablet computer, a smartphone, a workstation, a server computer, or a general-purpose computer, or a computer system composed of multiple computers connected together. In addition, such a computer system is also a computer in a broad sense. In addition, the evaluation device 10 can be installed using one or more virtual computer environments that can be executed within the computer. Alternatively, the evaluation device 10 can be a dedicated computer designed for evaluating the evaluation object, or it can be dedicated hardware implemented by dedicated circuits. In addition, if the evaluation device 10 can be connected to the Internet, the evaluation device 10 can be implemented through cloud computing.

[0032] The evaluation device 10 has an environment data acquisition section 100, an achievement data acquisition section 110, a data storage section 120, an input section 130, a preprocessing section 140, a learning section 150, a learning model storage section 160, an estimation section 170, an evaluation section 180, and an output section 190. Note that the above-described modules can be functional modules in which functions are separated from each other, and do not necessarily coincide with actual device structures. That is, in the present drawing, one module is illustrated, but it is not necessarily configured by one device. In addition, in the present drawing, different modules are illustrated, but it is not necessarily configured by different devices.

[0033] The environment data acquisition section 100 acquires environment data indicating an operation environment of the device 200. For example, the environment data acquisition section 100 can acquire data indicating an external air condition as the environment data. In addition, the environment data acquisition section 100 can acquire data indicating an operation state of the facility 20 in which the device 200 is installed as the environment data. The environment data acquisition section 100 acquires such environment data from the facility 20 in real time, for example, via a network. However, it is not limited thereto. The environment data acquisition section 100 can also acquire such environment data in batch units via an operator, various storage devices. The environment data acquisition section 100 supplies the acquired environment data to the data storage section 120.

[0034] The achievement data acquisition section 110 acquires achievement data indicating an operation achievement of the device 200. For example, the achievement data acquisition section 110 can acquire data indicating at least any one of a fuel usage amount of the device 200 and a power consumption amount of the device 200 when the device 200 is operated as the achievement data. The achievement data acquisition section 110 acquires such achievement data from the facility 20 in real time, for example, via a network. However, it is not limited thereto. The achievement data acquisition section 110 can also acquire such achievement data in batch units via an operator, various storage devices. The achievement data acquisition section 110 supplies the acquired achievement data to the data storage section 120.

[0035] The data storage section 120 stores the environment data and the achievement data. For example, the data storage section 120 stores the environment data supplied from the environment data acquisition section 100 and the achievement data supplied from the achievement data acquisition section 110 in association with dates. Further, the data storage section 120 supplies the environment data and the achievement data of a period to the preprocessing section 140 when supplied with information indicating the period.

[0036] The input section 130 receives information of a specified period. For example, the input section 130 can be a user interface that receives information of a period specified by a user via a keyboard, a mouse, or the like. However, the present application is not limited thereto. The input section 130 can acquire information of a period specified via a network, or via various storage devices. The input section 130 receives information of a learning target period that is a period for learning a relationship between an operation environment and an operation result, and information of an evaluation target period that is a period for evaluating an operation result, respectively. That is, the input section 130 receives information of two periods, a period that becomes a reference for comparison and a period that becomes an object for comparison. The input section 130 supplies the received information of a specified period to the data storage section 120 and the learning model storage section 160.

[0037] The pre-processing section 140 performs pre-processing on environment data and result data. For example, the pre-processing section 140 performs pre-processing on environment data and result data supplied from the data storage section 120. The pre-processing section 140 supplies the environment data and the result data that have been pre-processed to the learning section 150, the estimation section 170, and the evaluation section 180.

[0038] The learning section 150 generates a learning model. For example, if the learning section 150 is supplied with environment data and result data of a learning target period from the pre-processing section 140, the learning section 150 learns a relationship between an operation environment and an operation result using the data as learning data. Furthermore, the learning section 150 generates a learning model that performs machine learning in a manner that outputs an operation result corresponding to an operation environment. The learning section 150 supplies the generated learning model to the learning model storage section 160.

[0039] The learning model storage section 160 stores a learning model in association with each learning target period. For example, the learning model storage section 160 stores a learning model generated by the learning section 150 based on environment data and result data of a learning target period in association with each of the learning target periods. Furthermore, in the above description, a case where the learning model storage section 160 stores a learning model generated by the learning section 150 inside the evaluation device 10 is shown as an example. However, the present application is not limited thereto. The learning model storage section 160 can also store a learning model generated outside the evaluation device 10. That is, the learning section 150 can be provided instead of, or in addition to, inside the evaluation device 10.

[0040] Based on the environmental data and performance data from the learning period, the estimation unit 170 estimates the operational performance based on the operations during the learning period under the operating environment of the evaluation period. For example, the estimation unit 170 estimates the operational performance based on the operations during the learning period under the operating environment of the evaluation period based on the output of a learning model that has been machine-learned using the environmental data and performance data from the learning period as learning data and outputs operational performance corresponding to the operating environment. The estimation unit 170 supplies the estimated value of the operational performance to the evaluation unit 180.

[0041] Evaluation unit 180 calculates an index that evaluates the actual measured values ​​of the operating results during the evaluation period relative to the estimated values ​​of the estimated operating results. Specifically, evaluation unit 180 calculates an index that evaluates the actual measured values ​​of the operating results during the evaluation period using the estimated values ​​of the operating results supplied from estimating unit 170 as a reference. Evaluation unit 180 supplies the calculated index to output unit 190.

[0042] The output unit 190 outputs the indicators. For example, the output unit 190 displays and outputs the indicators provided by the evaluation unit 180 on a monitor. However, this is not limited to this. The output unit 190 may also output the indicators by voice, print, or transmit them to other functional units or other devices.

[0043] Figure 2 This figure illustrates an example of an evaluation process performed by the evaluation device 10 according to this embodiment. Furthermore, this figure illustrates an example in which the evaluation device 10 evaluates the energy usage of an air conditioning system (an example of equipment 200) installed in a workshop (an example of facility 20) as a result, thereby illustrating an example in which the energy-saving effect of the air conditioning system is output as an indicator.

[0044] In step 200, the evaluation device 10 acquires environmental data. For example, the environmental data acquisition unit 100 acquires environmental data representing the operating environment of the equipment 200 from the facility 20 via a network in real time. In this case, the environmental data acquisition unit 100 may acquire data representing external air conditions as the environmental data. As an example, the environmental data acquisition unit 100 may acquire data such as the external air temperature and humidity surrounding the facility 20 as the environmental data.

[0045] Furthermore, the environmental data acquisition unit 100 may acquire, as environmental data, data indicating the operating status of the facility 20 in which the equipment 200 is installed. For example, the environmental data acquisition unit 100 may acquire, as environmental data, data such as a production plan of a workshop or power consumption data of a group of devices operating in the workshop in relation to the production plan.

[0046] In this way, the environmental data acquisition unit 100 can acquire, as environmental data, data indicating the external air condition and data indicating the operating state of the facility 20. The environmental data acquisition unit 100 supplies the acquired environmental data to the data storage unit 120.

[0047] In step 210, the evaluation device 10 acquires performance data. For example, the performance data acquisition unit 110 acquires performance data indicating the operating performance of the equipment 200 in real time from the facility 20 via the network. In this case, the performance data acquisition unit 110 may acquire data indicating the amount of fuel used by the facility 200 as the performance data. For example, the performance data acquisition unit 110 may acquire data such as the amount of LPG (Liquified Petroleum Gas) or city gas used by the boiler to generate steam for heating and humidifying the interior of the facility 20.

[0048] Furthermore, the performance data acquisition unit 110 may acquire, as performance data, data indicating at least one of the power consumption of the equipment 200. For example, the performance data acquisition unit 110 may acquire, as performance data, data such as the power consumed by the chiller to obtain cooling water for cooling and dehumidifying the facility 20.

[0049] In this way, the performance data acquisition unit 110 can acquire data indicating at least one of the fuel usage and the power consumption of the equipment 200 as performance data. The performance data acquisition unit 110 supplies the acquired performance data to the data storage unit 120.

[0050] In step 220, the evaluation device 10 stores the environmental data and the achievement data. For example, the data storage unit 120 stores the environmental data acquired in step 200 and the achievement data acquired in step 210 in association with dates.

[0051] In step 230, the evaluation device 10 receives information specifying a period. For example, the input unit 130 receives information specifying a period input by the user via a keyboard, mouse, or the like. As an example, the input unit 130 receives information specifying a date range for the second half of the fiscal year 2018 (FY2018) (i.e., October 1, 2018, to March 31, 2019) as the information specifying the learning period. Furthermore, the input unit 130 receives, as information specifying the evaluation target period, information specifying the date range of the first half of FY2019 (i.e., April 1, 2019, to September 30, 2019), information specifying the date range of the second half of FY2019 (October 1, 2019, to March 31, 2020), information specifying the date range of the first half of FY2020 (April 1, 2020, to September 30, 2020), and information specifying the date range of the second half of FY2020 (October 1, 2020, to March 31, 2021). In this way, the input unit 130 can simultaneously receive information specifying multiple evaluation target periods.

[0052] In step 235, the evaluation device 10 determines whether the learning model corresponding to the target learning period has been saved. For example, the input unit 130 accesses the learning model storage unit 160 and determines whether the learning model corresponding to the target learning period received in step 230 (here, the second half of FY2018) has been saved. If it is determined in step 235 that the learning model has been saved (if yes), the evaluation device 10 proceeds to step 245.

[0053] On the other hand, if it is determined in step 235 that the data has not been saved (No), the input unit 130 supplies information specifying the learning period to the data storage unit 120. In response, the data storage unit 120 supplies the environment data and achievement data of the specified learning period to the pre-processing unit 140.

[0054] In step 240, the evaluation device 10 preprocesses the learning data. For example, the preprocessing unit 140 preprocesses the environmental data and performance data for the learning target period (here, the second half of FY2018) supplied from the data storage unit 120 in step 235. As an example, if data on outside air temperature and humidity at multiple locations around a workshop is obtained as environmental data, the preprocessing unit 140 can calculate the specific enthalpy of the outside air at each location based on this data, and calculate statistical values ​​(e.g., average values, median values, etc.) of the specific enthalpy of the outside air at multiple locations. Alternatively, if multiple boilers are operating in an air conditioning system and the LPG usage of each boiler is obtained as performance data, the preprocessing unit 140 can calculate the total LPG usage of the multiple boilers. The preprocessing unit 140 supplies the preprocessed environmental data and performance data to the learning unit 150.

[0055] In step 250, the evaluation device 10 generates a learning model. For example, the learning unit 150 uses the environmental data and performance data from the learning target period that were pre-processed in step 240 as learning data to learn the relationship between the operating environment and operating performance. Furthermore, the learning unit 150 generates a learning model that has been machine-learned to output operating performance corresponding to the operating environment. In this manner, if a learning model corresponding to the specified learning target period is not stored, the learning unit 150 uses the environmental data and performance data from the specified learning target period as learning data to generate a learning model corresponding to the specified learning target period. Furthermore, the learning unit 150 can select any machine-learned model at this time. For example, the learning unit 150 can generate a multivariate regression model for predicting LPG usage by performing linear regression using the specific enthalpy of outside air and the operating status of the vehicle as explanatory variables. The learning model generated in this manner can be interpreted as a model that can predict operating performance based on the operating environment within the date range used as learning data, using environmental data from other operating environments not used as learning data. The learning unit 150 supplies the generated learning model to the learning model storage unit 160.

[0056] In step 260 , the evaluation device 10 stores the learning model. For example, the learning model storage unit 160 stores the learning model generated in step 250 in association with each learning target period. Then, the evaluation device 10 advances the process to step 245 .

[0057] In step 245, the evaluation device 10 pre-processes the evaluation data. For example, the input unit 130 supplies the information received in step 230 for the designated evaluation period (here, the first half of FY2019, the second half of FY2019, the first half of FY2020, and the second half of FY2020) to the data storage unit 120. In response, the data storage unit 120 supplies the environmental data and results data for the evaluation period to the pre-processing unit 140. Furthermore, the pre-processing unit 140 pre-processes the environmental data and results data for the evaluation period supplied from the data storage unit 120. The specific method of pre-processing is the same as that of step 240, so its description is omitted here. The pre-processing unit 140 supplies the pre-processed environmental data to the estimation unit 170. Furthermore, the pre-processing unit 140 supplies the pre-processed results data to the evaluation unit 180. The results data supplied here can be referred to as the actual measured values ​​of the results data for the evaluation period.

[0058] In step 270, the evaluation device 10 estimates the operating results. For example, the estimation unit 170 estimates the operating results based on the operation during the learning object period under the operating environment of the evaluation object period based on the environmental data and the results data of the learning object period. More specifically, the input unit 130 supplies the information of the designated learning object period received in step 230 to the learning model storage unit 160. In response, the learning model storage unit 160 supplies the learning model corresponding to the learning object period to the estimation unit 170. In addition, this learning model can be a model that has been saved in advance or a model newly generated in step 250. Moreover, the estimation unit 170 estimates the operating results based on the operation during the learning object period under the operating environment of the evaluation object period using the value output from the learning model corresponding to the input of the environmental data pre-processed in step 245 to the learning model as the estimated value.

[0059] In this case, when evaluating the previous period of FY2019, the estimation unit 170 inputs the environmental data from the previous period of FY2019 into the learning model and uses the value output from the learning model as the estimated value. Specifically, the estimation unit 170 estimates the operational results for the next period of FY2018 if the operating environment of the next period of FY2018 were replaced with the same environment as the previous period of FY2019 when evaluating the previous period of FY2019. Similarly, when evaluating the next period of FY2019, the estimation unit 170 inputs the environmental data from the next period of FY2019 into the learning model and uses the value output from the learning model as the estimated value. Specifically, the estimation unit 170 estimates the operational results for the next period of FY2018 if the operating environment of the next period of FY2018 were replaced with the same environment as the next period of FY2019 when evaluating the next period of FY2019. The same applies to the first and second periods of FY2020. Thus, based on the output of a learning model that has been machine-learned using environmental data and performance data from the learning period as learning data to output performance corresponding to the operating environment, the estimation unit 170 estimates the operating performance based on operations during the learning period under the operating environment of the evaluation period. The estimation unit 170 supplies the estimated value of the estimated operating performance to the evaluation unit 180.

[0060] In step 280, the evaluation device 10 calculates an indicator. For example, the evaluation unit 180 calculates an indicator that evaluates the actual measurement value of the operating results during the evaluation object period relative to the estimated value of the operating results. As an example, the evaluation unit 180 divides the estimated value of the operating results supplied in step 270 by the actual measurement value of the operating results supplied in step 245 to calculate the indicator. Here, such an indicator is defined as a KPI (Key Performance Indicators). Such a KPI indicates that the higher the score (for example, energy-saving effect) is, the higher the score is (for example, energy-saving effect) is, and the lower the score is (for example, there is no energy-saving effect when it is lower than 1) is. Thus, the evaluation unit 180 calculates an indicator that evaluates the actual measurement value of the operating results during the evaluation object period based on the estimated value of the operating results supplied from the estimation unit 170. The evaluation unit 180 supplies the calculated indicator to the output unit 190.

[0061] In step 290, the evaluation device 10 outputs the indicator. For example, the output unit 190 displays and outputs the indicator calculated in step 280 on a monitor. Thus, the evaluation device 10 according to this embodiment estimates the results of past operations when attempting to replace the previous environment with the same environment as the evaluation period. Furthermore, the evaluation device 10 evaluates the actual measured operating results for the evaluation period relative to the estimated past operating results. Next, an example of the output of evaluation results by the evaluation device 10 according to this embodiment will be described in detail using the accompanying drawings.

[0062] Figure 3 The figure shows an output example of the evaluation results of the evaluation device 10 involved in this embodiment. In this figure, the horizontal axis represents the year, and the vertical axis represents the KPI. In addition, in this figure, the horizontal axis intersects the vertical axis at the position of KPI=1. That is, the area above the horizontal axis represents KPI>1, and the area below the horizontal axis represents KPI<1. In this figure, starting from the left, the KPIs of the four evaluation target periods, namely the first half of FY2019, the second half of FY2019, the first half of FY2020, and the second half of FY2020, are shown in bar graphs. In this way, the output unit 190 can output indicators for each of the multiple evaluation target periods. In addition, as mentioned above, the KPIs in this figure are calculated based on the second half of FY2018 as the learning target period. Here, the external air conditions in the first half and the second half vary depending on the season. In addition, even if the season is the same, the external air conditions vary due to climate changes. Moreover, if the year or period changes, the production plan of the workshop will naturally also change. However, according to the evaluation device 10 according to the present embodiment, the results of different periods are evaluated under the same conditions, and therefore the influence of the difference in the operating environment on the operating results can be offset.

[0063] As shown in this figure, the KPI is greater than 1 for the first half of FY2019, the second half of FY2019, and the second half of FY2020, indicating that energy savings were achieved relative to the second half of FY2018. Furthermore, the energy savings increase in the order of second half of FY2020 > first half of FY2019 > second half of FY2019. On the other hand, the KPI for the first half of FY2020 is less than 1, indicating that energy savings were not achieved relative to the second half of FY2018 (energy was wasted in the first half of FY2020 compared to the second half of FY2018). Users who see this display can, for example, determine that energy savings decreased from the first half of FY2019 to the first half of FY2020 and that energy savings improved in the second half of FY2020. Thus, for example, users can conclude that energy-saving activities implemented in the second half of FY2020 have taken effect and achieved improved results.

[0064] Figure 4Another example of output of the evaluation result of the evaluation device 10 according to this embodiment is shown. For example, the evaluation device 10 can be used with the user to select Figure 3 One of the multiple evaluation target periods shown in Figure 4 . Here, the evaluation object period of KPI<1, that is, the last period of FY2020, is set as the period selected by the user. In this figure, the horizontal axis represents the actual measured value of the results of the evaluation object period, and the vertical axis represents the estimated value of the operation results based on the operation of the learning object period under the operating environment of the evaluation object period. In addition, in this figure, the dotted line represents KPI=1, that is, the actual measured value is equal to the estimated value. That is, the area above the dotted line represents KPI>1, and the area below the dotted line represents KPI<1. In this figure, the last period of FY2020, which is the selected evaluation object period, is distinguished for each month, and the KPIs of the six months are represented by a scatter plot. In this way, the output unit 190 can output the indicators of each of the multiple periods distinguished for one of the evaluation object periods. In addition, when performing such output, the evaluation device 10 only needs to use the period for which the evaluation object period is distinguished, here the six "months" for distinguishing the half-period as the new evaluation object period, and execute it again. Figure 2 The indicators for each month can be calculated by following steps 245, 270, 280, and 290 in the process.

[0065] As shown in this figure, for October, November, February, and March, the KPI is greater than 1, so it can be said that there is an energy-saving effect relative to the second half of FY2018. In addition, in the area above the dotted line, the farther the distance from the dotted line, the greater the KPI, so it can be said that the energy-saving effect increases in the order of March > October > November > February. On the other hand, for December and January, the KPI is less than 1, so it can be said that there is no energy-saving effect relative to the second half of FY2018. In addition, in the area below the dotted line, the farther the distance from the dotted line, the smaller the KPI, so it can be said that December consumed more energy in vain than January. In other words, an improvement effect was achieved in four months of the half (October, November, February, and March), but the total deterioration in energy-saving effect in the remaining two months (December and January) was greater than the sum of these improvements, so it can be understood that there was no energy-saving effect in the entire first half of FY2020 relative to the second half of FY2018. Furthermore, a user viewing this display can determine, for example, that the deterioration in energy-saving performance following the first half of 2019 continued in October and November 2020, peaked in December 2020, began to improve in January 2021, and that energy-saving performance has been demonstrated since February 2021 compared to the second half of FY2018. Thus, for example, the user can determine that energy-saving activities implemented since January 2021 have been effective and have resulted in improvements.

[0066] Typically, efforts are continuously being made to reduce energy usage in air-conditioning equipment in workshops, etc., such as by remodeling equipment. However, when energy-saving effects are predicted based on rated power, for example, it is impossible to accurately predict the extent of energy reduction that can be achieved through energy-saving activities. In addition, in air-conditioning equipment in workshops, etc., the operating status of the entire workshop may also affect the operating rate of the air-conditioning equipment. Therefore, when the energy-saving effect is calculated based only on the operating status of the device of interest without considering the state of the air-requiring side, the calculated energy-saving effect lacks reliability. Therefore, it is expected that the energy-saving effect of workshops, etc. can be calculated more accurately.

[0067] In contrast, the evaluation device 10 according to the present embodiment estimates the operational results based on the operation during the learning object period under the operating environment of the evaluation object period based on the environmental data and the results data during the learning object period. Furthermore, the evaluation device 10 relatively evaluates the actual measured value of the operational results during the evaluation object period relative to the estimated value of the estimated operational results. Thus, according to the evaluation device 10 according to the present embodiment, the previous operational results are estimated when the previous environment is replaced with the same environment as the evaluation object period. Furthermore, the evaluation device 10 relatively evaluates the actual measured operational results during the evaluation object period relative to the estimated previous operational results. Therefore, the evaluation device 10 evaluates the results of different periods under the same conditions, and is therefore able to accurately evaluate the operational results by offsetting the influence of the different operating environments on the operational results.

[0068] Furthermore, the evaluation device 10 according to this embodiment estimates operating results based on the output of a learning model that has been machine-learned using environmental data and performance data from the learning period as learning data. Thus, the evaluation device 10 according to this embodiment estimates operating results based on the relationship between the operating environment and operating results learned from actual past data, making it possible to estimate operating results based on objective evidence.

[0069] The evaluation device 10 according to this embodiment further includes a learning unit that generates a learning model. Therefore, the evaluation device 10 according to this embodiment can provide a function of evaluating an evaluation target and a learning function for evaluating the evaluation target as an integrated device.

[0070] In addition, the evaluation device 10 according to the present embodiment generates a learning model corresponding to a designated learning target period in a case where the learning model is not stored. Thus, according to the evaluation device 10 according to the present embodiment, it is possible to avoid generation of a learning model that has been completed or obtained from another device, and thus it is possible to reduce the processing load involved in learning.

[0071] In addition, the evaluation device 10 according to the present embodiment can output indexes for each of a plurality of evaluation target periods, and can output indexes for each of a plurality of periods distinguished from one evaluation target period. Thus, according to the evaluation device 10 according to the present embodiment, it is possible to output evaluation results in various manners, and thus it is possible to provide opportunities for various analyses to a user.

[0072] In addition, the evaluation device 10 according to the present embodiment evaluates an evaluation target using data indicating an external air condition and an operation state of a facility as environment data, and using data indicating an energy usage amount of a device as achievement data. Thus, according to the evaluation device 10 according to the present embodiment, it is possible to more accurately evaluate an operation achievement, taking into account an external air condition and an operation state of a facility that have a particularly strong correlation with an energy usage amount.

[0073] Figure 5 One example of a block diagram of the control device 500 according to the present embodiment is shown in FIG. 5. In the control device 500 according to the present embodiment, components having the same functions and structures as those of the control device 500 according to the first embodiment are denoted by the same reference numerals, and the description thereof is omitted except for the following differences. Figure 5 In the control device 500 according to the present embodiment, the evaluation device 10 according to the first embodiment is connected to the control device 500. The control device 500 according to the present embodiment is connected to the device 200. Figure 1 In the control device 500 according to the present embodiment, the evaluation device 10 according to the first embodiment is connected to the control device 500. The control device 500 according to the present embodiment is connected to the device 200.

[0074] The status data acquisition unit 510 acquires status data representing the status of the facility 20 in which the device 200 is installed. For example, the status data acquisition unit 510 acquires the measured value (PV: Process Variable) representing the operating status of the result of controlling the device 200 to be controlled as status data from the facility 20 in real time via the network. Here, such a measured value can be the output of the device 200 to be controlled, that is, the control quantity, or various physical quantities that change according to the control quantity. In addition, in the above description, as an example, the situation where the status data acquisition unit 510 acquires such status data via the network is shown, but it is not limited to this. The status data acquisition unit 510 can also acquire such status data in batches via an operator or various storage devices. The status data acquisition unit 510 supplies the acquired status data to the control unit 520.

[0075] The control unit 520 controls the device 200 based on the indicator. For example, the control unit 520 can use the status data supplied by the status data acquisition unit 510 and the operation amount data indicating the operation amount applied by the control unit to the device 200 to control the device 200 as the control target using a control model that has undergone machine learning to output the operation amount corresponding to the status of the device 200. In other words, the control unit 520 can function as a so-called AI (Artificial Intelligence) controller.

[0076] At this time, the control unit 520 controls the device 200 based on the output of a control model that has been machine-learned to output the amount of operation to be applied to the device 200, using the indicator output from the output unit 190. The control unit 520 may previously store such a control model. Alternatively, or in addition to this, the control unit 520 may generate such a control model by performing reinforcement learning such that the higher the reward for an operation that at least partially includes the indicator, the more recommended the operation amount.

[0077] Typically, if an agent observes a state and chooses an action, the state changes based on that action. In reinforcement learning, a reward is assigned along with this change in state, allowing the agent to learn to choose better actions (determined decisions). Furthermore, in reinforcement learning, agents typically perform value evaluations and learn by selecting actions that maximize the total value of future rewards. Thus, in reinforcement learning, agents learn appropriate actions based on the interactions between actions and states, i.e., actions that maximize future rewards, by learning actions.

[0078] Therefore, the control section 520 can perform reinforcement learning in a manner that the operation amount that is higher as the reward of the index included as at least a part is higher is output as the operation amount that is more recommended, thereby generating the control model. As described above, the evaluation section 180 calculates the index by dividing the estimated value of the operation result by the actually measured value of the operation result. Therefore, the control section 520 can perform reinforcement learning using a reward function that makes the reward higher as the index (KPI) output from the output section 190 is larger and makes the reward lower as the index is smaller, thereby generating such a control model. Thereby, the control section 520 can generate a control model that outputs the operation amount that is higher as the KPI is larger as the operation amount that is more recommended. Also, the control section 520 can control the device 200 in a manner that makes the KPI larger, that is, in a manner that makes the energy saving effect amount larger, by giving the output of such a control model as the operation amount to the device 200.

[0079] Thus, the control device 500 according to the present embodiment controls the device 200 that is a control target by the control model that has performed reinforcement learning based on the KPI. Thereby, according to the control device 500 according to the present embodiment, it is possible to control the device 200 in a manner that maximizes the energy saving effect amount, for example. Therefore, the control device 500 according to the present embodiment can accurately evaluate the operation of the device 200 and can control the device 200 based on the result of the accurate evaluation, and thus it is possible to provide a structure that maximizes the energy saving effect amount, for example.

[0080] Further, in the above description, the control device 500 according to the present embodiment is illustrated as having a function of controlling the device 200 based on the calculated index, as one example, on the basis of all the functions of the above-described evaluation device 10. However, it is not limited thereto. At least a part of the functions of the above-described evaluation device 10 can be provided outside the control device 500 (on a cloud, for example). As one example, the function of estimating the operation result based on the past can be provided outside the control device 500. In this case, the control device 500 can have a result of estimation acquisition section that acquires the estimated value of the operation result estimated by the estimation section 170 provided outside, for example, instead of the environment data acquisition section 100, the result data acquisition section 110, the data storage section 120, the input section 130, the pre-processing section 140, the learning section 150, the learning model storage section 160, and the estimation section 170. Also, the function of relatively evaluating the result of the period to be evaluated with respect to the past based on the operation can also be provided outside the control device 500. In this case, the control device 500 can have an index acquisition section that acquires the index calculated by the evaluation section 180 provided outside, for example, instead of the evaluation section 180.

[0081] Various embodiments of the present invention may be described with reference to flowcharts and block diagrams, where a module may represent (1) a stage of a process for performing an operation, or (2) a portion of a device having the function of performing an operation. Specific stages and portions may be implemented using dedicated circuits, programmable circuits supplied with computer-readable instructions stored on a computer-readable medium, and / or processors supplied with computer-readable instructions stored on a computer-readable medium. Dedicated circuits may include digital and / or analog hardware circuits, as well as integrated circuits (ICs) and / or discrete circuits. Programmable circuits may include reconfigurable hardware circuits such as logical AND, logical OR, logical XOR, logical NAND, logical NOR, and other logical operations, flip-flop circuits, registers, memory elements such as field programmable gate arrays (FPGAs), programmable logic arrays (PLAs), and the like.

[0082] The computer-readable medium may include any tangible device capable of storing instructions for execution using an appropriate device. As a result, the computer-readable medium having the instructions stored therein has a product containing executable instructions for making a means for performing the operations specified in the flowchart or block diagram. Examples of computer-readable media include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable media include Floppy (registered trademark) floppy disks, floppy disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disk (DVD), Blu-ray (RTM) disks, memory sticks, integrated circuit cards, etc.

[0083] Computer-readable instructions may include any source code or object code described in assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or any combination of one or more programming languages ​​including object-oriented programming languages ​​such as Smalltalk (registered trademark), JAVA (registered trademark), C++, and existing procedural programming languages ​​such as the "C" programming language or similar programming languages.

[0084] Computer-readable instructions may be provided to a processor or programmable circuit of a general-purpose computer, a special-purpose computer, or other programmable data processing device via a local area network (LAN), a wide area network (WAN), or the like, and the computer-readable instructions may be executed to create a means for performing the operations specified in the flowchart or block diagram. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, and microcontrollers.

[0085] Figure 6 This figure illustrates an example of a computer 9900 that can embody all or part of the various aspects of the present invention. The program installed in computer 9900 can cause computer 9900 to function as an apparatus or one or more components of an apparatus according to an embodiment of the present invention, or to execute such an operation or one or more components, and / or can cause computer 9900 to perform a process or a stage of such a process according to an embodiment of the present invention. Such a program can be executed by CPU 9912 to cause computer 9900 to perform specific operations associated with some or all of the modules in the flowcharts and block diagrams described in this specification.

[0086] The computer 9900 according to this embodiment includes a CPU 9912, a RAM 9914, a graphics controller 9916, and a display device 9918, and these components are interconnected via a main controller 9910. Furthermore, the computer 9900 includes input / output units such as a communication interface 9922, a hard disk drive 9924, a DVD-ROM drive 9926, and an IC card drive, and these components are connected to the main controller 9910 via an input / output controller 9920. Furthermore, the computer includes a ROM 9930 and conventional input / output units such as a keyboard 9942, and these components are connected to the input / output controller 9920 via an input / output chip 9940.

[0087] The CPU 9912 controls each unit by executing operations according to programs stored in the ROM 9930 and the RAM 9914. The graphics controller 9916 obtains image data generated by the CPU 9912 in a frame buffer or the like or in the graphics controller itself and supplies it to the RAM 9914, and displays the image data on the display device 9918.

[0088] The communication interface 9922 communicates with other electronic devices via a network. The hard disk drive 9924 stores programs and data used by the CPU 9912 in the computer 9900. The DVD-ROM drive 9926 reads programs and data from the DVD-ROM 9901 and provides the programs and data to the hard disk drive 9924 via the RAM 9914. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0089] The ROM 9930 stores therein a startup program and the like executed by the computer 9900 upon activation, and / or programs that depend on the hardware of the computer 9900. In addition, the input / output chip 9940 connects various input / output units to the input / output controller 9920 via a parallel port, a serial port, a keyboard port, a mouse port, and the like.

[0090] The program is provided on a computer-readable medium such as a DVD-ROM 9901 or an IC card. The program is read from the computer-readable medium and installed in a hard disk drive 9924, a RAM 9914, or a ROM 9930, also examples of computer-readable media, and then executed by the CPU 9912. The information processing described in the program is read by the computer 9900, and the program and the various types of hardware resources described above are coordinated. By using the computer 9900 to implement information manipulation or processing, an apparatus or method can be constructed.

[0091] For example, when communication is performed between the computer 9900 and an external device, the CPU 9912 can execute a communication program loaded in the RAM 9914 and, based on the processing described in the communication program, issue communication processing instructions to the communication interface 9922. Under the control of the CPU 9912, the communication interface 9922 reads transmission data stored in a transmission buffer area provided in the RAM 9914, the hard disk drive 9924, the DVD-ROM 9901, or a recording medium such as an IC card, and transmits the read transmission data to the network or writes reception data received from the network to a reception buffer area provided on the recording medium.

[0092] Furthermore, the CPU 9912 can read all or a required portion of a file or database stored in an external recording medium such as the hard disk drive 9924, DVD-ROM drive 9926 (DVD-ROM 9901), or IC card into the RAM 9914, and perform various types of processing on the data in the RAM 9914. The CPU 9912 then writes the processed data back to the external recording medium.

[0093] Various types of information such as various types of programs, data, tables, and databases can be stored in a recording medium and subjected to information processing. The CPU 9912 can perform various types of processing, including various types of operations, information processing, conditional judgments, conditional branches, unconditional branches, information retrieval / replacement, etc., specified by the instruction sequence of the program recorded at any location in the present disclosure, on the data read from the RAM 9914, and write back the results to the RAM 9914. In addition, the CPU 9912 can retrieve information from files, databases, etc. in the recording medium. For example, in the case where a plurality of entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored in the recording medium, the CPU 9912 can retrieve an entry that specifies an attribute value of the first attribute that is consistent with the condition from the plurality of records, read the attribute value of the second attribute stored in the entry, and obtain the attribute value of the second attribute associated with the first attribute that satisfies the condition predetermined thereby.

[0094] The program or software module described above can be stored in a computer-readable medium on or near the computer 9900. In addition, a recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can be used as a computer-readable medium, thereby providing the program to the computer 9900 via the network.

[0095] The present invention has been described above using the embodiments. However, the technical scope of the present invention is not limited to the scope described in the above embodiments. Those skilled in the art will appreciate that various modifications or improvements may be made to the above embodiments. As is clear from the claims, embodiments incorporating such modifications or improvements are also within the technical scope of the present invention.

[0096] Regarding the order in which actions, sequences, steps, and stages, etc., of the apparatus, system, program, and method described in the claims, specifications, and drawings are executed, it should be noted that unless otherwise expressly indicated as "before," "before," or the like, and unless the output of a previous process is used in a subsequent process, the execution order may be any order. Even if the flow of actions in the claims, specifications, and drawings is described using the phrases "first," "next," or the like for convenience, it does not necessarily mean that the actions must be executed in that order.

[0097] Description of the label

[0098] 10 Evaluation device

[0099] 100 Environmental Data Acquisition Department

[0100] 110 Results Data Acquisition Department

[0101] 120 Data Storage Department

[0102] 130 input section

[0103] 140 preprocessing section

[0104] 150 learning section

[0105] 160 learning model storage section

[0106] 170 estimation section

[0107] 180 evaluation section

[0108] 190 output section

[0109] 200 apparatus

[0110] 500 control device

[0111] 510 state data acquisition section

[0112] 520 control section

[0113] 9900 computer

[0114] 9901 DVD-ROM

[0115] 9910 main controller

[0116] 9912 CPU

[0117] 9914 RAM

[0118] 9916 graphics controller

[0119] 9918 display device

[0120] 9920 input / output controller

[0121] 9922 communication interface

[0122] 9924 hard disk drive

[0123] 9926 DVD drive

[0124] 9930 ROM

[0125] 9940 input / output chip

[0126] 9942 keyboard

Claims

1. A control device, wherein: The control device has: an environmental data acquisition unit that acquires environmental data representing an operating environment of the device; a result data acquisition unit that acquires result data indicating the operation result of the equipment; an estimating unit that inputs the operating environment of the evaluation target period into a learning model and estimates an operating result based on the operation during the learning target period under the operating environment of the evaluation target period, wherein the learning model is generated by learning the relationship between the operating environment and the operating result during the learning target period and outputs an estimated value of the operating result based on the operation during the learning target period when the operating environment is input; an evaluation unit that calculates an index for evaluating an actual measurement value of the operating result during the evaluation target period relative to an estimated value of the estimated operating result; an output unit configured to output the indicator; a status data acquisition unit that acquires status data indicating the status of the device; and a control unit that controls the device by outputting an operation amount corresponding to the state of the device represented by the state data to the device, The control unit generates a control model that outputs an operation amount according to the state by performing reinforcement learning so that an operation amount having a higher reward that includes the indicator as at least a part thereof is output as a more recommended operation amount.

2. The control device according to claim 1, wherein: The control device further includes a learning unit that generates the learning model.

3. The control device according to claim 1 or 2, wherein: The control device further includes a learning model storage unit that stores the learning model in association with each of the learning target periods.

4. The control device according to claim 1, wherein: The control device also has: a learning unit that generates the learning model; and a learning model storage unit that stores the learning model in association with each of the learning target periods; The learning unit generates a learning model corresponding to the designated learning period using the environment data and the achievement data of the designated learning period as learning data when no learning model corresponding to the designated learning period is stored.

5. The control device according to claim 1 or 2, wherein: The output unit outputs the indicator for each of the plurality of evaluation target periods.

6. The control device according to claim 1 or 2, wherein: The output unit outputs the index for each of a plurality of periods obtained by dividing one evaluation target period.

7. The control device according to claim 1 or 2, wherein: The environmental data acquisition unit acquires data indicating an external air condition as the environmental data.

8. The control device according to claim 1 or 2, wherein: The environmental data acquisition unit acquires data indicating an operating state of a facility where the equipment is installed as the environmental data.

9. The control device according to claim 1 or 2, wherein: The performance data acquisition unit acquires data indicating at least one of a fuel usage amount of the equipment and a power consumption amount of the equipment as the performance data.

10. A control method, wherein: The control method has the following steps: obtaining environmental data representing an operating environment of the device; Acquiring performance data indicating performance of the device; Inputting the operating environment during the evaluation period into a learning model, estimating an operating result based on the operation during the learning period under the operating environment during the evaluation period, wherein the learning model is generated by learning the relationship between the operating environment and the operating result during the learning period, and outputting an estimated value of the operating result based on the operation during the learning period when the operating environment is input; calculating an index that relatively evaluates an actual measured value of the operating result during the evaluation target period with respect to an estimated value of the estimated operating result; outputting the indicator; obtaining status data representing a status of the device; controlling the device by outputting an operation amount corresponding to the state of the device indicated by the state data to the device; and Reinforcement learning is performed such that an operation amount having a higher reward including at least a part of the indicator is output as a more recommended operation amount, thereby generating a control model that outputs an operation amount according to the state.

11. A recording medium having a control program recorded thereon, wherein: The control program is executed by a computer, and causes the computer to function as the following functional unit: an environmental data acquisition unit that acquires environmental data representing an operating environment of the device; a result data acquisition unit that acquires result data indicating the operation result of the equipment; an estimating unit that inputs the operating environment of the evaluation target period into a learning model and estimates an operating result based on the operation during the learning target period under the operating environment of the evaluation target period, wherein the learning model is generated by learning the relationship between the operating environment and the operating result during the learning target period and outputs an estimated value of the operating result based on the operation during the learning target period when the operating environment is input; an evaluation unit that calculates an index for evaluating an actual measurement value of the operating result during the evaluation target period relative to an estimated value of the estimated operating result; an output unit configured to output the indicator; a status data acquisition unit that acquires status data indicating the status of the device; and a control unit that controls the device by outputting an operation amount corresponding to the state of the device represented by the state data to the device, The control unit generates a control model that outputs an operation amount according to the state by performing reinforcement learning so that an operation amount having a higher reward that includes the indicator as at least a part thereof is output as a more recommended operation amount.

Images